Compare commits

..

250 Commits

Author SHA1 Message Date
ado
53c5b779d4 [skip ci] Update version 2024-03-14 19:15:32 +01:00
ado
107a122718 [skip ci] Update README 2024-03-14 19:13:03 +01:00
red0124
55d0a4e598
Updated and added new functions related to headers, resolved ODR issues, resolved clang-tidy warnings (#50)
* Bugfix/odr violations (#47)

* Make common non-member functions inline, remove unreachable line from get_line_buffer

* [skip ci] Fix namespace comments

* Resolve clang-tidy warnings (#48)

* Resolve clang-tidy warnings, update single_header_generator.py

* Update single header test, resolve additional clang-tidy warnings

* Add header and raw_header methods,  update header usage methods error handling, write new and update existing unit tests

* Update parser error messages, fix parser tests

* Add [[nodiscard]] where fitting, update unit tests (#49)

* Add const where fitting, make splitter class members private, add #pragma once to ssp.hpp

* Modify header parsing for empty headers, update old and add new tests for header parsing

* Enable the parser to accept a header with one empty field, update unit tests

* Fix test CMakeLists.txt typo
2024-03-14 17:22:57 +01:00
red0124
1b9a01f787
Feature/fuzz (#44)
* Add fuzzing ci, add bedge to README
2024-03-03 20:46:12 +01:00
red0124
f5b750dd93
Merge pull request #43 from red0124/bugfix/ftell_slowdown
Remove usage of ftell when updating cursor position value
2024-03-02 02:20:05 +01:00
ado
7f53b585f9 Remove usage of ftell when updating cursor position value 2024-03-02 00:34:19 +01:00
ado
67ef6651c1 Fix README typos 2024-03-01 17:23:26 +01:00
ado
fa4ec324de Update version 2024-03-01 16:22:45 +01:00
red0124
f229de61d6
Merge pull request #42 from red0124/dev
Merge with development
2024-03-01 16:17:16 +01:00
red0124
df2beab6c3
Fix buffer overflow on multiline restricted with unterminated quote and multiple empty lines (#41) 2024-03-01 15:46:34 +01:00
red0124
27bd60b5ce
Fix bug with get_line_buffer when used with data buffer that is not null terminated and does not end with \n (#40) 2024-03-01 02:47:04 +01:00
red0124
c5b50f2b47
Fix compile issues for c++20 (#39) 2024-03-01 00:52:00 +01:00
red0124
d8dcce7f2a
Fix buffer overflow on multiline csv data containing null characters (#38) 2024-02-29 22:03:20 +01:00
red0124
126329608c
Add macOS ci (#36)
* Add macOS ci, update README
2024-02-28 22:20:26 +01:00
ado
ddaa446819 Update version 2024-02-28 00:58:14 +01:00
red0124
8bad2d72ea
Merge pull request #35 from red0124/feature/csv_buffer
Feature/csv buffer
2024-02-28 00:13:20 +01:00
ado
899a6e6f5e [skip ci] Update README 2024-02-28 00:08:04 +01:00
ado
0d3d8fa83e [skip ci] Update README 2024-02-28 00:04:59 +01:00
ado
7bbe2879cd [skip ci] Update README 2024-02-28 00:02:58 +01:00
ado
063d56fad9 [skip ci] Update README 2024-02-28 00:01:37 +01:00
ado
df78865f04 [skip ci] Update README 2024-02-27 23:56:13 +01:00
ado
852481d233 Fix converter unit tests 2024-02-27 02:49:50 +01:00
ado
c516a6f826 Fix extraction tests 2024-02-26 02:37:30 +01:00
ado
b660310acf [skip ci] Merge with master 2024-02-25 18:27:26 +01:00
ado
0a695cf09e Add ss::uint8 and ss::int8, add unit tests for them 2024-02-25 17:46:35 +01:00
ado
f8e14b1fcf [skip ci] Add std:: to invoked C std lib functions 2024-02-25 13:03:52 +01:00
ado
0ebbee1174 [skip ci] Remove obsolete check from get_line_buffer 2024-02-25 12:10:46 +01:00
ado
b3f3bdf8d1 [skip ci] Update ssp.hpp 2024-02-25 10:54:56 +01:00
ado
f4a06d40e7 Fix non-POSIX get_line 2024-02-25 10:53:21 +01:00
ado
f2ff40a625 Add strict_realloc 2024-02-25 10:42:11 +01:00
ado
110ee840cc Fix header usage functionality 2024-02-25 03:54:33 +01:00
ado
05f87bc78b [skip ci] Fix line method 2024-02-25 02:57:46 +01:00
ado
88e711a5f7 Add positions method to parser, write unit tests for it, update other parser tests 2024-02-25 02:06:48 +01:00
ado
383de57f9a Make ssize_t equal to intptr_t for non-POSIX environments 2024-02-24 19:09:41 +01:00
ado
c6f6ba9821 [skip ci] Reduce number of runs for no new line at end of data test 2024-02-24 14:37:26 +01:00
ado
c5e491041d [skip ci] Disable file mode for no new line at end of data test 2024-02-24 14:17:23 +01:00
ado
21b543ea4f [skip ci] Disable buffer mode no new line at end of data test 2024-02-24 13:55:31 +01:00
ado
8881649aca Update get_line_buffer, update new version of get_line_file to work with data that has no new line at eof 2024-02-23 23:25:03 +01:00
red0124
a27fd121a1
Merge pull request #34 from red0124/improvement/getline_update
Fix reallocation issues with non-POSIX get_line
2024-02-23 23:04:19 +01:00
ado
c0d3087f85 Fix reallocation issues with non-POSIX get_line 2024-02-23 21:56:44 +01:00
ado
273e8ad950 Merge with master, resolve conflicts 2024-02-23 02:49:13 +01:00
red0124
f8fdb97151
Merge pull request #33 from red0124/improvement/getline_update
Improvement/getline update
2024-02-23 01:53:47 +01:00
ado
c0ee100f99 Update ssp.hpp 2024-02-23 01:04:40 +01:00
ado
09e628020d Fix typo in common.hpp 2024-02-23 01:01:31 +01:00
ado
ea21b9ba04 Fix get_line possible leak 2024-02-23 00:59:58 +01:00
ado
230da6a3f2 Update getline implementation for non-POSIX systems 2024-02-22 23:59:58 +01:00
ado
3ea8adedfd Add extended tests with no new line at eof 2024-02-22 00:30:05 +01:00
ado
57ba23c574 [skip ci] Restore extended tests 2024-02-21 21:53:01 +01:00
ado
6516c6cc94 [skip ci] Restore CMAKE_GITHUB_CI, disable some extended tests 2024-02-21 21:38:27 +01:00
ado
cbb0a1ad8e [skip ci] Fix coverage.yml typo, disable extended tests 2024-02-21 21:23:38 +01:00
ado
1798b4c6f3 [skip ci] Update coverage.yml, fix lcov usage error 2024-02-21 20:52:34 +01:00
ado
90a116ac7b [skip ci] Update coverage.yml, add lcov filters 2024-02-21 20:19:42 +01:00
ado
cbbe0acb25 [skip ci] Fix ci_install_lcov.sh 2024-02-21 20:06:32 +01:00
ado
b993eb8852 [skip ci] Update coverage.yml to work with lcov2.0 2024-02-21 19:41:06 +01:00
ado
5e32d722e8 [skip ci] Update coverage.yml, add unit test for files and buffers without new line at the end of the file 2024-02-21 02:19:14 +01:00
ado
59f6591da3 [skip ci] Update coverage.yml 2024-02-21 01:25:53 +01:00
ado
9d96a7d47f [skip ci] Update coverage.yml 2024-02-19 01:59:10 +01:00
ado
d4fc2ee561 [skip ci] Update .gitignore 2024-02-19 01:23:09 +01:00
ado
d422667477 [skip ci] Remove some magic numbers 2024-02-19 01:16:19 +01:00
ado
aaa22046a5 Add null data buffer error handler and unit test, resolve TODOs 2024-02-19 01:00:42 +01:00
ado
775b8c93e2 Fix parser tests part 2 segment issues 2024-02-19 00:16:22 +01:00
ado
e4fba8a918 Split parser tests part 2 into additional segments 2024-02-19 00:04:09 +01:00
ado
11d57bd073 [skip ci] Try fix MINGW ci 2024-02-18 23:40:53 +01:00
ado
45b840a30a [skip ci] Try fix MINGW ci 2024-02-18 23:22:37 +01:00
ado
8bb773625b [skip ci] Try fix MINGW ci 2024-02-18 23:16:10 +01:00
ado
0466c7234c [skip ci] Try fix MINGW ci 2024-02-18 23:13:16 +01:00
ado
d21c387a33 [skip ci] Try fix MINGW ci 2024-02-18 23:07:26 +01:00
ado
d019edb2bf [skip ci] Update unit test CMakeList for MINGW 2024-02-18 21:58:06 +01:00
ado
a2666816de Add maybe_unused attribute to test helper functions 2024-02-18 19:58:25 +01:00
ado
417a03a8a4 Fix unit tests build 2024-02-18 19:36:41 +01:00
ado
baf4317ffa Add more unit tests for buffer mode 2024-02-18 18:58:51 +01:00
ado
63a618957b Add more unit tests for buffer mode 2024-02-18 18:46:26 +01:00
ado
dbaa8131e7 Update ssp.hpp 2024-02-18 15:29:03 +01:00
ado
ef8cf30919 Merge remote-tracking branch 'origin/master' into feature/csv_buffer 2024-02-18 15:27:46 +01:00
ado
42d618ed64 Make CI unit tests run sequentially 2024-02-18 14:12:10 +01:00
ado
8b07f7d6cb Remove tmpnam usage for random file generation 2024-02-18 13:55:14 +01:00
ado
82f8ed12b4 Fix out of bounds reading for get_line_buffer, write more buffer mode unit tests 2024-02-18 11:20:36 +01:00
ado
e89e268280 Update test helpers random file name generator 2024-02-17 22:21:37 +01:00
ado
aacf690640 Fix unit test CMakeLists typo 2024-02-17 21:10:25 +01:00
ado
f8c5757d99 Fix unit test CMakeLists 2024-02-17 21:08:24 +01:00
ado
ce03c371ae Split parser tests into multiple files, add more tests for buffer mode 2024-02-17 21:05:30 +01:00
ado
6c859959d6 Add more unit tests for buffer mode 2024-02-17 17:43:33 +01:00
ado
4434db29f6 Merge remote-tracking branch 'origin/master' into feature/csv_buffer 2024-02-17 03:39:11 +01:00
red0124
fdd153b63e
Merge pull request #32 from red0124/bugfix/header_usage_on_empty_file
Bugfix/header usage on empty file
2024-02-17 03:27:17 +01:00
ado
c05ab35a33 [skip ci] Update version 2024-02-17 02:41:02 +01:00
ado
8d1637cad1 Update single-header.yml 2024-02-17 02:34:51 +01:00
ado
7530be1c44 Fix undefined behavior when trying to fetch header information on an empty file 2024-02-17 02:26:19 +01:00
ado
7062888d72 Fix msvc build 2024-02-17 01:07:29 +01:00
ado
f04ede3a49 Add option to read csv data from a buffer, add some unit tests for the new feature 2024-02-17 00:55:36 +01:00
ado
4bedc32b63 [skip ci] Update README 2023-08-08 17:26:48 +02:00
ado
7822351a0b [skip ci] Update coverage badge 2023-08-08 17:23:36 +02:00
ado
a755f6c455 [skip ci] Update version 2023-08-08 16:40:21 +02:00
red0124
4db88c0490
Merge pull request #30 from red0124/feature/coverage_ci
Feature/coverage ci
2023-08-08 16:37:11 +02:00
ado
8b72deb1ed Remove some comments, update README, update ssp.hpp 2023-08-08 14:38:49 +02:00
ado
672b89b213 Add ability to convert larger numbers without fast_float, write unit tests 2023-08-08 14:11:51 +02:00
ado
b9d2c2aad9 Fix big number extraction test 2023-08-08 12:57:15 +02:00
ado
32cbfe1d17 Update extraction tests, update coverage-ci, update ssp.hpp 2023-08-08 12:43:56 +02:00
ado
848689451c Add big float extraction test 2023-08-08 12:29:04 +02:00
ado
b7e5dd28b8 Update splitter tests 2023-08-08 12:10:20 +02:00
ado
e316558a7b Add parser test for invalid rows with header 2023-08-08 11:07:26 +02:00
ado
1c6eacad30 Update coverage-ci badge 2023-08-08 10:05:07 +02:00
ado
5ac506d8f0 Add coverage-ci badge 2023-08-08 10:01:47 +02:00
ado
6e27f35209 Update coverage ci 2023-08-08 02:08:31 +02:00
ado
49a3a20e68 Update coverage ci 2023-08-08 02:03:42 +02:00
ado
1d0911ab3c Update coverage ci 2023-08-08 01:46:20 +02:00
ado
7d3d02f11d Update coverage ci 2023-08-08 01:33:59 +02:00
ado
17c21e260f Update coverage ci 2023-08-08 01:25:43 +02:00
ado
2504c3574d Update coverage ci 2023-08-08 01:19:21 +02:00
ado
0bd78120f3 Update coverage ci 2023-08-08 01:15:57 +02:00
ado
ed8f4e3147 Update coverage ci 2023-08-08 01:12:05 +02:00
ado
d4c9227830 Update coverage ci 2023-08-08 01:06:27 +02:00
ado
f232c7d995 Add coverage ci 2023-08-08 00:43:02 +02:00
ado
14ccb88664 Add single-header-ci badge 2023-08-07 19:58:56 +02:00
red0124
34833837ba
Merge pull request #29 from red0124/improvement/ci_update
Improvement/ci update
2023-08-07 19:56:12 +02:00
ado
5f27595ec1 Add workflow_dispatch to all workflows 2023-08-07 18:48:21 +02:00
ado
65371d05f3 Update single-header.yml 2023-08-07 18:40:54 +02:00
ado
31b924736f [skip ci] Update single-header.yml 2023-08-07 18:39:02 +02:00
ado
ffca94d47d [skip ci] Update single-header.yml 2023-08-07 18:37:58 +02:00
ado
7ba66ff99d Update single-header.yml 2023-08-07 18:36:38 +02:00
ado
f1e127dd2b [skip ci] Update single-header.yml 2023-08-07 18:35:30 +02:00
ado
3e3eb1b61c [skip ci] Update single-header.yml 2023-08-07 18:33:38 +02:00
ado
b9d8eb860e [skip ci] Restore ssp.hpp 2023-08-07 18:30:35 +02:00
ado
4139b50cd9 Test single-header-ci 2023-08-07 18:04:38 +02:00
ado
8924ad12e5 Update single-header.yml 2023-08-07 18:00:28 +02:00
ado
254bd24bbd Update single-header.yml 2023-08-07 17:57:26 +02:00
ado
55d1bbcf86 Update single-header.yml 2023-08-07 17:46:20 +02:00
ado
3e3c922624 [skip ci] Update single-header.yml 2023-08-07 17:45:03 +02:00
ado
c0e6e56364 Update single-header.yml 2023-08-07 17:40:14 +02:00
ado
66f57ba66a Update single-header.yml 2023-08-07 17:34:28 +02:00
ado
d37ec12bb5 [skip ci] Add single-header.yml 2023-08-07 17:28:53 +02:00
ado
9d7441b178 Update ci workflows 2023-08-07 16:53:04 +02:00
ado
dcf7e924ad Update ci workflows 2023-08-07 16:05:51 +02:00
ado
41c4bf9d35 Merge remote-tracking branch 'origin/master' into improvement/ci_update 2023-08-07 15:56:32 +02:00
ado
7b16f24d03 [skip ci] Update version and spp.hpp 2023-08-07 15:29:50 +02:00
red0124
db582709d4
Merge pull request #28 from red0124/improvement/integral_conversion_from_chars
Update conversion for integral values to use std::from_chars
2023-08-07 15:23:34 +02:00
ado
deee577c1f Update conversion for integral values to use std::from_chars 2023-08-07 14:11:06 +02:00
ado
535138d9b8 Update ci workflows 2023-08-07 02:16:48 +02:00
ado
f106889ada Update ci workflows 2023-08-07 01:49:49 +02:00
ado
e7045ce437 Update ci workflows 2023-08-07 01:33:50 +02:00
ado
d86c8e9fe8 Update ci workflows 2023-08-07 01:26:40 +02:00
ado
c0ef691889 Update ci workflows 2023-08-07 01:21:05 +02:00
ado
80c189f9c5 Update ci workflows 2023-08-07 01:16:45 +02:00
ado
07373ea043 Update ci workflows 2023-08-07 01:09:37 +02:00
ado
236b5da9c2 Update msys2-clang and msys2-gcc ci 2023-08-07 00:56:44 +02:00
ado
a1f01ec5cb Update clang ci 2023-08-07 00:44:43 +02:00
ado
cd6c2df359 Update msys2-clang ci 2023-08-06 23:49:55 +02:00
ado
9afe24785b Fix clang ci yaml 2023-08-06 23:41:44 +02:00
ado
479cf4bbe7 Update clang and gcc ci 2023-08-06 23:23:32 +02:00
ado
6d1aa7c7a9 [skip ci] Update version 2023-08-06 22:39:35 +02:00
ado
5fd6b561ed [skip ci] Update README 2023-08-06 22:36:59 +02:00
ado
1176c42fba Update README 2023-08-06 21:42:57 +02:00
red0124
e3c7db53f7
Merge pull request #27 from red0124/improvement/throw_on_error
Improvement/throw_on_error
2023-08-06 21:34:05 +02:00
ado
b913344722 Update README and ssp.hpp 2023-08-06 20:02:08 +02:00
ado
7df96d95c7 Make line method work on errors too 2023-08-06 19:56:28 +02:00
ado
6b4f456654 Update ssp.hpp 2023-08-06 19:44:26 +02:00
ado
9c11f22d79 Update rethrow error message line number 2023-08-06 19:43:41 +02:00
ado
975ef2e4a5 Add invalid multiline parser tests 2023-08-06 17:39:09 +02:00
ado
2c7bc763a5 Update ssp.hpp, add more parser unit tests 2023-08-06 16:55:52 +02:00
ado
a152a8cb4a Apply minor change to iterator, add handling of invalid header 2023-08-06 15:55:23 +02:00
ado
a378998e34 Update parser exception handling for invalid split/conversion 2023-08-06 14:04:21 +02:00
ado
77a69fbd6d [skip ci] Update README 2023-08-06 12:52:40 +02:00
ado
6a3ba48a2c [skip ci] Update README 2023-08-06 12:51:11 +02:00
ado
911d81248b [skip ci] Update README 2023-08-06 12:48:53 +02:00
ado
b6cef9577d [skip ci] Update README 2023-08-06 12:35:53 +02:00
ado
48117c038a [skip ci] Update README 2023-08-06 12:33:43 +02:00
ado
3c0a440fe9 [skip ci] Update README 2023-08-06 12:33:05 +02:00
ado
974257e099 [skip ci] Update README 2023-08-06 12:26:54 +02:00
ado
6a832ba11a [skip ci] Update README 2023-08-06 00:16:43 +02:00
ado
0e184d07ce [skip ci] Update README 2023-08-05 23:57:57 +02:00
ado
f17a3bdc59 Update ssp.hpp 2023-08-05 22:04:19 +02:00
ado
952ff236ff Fix header generation on empty line buffer 2023-08-05 20:48:01 +02:00
ado
7d44d503d9 Update the remainder of the unit tests to handle throw_on_error, make parser not set error if all lines are empty 2023-08-05 19:58:00 +02:00
ado
2e1c4c97ec Fix handling of invalid conversion/split within parser, update parser tests 2023-08-05 18:11:13 +02:00
ado
7c9ba953ad Fix typo 2023-08-05 13:52:31 +02:00
ado
3170cb661c Minor test changes 2023-08-05 13:49:10 +02:00
ado
03870aa1ba Update header parsing 2023-08-05 13:30:14 +02:00
ado
2b132bc33a Apply minor changes to tests 2023-08-05 12:05:17 +02:00
ado
a7ea9e42e5 Update error handling methods 2023-08-05 11:45:31 +02:00
ado
81484f737f Remove std::from_chars from extract.hpp 2023-08-04 22:31:29 +02:00
ado
634abdd38b Add invalid header usage tests 2023-08-04 21:22:23 +02:00
ado
c981ed6644 Write additional parser tests, update some of the existing tests to work with throw_on_error 2023-08-04 16:48:07 +02:00
ado
5173e7afbc Update test file names 2023-08-03 23:09:25 +02:00
ado
5963df5ffd Update test meson.build 2023-08-03 23:04:05 +02:00
ado
19538597e5 Make test_parser2 segments creage unique file names with respect to each other 2023-08-03 21:53:40 +02:00
ado
7656ba0a55 Fix CMake typo 2023-08-03 21:32:35 +02:00
ado
761445bd36 Split test_parser2 into multiple segments 2023-08-03 21:28:41 +02:00
ado
62e19df343 Update test_parser2 to include ignore_header and ignore_empty options 2023-08-03 17:04:06 +02:00
ado
0858369c7d Fix test CMakeList.txt 2023-07-31 23:30:26 +02:00
ado
2fe2c391fe WIP, Continue writing additional parser tests, add big object flag to test cmake for mingw 2023-07-31 23:25:47 +02:00
red0124
57abdb3923
Merge pull request #26 from red0124/bugfix/terminated_escape_on_quoted_multiline
Fix bug where the parsing of data containing escaped new lines within…
2023-07-31 23:05:52 +02:00
ado
5f46259914 Merge with bugfix 2023-07-31 21:52:08 +02:00
ado
fbe19de70c Merge remote-tracking branch 'origin/bugfix/terminated_escape_on_quoted_multiline' into improvement/throw_on_error 2023-07-31 21:50:45 +02:00
ado
27ef7f2e21 Fix bug where the parsing of data containing escaped new lines within quotes would produce bad outputs 2023-07-31 21:49:15 +02:00
ado
1d94989a4d Remove big obj flag for compilers except MSVC within test cmake 2023-07-30 23:06:04 +02:00
ado
84a7d46cbf WIP, Continue writing additional parser tests, add big object flag to test cmake 2023-07-30 22:59:37 +02:00
ado
9bf1dd6041 WIP, Continue writing additional parser tests 2023-07-30 19:50:29 +02:00
ado
ed71d963e0 WIP, Continue writing additional parser tests 2023-07-30 18:29:45 +02:00
ado
d9384f5d84 WIP, Continue writing additional parser tests 2023-07-30 17:29:26 +02:00
ado
a7608d825e WIP, Continue writing additional parser tests 2023-07-29 23:24:09 +02:00
ado
ba45fe1839 WIP, Continue writing additional parser tests 2023-07-29 22:57:33 +02:00
ado
37f1fb534f WIP, Continue writing additional parser tests 2023-07-29 22:51:32 +02:00
ado
e32905b59d WIP, Continue writing additional parser tests 2023-07-29 22:44:01 +02:00
ado
d6cf9bd006 WIP, Move additional parser tests to separate file 2023-07-29 20:41:31 +02:00
ado
f28f000035 WIP, Continue writing additional parser tests 2023-07-29 19:23:24 +02:00
ado
560bf815dc Merge with master 2023-07-29 16:53:24 +02:00
ado
f58091cf02 Fix test_parser.cpp formating 2023-07-29 16:51:34 +02:00
ado
a9903f7441 Add TODO comments 2023-07-29 16:47:50 +02:00
red0124
78c75abec7
Merge pull request #25 from red0124/feature/win_msvc_ci
Feature/win msvc ci
2023-07-29 16:25:26 +02:00
ado
76091263f2 Fix win-msvc ci badge 2023-07-29 16:11:27 +02:00
ado
d99cba274e Add win-msvc ci badge 2023-07-29 16:08:33 +02:00
ado
e70fc05646 Fix msvc ci configure step 2023-07-29 15:39:51 +02:00
ado
21d1096edf Fix msvc ci yaml issues 2023-07-29 15:37:53 +02:00
ado
72b74a3b64 Fix msvc ci configure step 2023-07-29 15:35:29 +02:00
ado
b7f6009eb9 Fix msvc ci yaml issues 2023-07-29 15:30:05 +02:00
ado
bdfd896861 Fix msvc vs 16 ci 2023-07-29 15:17:39 +02:00
ado
d6bc8086b2 Fix msvc ci configure step 2023-07-29 15:03:25 +02:00
ado
f762736da2 Fix msvc ci configure step 2023-07-29 14:59:49 +02:00
ado
739077a8db Fix msvc ci vs version names 2023-07-29 14:54:25 +02:00
ado
3dcb2ce2ef Add vs16 for msvc ci 2023-07-29 14:22:41 +02:00
ado
e4d9e10ac3 Update msvc ci 2023-07-29 14:14:02 +02:00
ado
a98742945b Update test helpers 2023-07-29 13:56:00 +02:00
ado
5d6c2c4af5 Remove move count unit test from parser tests 2023-07-29 01:44:19 +02:00
ado
58857fff2d Add config option for ctest within win-msvc.yml 2023-07-29 01:18:11 +02:00
ado
e6a7e09975 Add config option for ctest within win-msvc.yml 2023-07-29 01:18:01 +02:00
ado
705ce422f0 Add constexpr values to lambda to handle msvc issue with respect to captured constexpr values 2023-07-29 01:11:29 +02:00
ado
75f5ee2a55 Add _CRT_SECURE_NO_WARNINGS define to remove msvc warnings 2023-07-29 00:56:08 +02:00
ado
6baeb2d598 Add _CRT_SECURE_NO_WARNINGS define to remove msvc warnings 2023-07-29 00:53:04 +02:00
ado
7b1f49d304 Add _CRT_SECURE_NO_WARNINGS define to remove msvc warnings 2023-07-29 00:49:06 +02:00
ado
515ddad997 Add _CRT_SECURE_NO_WARNINGS define to remove msvc warnings 2023-07-29 00:45:33 +02:00
ado
fd39b5eef2 Try msvc ci 2023-07-29 00:38:27 +02:00
red0124
77185cb366
Merge pull request #24 from red0124/improvement/markdown_cicd_badge_link
Add hyperlink for cicd badges
2023-07-28 21:30:10 +02:00
ado
ad991d6a7d Add hyperlink for cicd badges 2023-07-28 21:28:32 +02:00
ado
31cbd20f8f WIP, Continue writing additional parser tests 2023-07-28 00:17:22 +02:00
red0124
304ca6ef0f
Merge pull request #23 from red0124/bugfix/single_header_include_update
Update ssp.hpp to handle issue fixed in the previous commit
2023-07-25 11:24:54 +02:00
ado
103ff33f21 Update ssp.hpp to handle issue fixed in the previous commit 2023-07-25 11:13:32 +02:00
ado
6dfc30d5c9 WIP, Fix merge compile error 2023-07-25 01:01:06 +02:00
ado
a6db4a7ad2 WIP, Write additional parser tests 2023-07-25 00:56:38 +02:00
ado
77631b8c0d Merge remote-tracking branch 'origin/master' into improvement/throw_on_error 2023-07-25 00:53:51 +02:00
ado
7d0a5598a8 Merge branch 'improvement/throw_on_error' of https://github.com/red0124/ssp into improvement/throw_on_error 2023-07-23 13:12:37 +02:00
ado
29c471e33a Update valid method for throw_on_error 2023-07-16 20:26:09 +02:00
ado
0ff86630f8 Add comment 2023-07-16 18:26:24 +02:00
ado
5ab41c0315 Fix bug where header reading corrupted the current line buffer 2023-07-16 18:25:15 +02:00
ado
03f5b839fc Minor changes 2023-07-13 22:29:49 +02:00
ado
9f4bcb03e1 Add buffer_size_ to reader move constructor and move operator=, apply minor changes 2023-07-12 23:33:33 +02:00
ado
567995aafb Remove deprecated code, add exception.hpp to single_header_generator.py script 2023-07-10 23:39:08 +02:00
ado
956156b412 Fix resplit segmentation fault, fix issues when working with empty files 2023-07-10 22:21:07 +02:00
ado
a7a97b3ba8 Make throw_on_error and string_error separate options, update parser to have line reading a separate function with respect to splitting 2023-07-10 02:39:24 +02:00
ado
41b89d1d3d Add converter tests with throw_on_error 2023-07-09 17:11:52 +02:00
ado
f3225b8b16 Add splitter tests with throw_on_error 2023-07-09 12:54:39 +02:00
ado
eeac30651a Implement throw_on_error functionality, rename some template parameters 2023-06-29 23:41:03 +02:00
52 changed files with 6961 additions and 2927 deletions

7
.github/fuzz/makefile vendored Normal file
View File

@ -0,0 +1,7 @@
EXE=ssp_fuzz
all:
clang++ ${CXXFLAGS} ssp_fuzz.cpp -fsanitize=fuzzer -std=c++17 -o ${EXE}
run:
./${EXE} -max_total_time=900

81
.github/fuzz/ssp_fuzz.cpp vendored Normal file
View File

@ -0,0 +1,81 @@
#include "../../ssp.hpp"
#include <filesystem>
#include <iostream>
#include <unistd.h>
template <typename... Ts>
void test_ssp_file_mode(const uint8_t* data, size_t size,
std::string delim = ss::default_delimiter) {
std::string file_name = std::filesystem::temp_directory_path().append(
"ss_fuzzer" + std::to_string(getpid()) + ".csv");
FILE* file = std::fopen(file_name.c_str(), "wb");
if (!file) {
std::exit(1);
}
std::fwrite(data, size, 1, file);
std::fclose(file);
ss::parser<Ts...> p{file_name.c_str(), delim};
while (!p.eof()) {
try {
const auto& [s0, s1] =
p.template get_next<std::string, std::string>();
if (s0.size() == 10000) {
std::cout << s0.size() << std::endl;
}
} catch (ss::exception& e) {
continue;
}
}
std::remove(file_name.c_str());
}
template <typename... Ts>
void test_ssp_buffer_mode(const uint8_t* data, size_t size,
std::string delim = ss::default_delimiter) {
ss::parser<Ts...> p{(const char*)data, size, delim};
while (!p.eof()) {
try {
const auto& [s0, s1] =
p.template get_next<std::string, std::string>();
if (s0.size() == 10000) {
std::cout << s0.size() << std::endl;
}
} catch (ss::exception& e) {
continue;
}
}
}
template <typename... Ts>
void test_ssp(const uint8_t* data, size_t size) {
test_ssp_file_mode<Ts...>(data, size);
test_ssp_file_mode<Ts..., ss::throw_on_error>(data, size);
test_ssp_file_mode<Ts...>(data, size, ":::");
test_ssp_file_mode<Ts..., ss::throw_on_error>(data, size, ":::");
test_ssp_buffer_mode<Ts...>(data, size);
test_ssp_buffer_mode<Ts..., ss::throw_on_error>(data, size);
test_ssp_buffer_mode<Ts...>(data, size, ":::");
test_ssp_buffer_mode<Ts..., ss::throw_on_error>(data, size, ":::");
}
extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) {
using escape = ss::escape<'\\'>;
using quote = ss::quote<'"'>;
using trim = ss::trim<' ', '\t'>;
using multiline_r = ss::multiline_restricted<5>;
test_ssp<>(data, size);
test_ssp<escape>(data, size);
test_ssp<quote>(data, size);
test_ssp<trim>(data, size);
test_ssp<quote, escape>(data, size);
test_ssp<escape, quote, multiline_r, trim>(data, size);
test_ssp<escape, quote, multiline_r, trim, ss::ignore_empty>(data, size);
return 0;
}

74
.github/workflows/coverage.yml vendored Normal file
View File

@ -0,0 +1,74 @@
name: coverage-ci
on:
workflow_dispatch:
push:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
pull_request:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
jobs:
test_coverage:
if: >-
! contains(toJSON(github.event.commits.*.message), '[skip ci]') &&
! contains(toJSON(github.event.commits.*.message), '[skip github]')
runs-on: ubuntu-latest
name: "Coverage"
container:
image: gcc:latest
options: -v /usr/local:/host_usr_local
steps:
- uses: actions/checkout@v1
- uses: friendlyanon/fetch-core-count@v1
id: cores
- name: CMake
run: echo "/host_usr_local/bin" >> $GITHUB_PATH
- name: Install dependencies
run: script/ci_install_deps.sh
- name: Install test coverage tools
run: |
apt update
apt install -y gcovr
- name: Install lcov2.0
run: script/ci_install_lcov.sh
- name: Configure
run: cmake -S test -B build -D CMAKE_BUILD_TYPE=Debug -D CMAKE_CXX_FLAGS="-Wall -fprofile-arcs -ftest-coverage --coverage -fno-elide-constructors -fno-default-inline"
- name: Build
run: cmake --build build -j ${{steps.cores.outputs.count}}
- name: Run
working-directory: build
run: ctest --output-on-failure
- name: Generate coverage report
run: |
lcov -d . -c -o out.info --rc branch_coverage=1 --no-external --filter branch --filter line --ignore-errors mismatch
lcov -e out.info '*include/ss*hpp' -o filtered.info
- name: Invoke coveralls
uses: coverallsapp/github-action@v2
with:
github-token: ${{secrets.GITHUB_TOKEN}}
file: filtered.info
format: lcov

43
.github/workflows/fuzz.yml vendored Normal file
View File

@ -0,0 +1,43 @@
name: fuzz-ci
on:
workflow_dispatch:
push:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
pull_request:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
jobs:
clang_tests:
if: >-
! contains(toJSON(github.event.commits.*.message), '[skip ci]') &&
! contains(toJSON(github.event.commits.*.message), '[skip github]')
runs-on: ubuntu-latest
name: "Fuzzing"
container:
image: silkeh/clang:15
options: -v /usr/local:/host_usr_local
steps:
- uses: actions/checkout@v1
- name: Build
working-directory: .github/fuzz
run: make
- name: Run
working-directory: .github/fuzz
run: make run

55
.github/workflows/macos-apple-clang.yml vendored Normal file
View File

@ -0,0 +1,55 @@
name: macos-apple-clang-ci
on:
workflow_dispatch:
push:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
pull_request:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
jobs:
clang_tests:
if: >-
! contains(toJSON(github.event.commits.*.message), '[skip ci]') &&
! contains(toJSON(github.event.commits.*.message), '[skip github]')
strategy:
matrix:
xcode: ['13.4.1', '14.1']
type: [Release, Debug]
runs-on: macos-12
env:
DEVELOPER_DIR: /Applications/Xcode_${{matrix.xcode}}.app/Contents/Developer
name: "Xcode ${{matrix.xcode}}: ${{matrix.type}}"
steps:
- uses: actions/checkout@v3
- uses: friendlyanon/fetch-core-count@v1
id: cores
- name: Install dependencies
run: script/ci_install_deps.sh
- name: Configure
run: cmake -S test -B build -DCMAKE_BUILD_TYPE=${{matrix.type}}
- name: Build
run: cmake --build build -j ${{steps.cores.outputs.count}}
- name: Run
working-directory: build
run: ctest --output-on-failure

51
.github/workflows/single-header.yml vendored Normal file
View File

@ -0,0 +1,51 @@
name: single-header-ci
on:
workflow_dispatch:
push:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
pull_request:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
jobs:
single_header_tests:
if: >-
! contains(toJSON(github.event.commits.*.message), '[skip ci]') &&
! contains(toJSON(github.event.commits.*.message), '[skip github]')
runs-on: ubuntu-latest
name: "Single Header Test"
container:
image: gcc:latest
options: -v /usr/local:/host_usr_local
steps:
- uses: actions/checkout@v1
- uses: friendlyanon/fetch-core-count@v1
id: cores
- name: Install dependencies
run: |
apt-get update
apt-get install -y git
- name: Single header update check
run: |
script/single_header_generator.py > tmp.hpp
diff ssp.hpp tmp.hpp
- name: Single header compile check
run: ./test/test_single_header.sh

View File

@ -1,6 +1,8 @@
name: ubuntu-latest-clang-ci
on:
workflow_dispatch:
push:
branches:
- master
@ -23,20 +25,22 @@ jobs:
strategy:
matrix:
version: [11, 10, 9, 8, 7]
# cmake clang12 is not able to compile a simple test program.
# /usr/bin/ld: cannot find -lunwind
version: ['latest', '15', '14', '13', '11', '10', '9', '8', '7']
type: [Release, Debug]
runs-on: ubuntu-latest
name: Clang ${{ matrix.version }}
name: "Clang ${{matrix.version}}: ${{matrix.type}}"
container:
image: teeks99/clang-ubuntu:${{ matrix.version }}
image: silkeh/clang:${{matrix.version}}
options: -v /usr/local:/host_usr_local
env:
CC: clang-${{ matrix.version }}
CXX: clang++-${{ matrix.version }}
CC: clang
CXX: clang++
CXXFLAGS: -stdlib=libc++
steps:
@ -55,11 +59,11 @@ jobs:
script/ci_install_deps.sh
- name: Configure
run: cmake -S test -B build -D CMAKE_BUILD_TYPE=Debug
run: cmake -S test -B build -DCMAKE_BUILD_TYPE=${{matrix.type}}
- name: Build
run: cmake --build build -j ${{steps.cores.outputs.count}}
- name: Run
working-directory: build
run: ctest --output-on-failure -j ${{ steps.cores.outputs.count }}
run: ctest --output-on-failure

View File

@ -1,6 +1,8 @@
name: ubuntu-latest-gcc-ci
on:
workflow_dispatch:
push:
branches:
- master
@ -23,15 +25,15 @@ jobs:
strategy:
matrix:
version: [10, 9, 8]
version: ['latest', '12', '11', '10', '9', '8']
type: [Release, Debug]
runs-on: ubuntu-latest
name: GCC ${{ matrix.version }}
name: "GCC ${{matrix.version}}: ${{matrix.type}}"
container:
image: gcc:${{matrix.version}}
options: -v /usr/local:/host_usr_local
steps:
@ -47,11 +49,11 @@ jobs:
run: script/ci_install_deps.sh
- name: Configure
run: cmake -S test -B build -D CMAKE_BUILD_TYPE=Debug
run: cmake -S test -B build -D CMAKE_BUILD_TYPE=${{matrix.type}}
- name: Build
run: cmake --build build -j ${{steps.cores.outputs.count}}
- name: Run
working-directory: build
run: ctest --output-on-failure -j ${{ steps.cores.outputs.count }}
run: ctest --output-on-failure

View File

@ -1,6 +1,8 @@
name: ubuntu-latest-icc-ci
on:
workflow_dispatch:
push:
branches:
- master
@ -27,6 +29,7 @@ jobs:
! contains(toJSON(github.event.commits.*.message), '[skip github]')
runs-on: ubuntu-latest
defaults:
run:
shell: bash
@ -50,7 +53,8 @@ jobs:
compiler-${{hashFiles('**/scripts/cache_exclude_linux.sh')}}
- name: Install icc
run: script/ci_install_icc.sh $LINUX_HPCKIT_URL $LINUX_CPP_COMPONENTS_WEB
run: >-
script/ci_install_icc.sh $LINUX_HPCKIT_URL $LINUX_CPP_COMPONENTS_WEB
- name: CMake
run: echo "/host_usr_local/bin" >> $GITHUB_PATH
@ -70,4 +74,4 @@ jobs:
- name: Run
working-directory: build
run: ctest --output-on-failure -j ${{ steps.cores.outputs.count }}
run: ctest --output-on-failure

65
.github/workflows/win-msvc.yml vendored Normal file
View File

@ -0,0 +1,65 @@
name: win-msvc-ci
on:
workflow_dispatch:
push:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
pull_request:
branches:
- master
- feature/**
- improvement/**
- bugfix/**
jobs:
ci:
if: >-
! contains(toJSON(github.event.commits.*.message), '[skip ci]') &&
! contains(toJSON(github.event.commits.*.message), '[skip github]')
defaults:
run:
shell: bash
strategy:
fail-fast: false
matrix:
config:
- os: windows-2019
vs: "Visual Studio 16 2019"
- os: windows-latest
vs: "Visual Studio 17 2022"
build: [Debug, Release]
platform: [Win32, x64]
runs-on: ${{matrix.config.os}}
name: "${{matrix.config.vs}}: ${{matrix.platform}}: ${{matrix.build}}"
steps:
- name: checkout
uses: actions/checkout@v2
- name: Install dependencies
run: script/ci_install_deps.sh
- name: Configure
run: >-
cmake -S test -B build -D CMAKE_BUILD_TYPE=${{matrix.build}}
-G "${{matrix.config.vs}}" -A ${{matrix.platform}}
- name: Build
run: cmake --build build -j ${{steps.cores.outputs.count}}
- name: Run
working-directory: build
run: >-
ctest -C Debug --output-on-failure

View File

@ -1,6 +1,8 @@
name: win-msys2-clang-ci
on:
workflow_dispatch:
push:
branches:
- master
@ -21,45 +23,42 @@ jobs:
! contains(toJSON(github.event.commits.*.message), '[skip ci]') &&
! contains(toJSON(github.event.commits.*.message), '[skip github]')
name: ${{ matrix.msystem }}
runs-on: windows-latest
defaults:
run:
shell: msys2 {0}
strategy:
fail-fast: false
matrix:
include:
os: [windows-2019, windows-latest]
type: [Release, Debug]
config:
- msystem: "MINGW64"
install: >-
git mingw-w64-x86_64-cmake mingw-w64-x86_64-ninja
mingw-w64-x86_64-clang
type: Release
- msystem: "MINGW32"
install: >-
git mingw-w64-i686-cmake mingw-w64-i686-ninja
mingw-w64-i686-clang
type: Release
- msystem: "MINGW64"
install: >-
git mingw-w64-x86_64-cmake mingw-w64-x86_64-ninja
mingw-w64-x86_64-clang
type: Debug
- msystem: "MINGW32"
install: >-
git mingw-w64-i686-cmake mingw-w64-i686-ninja
mingw-w64-i686-clang
type: Debug
runs-on: ${{matrix.os}}
name: "${{matrix.config.msystem}}: ${{matrix.os}}: ${{matrix.type}}"
env:
CMAKE_GENERATOR: Ninja
steps:
- uses: actions/checkout@v2
- uses: msys2/setup-msys2@v2
with:
update: true
msystem: ${{ matrix.msystem }}
install: ${{ matrix.install }}
msystem: ${{matrix.config.msystem}}
install: ${{matrix.config.install}}
- name: Install dependencies
run: script/ci_install_deps.sh
@ -67,11 +66,11 @@ jobs:
- name: Configure
run: >-
cmake -DCMAKE_CXX_COMPILER=clang++ -S test -B build
-DCMAKE_BUILD_TYPE=Debug
-DCMAKE_BUILD_TYPE=${{matrix.type}}
- name: Build
run: cmake --build build -j ${{steps.cores.outputs.count}}
- name: Run
working-directory: build
run: ctest --output-on-failure -j ${{ steps.cores.outputs.count }}
run: ctest --output-on-failure

View File

@ -1,6 +1,8 @@
name: win-msys2-gcc-ci
on:
workflow_dispatch:
push:
branches:
- master
@ -21,55 +23,52 @@ jobs:
! contains(toJSON(github.event.commits.*.message), '[skip ci]') &&
! contains(toJSON(github.event.commits.*.message), '[skip github]')
name: ${{ matrix.msystem }}
runs-on: windows-latest
defaults:
run:
shell: msys2 {0}
strategy:
fail-fast: false
matrix:
include:
os: [windows-2019, windows-latest]
type: [Release, Debug]
config:
- msystem: "MINGW64"
install: >-
git mingw-w64-x86_64-cmake mingw-w64-x86_64-ninja
mingw-w64-x86_64-gcc
type: Release
- msystem: "MINGW32"
install: >-
git mingw-w64-i686-cmake mingw-w64-i686-ninja
mingw-w64-i686-gcc
type: Release
- msystem: "MINGW64"
install: >-
git mingw-w64-x86_64-cmake mingw-w64-x86_64-ninja
mingw-w64-x86_64-gcc
type: Debug
- msystem: "MINGW32"
install: >-
git mingw-w64-i686-cmake mingw-w64-i686-ninja
mingw-w64-i686-gcc
type: Debug
runs-on: ${{matrix.os}}
name: "${{matrix.config.msystem}}: ${{matrix.os}}: ${{matrix.type}}"
env:
CMAKE_GENERATOR: Ninja
steps:
- uses: actions/checkout@v2
- uses: msys2/setup-msys2@v2
with:
update: true
msystem: ${{ matrix.msystem }}
install: ${{ matrix.install }}
msystem: ${{matrix.config.msystem}}
install: ${{matrix.config.install}}
- name: Install dependencies
run: script/ci_install_deps.sh
- name: Configure
run: cmake -S test -B build -D CMAKE_BUILD_TYPE=Debug
run: cmake -S test -B build -D CMAKE_BUILD_TYPE=${{matrix.type}}
- name: Build
run: cmake --build build -j ${{steps.cores.outputs.count}}
- name: Run
working-directory: build
run: ctest --output-on-failure -j ${{ steps.cores.outputs.count }}
run: ctest --output-on-failure

6
.gitignore vendored
View File

@ -1,10 +1,10 @@
compile_commands.json
.clang-format
.ccls-cache/*
.clang-tidy
.ccls-cache/
.cache/
experiment/
build/
hbuild/
subprojects/*
!subprojects/*.wrap
ssp.cpp
ssp.bin

View File

@ -2,7 +2,7 @@ cmake_minimum_required(VERSION 3.14)
project(
ssp
VERSION 1.4.0
VERSION 1.8.0
DESCRIPTION "csv parser"
HOMEPAGE_URL "https://github.com/red0124/ssp"
LANGUAGES CXX

253
README.md
View File

@ -8,25 +8,29 @@
```
[![License](https://img.shields.io/badge/license-MIT-blue.svg)](https://opensource.org/licenses/MIT)
![ubuntu-latest-gcc](https://github.com/red0124/ssp/workflows/ubuntu-latest-gcc-ci/badge.svg)
![ubuntu-latest-clang](https://github.com/red0124/ssp/workflows/ubuntu-latest-clang-ci/badge.svg)
![ubuntu-latest-icc](https://github.com/red0124/ssp/workflows/ubuntu-latest-icc-ci/badge.svg)
![windows-msys2-gcc](https://github.com/red0124/ssp/workflows/win-msys2-gcc-ci/badge.svg)
![windows-msys2-clang](https://github.com/red0124/ssp/workflows/win-msys2-clang-ci/badge.svg)
[![coverage](https://coveralls.io/repos/github/red0124/ssp/badge.svg?branch=master)](https://coveralls.io/github/red0124/ssp?branch=master)
[![fuzz](https://github.com/red0124/ssp/workflows/fuzz-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/fuzz.yml)
[![single-header](https://github.com/red0124/ssp/workflows/single-header-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/single-header.yml)
[![ubuntu-latest-gcc](https://github.com/red0124/ssp/workflows/ubuntu-latest-gcc-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/ubuntu-latest-gcc.yml)
[![ubuntu-latest-clang](https://github.com/red0124/ssp/workflows/ubuntu-latest-clang-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/ubuntu-latest-clang.yml)
[![ubuntu-latest-icc](https://github.com/red0124/ssp/workflows/ubuntu-latest-icc-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/ubuntu-latest-icc.yml)
[![windows-msys2-gcc](https://github.com/red0124/ssp/workflows/win-msys2-gcc-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/win-msys2-gcc.yml)
[![windows-msys2-clang](https://github.com/red0124/ssp/workflows/win-msys2-clang-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/win-msys2-clang.yml)
[![windows-msvc](https://github.com/red0124/ssp/workflows/win-msvc-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/win-msvc.yml)
[![macos-apple-clang](https://github.com/red0124/ssp/workflows/macos-apple-clang-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/macos-apple-clang.yml)
A header only "csv" parser which is fast and versatile with modern C++ api. Requires compiler with C++17 support. [Can also be used to convert strings to specific types.](#The-converter)
A header only CSV parser which is fast and versatile with modern C++ API. Requires compiler with C++17 support. [Can also be used to efficiently convert strings to specific types.](#the-converter)
Conversion for floating point values invoked using [fast-float](https://github.com/fastfloat/fast_float) .
Function traits taken from [qt-creator](https://code.woboq.org/qt5/qt-creator/src/libs/utils/functiontraits.h.html) .
Conversion for floating point values invoked using [fast-float](https://github.com/fastfloat/fast_float) . \
Function traits taken from *qt-creator* .
# Example
Lets say we have a csv file containing students in a given format '$name,$age,$grade' and we want to parse and print all the valid values:
Lets say we have a CSV file containing students in a given format (Id,Age,Grade) and we want to parse and print all the valid values:
```shell
$ cat students.csv
James Bailey,65,2.5
Brian S. Wolfe,40,1.9
Nathan Fielder,37,Really good grades
Bill (Heath) Gates,65,3.3
```
```cpp
@ -34,12 +38,10 @@ Bill (Heath) Gates,65,3.3
#include <ss/parser.hpp>
int main() {
ss::parser p{"students.csv", ","};
ss::parser<ss::throw_on_error> p{"students.csv"};
for(const auto& [name, age, grade] : p.iterate<std::string, int, float>()) {
if (p.valid()) {
std::cout << name << ' ' << age << ' ' << grade << std::endl;
}
for (const auto& [id, age, grade] : p.iterate<std::string, int, float>()) {
std::cout << id << ' ' << age << ' ' << grade << std::endl;
}
return 0;
@ -55,14 +57,15 @@ Bill (Heath) Gates 65 3.3
# Features
* [Works on any type](#custom-conversions)
* Easy to use
* No exceptions
* Can work without exceptions
* [Works with headers](#headers)
* [Works with quotes, escapes and spacings](#setup)
* [Works with CSV data stored in buffers](#buffer-mode)
* [Works with values containing new lines](#multiline)
* [Columns and rows can be ignored](#special-types)
* Works with any type of delimiter
* [Works with any type of delimiter](#delimiter)
* Can return whole objects composed of converted values
* [Descriptive error handling can be enabled](#error-handling)
* [Error handling can be configured](#error-handling)
* [Restrictions can be added for each column](#restrictions)
* [Works with `std::optional` and `std::variant`](#special-types)
* Works with **`CRLF`** and **`LF`**
@ -71,7 +74,7 @@ Bill (Heath) Gates 65 3.3
# Single header
The library can be used with a single header file **`ssp.hpp`**, but it sufferes a slight performance loss when converting floating point values since the **`fast_float`** library is not present within the file.
The library can be used with a single header file **`ssp.hpp`**, but it suffers a significant performance loss when converting floating point values since the **`fast_float`** library is not present within the file.
# Installation
@ -82,28 +85,28 @@ $ cmake --configure .
$ sudo make install
```
*Note, this will also install the fast_float library*
*Note, this will also install the fast_float library.*\
The library supports [CMake](#Cmake) and [meson](#Meson) build systems
# Usage
## Headers
The parser can be told to use only certain columns by parsing the header. This can be done by using the **`use_fields`** method. It accepts any number of string-like arguments or even an **`std::vector<std::string>`** with the field names. If any of the fields are not found within the header or if any fields are defined multiple times it will result in an error.
The parser can be told to use only certain columns by parsing the header. This can be done with the **`use_fields`** method. It accepts any number of string-like arguments or even an **`std::vector<std::string>`** with the field names. If any of the fields are not found within the header or if any fields are defined multiple times it will result in an error.
```shell
$ cat students_with_header.csv
Name,Age,Grade
Id,Age,Grade
James Bailey,65,2.5
Brian S. Wolfe,40,1.9
Bill (Heath) Gates,65,3.3
```
```cpp
// ...
ss::parser p{"students.csv", ","};
p.use_fields("Name", "Grade");
ss::parser<ss::throw_on_error> p{"students_with_header.csv"};
p.use_fields("Id", "Grade");
for(const auto& [name, grade] : p.iterate<std::string, float>()) {
std::cout << name << ' ' << grade << std::endl;
for(const auto& [id, grade] : p.iterate<std::string, float>()) {
std::cout << id << ' ' << grade << std::endl;
}
// ...
```
@ -113,62 +116,88 @@ James Bailey 2.5
Brian S. Wolfe 1.9
Bill (Heath) Gates 3.3
```
The header can be ignored using the **`ss::ignore_header`** [setup](#Setup) option or by calling the **`ignore_next`** metod after the parser has been constructed.
The header can be ignored using the **`ss::ignore_header`** [setup](#Setup) option or by calling the **`ignore_next`** method after the parser has been constructed. If the header has been ignored calling any method related to header usage will result in a compilation error.
```cpp
ss::parser<ss::ignore_header> p{file_name};
```
The fields with which the parser works with can be modified at any given time. The praser can also check if a field is present within the header by using the **`field_exists`** method.
The fields with which the parser works with can be modified at any given time. The parser can also check if a field is present within the header by using the **`field_exists`** method.
```cpp
// ...
ss::parser p{"students.csv", ","};
p.use_fields("Name", "Grade");
ss::parser<ss::throw_on_error> p{"students_with_header.csv"};
p.use_fields("Grade");
const auto& [name, grade] = p.get_next<std::string, float>();
std::cout << name << ' ' << grade << std::endl;
const auto& grade = p.get_next<std::string>();
std::cout << grade << std::endl;
if (p.field_exists("Age")) {
p.use_fields("Grade", "Name", "Age");
for (const auto& [grade, name, age] :
p.iterate<float, std::string, int>()) {
std::cout << grade << ' ' << name << ' ' << age << std::endl;
if (p.field_exists("Id")) {
p.use_fields("Grade", "Id");
for (const auto& [grade, id] : p.iterate<float, std::string>()) {
std::cout << grade << ' ' << id << std::endl;
}
}
// ...
```
```shell
$ ./a.out
James Bailey 2.5
1.9 Brian S. Wolfe 40
3.3 Bill (Heath) Gates 65
2.5
1.9 Brian S. Wolfe
3.3 Bill (Heath) Gates
```
The header is parsed with the same rules as other rows, the only difference is that **`multiline`** will be disabled when parsing the header. To get the data that is
present in the header as a **`std::vector<std::string>`**, the **`header`** method can be used, and to get the header line before it has been parsed, the **`raw_header`** method can be used:
```cpp
// ...
ss::parser<ss::throw_on_error> p{"students_with_header.csv"};
std::cout << p.raw_header() << std::endl;
for (const auto& field: p.header()) {
std::cout << "> " << field << std::endl;
}
// ...
```
```shell
$ ./a.out
Id,Age,Grade
> Id
> Age
> Grade
```
Methods related to headers can also fail, the error handling of these is done in the same way as for other methods.
## Conversions
An alternate loop to the example above would look like:
```cpp
// ...
ss::parser p{"students.csv"};
while (!p.eof()) {
auto [name, age, grade] = p.get_next<std::string, int, float>();
const auto& [id, age, grade] = p.get_next<std::string, int, float>();
if (p.valid()) {
std::cout << name << ' ' << age << ' ' << grade << std::endl;
std::cout << id << ' ' << age << ' ' << grade << std::endl;
}
}
// ...
```
The alternate example will be used to show some of the features of the library. The **`get_next`** method returns a tuple of objects specified inside the template type list.
The alternate example with exceptions disabled will be used to show some of the features of the library. The **`get_next`** method returns a tuple of objects specified inside the template type list.
If a conversion could not be applied, the method would return a tuple of default constructed objects, and the **`valid`** method would return **`false`**, for example if the third (grade) column in our csv could not be converted to a float the conversion would fail.
If a conversion could not be applied, the method would return a tuple of default constructed objects, and the **`valid`** method would return **`false`**, for example if the third (grade) column in our CSV could not be converted to a float the conversion would fail.
If **`get_next`** is called with a **`tuple`** as template parameter it would behave identically to passing the same tuple parameters to **`get_next`**:
```cpp
using student = std::tuple<std::string, int, float>;
// returns std::tuple<std::string, int, float>
auto [name, age, grade] = p.get_next<student>();
auto [id, age, grade] = p.get_next<student>();
```
*Note, it does not always return a student tuple since the returned tuples parameters may be altered as explained below (no void, no restrictions, ...)*
*Note, it does not always return the specified tuple since the returned tuples parameters may be altered as explained below (no void, no restrictions, ...)*
Whole objects can be returned using the **`get_object`** function which takes the tuple, created in a similar way as **`get_next`** does it, and creates an object out of it:
```cpp
struct student {
std::string name;
std::string id;
int age;
float grade;
};
@ -179,34 +208,46 @@ auto student = p.get_object<student, std::string, int, float>();
```
This works with any object if the constructor could be invoked using the template arguments given to **`get_object`**:
```cpp
// returns std::vector<std::string> containing 3 elements
auto vec = p.get_object<std::vector<std::string>, std::string, std::string,
std::string>();
// returns std::vector<std::string> containing 2 elements
auto vec = p.get_object<std::vector<std::string>, std::string, std::string>();
```
An iteration loop as in the first example which returns objects would look like:
An iterator loop as in the first example which returns objects would look like:
```cpp
for(const auto& student : p.iterate_object<student, std::string, int, float>()) {
for (const student& s : p.iterate_object<student, std::string, int, float>()) {
// ...
}
```
And finally, using something I personally like to do, a struct (class) with a **`tied`** method which returns a tuple of references to to the members of the struct.
```cpp
struct student {
std::string name;
std::string id;
int age;
float grade;
auto tied() { return std::tie(name, age, grade); }
auto tied() { return std::tie(id, age, grade); }
};
```
The method can be used to compare the object, serialize it, deserialize it, etc. Now **`get_next`** can accept such a struct and deduce the types to which to convert the csv.
The method can be used to compare the object, serialize it, deserialize it, etc. Now **`get_next`** can accept such a struct and deduce the types to which to convert the CSV.
```cpp
// returns student
auto s = p.get_next<student>();
```
This works with the iteration loop too.
*Note, the order in which the members of the tied method are returned must match the order of the elements in the csv*.
*Note, the order in which the members of the tied method are returned must match the order of the elements in the CSV*.
## Buffer mode
The parser also works with buffers containing CSV data instead of files. To parse buffer data with the parser simply create the parser by giving it the buffer, as **`const char*`**, and its size. The initial example using a buffer instead of a file would look similar to this:
```cpp
std::string buffer = "James Bailey,65,2.5\nBrian S. Wolfe,40,1.9\n";
ss::parser<ss::throw_on_error> p{buffer.c_str(), buffer.size()};
for (const auto& [id, age, grade] : p.iterate<std::string, int, float>()) {
std::cout << id << ' ' << age << ' ' << grade << std::endl;
}
return 0;
```
## Setup
By default, many of the features supported by the parser are disabled. They can be enabled within the template parameters of the parser. For example, to enable quoting and escaping the parser would look like:
```cpp
@ -224,14 +265,21 @@ using my_setup = ss::setup<ss::escape<'\\'>, ss::quote<'"'>>;
ss::parser<my_setup> p2{file_name};
```
Invalid setups will be met with **`static_asserts`**.
*Note, each setup parameter defined comes with a slight performance loss, so use them only if needed.*
*Note, most setup parameters defined come with a slight performance loss, so use them only if needed.*
### Delimiter
By default, **`,`** is used as the delimiter, a custom delimiter can be specified as the second constructor parameter.
```cpp
ss::parser p{file_name, "--"};
```
*Note, the delimiter can consist of multiple characters but the parser is slightly faster when using single character delimiters.*
### Empty lines
Empty lines can be ignored by defining **`ss::ignore_empty`** within the setup parameters:
```cpp
ss::parser<ss::ignore_empty> p{file_name};
```
If this setup option is not set then reading an empty line will result in an error (unless only one column is present within the csv).
If this setup option is not set then reading an empty line will result in an error (unless only one column is present within the CSV).
### Quoting
Quoting can be enabled by defining **`ss::quote`** within the setup parameters. A single character can be defined as the quoting character, for example to use **`"`** as a quoting character:
@ -280,7 +328,7 @@ Escaping and quoting can be used to leave the space if needed.
```
### Multiline
Multiline can be enabled by defining **`ss::multilne`** within the setup parameters. It enables the possibility to have the new line characters within rows. The new line character needs to be either escaped or within quotes so either **`ss::escape`** or **`ss::quote`** need to be enabled. There is a specific problem when using multiline, for example, if a row had an unterminated quote, the parser would assume it to be a new line within the row, so until another quote is found, it will treat it as one line which is fine usually, but it can cause the whole csv file to be treated as a single line by mistake. To prevent this **`ss::multiline_restricted`** can be used which accepts an unsigned number representing the maximum number of lines which can be allowed as a single multiline. Examples:
Multiline can be enabled by defining **`ss::multilne`** within the setup parameters. It enables the possibility to have the new line characters within rows. The new line character needs to be either escaped or within quotes so either **`ss::escape`** or **`ss::quote`** need to be enabled. There is a specific problem when using multiline, for example, if a row had an unterminated quote, the parser would assume it to be a new line within the row, so until another quote is found, it will treat it as one line which is fine usually, but it can cause the whole CSV file to be treated as a single line by mistake. To prevent this **`ss::multiline_restricted`** can be used which accepts an unsigned number representing the maximum number of lines which can be allowed as a single multiline. Examples:
```cpp
ss::parser<ss::multiline, ss::quote<'\"'>, ss::escape<'\\'>> p{file_name};
@ -306,14 +354,11 @@ ss::parser<ss::escape<'\\'>,
ss::trim<' ', '\t'>,
ss::multiline_restricted<5>> p{file_name};
while(!p.eof()) {
auto [name, age, grade] = p.get_next<std::string, int, float>();
if(!p.valid()) {
continue;
for (const auto& [id, age, grade] : p.iterate<std::string, int, float>()) {
if (p.valid()) {
std::cout << "'" << id << ' ' << age << ' ' << grade << "'\n";
}
std::cout << "'" << name << ' ' << age << ' ' << grade << "'" << std::endl;
}
```
input:
```
@ -334,22 +379,22 @@ Gates 65 3.3'
```
## Special types
Passing **`void`** makes the parser ignore a column. In the given example **`void`** could be given as the second template parameter to ignore the second (age) column in the csv, a tuple of only 2 parameters would be retuned:
Passing **`void`** makes the parser ignore a column. In the initial example **`void`** could be given as the second template parameter to ignore the second (age) column in the CSV, a tuple of only 2 parameters would be retuned:
```cpp
// returns std::tuple<std::string, float>
auto [name, grade] = p.get_next<std::string, void, float>();
auto [id, grade] = p.get_next<std::string, void, float>();
```
Works with different types of conversions too:
```cpp
using student = std::tuple<std::string, void, float>;
// returns std::tuple<std::string, float>
auto [name, grade] = p.get_next<student>();
auto [id, grade] = p.get_next<student>();
```
Values can also be converted to **`std::string_view`**. It is more efficient then converting values to **`std::string`** but one must be careful with the lifetime of it.
```cpp
// string_view name stays valid until the next line is read
auto [name, age, grade] = p.get_next<std::string_view, int, float>();
// string_view id stays valid until the next line is read
auto [id, age, grade] = p.get_next<std::string_view, int, float>();
```
To ignore a whole row, **`ignore_next`** could be used, returns **`false`** if **`eof`**:
@ -360,30 +405,36 @@ bool parser::ignore_next();
```cpp
// returns std::tuple<std::string, int, std::optional<float>>
auto [name, age, grade] = p.get_next<std::string, int, std::optional<float>();
auto [id, age, grade] = p.get_next<std::string, int, std::optional<float>>();
if (grade) {
// do something with grade
std::cout << grade.value() << std::endl;
}
```
Similar to **`std::optional`**, **`std::variant`** could be used to try other conversions if the previous failed _(Note, conversion to std::string will always pass)_:
```cpp
// returns std::tuple<std::string, int, std::variant<float, char>>
auto [name, age, grade] =
p.get_next<std::string, int, std::variant<float, char>();
auto [id, age, grade] =
p.get_next<std::string, int, std::variant<float, char>>();
if (std::holds_alternative<float>(grade)) {
// grade set as float
} else if (std::holds_alternative<char>(grade)) {
// grade set as char
}
```
Passing **`char`** and types that are aliases to it such as **`uint8_t`** and **`int8_t`** make the parser interpret the input data as a single character in a similar way to how **`std::cin`** does it. To read numeric values into something like **`uint8_t`** the **`ss::uint8`** and **`ss::int8`** types can be used. These are wrappers around the corresponding char aliases and can be implicitly converted to and from them. When these types are given to the parser he will try to read the given data and store it in the underlying element, but this time as a numeric value instead of a single character.
```cpp
// returns std::tuple<std::string, ss::uint8, float>
auto [id, age, grade] = p.get_next<std::string, ss::uint8, float>();
uint8_t age_copy = age;
```
## Restrictions
Custom **`restrictions`** can be used to narrow down the conversions of unwanted values. **`ss::ir`** (in range) and **`ss::ne`** (none empty) are some of those:
```cpp
// ss::ne makes sure that the name is not empty
// ss::ne makes sure that the id is not empty
// ss::ir makes sure that the grade will be in range [0, 10]
// returns std::tuple<std::string, int, float>
auto [name, age, grade] =
auto [id, age, grade] =
p.get_next<ss::ne<std::string>, int, ss::ir<float, 0, 10>>();
```
If the restrictions are not met, the conversion will fail. Other predefined restrictions are **`ss::ax`** (all except), **`ss::nx`** (none except) and **`ss::oor`** (out of range), **`ss::lt`** (less than), ...(see *restrictions.hpp*):
@ -410,9 +461,11 @@ struct even {
};
```
```cpp
// only even numbers will pass
// ...
// only even numbers will pass without invoking error handling
// returns std::tuple<std::string, int>
auto [name, age] = p.get_next<std::string, even<int>, void>();
const auto& [id, age] = p.get_next<std::string, even<int>, void>();
// ...
```
## Custom conversions
@ -439,21 +492,35 @@ The shape enum will be used in an example below. The **`inline`** is there just
## Error handling
By default, the parser handles errors only using the **`valid`** method which would return **`false`** if the file could not be opened, or if the conversion could not be made (invalid types, invalid number of columns, ...).\
The **`eof`** method can be used to detect if the end of the file was reached.
Detailed error messages can be accessed via the **`error_msg`** method, and to enable them **`ss::string_error`** needs to be included in the setup. If **`ss::string_error`** is not defined, the **`error_msg`** method will not be defined either.
The line number can be fetched using the **`line`** method.
The cursor position can be fetched using the **`position`** method.
```cpp
const std::string& parser::error_msg();
bool parser::valid();
bool parser::eof();
const std::string& parser::error_msg() const;
bool parser::valid() const;
bool parser::eof() const;
size_t parser::line() const;
size_t parser::position() const;
// ...
ss::parser<ss::string_error> parser;
```
An error can be detected using the **`valid`** method which would return **`false`** if the file could not be opened, or if the conversion could not be made (invalid types, invalid number of columns, ...). The **`eof`** method can be used to detect if the end of the file was reached.
The above two methods are preferable if invalid inputs are expected and allows for fast handling, but the parser can also be forced to throw an exception in case of an invalid input using the **`ss::throw_on_error`** setup option. Enabling exceptions also makes the **`valid`** method always return **`true`**.
```cpp
ss::parser<ss::throw_on_error> parser;
```
*Note, enabling this option will also make the parser throw if the constructor fails.*
## Substitute conversions
The parser can also be used to effectively parse files whose rows are not always in the same format (not a classical csv but still csv-like). A more complicated example would be the best way to demonstrate such a scenario.
The parser can also be used to effectively parse files whose rows are not always in the same format (not a classical CSV but still CSV-like). A more complicated example would be the best way to demonstrate such a scenario.\
***Important, substitute conversions do not work when throw_on_error is enabled.***
Supposing we have a file containing different shapes in given formats:
* circle RADIUS
@ -487,6 +554,11 @@ while (!p.eof()) {
.or_else<ss::nx<shape, shape::triangle>, udbl, udbl, udbl>()
.values();
if (!p.valid()) {
// handle error
continue;
}
if (circle_or_square) {
auto& [s, x] = circle_or_square.value();
double area = (s == shape::circle) ? x * x * M_PI : x * x;
@ -530,6 +602,7 @@ Each of those **`composite`** conversions can accept a lambda (or anything calla
// non negative double
using udbl = ss::gte<double, 0>;
while (!p.eof()) {
p.try_next<ss::nx<shape, shape::circle, shape::square>, udbl>(
[&](const auto& data) {
const auto& [s, x] = data;
@ -537,24 +610,26 @@ p.try_next<ss::nx<shape, shape::circle, shape::square>, udbl>(
shapes.emplace_back(s, area);
})
.or_else<ss::nx<shape, shape::rectangle>, udbl, udbl>(
[&](const shape s, const double a, const double b) {
shapes.emplace_back(s, a * b);
})
[&](shape s, double a, double b) { shapes.emplace_back(s, a * b); })
.or_else<ss::nx<shape, shape::triangle>, udbl, udbl, udbl>(
[&](auto&& s, auto& a, const double& b, double& c) {
[&](auto s, auto a, auto b, auto c) {
double sh = (a + b + c) / 2;
if (sh >= a && sh >= b && sh >= c) {
double area = sqrt(sh * (sh - a) * (sh - b) * (sh - c));
shapes.emplace_back(s, area);
}
})
.on_error([] {
// handle error
});
}
```
It is a bit less readable, but it removes the need to check which conversion was invoked. The **`composite`** also has an **`on_error`** method which accepts a lambda which will be invoked if no previous conversions were successful. The lambda can take no arguments or just one argument, an **`std::string`**, in which the error message is stored if **`string_error`** is enabled:
```cpp
p.try_next<int>()
.on_error([](const std::string& e) { /* int conversion failed */ })
.or_object<x, double>()
.on_error([] { /* int and x (all) conversions failed */ });
.on_error([] { /* int and x conversions failed (all previous failed) */ });
```
*See unit tests for more examples.*
@ -583,7 +658,7 @@ if (c.valid()) {
All setup parameters, special types and restrictions work on the converter too.
Error handling is also identical to error handling of the parser.
The converter has also the ability to just split the line, ~~tho it does not change it (kinda statically), hence the name of the library~~ and depending if either quoting or escaping are enabled it may change the line, rather than creating a copy, for performance reasons (the name of the library does not apply anymore, I may change it). It returns an **`std::vector`** of **`std::pair`**s of pointers, begin and end, each pair representing a split segment (column) of the whole string. The vector can then be used in a overloaded **`convert`** method. This allows the reuse of the same line without splitting it on every conversion.
The converter has also the ability to just split the line, and depending if either quoting or escaping are enabled it may change the line, rather than creating a copy, for performance reasons. It returns an **`std::vector`** of **`std::pair`**s of pointers, begin and end, each pair representing a split segment (column) of the whole string. The vector can then be used in a overloaded **`convert`** method. This allows the reuse of the same line without splitting it on every conversion.
```cpp
ss::converter c;
auto split_line = c.split("circle 10", " ");

View File

@ -1,9 +1,15 @@
#pragma once
#include <cstdint>
#include <cerrno>
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <vector>
#if !__unix__
#include <array>
#include <cstdint>
#endif
namespace ss {
struct none {};
@ -12,67 +18,134 @@ using string_range = std::pair<const char*, const char*>;
using split_data = std::vector<string_range>;
constexpr inline auto default_delimiter = ",";
constexpr inline auto get_line_initial_buffer_size = 128;
template <bool StringError>
inline void assert_string_error_defined() {
void assert_string_error_defined() {
static_assert(StringError,
"'string_error' needs to be enabled to use 'error_msg'");
}
template <bool ThrowOnError>
void assert_throw_on_error_not_defined() {
static_assert(!ThrowOnError, "cannot handle errors manually if "
"'throw_on_error' is enabled");
}
[[nodiscard]] inline void* strict_realloc(void* ptr, size_t size) {
ptr = std::realloc(ptr, size);
if (!ptr) {
throw std::bad_alloc{};
}
return ptr;
}
#if __unix__
inline ssize_t get_line(char** lineptr, size_t* n, FILE* stream) {
return getline(lineptr, n, stream);
[[nodiscard]] inline ssize_t get_line_file(char*& lineptr, size_t& n,
FILE* file) {
return getline(&lineptr, &n, file);
}
#else
using ssize_t = int64_t;
inline ssize_t get_line(char** lineptr, size_t* n, FILE* stream) {
size_t pos;
int c;
if (lineptr == nullptr || stream == nullptr || n == nullptr) {
errno = EINVAL;
return -1;
using ssize_t = intptr_t;
[[nodiscard]] inline ssize_t get_line_file(char*& lineptr, size_t& n,
FILE* file) {
std::array<char, get_line_initial_buffer_size> buff;
if (lineptr == nullptr || n < sizeof(buff)) {
const size_t new_n = sizeof(buff);
lineptr = static_cast<char*>(strict_realloc(lineptr, new_n));
n = new_n;
}
c = getc(stream);
if (c == EOF) {
return -1;
lineptr[0] = '\0';
size_t line_used = 0;
while (std::fgets(buff.data(), sizeof(buff), file) != nullptr) {
line_used = std::strlen(lineptr);
size_t buff_used = std::strlen(buff.data());
if (n <= buff_used + line_used) {
const size_t new_n = n * 2;
lineptr = static_cast<char*>(strict_realloc(lineptr, new_n));
n = new_n;
}
if (*lineptr == nullptr) {
*lineptr = static_cast<char*>(malloc(128));
if (*lineptr == nullptr) {
return -1;
std::memcpy(lineptr + line_used, buff.data(), buff_used);
line_used += buff_used;
lineptr[line_used] = '\0';
if (lineptr[line_used - 1] == '\n') {
return line_used;
}
*n = 128;
}
pos = 0;
while (c != EOF) {
if (pos + 1 >= *n) {
size_t new_size = *n + (*n >> 2);
if (new_size < 128) {
new_size = 128;
}
char* new_ptr = static_cast<char*>(
realloc(static_cast<void*>(*lineptr), new_size));
if (new_ptr == nullptr) {
return -1;
}
*n = new_size;
*lineptr = new_ptr;
return (line_used != 0) ? line_used : -1;
}
(*lineptr)[pos++] = c;
if (c == '\n') {
break;
}
c = getc(stream);
}
(*lineptr)[pos] = '\0';
return pos;
}
#endif
} /* ss */
[[nodiscard]] inline ssize_t get_line_buffer(char*& lineptr, size_t& n,
const char* const csv_data_buffer,
size_t csv_data_size,
size_t& curr_char) {
if (curr_char >= csv_data_size) {
return -1;
}
if (lineptr == nullptr || n < get_line_initial_buffer_size) {
auto* new_lineptr = static_cast<char*>(
strict_realloc(lineptr, get_line_initial_buffer_size));
lineptr = new_lineptr;
n = get_line_initial_buffer_size;
}
size_t line_used = 0;
while (curr_char < csv_data_size) {
if (line_used + 1 >= n) {
const size_t new_n = n * 2;
char* new_lineptr =
static_cast<char*>(strict_realloc(lineptr, new_n));
n = new_n;
lineptr = new_lineptr;
}
auto c = csv_data_buffer[curr_char++];
lineptr[line_used++] = c;
if (c == '\n') {
lineptr[line_used] = '\0';
return line_used;
}
}
lineptr[line_used] = '\0';
return line_used;
}
[[nodiscard]] inline std::tuple<ssize_t, bool> get_line(
char*& buffer, size_t& buffer_size, FILE* file,
const char* const csv_data_buffer, size_t csv_data_size,
size_t& curr_char) {
ssize_t ssize = 0;
if (file) {
ssize = get_line_file(buffer, buffer_size, file);
curr_char += ssize;
} else {
ssize = get_line_buffer(buffer, buffer_size, csv_data_buffer,
csv_data_size, curr_char);
}
if (ssize == -1) {
if (errno == ENOMEM) {
throw std::bad_alloc{};
}
return {ssize, true};
}
return {ssize, false};
}
} /* namespace ss */

View File

@ -1,4 +1,5 @@
#pragma once
#include "exception.hpp"
#include "extract.hpp"
#include "function_traits.hpp"
#include "restrictions.hpp"
@ -95,11 +96,12 @@ constexpr bool tied_class_v = tied_class<Ts...>::value;
// converter
////////////////
template <typename... Matchers>
template <typename... Options>
class converter {
using line_ptr_type = typename splitter<Matchers...>::line_ptr_type;
using line_ptr_type = typename splitter<Options...>::line_ptr_type;
constexpr static auto string_error = setup<Matchers...>::string_error;
constexpr static auto string_error = setup<Options...>::string_error;
constexpr static auto throw_on_error = setup<Options...>::throw_on_error;
constexpr static auto default_delimiter = ",";
using error_type = std::conditional_t<string_error, std::string, bool>;
@ -108,29 +110,34 @@ public:
// parses line with given delimiter, returns a 'T' object created with
// extracted values of type 'Ts'
template <typename T, typename... Ts>
T convert_object(line_ptr_type line,
const std::string& delim = default_delimiter) {
[[nodiscard]] T convert_object(
line_ptr_type line, const std::string& delim = default_delimiter) {
return to_object<T>(convert<Ts...>(line, delim));
}
// parses line with given delimiter, returns tuple of objects with
// extracted values of type 'Ts'
template <typename... Ts>
no_void_validator_tup_t<Ts...> convert(
[[nodiscard]] no_void_validator_tup_t<Ts...> convert(
line_ptr_type line, const std::string& delim = default_delimiter) {
split(line, delim);
return convert<Ts...>(splitter_.split_data_);
if (splitter_.valid()) {
return convert<Ts...>(splitter_.get_split_data());
} else {
handle_error_bad_split();
return {};
}
}
// parses already split line, returns 'T' object with extracted values
template <typename T, typename... Ts>
T convert_object(const split_data& elems) {
[[nodiscard]] T convert_object(const split_data& elems) {
return to_object<T>(convert<Ts...>(elems));
}
// same as above, but uses cached split line
template <typename T, typename... Ts>
T convert_object() {
[[nodiscard]] T convert_object() {
return to_object<T>(convert<Ts...>());
}
@ -139,11 +146,12 @@ public:
// one argument is given which is a class which has a tied
// method which returns a tuple, returns that type
template <typename T, typename... Ts>
no_void_validator_tup_t<T, Ts...> convert(const split_data& elems) {
[[nodiscard]] no_void_validator_tup_t<T, Ts...> convert(
const split_data& elems) {
if constexpr (sizeof...(Ts) == 0 && is_instance_of_v<std::tuple, T>) {
return convert_impl(elems, static_cast<T*>(nullptr));
} else if constexpr (tied_class_v<T, Ts...>) {
using arg_ref_tuple = std::result_of_t<decltype (&T::tied)(T)>;
using arg_ref_tuple = std::invoke_result_t<decltype(&T::tied), T>;
using arg_tuple = apply_trait_t<std::decay, arg_ref_tuple>;
return to_object<T>(
@ -155,24 +163,26 @@ public:
// same as above, but uses cached split line
template <typename T, typename... Ts>
no_void_validator_tup_t<T, Ts...> convert() {
return convert<T, Ts...>(splitter_.split_data_);
[[nodiscard]] no_void_validator_tup_t<T, Ts...> convert() {
return convert<T, Ts...>(splitter_.get_split_data());
}
bool valid() const {
[[nodiscard]] bool valid() const {
if constexpr (string_error) {
return error_.empty();
} else if constexpr (throw_on_error) {
return true;
} else {
return !error_;
}
}
const std::string& error_msg() const {
[[nodiscard]] const std::string& error_msg() const {
assert_string_error_defined<string_error>();
return error_;
}
bool unterminated_quote() const {
[[nodiscard]] bool unterminated_quote() const {
return splitter_.unterminated_quote();
}
@ -180,9 +190,9 @@ public:
// contain the beginnings and the ends of each column of the string
const split_data& split(line_ptr_type line,
const std::string& delim = default_delimiter) {
splitter_.split_data_.clear();
splitter_.clear_split_data();
if (line[0] == '\0') {
return splitter_.split_data_;
return splitter_.get_split_data();
}
return splitter_.split(line, delim);
@ -198,7 +208,7 @@ private:
return splitter_.resplit(new_line, new_size, delim);
}
size_t size_shifted() {
[[nodiscard]] size_t size_shifted() {
return splitter_.size_shifted();
}
@ -214,9 +224,11 @@ private:
}
}
std::string error_sufix(const string_range msg, size_t pos) const {
[[nodiscard]] std::string error_sufix(const string_range msg,
size_t pos) const {
constexpr static auto reserve_size = 32;
std::string error;
error.reserve(32);
error.reserve(reserve_size);
error.append("at column ")
.append(std::to_string(pos + 1))
.append(": \'")
@ -225,98 +237,112 @@ private:
return error;
}
void set_error_unterminated_quote() {
void handle_error_bad_split() {
if constexpr (string_error) {
error_.clear();
error_.append(splitter_.error_msg());
} else {
} else if constexpr (!throw_on_error) {
error_ = true;
}
}
void set_error_unterminated_escape() {
void handle_error_unterminated_escape() {
if constexpr (string_error) {
error_.clear();
splitter_.set_error_unterminated_escape();
splitter_.handle_error_unterminated_escape();
error_.append(splitter_.error_msg());
} else if constexpr (throw_on_error) {
splitter_.handle_error_unterminated_escape();
} else {
error_ = true;
}
}
void set_error_multiline_limit_reached() {
void handle_error_unterminated_quote() {
if constexpr (string_error) {
error_.clear();
error_.append("multiline limit reached.");
splitter_.handle_error_unterminated_quote();
error_.append(splitter_.error_msg());
} else if constexpr (throw_on_error) {
splitter_.handle_error_unterminated_quote();
} else {
error_ = true;
}
}
void set_error_invalid_conversion(const string_range msg, size_t pos) {
void handle_error_multiline_limit_reached() {
constexpr static auto error_msg = "multiline limit reached";
splitter_.unterminated_quote_ = false;
if constexpr (string_error) {
error_.clear();
error_.append("invalid conversion for parameter ")
.append(error_sufix(msg, pos));
error_.append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg};
} else {
error_ = true;
}
}
void set_error_validate(const char* const error, const string_range msg,
size_t pos) {
void handle_error_invalid_conversion(const string_range msg, size_t pos) {
constexpr static auto error_msg = "invalid conversion for parameter ";
if constexpr (string_error) {
error_.clear();
error_.append(error_msg).append(error_sufix(msg, pos));
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg + error_sufix(msg, pos)};
} else {
error_ = true;
}
}
void handle_error_validation_failed(const char* const error,
const string_range msg, size_t pos) {
if constexpr (string_error) {
error_.clear();
error_.append(error).append(" ").append(error_sufix(msg, pos));
} else if constexpr (throw_on_error) {
throw ss::exception{error + (" " + error_sufix(msg, pos))};
} else {
error_ = true;
}
}
void set_error_number_of_columns(size_t expected_pos, size_t pos) {
void handle_error_number_of_columns(size_t expected_pos, size_t pos) {
constexpr static auto error_msg1 =
"invalid number of columns, expected: ";
constexpr static auto error_msg2 = ", got: ";
if constexpr (string_error) {
error_.clear();
error_.append("invalid number of columns, expected: ")
error_.append(error_msg1)
.append(std::to_string(expected_pos))
.append(", got: ")
.append(error_msg2)
.append(std::to_string(pos));
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg1 + std::to_string(expected_pos) +
error_msg2 + std::to_string(pos)};
} else {
error_ = true;
}
}
void set_error_incompatible_mapping(size_t argument_size,
void handle_error_incompatible_mapping(size_t argument_size,
size_t mapping_size) {
constexpr static auto error_msg1 =
"number of arguments does not match mapping, expected: ";
constexpr static auto error_msg2 = ", got: ";
if constexpr (string_error) {
error_.clear();
error_
.append(
"number of arguments does not match mapping, expected: ")
error_.append(error_msg1)
.append(std::to_string(mapping_size))
.append(", got: ")
.append(error_msg2)
.append(std::to_string(argument_size));
} else {
error_ = true;
}
}
void set_error_invalid_mapping() {
if constexpr (string_error) {
error_.clear();
error_.append("received empty mapping");
} else {
error_ = true;
}
}
void set_error_mapping_out_of_range(size_t maximum_index,
size_t number_of_columnts) {
if constexpr (string_error) {
error_.clear();
error_.append("maximum index: ")
.append(std::to_string(maximum_index))
.append(", greater then number of columns: ")
.append(std::to_string(number_of_columnts));
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg1 + std::to_string(mapping_size) +
error_msg2 + std::to_string(argument_size)};
} else {
error_ = true;
}
@ -327,40 +353,39 @@ private:
////////////////
template <typename... Ts>
no_void_validator_tup_t<Ts...> convert_impl(const split_data& elems) {
[[nodiscard]] no_void_validator_tup_t<Ts...> convert_impl(
const split_data& elems) {
clear_error();
using return_type = no_void_validator_tup_t<Ts...>;
if (!splitter_.valid()) {
set_error_unterminated_quote();
no_void_validator_tup_t<Ts...> ret{};
return ret;
handle_error_bad_split();
return {};
}
if (!columns_mapped()) {
if (sizeof...(Ts) != elems.size()) {
set_error_number_of_columns(sizeof...(Ts), elems.size());
return return_type{};
handle_error_number_of_columns(sizeof...(Ts), elems.size());
return {};
}
} else {
if (sizeof...(Ts) != column_mappings_.size()) {
set_error_incompatible_mapping(sizeof...(Ts),
handle_error_incompatible_mapping(sizeof...(Ts),
column_mappings_.size());
return return_type{};
return {};
}
if (elems.size() != number_of_columns_) {
set_error_number_of_columns(number_of_columns_, elems.size());
return return_type{};
handle_error_number_of_columns(number_of_columns_,
elems.size());
return {};
}
}
return extract_tuple<Ts...>(elems);
}
// do not know how to specialize by return type :(
template <typename... Ts>
no_void_validator_tup_t<std::tuple<Ts...>> convert_impl(
[[nodiscard]] no_void_validator_tup_t<std::tuple<Ts...>> convert_impl(
const split_data& elems, const std::tuple<Ts...>*) {
return convert_impl<Ts...>(elems);
}
@ -369,31 +394,21 @@ private:
// column mapping
////////////////
bool columns_mapped() const {
return column_mappings_.size() != 0;
[[nodiscard]] bool columns_mapped() const {
return !column_mappings_.empty();
}
size_t column_position(size_t tuple_position) const {
[[nodiscard]] size_t column_position(size_t tuple_position) const {
if (!columns_mapped()) {
return tuple_position;
}
return column_mappings_[tuple_position];
}
// assumes positions are valid and the vector is not empty
void set_column_mapping(std::vector<size_t> positions,
size_t number_of_columns) {
if (positions.empty()) {
set_error_invalid_mapping();
return;
}
auto max_index = *std::max_element(positions.begin(), positions.end());
if (max_index >= number_of_columns) {
set_error_mapping_out_of_range(max_index, number_of_columns);
return;
}
column_mappings_ = positions;
column_mappings_ = std::move(positions);
number_of_columns_ = number_of_columns;
}
@ -414,21 +429,22 @@ private:
}
if constexpr (std::is_same_v<T, std::string>) {
extract(msg.first, msg.second, dst);
static_cast<void>(extract(msg.first, msg.second, dst));
return;
}
if (!extract(msg.first, msg.second, dst)) {
set_error_invalid_conversion(msg, pos);
handle_error_invalid_conversion(msg, pos);
return;
}
if constexpr (has_m_ss_valid_t<T>) {
if (T validator; !validator.ss_valid(dst)) {
if constexpr (has_m_error_t<T>) {
set_error_validate(validator.error(), msg, pos);
handle_error_validation_failed(validator.error(), msg, pos);
} else {
set_error_validate("validation error", msg, pos);
handle_error_validation_failed("validation error", msg,
pos);
}
return;
}
@ -459,7 +475,8 @@ private:
}
template <typename... Ts>
no_void_validator_tup_t<Ts...> extract_tuple(const split_data& elems) {
[[nodiscard]] no_void_validator_tup_t<Ts...> extract_tuple(
const split_data& elems) {
static_assert(!all_of_v<std::is_void, Ts...>,
"at least one parameter must be non void");
no_void_validator_tup_t<Ts...> ret{};
@ -472,13 +489,13 @@ private:
////////////////
error_type error_{};
splitter<Matchers...> splitter_;
splitter<Options...> splitter_;
template <typename...>
friend class parser;
std::vector<size_t> column_mappings_;
size_t number_of_columns_;
size_t number_of_columns_{0};
};
} /* ss */
} /* namespace ss */

23
include/ss/exception.hpp Normal file
View File

@ -0,0 +1,23 @@
#pragma once
#include <exception>
#include <string>
namespace ss {
////////////////
// exception
////////////////
class exception : public std::exception {
std::string msg_;
public:
exception(std::string msg) : msg_{std::move(msg)} {
}
[[nodiscard]] char const* what() const noexcept override {
return msg_.c_str();
}
};
} /* namespace ss */

View File

@ -1,9 +1,9 @@
#pragma once
#include "type_traits.hpp"
#include <charconv>
#include <cstdint>
#include <cstring>
#include <functional>
#include <limits>
#include <optional>
#include <string>
#include <string_view>
@ -12,7 +12,9 @@
#ifndef SSP_DISABLE_FAST_FLOAT
#include <fast_float/fast_float.h>
#else
#include <charconv>
#include <algorithm>
#include <array>
#include <cstdlib>
#endif
namespace ss {
@ -24,8 +26,8 @@ namespace ss {
#ifndef SSP_DISABLE_FAST_FLOAT
template <typename T>
std::enable_if_t<std::is_floating_point_v<T>, std::optional<T>> to_num(
const char* const begin, const char* const end) {
[[nodiscard]] std::enable_if_t<std::is_floating_point_v<T>, std::optional<T>>
to_num(const char* const begin, const char* const end) {
T ret;
auto [ptr, ec] = fast_float::from_chars(begin, end, ret);
@ -38,7 +40,81 @@ std::enable_if_t<std::is_floating_point_v<T>, std::optional<T>> to_num(
#else
template <typename T>
std::enable_if_t<std::is_floating_point_v<T>, std::optional<T>> to_num(
[[nodiscard]] std::enable_if_t<std::is_floating_point_v<T>, std::optional<T>>
to_num(const char* const begin, const char* const end) {
static_assert(!std::is_same_v<T, long double>,
"Conversion to long double is disabled");
constexpr static auto buff_max = 64;
std::array<char, buff_max> short_buff;
const size_t string_range = std::distance(begin, end);
std::string long_buff;
char* buff = nullptr;
if (string_range > buff_max) {
long_buff = std::string{begin, end};
buff = long_buff.data();
} else {
buff = short_buff.data();
buff[string_range] = '\0';
std::copy_n(begin, string_range, buff);
}
T ret;
char* parse_end = nullptr;
if constexpr (std::is_same_v<T, float>) {
ret = std::strtof(buff, &parse_end);
} else if constexpr (std::is_same_v<T, double>) {
ret = std::strtod(buff, &parse_end);
}
if (parse_end != buff + string_range) {
return std::nullopt;
}
return ret;
}
#endif
////////////////
// numeric_wrapper
////////////////
template <typename T>
struct numeric_wrapper {
using type = T;
numeric_wrapper() = default;
numeric_wrapper(numeric_wrapper&&) noexcept = default;
numeric_wrapper(const numeric_wrapper&) = default;
numeric_wrapper& operator=(numeric_wrapper&&) noexcept = default;
numeric_wrapper& operator=(const numeric_wrapper&) = default;
~numeric_wrapper() = default;
numeric_wrapper(T other) : value{other} {
}
operator T() {
return value;
}
operator T() const {
return value;
}
T value;
};
using int8 = numeric_wrapper<int8_t>;
using uint8 = numeric_wrapper<uint8_t>;
template <typename T>
[[nodiscard]] std::enable_if_t<std::is_integral_v<T>, std::optional<T>> to_num(
const char* const begin, const char* const end) {
T ret;
auto [ptr, ec] = std::from_chars(begin, end, ret);
@ -49,196 +125,17 @@ std::enable_if_t<std::is_floating_point_v<T>, std::optional<T>> to_num(
return ret;
}
#endif
template <typename T>
[[nodiscard]] std::enable_if_t<is_instance_of_v<numeric_wrapper, T>,
std::optional<T>>
to_num(const char* const begin, const char* const end) {
T ret;
auto [ptr, ec] = std::from_chars(begin, end, ret.value);
inline std::optional<short> from_char(char c) {
if (c >= '0' && c <= '9') {
return c - '0';
}
if (ec != std::errc() || ptr != end) {
return std::nullopt;
}
#if defined(__clang__) && defined(__MINGW32__) && !defined(__MINGW64__)
#define MINGW32_CLANG
#endif
// mingw32 clang does not support some of the builtin functions
#if (defined(__clang__) || defined(__GNUC__) || defined(__GUNG__)) && \
!defined(MINGW32_CLANG)
////////////////
// mul overflow detection
////////////////
template <typename T>
bool mul_overflow(T& result, T operand) {
return __builtin_mul_overflow(result, operand, &result);
}
template <>
inline bool mul_overflow(int& result, int operand) {
return __builtin_smul_overflow(result, operand, &result);
}
template <>
inline bool mul_overflow(long& result, long operand) {
return __builtin_smull_overflow(result, operand, &result);
}
template <>
inline bool mul_overflow(long long& result, long long operand) {
return __builtin_smulll_overflow(result, operand, &result);
}
template <>
inline bool mul_overflow(unsigned int& result, unsigned int operand) {
return __builtin_umul_overflow(result, operand, &result);
}
template <>
inline bool mul_overflow(unsigned long& result, unsigned long operand) {
return __builtin_umull_overflow(result, operand, &result);
}
template <>
inline bool mul_overflow(unsigned long long& result,
unsigned long long operand) {
return __builtin_umulll_overflow(result, operand, &result);
}
////////////////
// addition overflow detection
////////////////
template <typename T>
inline bool add_overflow(T& result, T operand) {
return __builtin_add_overflow(result, operand, &result);
}
template <>
inline bool add_overflow(int& result, int operand) {
return __builtin_sadd_overflow(result, operand, &result);
}
template <>
inline bool add_overflow(long& result, long operand) {
return __builtin_saddl_overflow(result, operand, &result);
}
template <>
inline bool add_overflow(long long& result, long long operand) {
return __builtin_saddll_overflow(result, operand, &result);
}
template <>
inline bool add_overflow(unsigned int& result, unsigned int operand) {
return __builtin_uadd_overflow(result, operand, &result);
}
template <>
inline bool add_overflow(unsigned long& result, unsigned long operand) {
return __builtin_uaddl_overflow(result, operand, &result);
}
template <>
inline bool add_overflow(unsigned long long& result,
unsigned long long operand) {
return __builtin_uaddll_overflow(result, operand, &result);
}
////////////////
// substraction overflow detection
////////////////
template <typename T>
inline bool sub_overflow(T& result, T operand) {
return __builtin_sub_overflow(result, operand, &result);
}
template <>
inline bool sub_overflow(int& result, int operand) {
return __builtin_ssub_overflow(result, operand, &result);
}
template <>
inline bool sub_overflow(long& result, long operand) {
return __builtin_ssubl_overflow(result, operand, &result);
}
template <>
inline bool sub_overflow(long long& result, long long operand) {
return __builtin_ssubll_overflow(result, operand, &result);
}
template <>
inline bool sub_overflow(unsigned int& result, unsigned int operand) {
return __builtin_usub_overflow(result, operand, &result);
}
template <>
inline bool sub_overflow(unsigned long& result, unsigned long operand) {
return __builtin_usubl_overflow(result, operand, &result);
}
template <>
inline bool sub_overflow(unsigned long long& result,
unsigned long long operand) {
return __builtin_usubll_overflow(result, operand, &result);
}
template <typename T, typename F>
bool shift_and_add_overflow(T& value, T digit, F add_last_digit_owerflow) {
if (mul_overflow<T>(value, 10) || add_last_digit_owerflow(value, digit)) {
return true;
}
return false;
}
#else
template <typename T, typename U>
bool shift_and_add_overflow(T& value, T digit, U is_negative) {
digit = (is_negative) ? -digit : digit;
T old_value = value;
value = 10 * value + digit;
T expected_old_value = (value - digit) / 10;
if (old_value != expected_old_value) {
return true;
}
return false;
}
#endif
template <typename T>
std::enable_if_t<std::is_integral_v<T>, std::optional<T>> to_num(
const char* begin, const char* end) {
if (begin == end) {
return std::nullopt;
}
bool is_negative = false;
if constexpr (std::is_signed_v<T>) {
is_negative = *begin == '-';
if (is_negative) {
++begin;
}
}
#if (defined(__clang__) || defined(__GNUC__) || defined(__GUNG__)) && \
!defined(MINGW32_CLANG)
auto add_last_digit_owerflow =
(is_negative) ? sub_overflow<T> : add_overflow<T>;
#else
auto add_last_digit_owerflow = is_negative;
#endif
T value = 0;
for (auto i = begin; i != end; ++i) {
if (auto digit = from_char(*i);
!digit || shift_and_add_overflow<T>(value, digit.value(),
add_last_digit_owerflow)) {
return std::nullopt;
}
}
return value;
return ret;
}
////////////////
@ -250,12 +147,14 @@ template <typename T>
struct unsupported_type {
constexpr static bool value = false;
};
} /* namespace */
} /* namespace errors */
template <typename T>
std::enable_if_t<!std::is_integral_v<T> && !std::is_floating_point_v<T> &&
[[nodiscard]] std::enable_if_t<!std::is_integral_v<T> &&
!std::is_floating_point_v<T> &&
!is_instance_of_v<std::optional, T> &&
!is_instance_of_v<std::variant, T>,
!is_instance_of_v<std::variant, T> &&
!is_instance_of_v<numeric_wrapper, T>,
bool>
extract(const char*, const char*, T&) {
static_assert(error::unsupported_type<T>::value,
@ -264,7 +163,10 @@ extract(const char*, const char*, T&) {
}
template <typename T>
std::enable_if_t<std::is_integral_v<T> || std::is_floating_point_v<T>, bool>
[[nodiscard]] std::enable_if_t<std::is_integral_v<T> ||
std::is_floating_point_v<T> ||
is_instance_of_v<numeric_wrapper, T>,
bool>
extract(const char* begin, const char* end, T& value) {
auto optional_value = to_num<T>(begin, end);
if (!optional_value) {
@ -275,8 +177,8 @@ extract(const char* begin, const char* end, T& value) {
}
template <typename T>
std::enable_if_t<is_instance_of_v<std::optional, T>, bool> extract(
const char* begin, const char* end, T& value) {
[[nodiscard]] std::enable_if_t<is_instance_of_v<std::optional, T>, bool>
extract(const char* begin, const char* end, T& value) {
typename T::value_type raw_value;
if (extract(begin, end, raw_value)) {
value = raw_value;
@ -287,7 +189,8 @@ std::enable_if_t<is_instance_of_v<std::optional, T>, bool> extract(
}
template <typename T, size_t I>
bool extract_variant(const char* begin, const char* end, T& value) {
[[nodiscard]] bool extract_variant(const char* begin, const char* end,
T& value) {
using IthType = std::variant_alternative_t<I, std::decay_t<T>>;
IthType ithValue;
if (extract<IthType>(begin, end, ithValue)) {
@ -300,7 +203,7 @@ bool extract_variant(const char* begin, const char* end, T& value) {
}
template <typename T>
std::enable_if_t<is_instance_of_v<std::variant, T>, bool> extract(
[[nodiscard]] std::enable_if_t<is_instance_of_v<std::variant, T>, bool> extract(
const char* begin, const char* end, T& value) {
return extract_variant<T, 0>(begin, end, value);
}
@ -310,7 +213,8 @@ std::enable_if_t<is_instance_of_v<std::variant, T>, bool> extract(
////////////////
template <>
inline bool extract(const char* begin, const char* end, bool& value) {
[[nodiscard]] inline bool extract(const char* begin, const char* end,
bool& value) {
if (end == begin + 1) {
if (*begin == '1') {
value = true;
@ -320,10 +224,13 @@ inline bool extract(const char* begin, const char* end, bool& value) {
return false;
}
} else {
size_t size = end - begin;
if (size == 4 && strncmp(begin, "true", size) == 0) {
constexpr static auto true_size = 4;
constexpr static auto false_size = 5;
const size_t size = end - begin;
if (size == true_size && std::strncmp(begin, "true", size) == 0) {
value = true;
} else if (size == 5 && strncmp(begin, "false", size) == 0) {
} else if (size == false_size &&
std::strncmp(begin, "false", size) == 0) {
value = false;
} else {
return false;
@ -334,22 +241,24 @@ inline bool extract(const char* begin, const char* end, bool& value) {
}
template <>
inline bool extract(const char* begin, const char* end, char& value) {
[[nodiscard]] inline bool extract(const char* begin, const char* end,
char& value) {
value = *begin;
return (end == begin + 1);
}
template <>
inline bool extract(const char* begin, const char* end, std::string& value) {
[[nodiscard]] inline bool extract(const char* begin, const char* end,
std::string& value) {
value = std::string{begin, end};
return true;
}
template <>
inline bool extract(const char* begin, const char* end,
[[nodiscard]] inline bool extract(const char* begin, const char* end,
std::string_view& value) {
value = std::string_view{begin, static_cast<size_t>(end - begin)};
return true;
}
} /* ss */
} /* namespace ss */

View File

@ -2,7 +2,6 @@
#include <cstdlib>
#include <functional>
#include <tuple>
namespace ss {
@ -77,4 +76,4 @@ struct member_wrapper<R T::*> {
template <typename T> \
constexpr bool has_m_##method##_t = has_m_##method<T>::value;
} /* trait */
} /* namespace ss */

File diff suppressed because it is too large Load Diff

View File

@ -10,7 +10,7 @@ template <typename T, auto... Values>
struct ax {
private:
template <auto X, auto... Xs>
bool ss_valid_impl(const T& x) const {
[[nodiscard]] bool ss_valid_impl(const T& x) const {
if constexpr (sizeof...(Xs) != 0) {
return x != X && ss_valid_impl<Xs...>(x);
}
@ -18,11 +18,11 @@ private:
}
public:
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return ss_valid_impl<Values...>(value);
}
const char* error() const {
[[nodiscard]] const char* error() const {
return "value excluded";
}
};
@ -35,7 +35,7 @@ template <typename T, auto... Values>
struct nx {
private:
template <auto X, auto... Xs>
bool ss_valid_impl(const T& x) const {
[[nodiscard]] bool ss_valid_impl(const T& x) const {
if constexpr (sizeof...(Xs) != 0) {
return x == X || ss_valid_impl<Xs...>(x);
}
@ -43,11 +43,11 @@ private:
}
public:
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return ss_valid_impl<Values...>(value);
}
const char* error() const {
[[nodiscard]] const char* error() const {
return "value excluded";
}
};
@ -61,28 +61,28 @@ public:
template <typename T, auto N>
struct gt {
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return value > N;
}
};
template <typename T, auto N>
struct gte {
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return value >= N;
}
};
template <typename T, auto N>
struct lt {
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return value < N;
}
};
template <typename T, auto N>
struct lte {
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return value <= N;
}
};
@ -93,7 +93,7 @@ struct lte {
template <typename T, auto Min, auto Max>
struct ir {
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return value >= Min && value <= Max;
}
};
@ -104,7 +104,7 @@ struct ir {
template <typename T, auto Min, auto Max>
struct oor {
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return value < Min || value > Max;
}
};
@ -115,13 +115,13 @@ struct oor {
template <typename T>
struct ne {
bool ss_valid(const T& value) const {
[[nodiscard]] bool ss_valid(const T& value) const {
return !value.empty();
}
const char* error() const {
[[nodiscard]] const char* error() const {
return "empty field";
}
};
} /* ss */
} /* namespace ss */

View File

@ -109,7 +109,7 @@ struct get_matcher<Matcher, T, Ts...> {
struct is_matcher : is_instance_of_matcher<U, Matcher> {};
static_assert(count_v<is_matcher, T, Ts...> <= 1,
"the same matcher is cannot"
"the same matcher cannot "
"be defined multiple times");
using type = std::conditional_t<is_matcher<T>::value, T,
typename get_matcher<Matcher, Ts...>::type>;
@ -165,34 +165,40 @@ using get_multiline_t = typename get_multiline<Ts...>::type;
// string_error
////////////////
class string_error;
class string_error {};
////////////////
// ignore_header
////////////////
class ignore_header;
class ignore_header {};
////////////////
// ignore_empty
////////////////
class ignore_empty;
class ignore_empty {};
////////////////
// throw_on_error
////////////////
class throw_on_error {};
////////////////
// setup implementation
////////////////
template <typename... Ts>
template <typename... Options>
struct setup {
private:
template <typename T>
template <typename Option>
struct is_matcher
: std::disjunction<is_instance_of_matcher_t<T, quote>,
is_instance_of_matcher_t<T, escape>,
is_instance_of_matcher_t<T, trim>,
is_instance_of_matcher_t<T, trim_left>,
is_instance_of_matcher_t<T, trim_right>> {};
: std::disjunction<is_instance_of_matcher_t<Option, quote>,
is_instance_of_matcher_t<Option, escape>,
is_instance_of_matcher_t<Option, trim>,
is_instance_of_matcher_t<Option, trim_left>,
is_instance_of_matcher_t<Option, trim_right>> {};
template <typename T>
struct is_string_error : std::is_same<T, string_error> {};
@ -203,39 +209,48 @@ private:
template <typename T>
struct is_ignore_empty : std::is_same<T, ignore_empty> {};
constexpr static auto count_matcher = count_v<is_matcher, Ts...>;
template <typename T>
struct is_throw_on_error : std::is_same<T, throw_on_error> {};
constexpr static auto count_matcher = count_v<is_matcher, Options...>;
constexpr static auto count_multiline =
count_v<is_instance_of_multiline, Ts...>;
count_v<is_instance_of_multiline, Options...>;
constexpr static auto count_string_error = count_v<is_string_error, Ts...>;
constexpr static auto count_string_error =
count_v<is_string_error, Options...>;
constexpr static auto count_ignore_header =
count_v<is_ignore_header, Ts...>;
count_v<is_ignore_header, Options...>;
constexpr static auto count_ignore_empty = count_v<is_ignore_empty, Ts...>;
constexpr static auto count_throw_on_error =
count_v<is_throw_on_error, Options...>;
constexpr static auto count_ignore_empty =
count_v<is_ignore_empty, Options...>;
constexpr static auto number_of_valid_setup_types =
count_matcher + count_multiline + count_string_error +
count_ignore_header + count_ignore_empty;
count_ignore_header + count_ignore_empty + count_throw_on_error;
using trim_left_only = get_matcher_t<trim_left, Ts...>;
using trim_right_only = get_matcher_t<trim_right, Ts...>;
using trim_all = get_matcher_t<trim, Ts...>;
using trim_left_only = get_matcher_t<trim_left, Options...>;
using trim_right_only = get_matcher_t<trim_right, Options...>;
using trim_all = get_matcher_t<trim, Options...>;
public:
using quote = get_matcher_t<quote, Ts...>;
using escape = get_matcher_t<escape, Ts...>;
using quote = get_matcher_t<quote, Options...>;
using escape = get_matcher_t<escape, Options...>;
using trim_left =
std::conditional_t<trim_all::enabled, trim_all, trim_left_only>;
using trim_right =
std::conditional_t<trim_all::enabled, trim_all, trim_right_only>;
using multiline = get_multiline_t<Ts...>;
using multiline = get_multiline_t<Options...>;
constexpr static bool string_error = (count_string_error == 1);
constexpr static bool ignore_header = (count_ignore_header == 1);
constexpr static bool ignore_empty = (count_ignore_empty == 1);
constexpr static bool throw_on_error = (count_throw_on_error == 1);
private:
#define ASSERT_MSG "cannot have the same match character in multiple matchers"
@ -254,21 +269,31 @@ private:
static_assert(
!multiline::enabled ||
(multiline::enabled && (quote::enabled || escape::enabled)),
"to enable multiline either quote or escape need to be enabled");
"to enable multiline either quote or escape needs to be enabled");
static_assert(!(trim_all::enabled && trim_left_only::enabled) &&
!(trim_all::enabled && trim_right_only::enabled),
"ambiguous trim setup");
static_assert(count_multiline <= 1, "mutliline defined multiple times");
static_assert(count_string_error <= 1,
"string_error defined multiple times");
static_assert(number_of_valid_setup_types == sizeof...(Ts),
static_assert(count_throw_on_error <= 1,
"throw_on_error defined multiple times");
static_assert(count_throw_on_error + count_string_error <= 1,
"cannot define both throw_on_error and string_error");
static_assert(number_of_valid_setup_types == sizeof...(Options),
"one or multiple invalid setup parameters defined");
};
template <typename... Ts>
struct setup<setup<Ts...>> : setup<Ts...> {};
template <typename... Options>
struct setup<setup<Options...>> : setup<Options...> {};
} /* ss */
template <typename... Options>
struct setup<std::tuple<Options...>> : setup<Options...> {};
} /* namespace ss */

View File

@ -1,26 +1,26 @@
#pragma once
#include "common.hpp"
#include "exception.hpp"
#include "setup.hpp"
#include "type_traits.hpp"
#include <algorithm>
#include <cstdlib>
#include <cstring>
#include <memory>
#include <string>
#include <vector>
namespace ss {
template <typename... Ts>
template <typename... Options>
class splitter {
private:
using quote = typename setup<Ts...>::quote;
using trim_left = typename setup<Ts...>::trim_left;
using trim_right = typename setup<Ts...>::trim_right;
using escape = typename setup<Ts...>::escape;
using multiline = typename setup<Ts...>::multiline;
using quote = typename setup<Options...>::quote;
using trim_left = typename setup<Options...>::trim_left;
using trim_right = typename setup<Options...>::trim_right;
using escape = typename setup<Options...>::escape;
using multiline = typename setup<Options...>::multiline;
constexpr static auto string_error = setup<Ts...>::string_error;
constexpr static auto string_error = setup<Options...>::string_error;
constexpr static auto throw_on_error = setup<Options...>::throw_on_error;
constexpr static auto is_const_line = !quote::enabled && !escape::enabled;
using error_type = std::conditional_t<string_error, std::string, bool>;
@ -28,20 +28,22 @@ private:
public:
using line_ptr_type = std::conditional_t<is_const_line, const char*, char*>;
bool valid() const {
[[nodiscard]] bool valid() const {
if constexpr (string_error) {
return error_.empty();
} else if constexpr (throw_on_error) {
return true;
} else {
return !error_;
}
}
const std::string& error_msg() const {
[[nodiscard]] const std::string& error_msg() const {
assert_string_error_defined<string_error>();
return error_;
}
bool unterminated_quote() const {
[[nodiscard]] bool unterminated_quote() const {
return unterminated_quote_;
}
@ -53,13 +55,21 @@ public:
return split_impl_select_delim(delimiter);
}
[[nodiscard]] const split_data& get_split_data() const {
return split_data_;
}
void clear_split_data() {
split_data_.clear();
}
private:
////////////////
// resplit
////////////////
// number of characters the end of line is shifted backwards
size_t size_shifted() const {
[[nodiscard]] size_t size_shifted() const {
return escaped_;
}
@ -77,16 +87,16 @@ private:
// resplitting, continue from last slice
if (!quote::enabled || !multiline::enabled || split_data_.empty() ||
!unterminated_quote()) {
set_error_invalid_resplit();
handle_error_invalid_resplit();
return split_data_;
}
const auto [old_line, old_begin] = *std::prev(split_data_.end());
size_t begin = old_begin - old_line - 1;
const size_t begin = old_begin - old_line - 1;
// safety measure
if (new_size != -1 && static_cast<size_t>(new_size) < begin) {
set_error_invalid_resplit();
handle_error_invalid_resplit();
return split_data_;
}
@ -112,55 +122,75 @@ private:
void clear_error() {
if constexpr (string_error) {
error_.clear();
} else {
} else if constexpr (!throw_on_error) {
error_ = false;
}
unterminated_quote_ = false;
}
void set_error_empty_delimiter() {
void handle_error_empty_delimiter() {
constexpr static auto error_msg = "empty delimiter";
if constexpr (string_error) {
error_.clear();
error_.append("empt delimiter");
error_.append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg};
} else {
error_ = true;
}
}
void set_error_mismatched_quote(size_t n) {
void handle_error_mismatched_quote(size_t n) {
constexpr static auto error_msg = "mismatched quote at position: ";
if constexpr (string_error) {
error_.clear();
error_.append("mismatched quote at position: " + std::to_string(n));
error_.append(error_msg + std::to_string(n));
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg + std::to_string(n)};
} else {
error_ = true;
}
}
void set_error_unterminated_escape() {
void handle_error_unterminated_escape() {
constexpr static auto error_msg =
"unterminated escape at the end of the line";
if constexpr (string_error) {
error_.clear();
error_.append("unterminated escape at the end of the line");
error_.append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg};
} else {
error_ = true;
}
}
void set_error_unterminated_quote() {
unterminated_quote_ = true;
void handle_error_unterminated_quote() {
constexpr static auto error_msg = "unterminated quote";
if constexpr (string_error) {
error_.clear();
error_.append("unterminated quote");
error_.append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg};
} else {
error_ = true;
}
}
void set_error_invalid_resplit() {
unterminated_quote_ = false;
void handle_error_invalid_resplit() {
constexpr static auto error_msg =
"invalid resplit, new line must be longer"
"than the end of the last slice";
if constexpr (string_error) {
error_.clear();
error_.append("invalid resplit, new line must be longer"
"than the end of the last slice");
error_.append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg};
} else {
error_ = true;
}
@ -170,19 +200,19 @@ private:
// matching
////////////////
bool match(const char* const curr, char delim) {
[[nodiscard]] bool match(const char* const curr, char delim) {
return *curr == delim;
};
bool match(const char* const curr, const std::string& delim) {
return strncmp(curr, delim.c_str(), delim.size()) == 0;
[[nodiscard]] bool match(const char* const curr, const std::string& delim) {
return std::strncmp(curr, delim.c_str(), delim.size()) == 0;
};
size_t delimiter_size(char) {
[[nodiscard]] size_t delimiter_size(char) {
return 1;
}
size_t delimiter_size(const std::string& delim) {
[[nodiscard]] size_t delimiter_size(const std::string& delim) {
return delim.size();
}
@ -203,7 +233,7 @@ private:
}
template <typename Delim>
std::tuple<size_t, bool> match_delimiter(line_ptr_type begin,
[[nodiscard]] std::tuple<size_t, bool> match_delimiter(line_ptr_type begin,
const Delim& delim) {
line_ptr_type end = begin;
@ -235,7 +265,9 @@ private:
if constexpr (escape::enabled) {
if (escape::match(*curr)) {
if (curr[1] == '\0') {
set_error_unterminated_escape();
if constexpr (!multiline::enabled) {
handle_error_unterminated_escape();
}
done_ = true;
return;
}
@ -282,7 +314,7 @@ private:
clear_error();
switch (delimiter.size()) {
case 0:
set_error_empty_delimiter();
handle_error_empty_delimiter();
return split_data_;
case 1:
return split_impl(delimiter[0]);
@ -296,8 +328,9 @@ private:
trim_left_if_enabled(begin_);
for (done_ = false; !done_; read(delim))
;
for (done_ = false; !done_;) {
read(delim);
}
return split_data_;
}
@ -362,7 +395,9 @@ private:
if (end_[1] == '\0') {
// eol, unterminated escape
// eg: ... "hel\\0
set_error_unterminated_escape();
if constexpr (!multiline::enabled) {
handle_error_unterminated_escape();
}
done_ = true;
break;
}
@ -379,7 +414,10 @@ private:
// eg: ..."hell\0 -> quote not terminated
if (*end_ == '\0') {
shift_and_set_current();
set_error_unterminated_quote();
unterminated_quote_ = true;
if constexpr (!multiline::enabled) {
handle_error_unterminated_quote();
}
split_data_.emplace_back(line_, begin_);
done_ = true;
break;
@ -418,7 +456,7 @@ private:
} else {
// mismatched quote
// eg: ...,"hel"lo,... -> error
set_error_mismatched_quote(end_ - line_);
handle_error_mismatched_quote(end_ - line_);
split_data_.emplace_back(line_, begin_);
}
done_ = true;
@ -431,7 +469,6 @@ private:
// members
////////////////
public:
error_type error_{};
bool unterminated_quote_{false};
bool done_{true};
@ -448,4 +485,4 @@ public:
friend class converter;
};
} /* ss */
} /* namespace ss */

View File

@ -34,7 +34,11 @@ struct left_of_impl;
template <size_t N, typename T, typename... Ts>
struct left_of_impl {
static_assert(N < 128, "recursion limit reached");
private:
constexpr static auto recursion_limit = 128;
public:
static_assert(N < recursion_limit, "recursion limit reached");
static_assert(N != 0, "cannot take the whole tuple");
using type = tup_cat_t<T, typename left_of_impl<N - 1, Ts...>::type>;
};
@ -362,12 +366,12 @@ constexpr bool is_instance_of_v = is_instance_of<Template, Ts...>::value;
////////////////
template <class T, std::size_t... Is, class U>
T to_object_impl(std::index_sequence<Is...>, U&& data) {
[[nodiscard]] T to_object_impl(std::index_sequence<Is...>, U&& data) {
return {std::get<Is>(std::forward<U>(data))...};
}
template <class T, class U>
T to_object(U&& data) {
[[nodiscard]] T to_object(U&& data) {
using NoRefU = std::decay_t<U>;
if constexpr (is_instance_of_v<std::tuple, NoRefU>) {
return to_object_impl<
@ -378,4 +382,4 @@ T to_object(U&& data) {
}
}
} /* trait */
} /* namespace ss */

View File

@ -6,7 +6,7 @@ project(
'cpp_std=c++17',
'buildtype=debugoptimized',
'wrap_mode=forcefallback'],
version: '1.4.0',
version: '1.8.0',
meson_version:'>=0.54.0')
fast_float_dep = dependency('fast_float')

6
script/ci_install_lcov.sh Executable file
View File

@ -0,0 +1,6 @@
#!/usr/bin/env sh
echo yes | cpan DateTime Capture::Tiny
wget -qO- https://github.com/linux-test-project/lcov/releases/download/v2.0/lcov-2.0.tar.gz | tar xvz
(cd lcov-2.0 && make install)

View File

@ -2,6 +2,7 @@
headers_dir = 'include/ss/'
headers = ['type_traits.hpp',
'exception.hpp',
'function_traits.hpp',
'restrictions.hpp',
'common.hpp',
@ -13,14 +14,21 @@ headers = ['type_traits.hpp',
combined_file = []
includes = []
in_pp_block = False
for header in headers:
with open(headers_dir + header) as f:
for line in f.read().splitlines():
if '#if ' in line:
in_pp_block = True
if '#endif' in line:
in_pp_block = False
if '#include "' in line or '#include <fast_float' in line:
continue
if '#include <' in line:
if '#include <' in line and not in_pp_block:
includes.append(line)
continue
@ -29,6 +37,7 @@ for header in headers:
includes = sorted(set(includes))
print('#pragma once')
print('\n'.join(includes))
print('#define SSP_DISABLE_FAST_FLOAT')
print('\n'.join(combined_file))

1642
ssp.hpp

File diff suppressed because it is too large Load Diff

View File

@ -5,37 +5,43 @@ project(ssp_tests CXX)
# ---- Dependencies ----
include(FetchContent)
fetchcontent_declare(ssp SOURCE_DIR "${PROJECT_SOURCE_DIR}/..")
fetchcontent_makeavailable(ssp)
FetchContent_Declare(ssp SOURCE_DIR "${PROJECT_SOURCE_DIR}/..")
FetchContent_MakeAvailable(ssp)
if(CMAKE_CXX_COMPILER_ID MATCHES "GNU|Clang")
target_compile_options(ssp INTERFACE -Wall -Wextra)
endif()
if (MSVC)
add_compile_options(/bigobj)
elseif (MINGW)
add_compile_options(-Wa,-mbig-obj)
endif ()
include(FetchContent)
fetchcontent_declare(
FetchContent_Declare(
DOCTEST
GIT_REPOSITORY https://github.com/red0124/doctest
GIT_TAG origin/master
GIT_SHALLOW TRUE
)
GIT_SHALLOW TRUE)
fetchcontent_makeavailable(DOCTEST)
FetchContent_MakeAvailable(DOCTEST)
set(DOCTEST "${FETCHCONTENT_BASE_DIR}/doctest-src")
# ---- Test ----
enable_testing()
foreach(name IN ITEMS test_splitter test_parser test_converter test_extractions)
foreach(name IN ITEMS test_splitter test_parser1_1 test_parser1_2
test_parser1_3 test_parser1_4 test_parser1_5
test_converter test_extractions test_parser2_1
test_parser2_2 test_parser2_3 test_parser2_4
test_parser2_5 test_parser2_6
test_extractions_without_fast_float)
add_executable("${name}" "${name}.cpp")
target_link_libraries(
"${name}"
PRIVATE ssp::ssp fast_float doctest::doctest
)
target_link_libraries("${name}" PRIVATE ssp::ssp fast_float
doctest::doctest)
target_compile_definitions(
"${name}"
PRIVATE DOCTEST_CONFIG_IMPLEMENT_WITH_MAIN CMAKE_GITHUB_CI
)
"${name}" PRIVATE DOCTEST_CONFIG_IMPLEMENT_WITH_MAIN CMAKE_GITHUB_CI)
add_test(NAME "${name}" COMMAND "${name}")
endforeach()

View File

@ -1,19 +1,33 @@
test_sources = files([
'test_main.cpp',
'test_splitter.cpp',
'test_converter.cpp',
'test_parser.cpp',
'test_extractions.cpp',
'test_extractions_without_fast_float.cpp',
])
doctest_dep = dependency('doctest')
add_project_arguments('-DDOCTEST_CONFIG_IMPLEMENT_WITH_MAIN', language: 'cpp')
test_exe = executable(
'test_ssp',
sources: test_sources,
dependencies: [doctest_dep, ssp_dep],
cpp_args: '-lstdc++fs'
tests = [
'parser1_1',
'parser1_2',
'parser1_3',
'parser1_4',
'parser1_5',
'splitter',
'converter',
'extractions',
'parser2_1',
'parser2_2',
'parser2_3',
'parser2_4',
'parser2_5',
'parser2_6',
'extractions_without_fast_float',
]
foreach name : tests
test_name = 'test_' + name
exe = executable(
test_name,
test_name + '.cpp',
dependencies: [doctest_dep, ssp_dep]
)
test('test_ssp', test_exe)
test(test_name, exe, timeout: 60)
endforeach

View File

@ -1,9 +1,30 @@
#include "test_helpers.hpp"
#include <algorithm>
#include <ss/converter.hpp>
TEST_CASE("converter test split") {
ss::converter c;
for (const auto& [s, expected, delim] :
// clang-format off
{std::make_tuple("a,b,c,d", std::vector{"a", "b", "c", "d"}, ","),
{"", {}, " "},
{" x x x x | x ", {" x x x x ", " x "}, "|"},
{"a::b::c::d", {"a", "b", "c", "d"}, "::"},
{"x\t-\ty", {"x", "y"}, "\t-\t"},
{"x", {"x"}, ","}}
// clang-format on
) {
auto split = c.split(s, delim);
CHECK_EQ(split.size(), expected.size());
for (size_t i = 0; i < split.size(); ++i) {
auto s = std::string(split[i].first, split[i].second);
CHECK_EQ(s, expected[i]);
}
}
}
TEST_CASE("converter test split with exceptions") {
ss::converter<ss::throw_on_error> c;
try {
for (const auto& [s, expected, delim] :
// clang-format off
{std::make_tuple("a,b,c,d", std::vector{"a", "b", "c", "d"}, ","),
@ -20,92 +41,93 @@ TEST_CASE("converter test split") {
CHECK_EQ(s, expected[i]);
}
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE("converter test valid conversions") {
TEST_CASE_TEMPLATE("converter test valid conversions", T, int, ss::uint8) {
ss::converter c;
{
auto tup = c.convert<int>("5");
auto tup = c.convert<T>("5");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
}
{
auto tup = c.convert<int, void>("5,junk");
auto tup = c.convert<T, void>("5,junk");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
}
{
auto tup = c.convert<void, int>("junk,5");
auto tup = c.convert<void, T>("junk,5");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
}
{
auto tup = c.convert<int, void, void>("5\njunk\njunk", "\n");
auto tup = c.convert<T, void, void>("5\njunk\njunk", "\n");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
}
{
auto tup = c.convert<void, int, void>("junk 5 junk", " ");
auto tup = c.convert<void, T, void>("junk 5 junk", " ");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
}
{
auto tup = c.convert<void, void, int>("junk\tjunk\t5", "\t");
auto tup = c.convert<void, void, T>("junk\tjunk\t5", "\t");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
}
{
auto tup =
c.convert<void, void, std::optional<int>>("junk\tjunk\t5", "\t");
c.convert<void, void, std::optional<T>>("junk\tjunk\t5", "\t");
REQUIRE(c.valid());
REQUIRE(tup.has_value());
CHECK_EQ(tup, 5);
}
{
auto tup = c.convert<int, double, void>("5,6.6,junk");
auto tup = c.convert<T, double, void>("5,6.6,junk");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
}
{
auto tup = c.convert<int, void, double>("5,junk,6.6");
auto tup = c.convert<T, void, double>("5,junk,6.6");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
}
{
auto tup = c.convert<void, int, double>("junk;5;6.6", ";");
auto tup = c.convert<void, T, double>("junk;5;6.6", ";");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
}
{
auto tup =
c.convert<void, std::optional<int>, double>("junk;5;6.6", ";");
auto tup = c.convert<void, std::optional<T>, double>("junk;5;6.6", ";");
REQUIRE(c.valid());
REQUIRE(std::get<0>(tup).has_value());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
}
{
auto tup =
c.convert<void, std::optional<int>, double>("junk;5.4;6.6", ";");
c.convert<void, std::optional<T>, double>("junk;5.4;6.6", ";");
REQUIRE(c.valid());
REQUIRE_FALSE(std::get<0>(tup).has_value());
CHECK_EQ(tup, std::make_tuple(std::optional<int>{}, 6.6));
CHECK_EQ(tup, std::make_tuple(std::optional<T>{}, 6.6));
}
{
auto tup =
c.convert<void, std::variant<int, double>, double>("junk;5;6.6",
";");
c.convert<void, std::variant<T, double>, double>("junk;5;6.6", ";");
REQUIRE(c.valid());
REQUIRE(std::holds_alternative<int>(std::get<0>(tup)));
CHECK_EQ(tup, std::make_tuple(std::variant<int, double>{5}, 6.6));
REQUIRE(std::holds_alternative<T>(std::get<0>(tup)));
CHECK_EQ(tup, std::make_tuple(std::variant<T, double>{T(5)}, 6.6));
}
{
auto tup =
c.convert<void, std::variant<int, double>, double>("junk;5.5;6.6",
c.convert<void, std::variant<T, double>, double>("junk;5.5;6.6",
";");
REQUIRE(c.valid());
REQUIRE(std::holds_alternative<double>(std::get<0>(tup)));
CHECK_EQ(tup, std::make_tuple(std::variant<int, double>{5.5}, 6.6));
CHECK_EQ(tup, std::make_tuple(std::variant<T, double>{5.5}, 6.6));
}
{
auto tup = c.convert<void, std::string_view, double,
@ -116,81 +138,272 @@ TEST_CASE("converter test valid conversions") {
}
}
TEST_CASE("converter test invalid conversions") {
TEST_CASE_TEMPLATE("converter test valid conversions with exceptions", T, int,
ss::uint8) {
ss::converter<ss::throw_on_error> c;
try {
auto tup = c.convert<T>("5");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<T, void>("5,junk");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<void, T>("junk,5");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<T, void, void>("5\njunk\njunk", "\n");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<void, T, void>("junk 5 junk", " ");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<void, void, T>("junk\tjunk\t5", "\t");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup =
c.convert<void, void, std::optional<T>>("junk\tjunk\t5", "\t");
REQUIRE(c.valid());
REQUIRE(tup.has_value());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<T, double, void>("5,6.6,junk");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<T, void, double>("5,junk,6.6");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<void, T, double>("junk;5;6.6", ";");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<void, std::optional<T>, double>("junk;5;6.6", ";");
REQUIRE(c.valid());
REQUIRE(std::get<0>(tup).has_value());
CHECK_EQ(tup, std::make_tuple(5, 6.6));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup =
c.convert<void, std::optional<T>, double>("junk;5.4;6.6", ";");
REQUIRE(c.valid());
REQUIRE_FALSE(std::get<0>(tup).has_value());
CHECK_EQ(tup, std::make_tuple(std::optional<T>{}, 6.6));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup =
c.convert<void, std::variant<T, double>, double>("junk;5;6.6", ";");
REQUIRE(c.valid());
REQUIRE(std::holds_alternative<T>(std::get<0>(tup)));
CHECK_EQ(tup, std::make_tuple(std::variant<T, double>{T(5)}, 6.6));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup =
c.convert<void, std::variant<T, double>, double>("junk;5.5;6.6",
";");
REQUIRE(c.valid());
REQUIRE(std::holds_alternative<double>(std::get<0>(tup)));
CHECK_EQ(tup, std::make_tuple(std::variant<T, double>{5.5}, 6.6));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<void, std::string_view, double,
std::string_view>("junk;s1;6.6;s2", ";");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(std::string_view{"s1"}, 6.6,
std::string_view{"s2"}));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE_TEMPLATE("converter test invalid conversions", T, int, ss::uint8) {
ss::converter c;
c.convert<int>("");
std::ignore = c.convert<T>("");
REQUIRE_FALSE(c.valid());
c.convert<int>("10", "");
std::ignore = c.convert<T>("1", "");
REQUIRE_FALSE(c.valid());
c.convert<int, void>("");
std::ignore = c.convert<T>("10", "");
REQUIRE_FALSE(c.valid());
c.convert<int, void>(",junk");
std::ignore = c.convert<T, void>("");
REQUIRE_FALSE(c.valid());
c.convert<void, int>("junk,");
std::ignore = c.convert<T, void>(",junk");
REQUIRE_FALSE(c.valid());
c.convert<int>("x");
std::ignore = c.convert<void, T>("junk,");
REQUIRE_FALSE(c.valid());
c.convert<int, void>("x");
std::ignore = c.convert<T>("x");
REQUIRE_FALSE(c.valid());
c.convert<int, void>("x,junk");
std::ignore = c.convert<T, void>("x");
REQUIRE_FALSE(c.valid());
c.convert<void, int>("junk,x");
std::ignore = c.convert<T, void>("x,junk");
REQUIRE_FALSE(c.valid());
c.convert<void, std::variant<int, double>, double>("junk;.5.5;6", ";");
std::ignore = c.convert<void, T>("junk,x");
REQUIRE_FALSE(c.valid());
std::ignore =
c.convert<void, std::variant<T, double>, double>("junk;.5.5;6", ";");
REQUIRE_FALSE(c.valid());
}
TEST_CASE("converter test ss:ax restriction (all except)") {
TEST_CASE_TEMPLATE("converter test invalid conversions with exceptions", T, int,
ss::uint8) {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(std::ignore = c.convert<T>(""));
REQUIRE_EXCEPTION(std::ignore = c.convert<T>("1", ""));
REQUIRE_EXCEPTION(std::ignore = c.convert<T>("10", ""));
REQUIRE_EXCEPTION(std::ignore = c.convert<T, void>(""));
REQUIRE_EXCEPTION(std::ignore = c.convert<T, void>(",junk"));
REQUIRE_EXCEPTION(std::ignore = c.convert<void, T>("junk,"));
REQUIRE_EXCEPTION(std::ignore = c.convert<T>("x"));
REQUIRE_EXCEPTION(std::ignore = c.convert<T, void>("x"));
REQUIRE_EXCEPTION(std::ignore = c.convert<T, void>("x,junk"));
REQUIRE_EXCEPTION(std::ignore = c.convert<void, T>("junk,x"));
REQUIRE_EXCEPTION(
std::ignore =
c.convert<void, std::variant<T, double>, double>("junk;.5.5;6",
";"));
}
TEST_CASE_TEMPLATE("converter test ss:ax restriction (all except)", T, int,
ss::uint8) {
ss::converter c;
c.convert<ss::ax<int, 0>>("0");
std::ignore = c.convert<ss::ax<T, 0>>("0");
REQUIRE_FALSE(c.valid());
c.convert<ss::ax<int, 0, 1, 2>>("1");
std::ignore = c.convert<ss::ax<T, 0, 1, 2>>("1");
REQUIRE_FALSE(c.valid());
c.convert<void, char, ss::ax<int, 0, 1, 2>>("junk,c,1");
std::ignore = c.convert<void, char, ss::ax<T, 0, 1, 2>>("junk,c,1");
REQUIRE_FALSE(c.valid());
c.convert<ss::ax<int, 1>, char>("1,c");
std::ignore = c.convert<ss::ax<T, 1>, char>("1,c");
REQUIRE_FALSE(c.valid());
{
int tup = c.convert<ss::ax<int, 1>>("3");
T tup = c.convert<ss::ax<T, 1>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
std::tuple<char, int> tup = c.convert<char, ss::ax<int, 1>>("c,3");
std::tuple<char, T> tup = c.convert<char, ss::ax<T, 1>>("c,3");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple('c', 3));
}
{
std::tuple<int, char> tup = c.convert<ss::ax<int, 1>, char>("3,c");
std::tuple<T, char> tup = c.convert<ss::ax<T, 1>, char>("3,c");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(3, 'c'));
}
}
TEST_CASE_TEMPLATE(
"converter test ss:ax restriction (all except) with exceptions", T, int,
ss::uint8) {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ax<T, 0>>("0"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ax<T, 0, 1, 2>>("1"));
REQUIRE_EXCEPTION(
std::ignore = c.convert<void, char, ss::ax<T, 0, 1, 2>>("junk,c,1"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ax<T, 1>, char>("1,c"));
try {
{
T tup = c.convert<ss::ax<T, 1>>("3");
CHECK_EQ(tup, 3);
}
{
std::tuple<char, T> tup = c.convert<char, ss::ax<T, 1>>("c,3");
CHECK_EQ(tup, std::make_tuple('c', 3));
}
{
std::tuple<T, char> tup = c.convert<ss::ax<T, 1>, char>("3,c");
CHECK_EQ(tup, std::make_tuple(3, 'c'));
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE("converter test ss:nx restriction (none except)") {
ss::converter c;
c.convert<ss::nx<int, 1>>("3");
std::ignore = c.convert<ss::nx<int, 1>>("3");
REQUIRE_FALSE(c.valid());
c.convert<char, ss::nx<int, 1, 2, 69>>("c,3");
std::ignore = c.convert<char, ss::nx<int, 1, 2, 69>>("c,3");
REQUIRE_FALSE(c.valid());
c.convert<ss::nx<int, 1>, char>("3,c");
std::ignore = c.convert<ss::nx<int, 1>, char>("3,c");
REQUIRE_FALSE(c.valid());
{
@ -215,53 +428,123 @@ TEST_CASE("converter test ss:nx restriction (none except)") {
}
}
TEST_CASE("converter test ss:ir restriction (in range)") {
ss::converter c;
TEST_CASE("converter test ss:nx restriction (none except) with exceptions") {
ss::converter<ss::throw_on_error> c;
c.convert<ss::ir<int, 0, 2>>("3");
REQUIRE_FALSE(c.valid());
c.convert<char, ss::ir<int, 4, 69>>("c,3");
REQUIRE_FALSE(c.valid());
c.convert<ss::ir<int, 1, 2>, char>("3,c");
REQUIRE_FALSE(c.valid());
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::nx<int, 1>>("3"));
REQUIRE_EXCEPTION(std::ignore =
c.convert<char, ss::nx<int, 1, 2, 69>>("c,3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::nx<int, 1>, char>("3,c"));
try {
{
auto tup = c.convert<ss::ir<int, 1, 5>>("3");
auto tup = c.convert<ss::nx<int, 3>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::ir<int, 0, 2>>("2");
auto tup = c.convert<ss::nx<int, 0, 1, 2>>("2");
REQUIRE(c.valid());
CHECK_EQ(tup, 2);
}
{
auto tup = c.convert<char, void, ss::ir<int, 0, 1>>("c,junk,1");
auto tup = c.convert<char, void, ss::nx<int, 0, 1, 2>>("c,junk,1");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple('c', 1));
}
{
auto tup = c.convert<ss::ir<int, 1, 20>, char>("1,c");
auto tup = c.convert<ss::nx<int, 1>, char>("1,c");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(1, 'c'));
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE_TEMPLATE("converter test ss:ir restriction (in range)", T, int,
ss::uint8) {
ss::converter c;
std::ignore = c.convert<ss::ir<T, 0, 2>>("3");
REQUIRE_FALSE(c.valid());
std::ignore = c.convert<char, ss::ir<T, 4, 69>>("c,3");
REQUIRE_FALSE(c.valid());
std::ignore = c.convert<ss::ir<T, 1, 2>, char>("3,c");
REQUIRE_FALSE(c.valid());
{
auto tup = c.convert<ss::ir<T, 1, 5>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::ir<T, 0, 2>>("2");
REQUIRE(c.valid());
CHECK_EQ(tup, 2);
}
{
auto tup = c.convert<char, void, ss::ir<T, 0, 1>>("c,junk,1");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple('c', 1));
}
{
auto tup = c.convert<ss::ir<T, 1, 20>, char>("1,c");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(1, 'c'));
}
}
TEST_CASE_TEMPLATE(
"converter test ss:ir restriction (in range) with exceptions", T, int,
ss::uint8) {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ir<T, 0, 2>>("3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<char, ss::ir<T, 4, 69>>("c,3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ir<T, 1, 2>, char>("3,c"));
try {
{
auto tup = c.convert<ss::ir<T, 1, 5>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::ir<T, 0, 2>>("2");
REQUIRE(c.valid());
CHECK_EQ(tup, 2);
}
{
auto tup = c.convert<char, void, ss::ir<T, 0, 1>>("c,junk,1");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple('c', 1));
}
{
auto tup = c.convert<ss::ir<T, 1, 20>, char>("1,c");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(1, 'c'));
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE("converter test ss:oor restriction (out of range)") {
ss::converter c;
c.convert<ss::oor<int, 1, 5>>("3");
std::ignore = c.convert<ss::oor<int, 1, 5>>("3");
REQUIRE_FALSE(c.valid());
c.convert<ss::oor<int, 0, 2>>("2");
std::ignore = c.convert<ss::oor<int, 0, 2>>("2");
REQUIRE_FALSE(c.valid());
c.convert<char, ss::oor<int, 0, 1>, void>("c,1,junk");
std::ignore = c.convert<char, ss::oor<int, 0, 1>, void>("c,1,junk");
REQUIRE_FALSE(c.valid());
c.convert<ss::oor<int, 1, 20>, char>("1,c");
std::ignore = c.convert<ss::oor<int, 1, 20>, char>("1,c");
REQUIRE_FALSE(c.valid());
{
@ -283,7 +566,40 @@ TEST_CASE("converter test ss:oor restriction (out of range)") {
}
}
const std::vector<int> extracted_vector = {1, 2, 3};
TEST_CASE("converter test ss:oor restriction (out of range) with exceptions") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::oor<int, 1, 5>>("3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::oor<int, 0, 2>>("2"));
REQUIRE_EXCEPTION(
std::ignore = c.convert<char, ss::oor<int, 0, 1>, void>("c,1,junk"));
REQUIRE_EXCEPTION(std::ignore =
c.convert<ss::oor<int, 1, 20>, char>("1,c"));
try {
{
auto tup = c.convert<ss::oor<int, 0, 2>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<char, void, ss::oor<int, 4, 69>>("c,junk,3");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple('c', 3));
}
{
auto tup = c.convert<ss::oor<int, 1, 2>, char>("3,c");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(3, 'c'));
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
const inline std::vector<int> extracted_vector = {1, 2, 3};
// custom extract
template <>
@ -299,19 +615,19 @@ inline bool ss::extract(const char* begin, const char* end,
TEST_CASE("converter test ss:ne restriction (not empty)") {
ss::converter c;
c.convert<ss::ne<std::string>>("");
std::ignore = c.convert<ss::ne<std::string>>("");
REQUIRE_FALSE(c.valid());
c.convert<int, ss::ne<std::string>>("3,");
std::ignore = c.convert<int, ss::ne<std::string>>("3,");
REQUIRE_FALSE(c.valid());
c.convert<ss::ne<std::string>, int>(",3");
std::ignore = c.convert<ss::ne<std::string>, int>(",3");
REQUIRE_FALSE(c.valid());
c.convert<void, ss::ne<std::string>, int>("junk,,3");
std::ignore = c.convert<void, ss::ne<std::string>, int>("junk,,3");
REQUIRE_FALSE(c.valid());
c.convert<ss::ne<std::vector<int>>>("");
std::ignore = c.convert<ss::ne<std::vector<int>>>("");
REQUIRE_FALSE(c.valid());
{
@ -331,26 +647,58 @@ TEST_CASE("converter test ss:ne restriction (not empty)") {
}
}
TEST_CASE("converter test ss:ne restriction (not empty) with exceptions") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ne<std::string>>(""));
REQUIRE_EXCEPTION(std::ignore = c.convert<int, ss::ne<std::string>>("3,"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ne<std::string>, int>(",3"));
REQUIRE_EXCEPTION(std::ignore =
c.convert<void, ss::ne<std::string>, int>("junk,,3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::ne<std::vector<int>>>(""));
try {
{
auto tup = c.convert<ss::ne<std::string>>("s");
REQUIRE(c.valid());
CHECK_EQ(tup, "s");
}
{
auto tup =
c.convert<std::optional<int>, ss::ne<std::string>>("1,s");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple(1, "s"));
}
{
auto tup = c.convert<ss::ne<std::vector<int>>>("{1 2 3}");
REQUIRE(c.valid());
CHECK_EQ(tup, extracted_vector);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE(
"converter test ss:lt ss::lte ss::gt ss::gte restriction (in range)") {
ss::converter c;
c.convert<ss::lt<int, 3>>("3");
std::ignore = c.convert<ss::lt<int, 3>>("3");
REQUIRE_FALSE(c.valid());
c.convert<ss::lt<int, 2>>("3");
std::ignore = c.convert<ss::lt<int, 2>>("3");
REQUIRE_FALSE(c.valid());
c.convert<ss::gt<int, 3>>("3");
std::ignore = c.convert<ss::gt<int, 3>>("3");
REQUIRE_FALSE(c.valid());
c.convert<ss::gt<int, 4>>("3");
std::ignore = c.convert<ss::gt<int, 4>>("3");
REQUIRE_FALSE(c.valid());
c.convert<ss::lte<int, 2>>("3");
std::ignore = c.convert<ss::lte<int, 2>>("3");
REQUIRE_FALSE(c.valid());
c.convert<ss::gte<int, 4>>("3");
std::ignore = c.convert<ss::gte<int, 4>>("3");
REQUIRE_FALSE(c.valid());
{
@ -390,13 +738,70 @@ TEST_CASE(
}
}
TEST_CASE("converter test ss:lt ss::lte ss::gt ss::gte restriction (in range) "
"with exception") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::lt<int, 3>>("3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::lt<int, 2>>("3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::gt<int, 3>>("3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::gt<int, 4>>("3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::lte<int, 2>>("3"));
REQUIRE_EXCEPTION(std::ignore = c.convert<ss::gte<int, 4>>("3"));
try {
{
auto tup = c.convert<ss::lt<int, 4>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::gt<int, 2>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::lte<int, 4>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::lte<int, 3>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::gte<int, 2>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::gte<int, 3>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE("converter test error mode") {
ss::converter<ss::string_error> c;
c.convert<int>("junk");
std::ignore = c.convert<int>("junk");
CHECK_FALSE(c.valid());
CHECK_FALSE(c.error_msg().empty());
}
TEST_CASE("converter test throw on error mode") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(std::ignore = c.convert<int>("junk"));
}
TEST_CASE("converter test converter with quotes spacing and escaping") {
{
ss::converter c;
@ -444,6 +849,66 @@ TEST_CASE("converter test converter with quotes spacing and escaping") {
}
}
TEST_CASE("converter test converter with quotes spacing and escaping with "
"exceptions") {
try {
ss::converter<ss::throw_on_error> c;
auto tup = c.convert<std::string, std::string, std::string>(
R"("just","some","strings")");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple("\"just\"", "\"some\"", "\"strings\""));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>> c;
auto tup = c.convert<std::string, std::string, double, char>(
buff(R"("just",some,"12.3","a")"));
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple("just", "some", 12.3, 'a'));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::throw_on_error, ss::trim<' '>> c;
auto tup = c.convert<std::string, std::string, double, char>(
buff(R"( just , some , 12.3 ,a )"));
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple("just", "some", 12.3, 'a'));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::throw_on_error, ss::escape<'\\'>> c;
auto tup =
c.convert<std::string, std::string>(buff(R"(ju\,st,strings)"));
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple("ju,st", "strings"));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::throw_on_error, ss::escape<'\\'>, ss::trim<' '>,
ss::quote<'"'>>
c;
auto tup = c.convert<std::string, std::string, double, std::string>(
buff(R"( ju\,st , "so,me" , 12.34 , "str""ings")"));
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple("ju,st", "so,me", 12.34, "str\"ings"));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE("converter test invalid split conversions") {
ss::converter<ss::string_error, ss::escape<'\\'>, ss::trim<' '>,
ss::quote<'"'>>
@ -451,7 +916,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// mismatched quote
auto tup = c.convert<std::string, std::string, double, char>(
std::ignore = c.convert<std::string, std::string, double, char>(
buff(R"( "just , some , "12.3","a" )"));
CHECK_FALSE(c.valid());
CHECK_FALSE(c.unterminated_quote());
@ -460,7 +925,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated quote
auto tup = c.convert<std::string, std::string, double, std::string>(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"( ju\,st , "so,me" , 12.34 , "str""ings)"));
CHECK_FALSE(c.valid());
CHECK(c.unterminated_quote());
@ -469,7 +934,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated escape
auto tup = c.convert<std::string, std::string, double, std::string>(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"(just,some,2,strings\)"));
CHECK_FALSE(c.valid());
CHECK_FALSE(c.unterminated_quote());
@ -478,7 +943,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated escape while quoting
auto tup = c.convert<std::string, std::string, double, std::string>(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"(just,some,2,"strings\)"));
CHECK_FALSE(c.valid());
CHECK_FALSE(c.unterminated_quote());
@ -487,10 +952,46 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated escaped quote
auto tup = c.convert<std::string, std::string, double, std::string>(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"(just,some,2,"strings\")"));
CHECK_FALSE(c.valid());
CHECK(c.unterminated_quote());
CHECK_FALSE(c.error_msg().empty());
}
}
TEST_CASE("converter test invalid split conversions with exceptions") {
ss::converter<ss::escape<'\\'>, ss::trim<' '>, ss::quote<'"'>,
ss::throw_on_error>
c;
// mismatched quote
REQUIRE_EXCEPTION(std::ignore =
c.convert<std::string, std::string, double, char>(
buff(R"( "just , some , "12.3","a" )")));
CHECK_FALSE(c.unterminated_quote());
// unterminated quote
REQUIRE_EXCEPTION(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"( ju\,st , "so,me" , 12.34 , "str""ings)")));
CHECK(c.unterminated_quote());
// unterminated escape
REQUIRE_EXCEPTION(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"(just,some,2,strings\)")));
CHECK_FALSE(c.unterminated_quote());
// unterminated escape while quoting
REQUIRE_EXCEPTION(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"(just,some,2,"strings\)")));
CHECK_FALSE(c.unterminated_quote());
// unterminated escaped quote
REQUIRE_EXCEPTION(
std::ignore = c.convert<std::string, std::string, double, std::string>(
buff(R"(just,some,2,"strings\")")));
CHECK(c.unterminated_quote());
}

View File

@ -2,15 +2,41 @@
#include <algorithm>
#include <ss/extract.hpp>
namespace {
template <typename T>
struct numeric_limits : public std::numeric_limits<T> {};
template <typename T>
struct numeric_limits<ss::numeric_wrapper<T>> : public std::numeric_limits<T> {
};
template <typename T>
struct is_signed : public std::is_signed<T> {};
template <>
struct is_signed<ss::int8> : public std::true_type {};
template <typename T>
struct is_unsigned : public std::is_unsigned<T> {};
template <>
struct is_unsigned<ss::uint8> : public std::true_type {};
} /* anonymous namespace */
static_assert(is_signed<ss::int8>::value);
static_assert(is_unsigned<ss::uint8>::value);
TEST_CASE("testing extract functions for floating point values") {
CHECK_FLOATING_CONVERSION(123.456, float);
CHECK_FLOATING_CONVERSION(123.456, double);
CHECK_FLOATING_CONVERSION(69, float);
CHECK_FLOATING_CONVERSION(69, double);
CHECK_FLOATING_CONVERSION(59, float);
CHECK_FLOATING_CONVERSION(59, double);
CHECK_FLOATING_CONVERSION(420., float);
CHECK_FLOATING_CONVERSION(420., double);
CHECK_FLOATING_CONVERSION(4210., float);
CHECK_FLOATING_CONVERSION(4210., double);
CHECK_FLOATING_CONVERSION(0.123, float);
CHECK_FLOATING_CONVERSION(0.123, double);
@ -22,18 +48,18 @@ TEST_CASE("testing extract functions for floating point values") {
#define CHECK_DECIMAL_CONVERSION(input, type) \
{ \
std::string s = #input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
REQUIRE(t.has_value()); \
CHECK_EQ(t.value(), type(input)); \
type value; \
bool valid = ss::extract(s.c_str(), s.c_str() + s.size(), value); \
REQUIRE(valid); \
CHECK_EQ(value, type(input)); \
} \
{ \
/* check negative too */ \
if (std::is_signed_v<type>) { \
auto s = std::string("-") + #input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
REQUIRE(t.has_value()); \
CHECK_EQ(t.value(), type(-input)); \
} \
if (is_signed<type>::value) { \
std::string s = std::string("-") + #input; \
type value; \
bool valid = ss::extract(s.c_str(), s.c_str() + s.size(), value); \
REQUIRE(valid); \
CHECK_EQ(value, type(-input)); \
}
using us = unsigned short;
@ -43,6 +69,8 @@ using ll = long long;
using ull = unsigned long long;
TEST_CASE("extract test functions for decimal values") {
CHECK_DECIMAL_CONVERSION(12, ss::int8);
CHECK_DECIMAL_CONVERSION(12, ss::uint8);
CHECK_DECIMAL_CONVERSION(1234, short);
CHECK_DECIMAL_CONVERSION(1234, us);
CHECK_DECIMAL_CONVERSION(1234, int);
@ -54,6 +82,9 @@ TEST_CASE("extract test functions for decimal values") {
}
TEST_CASE("extract test functions for numbers with invalid inputs") {
// negative unsigned value for numeric_wrapper
CHECK_INVALID_CONVERSION("-12", ss::uint8);
// negative unsigned value
CHECK_INVALID_CONVERSION("-1234", ul);
@ -70,46 +101,38 @@ TEST_CASE("extract test functions for numbers with invalid inputs") {
CHECK_INVALID_CONVERSION("", int);
}
#define CHECK_OUT_OF_RANGE_CONVERSION(type) \
{ \
std::string s = std::to_string(std::numeric_limits<type>::max()); \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
CHECK(t.has_value()); \
for (auto& i : s) { \
if (i != '9' && i != '.') { \
i = '9'; \
break; \
} \
} \
t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
CHECK_FALSE(t.has_value()); \
} \
{ \
std::string s = std::to_string(std::numeric_limits<type>::min()); \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
CHECK(t.has_value()); \
for (auto& i : s) { \
if (std::is_signed_v<type> && i != '9' && i != '.') { \
i = '9'; \
break; \
} else if (std::is_unsigned_v<type>) { \
s = "-1"; \
break; \
} \
} \
t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
CHECK_FALSE(t.has_value()); \
TEST_CASE_TEMPLATE(
"extract test functions for numbers with out of range inputs", T, short, us,
int, ui, long, ul, ll, ull, ss::uint8) {
{
std::string s = std::to_string(numeric_limits<T>::max());
auto t = ss::to_num<T>(s.c_str(), s.c_str() + s.size());
CHECK(t.has_value());
for (auto& i : s) {
if (i != '9' && i != '.') {
i = '9';
break;
}
}
t = ss::to_num<T>(s.c_str(), s.c_str() + s.size());
CHECK_FALSE(t.has_value());
}
{
std::string s = std::to_string(numeric_limits<T>::min());
auto t = ss::to_num<T>(s.c_str(), s.c_str() + s.size());
CHECK(t.has_value());
for (auto& i : s) {
if (is_signed<T>::value && i != '9' && i != '.') {
i = '9';
break;
} else if (is_unsigned<T>::value) {
s = "-1";
break;
}
}
t = ss::to_num<T>(s.c_str(), s.c_str() + s.size());
CHECK_FALSE(t.has_value());
}
TEST_CASE("extract test functions for numbers with out of range inputs") {
CHECK_OUT_OF_RANGE_CONVERSION(short);
CHECK_OUT_OF_RANGE_CONVERSION(us);
CHECK_OUT_OF_RANGE_CONVERSION(int);
CHECK_OUT_OF_RANGE_CONVERSION(ui);
CHECK_OUT_OF_RANGE_CONVERSION(long);
CHECK_OUT_OF_RANGE_CONVERSION(ul);
CHECK_OUT_OF_RANGE_CONVERSION(ll);
CHECK_OUT_OF_RANGE_CONVERSION(ull);
}
TEST_CASE("extract test functions for boolean values") {
@ -142,12 +165,12 @@ TEST_CASE("extract test functions for char values") {
}
}
TEST_CASE("extract test functions for std::optional") {
for (const auto& [i, s] :
{std::pair<std::optional<int>, std::string>{1, "1"},
TEST_CASE_TEMPLATE("extract test functions for std::optional", T, int,
ss::int8) {
for (const auto& [i, s] : {std::pair<std::optional<T>, std::string>{1, "1"},
{69, "69"},
{-4, "-4"}}) {
std::optional<int> v;
std::optional<T> v;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), v));
REQUIRE(v.has_value());
CHECK_EQ(*v, i);
@ -164,7 +187,7 @@ TEST_CASE("extract test functions for std::optional") {
}
for (const std::string s : {"aa", "xxx", ""}) {
std::optional<int> v;
std::optional<T> v;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), v));
CHECK_FALSE(v.has_value());
}
@ -176,56 +199,57 @@ TEST_CASE("extract test functions for std::optional") {
}
}
TEST_CASE("extract test functions for std::variant") {
TEST_CASE_TEMPLATE("extract test functions for std::variant", T, int,
ss::uint8) {
{
std::string s = "22";
{
std::variant<int, double, std::string> var;
std::variant<T, double, std::string> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, double);
CHECK_NOT_VARIANT(var, std::string);
REQUIRE_VARIANT(var, 22, int);
REQUIRE_VARIANT(var, 22, T);
}
{
std::variant<double, int, std::string> var;
std::variant<double, T, std::string> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, std::string);
REQUIRE_VARIANT(var, 22, double);
}
{
std::variant<std::string, double, int> var;
std::variant<std::string, double, T> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "22", std::string);
}
{
std::variant<int> var;
std::variant<T> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, 22, int);
REQUIRE_VARIANT(var, 22, T);
}
}
{
std::string s = "22.2";
{
std::variant<int, double, std::string> var;
std::variant<T, double, std::string> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, std::string);
REQUIRE_VARIANT(var, 22.2, double);
}
{
std::variant<double, int, std::string> var;
std::variant<double, T, std::string> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, std::string);
REQUIRE_VARIANT(var, 22.2, double);
}
{
std::variant<std::string, double, int> var;
std::variant<std::string, double, T> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "22.2", std::string);
}
@ -233,45 +257,62 @@ TEST_CASE("extract test functions for std::variant") {
{
std::string s = "2.2.2";
{
std::variant<int, double, std::string> var;
std::variant<T, double, std::string> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "2.2.2", std::string);
}
{
std::variant<double, std::string, int> var;
std::variant<double, std::string, T> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "2.2.2", std::string);
}
{
std::variant<std::string, double, int> var;
std::variant<std::string, double, T> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "2.2.2", std::string);
}
{
std::variant<int, double> var;
std::variant<T, double> var;
REQUIRE_FALSE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, int{}, int);
REQUIRE_VARIANT(var, T{}, T);
CHECK_NOT_VARIANT(var, double);
}
{
std::variant<double, int> var;
std::variant<double, T> var;
REQUIRE_FALSE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, double{}, double);
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, T);
}
{
std::variant<int> var;
std::variant<T> var;
REQUIRE_FALSE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, int{}, int);
REQUIRE_VARIANT(var, T{}, T);
}
}
}
TEST_CASE("extract test with long number string") {
{
std::string string_num =
std::string(20, '1') + "." + std::string(20, '2');
CHECK_FLOATING_CONVERSION_LONG_NUMBER(string_num, float, stof);
CHECK_FLOATING_CONVERSION_LONG_NUMBER(string_num, double, stod);
}
{
std::string string_num =
std::string(50, '1') + "." + std::string(50, '2');
CHECK_FLOATING_CONVERSION_LONG_NUMBER(string_num, double, stod);
}
}

View File

@ -1,5 +1,4 @@
#include "test_helpers.hpp"
#include <algorithm>
#define SSP_DISABLE_FAST_FLOAT
#include <ss/extract.hpp>
@ -9,11 +8,11 @@ TEST_CASE(
CHECK_FLOATING_CONVERSION(123.456, float);
CHECK_FLOATING_CONVERSION(123.456, double);
CHECK_FLOATING_CONVERSION(69, float);
CHECK_FLOATING_CONVERSION(69, double);
CHECK_FLOATING_CONVERSION(59, float);
CHECK_FLOATING_CONVERSION(59, double);
CHECK_FLOATING_CONVERSION(420., float);
CHECK_FLOATING_CONVERSION(420., double);
CHECK_FLOATING_CONVERSION(4210., float);
CHECK_FLOATING_CONVERSION(4210., double);
CHECK_FLOATING_CONVERSION(0.123, float);
CHECK_FLOATING_CONVERSION(0.123, double);
@ -130,3 +129,20 @@ TEST_CASE("extract test functions for std::variant without fast float") {
}
}
}
TEST_CASE("extract test with long number string without fast float") {
{
std::string string_num =
std::string(20, '1') + "." + std::string(20, '2');
CHECK_FLOATING_CONVERSION_LONG_NUMBER(string_num, float, stof);
CHECK_FLOATING_CONVERSION_LONG_NUMBER(string_num, double, stod);
}
{
std::string string_num =
std::string(50, '1') + "." + std::string(50, '2');
CHECK_FLOATING_CONVERSION_LONG_NUMBER(string_num, double, stod);
}
}

View File

@ -1,6 +1,14 @@
#pragma once
#include <cstdlib>
#include <cstring>
#include <algorithm>
#include <ctime>
#include <filesystem>
#include <fstream>
#include <iomanip>
#include <ss/common.hpp>
#include <ss/setup.hpp>
#include <sstream>
#include <string>
#include <vector>
#ifdef CMAKE_GITHUB_CI
#include <doctest/doctest.h>
@ -8,45 +16,82 @@
#include <doctest.h>
#endif
namespace ss {
template <typename... Ts>
class parser;
} /* namespace ss */
namespace {
struct bool_error {};
template <typename T, typename U = bool_error>
struct config {
using BufferMode = T;
using ErrorMode = U;
constexpr static auto ThrowOnError = std::is_same_v<U, ss::throw_on_error>;
constexpr static auto StringError = std::is_same_v<U, ss::string_error>;
};
#define ParserOptionCombinations \
config<std::true_type>, config<std::true_type, ss::string_error>, \
config<std::true_type, ss::throw_on_error>, config<std::false_type>, \
config<std::false_type, ss::string_error>, \
config<std::false_type, ss::throw_on_error>
struct buffer {
char* data_{nullptr};
std::string data_;
char* operator()(const char* data) {
if (data_) {
delete[] data_;
}
data_ = new char[strlen(data) + 1];
strcpy(data_, data);
return data_;
char* operator()(const std::string& data) {
data_ = data;
return data_.data();
}
char* append(const char* data) {
if (data_) {
char* new_data_ = new char[strlen(data_) + strlen(data) + 1];
strcpy(new_data_, data_);
strcat(new_data_, data);
delete[] data_;
data_ = new_data_;
return data_;
} else {
return operator()(data);
}
char* append(const std::string& data) {
data_ += data;
return data_.data();
}
char* append_overwrite_last(const char* data, size_t size) {
data_[strlen(data_) - size] = '\0';
char* append_overwrite_last(const std::string& data, size_t size) {
data_.resize(data_.size() - size);
return append(data);
}
~buffer() {
if (data_) {
delete[] data_;
}
}
};
[[maybe_unused]] inline buffer buff;
[[maybe_unused]] std::string time_now_rand() {
std::srand(std::time(nullptr));
std::stringstream ss;
auto t = std::time(nullptr);
auto tm = *std::localtime(&t);
ss << std::put_time(&tm, "%d%m%Y%H%M%S");
std::srand(std::time(nullptr));
return ss.str() + std::to_string(rand());
}
struct unique_file_name {
static inline int i = 0;
std::string name;
unique_file_name(const std::string& test) {
do {
name = "ssp_test_" + test + "_" + std::to_string(i++) + "_" +
time_now_rand() + "_file.csv";
} while (std::filesystem::exists(name));
}
~unique_file_name() {
try {
std::filesystem::remove(name);
} catch (const std::filesystem::filesystem_error& e) {
std::cerr << e.what() << std::endl;
}
}
};
#define CHECK_FLOATING_CONVERSION(input, type) \
{ \
auto eps = std::numeric_limits<type>::min(); \
@ -64,6 +109,18 @@ struct buffer {
CHECK_LT(std::abs(t.value() - type(-input)), eps); \
}
#define CHECK_FLOATING_CONVERSION_LONG_NUMBER(STRING_NUMBER, TYPE, CONVERTER) \
{ \
auto begin = STRING_NUMBER.c_str(); \
auto end = begin + STRING_NUMBER.size(); \
\
auto number = ss::to_num<TYPE>(begin, end); \
REQUIRE(number.has_value()); \
\
auto expected_number = CONVERTER(STRING_NUMBER); \
CHECK_EQ(number.value(), expected_number); \
}
#define CHECK_INVALID_CONVERSION(input, type) \
{ \
std::string s = input; \
@ -79,3 +136,120 @@ struct buffer {
}
#define CHECK_NOT_VARIANT(var, type) CHECK(!std::holds_alternative<type>(var));
#define REQUIRE_EXCEPTION(...) \
try { \
__VA_ARGS__; \
FAIL("Expected exception"); \
} catch (ss::exception & e) { \
CHECK_FALSE(std::string{e.what()}.empty()); \
}
#define CHECK_EQ_ARRAY(first, second) \
{ \
const auto& first_ = (first); \
const auto& second_ = (second); \
CHECK_EQ(first_.size(), second_.size()); \
for (size_t i_ = 0; i_ < std::min(first_.size(), second_.size()); \
++i_) { \
CHECK_EQ(first_[i_], second_[i_]); \
} \
}
template <typename T>
[[maybe_unused]] std::vector<std::vector<T>> vector_combinations(
const std::vector<T>& v, size_t n) {
std::vector<std::vector<T>> ret;
if (n <= 1) {
for (const auto& i : v) {
ret.push_back({i});
}
return ret;
}
auto inner_combinations = vector_combinations(v, n - 1);
for (const auto& i : v) {
for (auto j : inner_combinations) {
j.insert(j.begin(), i);
ret.push_back(std::move(j));
}
}
return ret;
}
[[maybe_unused]] std::string merge_header(
const std::vector<std::string>& header,
const std::string& delimiter = ss::default_delimiter) {
std::string s;
if (!header.empty()) {
for (const auto& i : header) {
s.append(i);
s.append(delimiter);
}
for (size_t i = 0; i < delimiter.size(); ++i) {
s.pop_back();
}
}
return s;
};
[[maybe_unused]] std::string make_buffer(const std::string& file_name) {
std::ifstream in{file_name, std::ios::binary};
std::string tmp;
std::string out;
auto copy_if_whitespaces = [&] {
std::string matches = "\n\r\t ";
while (std::any_of(matches.begin(), matches.end(),
[&](auto c) { return in.peek() == c; })) {
if (in.peek() == '\r') {
out += "\r\n";
in.ignore(2);
} else {
out += std::string{static_cast<char>(in.peek())};
in.ignore(1);
}
}
};
// Evade small string optimization
out.reserve(sizeof(out) + 1);
copy_if_whitespaces();
while (in >> tmp) {
out += tmp;
copy_if_whitespaces();
}
return out;
}
template <bool buffer_mode, typename... Ts>
std::tuple<ss::parser<Ts...>, std::string> make_parser_impl(
const std::string& file_name, std::string delim = ss::default_delimiter) {
if (buffer_mode) {
auto buffer = make_buffer(file_name);
return {ss::parser<Ts...>{buffer.data(), buffer.size(), delim},
std::move(buffer)};
} else {
return {ss::parser<Ts...>{file_name, delim}, std::string{}};
}
}
template <bool buffer_mode, typename ErrorMode, typename... Ts>
[[maybe_unused]] std::enable_if_t<
!std::is_same_v<ErrorMode, bool_error>,
std::tuple<ss::parser<ErrorMode, Ts...>, std::string>>
make_parser(const std::string& file_name,
std::string delim = ss::default_delimiter) {
return make_parser_impl<buffer_mode, ErrorMode, Ts...>(file_name, delim);
}
template <bool buffer_mode, typename ErrorMode, typename... Ts>
[[maybe_unused]] std::enable_if_t<std::is_same_v<ErrorMode, bool_error>,
std::tuple<ss::parser<Ts...>, std::string>>
make_parser(const std::string& file_name,
std::string delim = ss::default_delimiter) {
return make_parser_impl<buffer_mode, Ts...>(file_name, delim);
}
} /* anonymous namespace */

File diff suppressed because it is too large Load Diff

116
test/test_parser1.hpp Normal file
View File

@ -0,0 +1,116 @@
#pragma once
#include "test_helpers.hpp"
#include <algorithm>
#include <filesystem>
#include <fstream>
#include <iomanip>
#include <iostream>
#include <ss/parser.hpp>
#include <sstream>
#include <unordered_map>
#include <unordered_set>
namespace {
#ifdef _WIN32
void replace_all(std::string& s, const std::string& from,
const std::string& to) {
if (from.empty()) return;
size_t start_pos = 0;
while ((start_pos = s.find(from, start_pos)) != std::string::npos) {
s.replace(start_pos, from.length(), to);
start_pos += to.length();
}
}
#endif
template <typename... Ts>
void expect_error_on_command(ss::parser<Ts...>& p,
const std::function<void()> command) {
if (ss::setup<Ts...>::throw_on_error) {
try {
command();
FAIL("expected exception");
} catch (const std::exception& e) {
CHECK_FALSE(std::string{e.what()}.empty());
}
} else {
command();
CHECK(!p.valid());
if constexpr (ss::setup<Ts...>::string_error) {
CHECK_FALSE(p.error_msg().empty());
}
}
}
[[maybe_unused]] void update_if_crlf(std::string& s) {
#ifdef _WIN32
replace_all(s, "\r\n", "\n");
#else
(void)(s);
#endif
}
struct X {
constexpr static auto delim = ",";
constexpr static auto empty = "_EMPTY_";
int i;
double d;
std::string s;
[[nodiscard]] std::string to_string() const {
if (s == empty) {
return "";
}
return std::to_string(i)
.append(delim)
.append(std::to_string(d))
.append(delim)
.append(s);
}
[[nodiscard]] auto tied() const {
return std::tie(i, d, s);
}
};
template <typename T>
[[nodiscard]] std::enable_if_t<ss::has_m_tied_t<T>, bool> operator==(
const T& lhs, const T& rhs) {
return lhs.tied() == rhs.tied();
}
template <typename T>
static void make_and_write(const std::string& file_name,
const std::vector<T>& data,
const std::vector<std::string>& header = {},
bool new_line_eof = true) {
std::ofstream out{file_name};
#ifdef _WIN32
std::vector<const char*> new_lines = {"\n"};
#else
std::vector<const char*> new_lines = {"\n", "\r\n"};
#endif
for (const auto& i : header) {
if (&i != &header.front()) {
out << T::delim;
}
out << i;
}
if (!header.empty()) {
out << new_lines.front();
}
for (size_t i = 0; i < data.size(); ++i) {
out << data[i].to_string();
if (new_line_eof || i + 1 < data.size()) {
out << new_lines[i % new_lines.size()];
}
}
}
} /* anonymous namespace */

675
test/test_parser1_1.cpp Normal file
View File

@ -0,0 +1,675 @@
#include "test_parser1.hpp"
TEST_CASE("test file not found") {
unique_file_name f{"file_not_found"};
{
ss::parser p{f.name, ","};
CHECK_FALSE(p.valid());
}
{
ss::parser<ss::string_error> p{f.name, ","};
CHECK_FALSE(p.valid());
CHECK_FALSE(p.error_msg().empty());
}
try {
ss::parser<ss::throw_on_error> p{f.name, ","};
FAIL("Expected exception...");
} catch (const std::exception& e) {
CHECK_FALSE(std::string{e.what()}.empty());
}
}
TEST_CASE("test null buffer") {
{
ss::parser p{nullptr, 10, ","};
CHECK_FALSE(p.valid());
}
{
ss::parser<ss::string_error> p{nullptr, 10, ","};
CHECK_FALSE(p.valid());
CHECK_FALSE(p.error_msg().empty());
}
try {
ss::parser<ss::throw_on_error> p{nullptr, 10, ","};
FAIL("Expected exception...");
} catch (const std::exception& e) {
CHECK_FALSE(std::string{e.what()}.empty());
}
}
struct Y {
constexpr static auto delim = ",";
std::string s1;
std::string s2;
std::string s3;
std::string to_string() const {
return std::string{}
.append(s1)
.append(delim)
.append(s2)
.append(delim)
.append(s3);
}
[[nodiscard]] auto tied() const {
return std::tie(s1, s2, s3);
}
};
TEST_CASE_TEMPLATE("test position method", T, ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"position_method"};
std::vector<Y> data = {{"1", "21", "x"}, {"321", "4", "y"},
{"54", "6", "zz"}, {"7", "876", "uuuu"},
{"910", "10", "v"}, {"10", "321", "ww"}};
make_and_write(f.name, data);
auto [p, buff] = make_parser<buffer_mode, ErrorMode>(f.name);
auto data_at = [&buff = buff, &f = f](auto n) {
if (!buff.empty()) {
return buff[n];
} else {
auto file = std::fopen(f.name.c_str(), "r");
std::fseek(file, n, SEEK_SET);
return static_cast<char>(std::fgetc(file));
}
};
while (!p.eof()) {
auto curr_char = p.position();
const auto& [s1, s2, s3] =
p.template get_next<std::string, std::string, std::string>();
auto s = s1 + "," + s2 + "," + s3;
for (size_t i = 0; i < s1.size(); ++i) {
CHECK_EQ(data_at(curr_char + i), s[i]);
}
auto last_char = data_at(curr_char + s.size());
CHECK((last_char == '\n' || last_char == '\r'));
}
}
TEST_CASE_TEMPLATE("test line method", T, ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"line_method"};
std::vector<Y> data = {{"1", "21", "x"}, {"321", "4", "y"},
{"54", "6", "zz"}, {"7", "876", "uuuu"},
{"910", "10", "v"}, {"10", "321", "ww"}};
make_and_write(f.name, data);
auto [p, buff] = make_parser<buffer_mode, ErrorMode>(f.name);
auto expected_line = 0;
CHECK_EQ(p.line(), expected_line);
while (!p.eof()) {
std::ignore =
p.template get_next<std::string, std::string, std::string>();
++expected_line;
CHECK_EQ(p.line(), expected_line);
}
CHECK_EQ(p.line(), data.size());
}
TEST_CASE_TEMPLATE("parser test various valid cases", T,
ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"various_valid_cases"};
std::vector<X> data = {{1, 2, "x"}, {3, 4, "y"}, {5, 6, "z"},
{7, 8, "u"}, {9, 10, "v"}, {11, 12, "w"}};
make_and_write(f.name, data);
auto csv_data_buffer = make_buffer(f.name);
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
ss::parser p0{std::move(p)};
p = std::move(p0);
std::vector<X> i;
auto [p2, __] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i2;
auto move_rotate = [&p = p, &p0 = p0] {
auto p1 = std::move(p);
p0 = std::move(p1);
p = std::move(p0);
};
while (!p.eof()) {
move_rotate();
auto a = p.template get_next<int, double, std::string>();
i.emplace_back(ss::to_object<X>(a));
}
for (const auto& a : p2.template iterate<int, double, std::string>()) {
i2.emplace_back(ss::to_object<X>(a));
}
CHECK_EQ(i, data);
CHECK_EQ(i2, data);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
auto [p2, __] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i2;
auto [p3, ___] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i3;
std::vector<X> expected = {std::begin(data) + 1, std::end(data)};
using tup = std::tuple<int, double, std::string>;
p.ignore_next();
while (!p.eof()) {
auto a = p.template get_next<tup>();
i.emplace_back(ss::to_object<X>(a));
}
p2.ignore_next();
for (const auto& a : p2.template iterate<tup>()) {
i2.emplace_back(ss::to_object<X>(a));
}
p3.ignore_next();
for (auto it = p3.template iterate<tup>().begin();
it != p3.template iterate<tup>().end(); ++it) {
i3.emplace_back(ss::to_object<X>(*it));
}
CHECK_EQ(i, expected);
CHECK_EQ(i2, expected);
CHECK_EQ(i3, expected);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
auto [p2, __] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i2;
while (!p.eof()) {
i.push_back(p.template get_object<X, int, double, std::string>());
}
for (auto&& a :
p2.template iterate_object<X, int, double, std::string>()) {
i2.push_back(std::move(a));
}
CHECK_EQ(i, data);
CHECK_EQ(i2, data);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
for (auto&& a :
p.template iterate_object<X, int, double, std::string>()) {
i.push_back(std::move(a));
}
CHECK_EQ(i, data);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
auto [p2, __] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i2;
using tup = std::tuple<int, double, std::string>;
while (!p.eof()) {
i.push_back(p.template get_object<X, tup>());
}
for (auto it = p2.template iterate_object<X, tup>().begin();
it != p2.template iterate_object<X, tup>().end(); it++) {
i2.push_back({it->i, it->d, it->s});
}
CHECK_EQ(i, data);
CHECK_EQ(i2, data);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
using tup = std::tuple<int, double, std::string>;
for (auto&& a : p.template iterate_object<X, tup>()) {
i.push_back(std::move(a));
}
CHECK_EQ(i, data);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
while (!p.eof()) {
i.push_back(p.template get_next<X>());
}
CHECK_EQ(i, data);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
for (auto&& a : p.template iterate<X>()) {
i.push_back(std::move(a));
}
CHECK_EQ(i, data);
}
{
constexpr int excluded = 3;
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
auto [p2, __] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i2;
while (!p.eof()) {
try {
auto a = p.template get_object<X, ss::ax<int, excluded>, double,
std::string>();
if (p.valid()) {
i.push_back(a);
}
} catch (...) {
// ignore
};
}
if (!T::ThrowOnError) {
for (auto&& a : p2.template iterate_object<X, ss::ax<int, excluded>,
double, std::string>()) {
if (p2.valid()) {
i2.push_back(std::move(a));
}
}
}
std::vector<X> expected;
for (auto& x : data) {
if (x.i != excluded) {
expected.push_back(x);
}
}
std::copy_if(data.begin(), data.end(), expected.begin(),
[&](const X& x) { return x.i != excluded; });
CHECK_EQ(i, expected);
if (!T::ThrowOnError) {
CHECK_EQ(i2, expected);
}
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
auto [p2, __] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i2;
while (!p.eof()) {
try {
auto a = p.template get_object<X, ss::nx<int, 3>, double,
std::string>();
if (p.valid()) {
i.push_back(a);
}
} catch (...) {
// ignore
}
}
if (!T::ThrowOnError) {
for (auto&& a : p2.template iterate_object<X, ss::nx<int, 3>,
double, std::string>()) {
if (p2.valid()) {
i2.push_back(std::move(a));
}
}
}
std::vector<X> expected = {{3, 4, "y"}};
CHECK_EQ(i, expected);
if (!T::ThrowOnError) {
CHECK_EQ(i2, expected);
}
}
{
unique_file_name empty_f{"various_valid_cases"};
std::vector<X> empty_data = {};
make_and_write(empty_f.name, empty_data);
auto [p, _] = make_parser<buffer_mode, ErrorMode>(empty_f.name, ",");
std::vector<X> i;
auto [p2, __] = make_parser<buffer_mode, ErrorMode>(empty_f.name, ",");
std::vector<X> i2;
while (!p.eof()) {
i.push_back(p.template get_next<X>());
}
for (auto&& a : p2.template iterate<X>()) {
i2.push_back(std::move(a));
}
CHECK(i.empty());
CHECK(i2.empty());
}
}
using test_tuple = std::tuple<double, char, double>;
struct test_struct {
int i;
double d;
char c;
auto tied() {
return std::tie(i, d, c);
}
};
static inline void expect_test_struct(const test_struct&) {
}
TEST_CASE_TEMPLATE("parser test composite conversion", BufferMode,
std::true_type, std::false_type) {
constexpr auto buffer_mode = BufferMode::value;
unique_file_name f{"composite_conversion"};
{
std::ofstream out{f.name};
for (auto& i :
{"10,a,11.1", "10,20,11.1", "junk", "10,11.1", "1,11.1,a", "junk",
"10,junk", "11,junk", "10,11.1,c", "10,20", "10,22.2,f"}) {
out << i << std::endl;
}
}
auto [p, _] = make_parser<buffer_mode, ss::string_error>(f.name, ",");
auto fail = [] { FAIL(""); };
auto expect_error = [](auto error) { CHECK(!error.empty()); };
auto ignore_error = [] {};
REQUIRE(p.valid());
REQUIRE_FALSE(p.eof());
{
constexpr static auto expectedData = std::tuple{10, 'a', 11.1};
auto [d1, d2, d3, d4] =
p.template try_next<int, int, double>(fail)
.template or_else<test_struct>(fail)
.template or_else<int, char, double>(
[&](auto&& data) { CHECK_EQ(data, expectedData); })
.on_error(fail)
.template or_else<test_tuple>(fail)
.values();
REQUIRE(p.valid());
REQUIRE_FALSE(d1);
REQUIRE_FALSE(d2);
REQUIRE(d3);
REQUIRE_FALSE(d4);
CHECK_EQ(*d3, expectedData);
}
{
REQUIRE(!p.eof());
constexpr static auto expectedData = std::tuple{10, 20, 11.1};
auto [d1, d2, d3, d4] =
p.template try_next<int, int, double>(
[&](auto& i1, auto i2, double d) {
CHECK_EQ(std::tie(i1, i2, d), expectedData);
})
.on_error(fail)
.template or_object<test_struct, int, double, char>(fail)
.on_error(fail)
.template or_else<test_tuple>(fail)
.on_error(fail)
.template or_else<int, char, double>(fail)
.values();
REQUIRE(p.valid());
REQUIRE(d1);
REQUIRE_FALSE(d2);
REQUIRE_FALSE(d3);
REQUIRE_FALSE(d4);
CHECK_EQ(*d1, expectedData);
}
{
REQUIRE(!p.eof());
auto [d1, d2, d3, d4, d5] =
p.template try_object<test_struct, int, double, char>(fail)
.on_error(expect_error)
.template or_else<int, char, char>(fail)
.template or_else<test_struct>(fail)
.template or_else<test_tuple>(fail)
.template or_else<int, char, double>(fail)
.values();
REQUIRE_FALSE(p.valid());
REQUIRE_FALSE(d1);
REQUIRE_FALSE(d2);
REQUIRE_FALSE(d3);
REQUIRE_FALSE(d4);
REQUIRE_FALSE(d5);
}
{
REQUIRE(!p.eof());
auto [d1, d2] =
p.template try_next<int, double>([](auto& i, auto& d) {
REQUIRE_EQ(std::tie(i, d), std::tuple{10, 11.1});
})
.template or_else<int, double>([](auto&, auto&) { FAIL(""); })
.values();
REQUIRE(p.valid());
REQUIRE(d1);
REQUIRE_FALSE(d2);
}
{
REQUIRE(!p.eof());
auto [d1, d2] =
p.template try_next<int, double>([](auto&, auto&) { FAIL(""); })
.template or_else<test_struct>(expect_test_struct)
.values();
REQUIRE(p.valid());
REQUIRE_FALSE(d1);
REQUIRE(d2);
CHECK_EQ(d2->tied(), std::tuple{1, 11.1, 'a'});
}
{
REQUIRE(!p.eof());
auto [d1, d2, d3, d4, d5] =
p.template try_next<int, int, double>(fail)
.template or_object<test_struct, int, double, char>()
.template or_else<test_struct>(expect_test_struct)
.template or_else<test_tuple>(fail)
.template or_else<std::tuple<int, double>>(fail)
.on_error(ignore_error)
.on_error(expect_error)
.values();
REQUIRE_FALSE(p.valid());
REQUIRE_FALSE(d1);
REQUIRE_FALSE(d2);
REQUIRE_FALSE(d3);
REQUIRE_FALSE(d4);
REQUIRE_FALSE(d5);
}
{
REQUIRE(!p.eof());
auto [d1, d2] =
p.template try_next<int, std::optional<int>>()
.on_error(ignore_error)
.on_error(fail)
.template or_else<std::tuple<int, std::string>>(fail)
.on_error(ignore_error)
.on_error(fail)
.on_error(ignore_error)
.values();
REQUIRE(p.valid());
REQUIRE(d1);
REQUIRE_FALSE(d2);
CHECK_EQ(*d1, std::tuple{10, std::nullopt});
}
{
REQUIRE_FALSE(p.eof());
auto [d1, d2] =
p.template try_next<int, std::variant<int, std::string>>()
.on_error(fail)
.template or_else<std::tuple<int, std::string>>(fail)
.on_error(fail)
.values();
REQUIRE(p.valid());
REQUIRE(d1);
REQUIRE_FALSE(d2);
CHECK_EQ(*d1, std::tuple{11, std::variant<int, std::string>{"junk"}});
}
{
REQUIRE(!p.eof());
auto [d1, d2] = p.template try_object<test_struct, int, double, char>()
.template or_else<int>(fail)
.values();
REQUIRE(p.valid());
REQUIRE(d1);
REQUIRE_FALSE(d2);
CHECK_EQ(d1->tied(), std::tuple{10, 11.1, 'c'});
}
{
REQUIRE_FALSE(p.eof());
auto [d1, d2, d3, d4] =
p.template try_next<int, int>([] { return false; })
.template or_else<int, double>([](auto&) { return false; })
.template or_else<int, int>()
.template or_else<int, int>(fail)
.values();
REQUIRE(p.valid());
REQUIRE_FALSE(d1);
REQUIRE_FALSE(d2);
REQUIRE(d3);
REQUIRE_FALSE(d4);
CHECK_EQ(d3.value(), std::tuple{10, 20});
}
{
REQUIRE(!p.eof());
auto [d1, d2, d3, d4] =
p.template try_object<test_struct, int, double, char>(
[] { return false; })
.template or_else<int, double>([](auto&) { return false; })
.template or_object<test_struct, int, double, char>()
.template or_else<int, int>(fail)
.values();
REQUIRE(p.valid());
REQUIRE_FALSE(d1);
REQUIRE_FALSE(d2);
REQUIRE(d3);
REQUIRE_FALSE(d4);
CHECK_EQ(d3->tied(), std::tuple{10, 22.2, 'f'});
}
CHECK(p.eof());
}
template <bool buffer_mode, typename... Ts>
void test_no_new_line_at_eof_impl(const std::vector<X>& data) {
unique_file_name f{"no_new_line_at_eof"};
make_and_write(f.name, data, {}, false);
auto [p, _] = make_parser<buffer_mode, Ts...>(f.name);
std::vector<X> parsed_data;
for (const auto& el : p.template iterate<X>()) {
parsed_data.push_back(el);
}
CHECK_EQ(data, parsed_data);
}
template <bool buffer_mode, typename... Ts>
void test_no_new_line_at_eof() {
test_no_new_line_at_eof_impl<buffer_mode, Ts...>({});
test_no_new_line_at_eof_impl<buffer_mode, Ts...>({{1, 2, "X"}});
test_no_new_line_at_eof_impl<buffer_mode, Ts...>({{1, 2, "X"}, {}});
test_no_new_line_at_eof_impl<buffer_mode, Ts...>(
{{1, 2, "X"}, {3, 4, "YY"}});
test_no_new_line_at_eof_impl<buffer_mode, Ts...>(
{{1, 2, "X"}, {3, 4, "YY"}, {}});
test_no_new_line_at_eof_impl<buffer_mode, Ts...>(
{{1, 2, "X"}, {3, 4, "YY"}, {5, 6, "ZZZ"}, {7, 8, "UUU"}});
for (size_t i = 0; i < 2 * ss::get_line_initial_buffer_size; ++i) {
test_no_new_line_at_eof_impl<buffer_mode, Ts...>(
{{1, 2, std::string(i, 'X')}});
for (size_t j = 0; j < 2 * ss::get_line_initial_buffer_size; j += 13) {
test_no_new_line_at_eof_impl<buffer_mode, Ts...>(
{{1, 2, std::string(i, 'X')}, {3, 4, std::string(j, 'Y')}});
test_no_new_line_at_eof_impl<buffer_mode, Ts...>(
{{1, 2, std::string(j, 'X')}, {3, 4, std::string(i, 'Y')}});
}
}
}
TEST_CASE_TEMPLATE("test no new line at end of data", T,
ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
test_no_new_line_at_eof<buffer_mode, ErrorMode>();
}

265
test/test_parser1_2.cpp Normal file
View File

@ -0,0 +1,265 @@
#include "test_parser1.hpp"
struct my_string {
char* data{nullptr};
my_string() = default;
~my_string() {
delete[] data;
}
// make sure no object is copied
my_string(const my_string&) = delete;
my_string& operator=(const my_string&) = delete;
my_string(my_string&& other) : data{other.data} {
other.data = nullptr;
}
my_string& operator=(my_string&& other) {
data = other.data;
return *this;
}
};
template <>
inline bool ss::extract(const char* begin, const char* end, my_string& s) {
size_t size = end - begin;
s.data = new char[size + 1];
strncpy(s.data, begin, size);
s.data[size] = '\0';
return true;
}
struct xyz {
my_string x;
my_string y;
my_string z;
auto tied() {
return std::tie(x, y, z);
}
};
TEST_CASE_TEMPLATE("test moving of parsed composite values", T,
config<std::true_type>, config<std::false_type>,
config<std::true_type, ss::string_error>,
config<std::false_type, ss::string_error>) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
// to compile is enough
return;
auto [p, _] = make_parser<buffer_mode, ErrorMode>("", "");
std::ignore =
p.template try_next<my_string, my_string, my_string>()
.template or_else<my_string, my_string, my_string, my_string>(
[](auto&&) {})
.template or_else<my_string>([](auto&) {})
.template or_else<xyz>([](auto&&) {})
.template or_object<xyz, my_string, my_string, my_string>(
[](auto&&) {})
.template or_else<std::tuple<my_string, my_string, my_string>>(
[](auto&, auto&, auto&) {});
}
TEST_CASE_TEMPLATE("parser test string error mode", BufferMode, std::true_type,
std::false_type) {
unique_file_name f{"string_error"};
{
std::ofstream out{f.name};
out << "junk" << std::endl;
out << "junk" << std::endl;
}
auto [p, _] = make_parser<BufferMode::value, ss::string_error>(f.name, ",");
REQUIRE_FALSE(p.eof());
std::ignore = p.template get_next<int>();
CHECK_FALSE(p.valid());
CHECK_FALSE(p.error_msg().empty());
}
TEST_CASE_TEMPLATE("parser throw on error mode", BufferMode, std::true_type,
std::false_type) {
unique_file_name f{"throw_on_error"};
{
std::ofstream out{f.name};
out << "junk" << std::endl;
out << "junk" << std::endl;
}
auto [p, _] =
make_parser<BufferMode::value, ss::throw_on_error>(f.name, ",");
REQUIRE_FALSE(p.eof());
try {
std::ignore = p.template get_next<int>();
FAIL("Expected exception...");
} catch (const std::exception& e) {
CHECK_FALSE(std::string{e.what()}.empty());
}
}
static inline std::string no_quote(const std::string& s) {
if (!s.empty() && s[0] == '"') {
return {std::next(begin(s)), std::prev(end(s))};
}
return s;
}
TEST_CASE_TEMPLATE("test quote multiline", T, ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"quote_multiline"};
std::vector<X> data = {{1, 2, "\"x\r\nx\nx\""},
{3, 4, "\"y\ny\r\ny\""},
{5, 6, "\"z\nz\""},
{7, 8, "\"u\"\"\""},
{9, 10, "v"},
{11, 12, "\"w\n\""}};
for (auto& [_, __, s] : data) {
update_if_crlf(s);
}
make_and_write(f.name, data);
for (auto& [_, __, s] : data) {
s = no_quote(s);
if (s[0] == 'u') {
s = "u\"";
}
}
auto [p, _] = make_parser<buffer_mode, ErrorMode, ss::multiline,
ss::quote<'"'>>(f.name, ",");
std::vector<X> i;
while (!p.eof()) {
auto a = p.template get_next<int, double, std::string>();
i.emplace_back(ss::to_object<X>(a));
}
for (auto& [_, __, s] : i) {
update_if_crlf(s);
}
CHECK_EQ(i, data);
auto [p_no_multiline, __] =
make_parser<buffer_mode, ErrorMode, ss::quote<'"'>>(f.name, ",");
while (!p.eof()) {
auto command = [&p_no_multiline = p_no_multiline] {
std::ignore =
p_no_multiline.template get_next<int, double, std::string>();
};
expect_error_on_command(p_no_multiline, command);
}
}
static inline std::string no_escape(std::string& s) {
s.erase(std::remove(begin(s), end(s), '\\'), end(s));
return s;
}
TEST_CASE_TEMPLATE("test escape multiline", T, ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"escape_multiline"};
std::vector<X> data = {{1, 2, "x\\\nx\\\r\nx"},
{5, 6, "z\\\nz\\\nz"},
{7, 8, "u"},
{3, 4, "y\\\ny\\\ny"},
{9, 10, "v\\\\"},
{11, 12, "w\\\n"}};
for (auto& [_, __, s] : data) {
update_if_crlf(s);
}
make_and_write(f.name, data);
for (auto& [_, __, s] : data) {
s = no_escape(s);
if (s == "v") {
s = "v\\";
}
}
auto [p, _] = make_parser<buffer_mode, ErrorMode, ss::multiline,
ss::escape<'\\'>>(f.name, ",");
std::vector<X> i;
while (!p.eof()) {
auto a = p.template get_next<int, double, std::string>();
i.emplace_back(ss::to_object<X>(a));
}
for (auto& [_, __, s] : i) {
update_if_crlf(s);
}
CHECK_EQ(i, data);
auto [p_no_multiline, __] =
make_parser<buffer_mode, ErrorMode, ss::escape<'\\'>>(f.name, ",");
while (!p.eof()) {
auto command = [&p_no_multiline = p_no_multiline] {
auto a =
p_no_multiline.template get_next<int, double, std::string>();
};
expect_error_on_command(p_no_multiline, command);
}
}
TEST_CASE_TEMPLATE("test quote escape multiline", T, ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"quote_escape_multiline"};
{
std::ofstream out{f.name};
out << "1,2,\"just\\\n\nstrings\"" << std::endl;
#ifndef _WIN32
out << "3,4,\"just\r\nsome\\\r\n\n\\\nstrings\"" << std::endl;
out << "5,6,\"just\\\n\\\r\n\r\n\nstrings" << std::endl;
#else
out << "3,4,\"just\nsome\\\n\n\\\nstrings\"" << std::endl;
out << "5,6,\"just\\\n\\\n\n\nstrings" << std::endl;
#endif
out << "7,8,\"just strings\"" << std::endl;
out << "9,10,just strings" << std::endl;
}
size_t bad_lines = 1;
auto num_errors = 0;
auto [p, _] = make_parser<buffer_mode, ErrorMode, ss::multiline,
ss::escape<'\\'>, ss::quote<'"'>>(f.name);
std::vector<X> i;
while (!p.eof()) {
try {
auto a = p.template get_next<int, double, std::string>();
if (p.valid()) {
i.emplace_back(ss::to_object<X>(a));
} else {
++num_errors;
}
} catch (const std::exception& e) {
++num_errors;
}
}
CHECK(bad_lines == num_errors);
std::vector<X> data = {{1, 2, "just\n\nstrings"},
#ifndef _WIN32
{3, 4, "just\r\nsome\r\n\n\nstrings"},
#else
{3, 4, "just\nsome\n\n\nstrings"},
#endif
{9, 10, "just strings"}};
for (auto& [_, __, s] : i) {
update_if_crlf(s);
}
CHECK_EQ(i, data);
}

314
test/test_parser1_3.cpp Normal file
View File

@ -0,0 +1,314 @@
#include "test_parser1.hpp"
TEST_CASE_TEMPLATE("test multiline restricted", T, ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"multiline_restricted"};
{
std::ofstream out{f.name};
out << "1,2,\"just\n\nstrings\"" << std::endl;
#ifndef _WIN32
out << "3,4,\"ju\n\r\n\nnk\"" << std::endl;
out << "5,6,just\\\n\\\r\nstrings" << std::endl;
#else
out << "3,4,\"ju\n\n\nnk\"" << std::endl;
out << "5,6,just\\\n\\\nstrings" << std::endl;
#endif
out << "7,8,ju\\\n\\\n\\\nnk" << std::endl;
out << "99,100,\"\n\n\n\n" << std::endl;
out << "9,10,\"just\\\n\nstrings\"" << std::endl;
out << "11,12,\"ju\\\n|\n\n\n\n\nk\"" << std::endl;
out << "13,14,\"ju\\\n\\\n15,16\"\\\n\\\\\n\nnk\"" << std::endl;
out << "17,18,\"ju\\\n\\\n\\\n\\\\\n\nnk\"" << std::endl;
out << "19,20,just strings" << std::endl;
}
auto bad_lines = 20;
auto num_errors = 0;
auto [p, _] =
make_parser<buffer_mode, ErrorMode, ss::multiline_restricted<2>,
ss::quote<'"'>, ss::escape<'\\'>>(f.name, ",");
std::vector<X> i;
while (!p.eof()) {
try {
auto a = p.template get_next<int, double, std::string>();
if (p.valid()) {
i.emplace_back(ss::to_object<X>(a));
} else {
++num_errors;
}
} catch (const std::exception& e) {
++num_errors;
}
}
CHECK(bad_lines == num_errors);
std::vector<X> data = {{1, 2, "just\n\nstrings"},
#ifndef _WIN32
{5, 6, "just\n\r\nstrings"},
#else
{5, 6, "just\n\nstrings"},
#endif
{9, 10, "just\n\nstrings"},
{19, 20, "just strings"}};
for (auto& [_, __, s] : i) {
update_if_crlf(s);
}
if (i.size() != data.size()) {
CHECK_EQ(i.size(), data.size());
}
CHECK_EQ(i, data);
}
template <typename T, typename... Ts>
void test_unterminated_line(const std::vector<std::string>& lines,
size_t bad_line) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"unterminated_line"};
std::ofstream out{f.name};
for (const auto& line : lines) {
out << line << std::endl;
}
out.close();
auto [p, _] = make_parser<buffer_mode, ErrorMode, Ts...>(f.name);
size_t line = 0;
while (!p.eof()) {
auto command = [&p = p] {
std::ignore = p.template get_next<int, double, std::string>();
};
if (line == bad_line) {
expect_error_on_command(p, command);
break;
} else {
command();
CHECK(p.valid());
++line;
}
}
}
TEST_CASE_TEMPLATE("parser test csv on multiline with errors", T,
ParserOptionCombinations) {
using multiline = ss::multiline_restricted<3>;
using escape = ss::escape<'\\'>;
using quote = ss::quote<'"'>;
// unterminated escape
{
const std::vector<std::string> lines{"1,2,just\\"};
test_unterminated_line<T, multiline, escape>(lines, 0);
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"1,2,just\\", "9,8,second"};
test_unterminated_line<T, multiline, escape>(lines, 0);
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"9,8,first", "1,2,just\\"};
test_unterminated_line<T, multiline, escape>(lines, 1);
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,first", "1,2,just\\",
"3,4,third"};
test_unterminated_line<T, multiline, escape>(lines, 1);
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,first",
"1,2,just\\\nstrings\\",
"3,4,th\\\nird"};
test_unterminated_line<T, multiline, escape>(lines, 1);
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,first", "3,4,second",
"1,2,just\\"};
test_unterminated_line<T, multiline, escape>(lines, 2);
test_unterminated_line<T, multiline, escape, quote>(lines, 2);
}
{
const std::vector<std::string> lines{"9,8,\\first", "3,4,second",
"1,2,jus\\t\\"};
test_unterminated_line<T, multiline, escape>(lines, 2);
test_unterminated_line<T, multiline, escape, quote>(lines, 2);
}
// unterminated quote
{
const std::vector<std::string> lines{"1,2,\"just"};
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
test_unterminated_line<T, multiline, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"1,2,\"just", "9,8,second"};
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
test_unterminated_line<T, multiline, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"9,8,first", "1,2,\"just"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
test_unterminated_line<T, multiline, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,first", "1,2,\"just",
"3,4,th\\,ird"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
test_unterminated_line<T, multiline, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,first", "3,4,second",
"1,2,\"just"};
test_unterminated_line<T, multiline, escape, quote>(lines, 2);
test_unterminated_line<T, multiline, quote>(lines, 2);
}
{
const std::vector<std::string> lines{"9,8,\"first\"",
"\"3\",4,\"sec,ond\"",
"1,2,\"ju\"\"st"};
test_unterminated_line<T, multiline, escape, quote>(lines, 2);
test_unterminated_line<T, multiline, quote>(lines, 2);
}
// unterminated quote and escape
{
const std::vector<std::string> lines{"1,2,\"just\\"};
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"1,2,\"just\\\n\\"};
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"1,2,\"just\n\\"};
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"9,8,first", "1,2,\"just\n\\"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,first", "1,2,\"just\n\\",
"4,3,thrid"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,f\\\nirst", "1,2,\"just\n\\",
"4,3,thrid"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,\"f\ni\nrst\"",
"1,2,\"just\n\\", "4,3,thrid"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
// multiline limmit reached escape
{
const std::vector<std::string> lines{"1,2,\\\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape>(lines, 0);
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"9,8,first",
"1,2,\\\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape>(lines, 1);
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,fi\\\nrs\\\nt",
"1,2,\\\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape>(lines, 1);
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,first",
"1,2,\\\n\\\n\\\n\\\njust",
"4,3,third"};
test_unterminated_line<T, multiline, escape>(lines, 1);
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
// multiline limmit reached quote
{
const std::vector<std::string> lines{"1,2,\"\n\n\n\n\njust\""};
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
test_unterminated_line<T, multiline, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"9,8,first",
"1,2,\"\n\n\n\n\njust\""};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
test_unterminated_line<T, multiline, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,\"fir\nst\"",
"1,2,\"\n\n\n\n\njust\""};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
test_unterminated_line<T, multiline, quote>(lines, 1);
}
// multiline limmit reached quote and escape
{
const std::vector<std::string> lines{"1,2,\"\\\n\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape, quote>(lines, 0);
}
{
const std::vector<std::string> lines{"9,8,first",
"1,2,\"\\\n\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,fi\\\nrst",
"1,2,\"\\\n\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,\"fi\nrst\"",
"1,2,\"\\\n\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
{
const std::vector<std::string> lines{"9,8,\"fi\nr\\\nst\"",
"1,2,\"\\\n\n\\\n\\\n\\\njust"};
test_unterminated_line<T, multiline, escape, quote>(lines, 1);
}
}

400
test/test_parser1_4.cpp Normal file
View File

@ -0,0 +1,400 @@
#include "test_parser1.hpp"
template <typename T, typename Tuple>
struct has_type;
template <typename T, typename... Us>
struct has_type<T, std::tuple<Us...>>
: std::disjunction<std::is_same<T, Us>...> {};
template <typename T, typename... Ts>
static void test_fields(const std::string file_name, const std::vector<X>& data,
const std::vector<std::string>& header,
const std::vector<std::string>& fields) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
using CaseType = std::tuple<Ts...>;
auto [p, _] = make_parser<buffer_mode, ErrorMode>(file_name, ",");
CHECK_FALSE(p.field_exists("Unknown"));
p.use_fields(fields);
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
std::vector<CaseType> i;
for (const auto& a : p.template iterate<CaseType>()) {
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
i.push_back(a);
}
CHECK_EQ(i.size(), data.size());
for (size_t j = 0; j < i.size(); ++j) {
if constexpr (has_type<int, CaseType>::value) {
CHECK_EQ(std::get<int>(i[j]), data[j].i);
}
if constexpr (has_type<double, CaseType>::value) {
CHECK_EQ(std::get<double>(i[j]), data[j].d);
}
if constexpr (has_type<std::string, CaseType>::value) {
CHECK_EQ(std::get<std::string>(i[j]), data[j].s);
}
}
}
TEST_CASE_TEMPLATE("test various cases with header", T,
ParserOptionCombinations) {
unique_file_name f{"various_cases_with_header"};
using str = std::string;
constexpr static auto Int = "Int";
constexpr static auto Dbl = "Double";
constexpr static auto Str = "String";
const std::vector<std::string> header{Int, Dbl, Str};
std::vector<X> data = {{1, 2, "x"}, {3, 4, "y"}, {5, 6, "z"},
{7, 8, "u"}, {9, 10, "v"}, {11, 12, "w"}};
make_and_write(f.name, data, header);
const auto& o = f.name;
const auto& d = data;
{
ss::parser<ss::string_error> p{f.name, ","};
std::vector<X> i;
for (const auto& a : p.iterate<int, double, std::string>()) {
CHECK_EQ(header, p.header());
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
i.emplace_back(ss::to_object<X>(a));
}
CHECK_NE(i, data);
}
{
ss::parser<ss::string_error> p{f.name, ","};
std::vector<X> i;
p.ignore_next();
for (const auto& a : p.iterate<int, double, std::string>()) {
CHECK_EQ(header, p.header());
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
i.emplace_back(ss::to_object<X>(a));
}
CHECK_EQ(i, data);
}
{
ss::parser<ss::string_error> p{f.name, ","};
CHECK_EQ(header, p.header());
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
p.use_fields(Int, Dbl);
CHECK_EQ(header, p.header());
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
{
auto [int_, double_] = p.get_next<int, double>();
CHECK_EQ(int_, data[0].i);
CHECK_EQ(double_, data[0].d);
}
p.use_fields(Dbl, Int);
CHECK_EQ(header, p.header());
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
{
auto [double_, int_] = p.get_next<double, int>();
CHECK_EQ(int_, data[1].i);
CHECK_EQ(double_, data[1].d);
}
p.use_fields(Str);
{
auto string_ = p.get_next<std::string>();
CHECK_EQ(string_, data[2].s);
}
p.use_fields(Str, Int, Dbl);
{
auto [string_, int_, double_] =
p.get_next<std::string, int, double>();
CHECK_EQ(double_, data[3].d);
CHECK_EQ(int_, data[3].i);
CHECK_EQ(string_, data[3].s);
}
}
/* python used to generate permutations
import itertools
header = {'str': 'Str',
'double': 'Dbl',
'int': 'Int'}
keys = ['str', 'int', 'double']
for r in range (1, 3):
combinations = list(itertools.permutations(keys, r = r))
for combination in combinations:
template_params = []
arg_params = []
for type in combination:
template_params.append(type)
arg_params.append(header[type])
call = 'testFields<' + ', '.join(template_params) + \
'>(o, d, header, {' + ', '.join(arg_params) + '});'
print(call)
*/
test_fields<T, str>(o, d, header, {Str});
test_fields<T, int>(o, d, header, {Int});
test_fields<T, double>(o, d, header, {Dbl});
test_fields<T, str, int>(o, d, header, {Str, Int});
test_fields<T, str, double>(o, d, header, {Str, Dbl});
test_fields<T, int, str>(o, d, header, {Int, Str});
test_fields<T, int, double>(o, d, header, {Int, Dbl});
test_fields<T, double, str>(o, d, header, {Dbl, Str});
test_fields<T, double, int>(o, d, header, {Dbl, Int});
test_fields<T, str, int, double>(o, d, header, {Str, Int, Dbl});
test_fields<T, str, double, int>(o, d, header, {Str, Dbl, Int});
test_fields<T, int, str, double>(o, d, header, {Int, Str, Dbl});
test_fields<T, int, double, str>(o, d, header, {Int, Dbl, Str});
test_fields<T, double, str, int>(o, d, header, {Dbl, Str, Int});
test_fields<T, double, int, str>(o, d, header, {Dbl, Int, Str});
}
template <typename T>
void test_invalid_fields(const std::vector<std::string>& lines,
const std::vector<std::string>& fields) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
auto check_header = [&lines](auto& p) {
if (lines.empty()) {
CHECK_EQ(p.header().size(), 1);
CHECK_EQ(p.header().at(0), "");
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
} else {
CHECK_EQ(lines[0], merge_header(p.header()));
CHECK_EQ(merge_header(p.header(), ","), p.raw_header());
}
CHECK(p.valid());
};
unique_file_name f{"invalid_fields"};
{
std::ofstream out{f.name};
for (const auto& line : lines) {
out << line << std::endl;
}
}
{
// No fields specified
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
auto command = [&p = p] { p.use_fields(); };
expect_error_on_command(p, command);
check_header(p);
}
{
// Unknown field
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
auto command = [&p = p] { p.use_fields("Unknown"); };
expect_error_on_command(p, command);
check_header(p);
}
{
// Field used multiple times
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
auto command = [&p = p, &fields = fields] {
p.use_fields(fields.at(0), fields.at(0));
};
if (!fields.empty()) {
expect_error_on_command(p, command);
}
check_header(p);
}
{
// Mapping out of range
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
auto command = [&p = p, &fields = fields] {
p.use_fields(fields.at(0));
std::ignore = p.template get_next<std::string, std::string>();
};
check_header(p);
if (!fields.empty()) {
expect_error_on_command(p, command);
}
check_header(p);
}
{
// Invalid header
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
auto command = [&p = p, &fields = fields] { p.use_fields(fields); };
check_header(p);
if (!fields.empty()) {
// Pass if there are no duplicates, fail otherwise
if (std::unordered_set<std::string>{fields.begin(), fields.end()}
.size() != fields.size()) {
expect_error_on_command(p, command);
} else {
command();
CHECK(p.valid());
if (!p.valid()) {
if constexpr (T::StringError) {
std::cout << p.error_msg() << std::endl;
}
}
}
}
check_header(p);
}
}
TEST_CASE_TEMPLATE("test invalid header fields usage", T,
ParserOptionCombinations) {
test_invalid_fields<T>({}, {});
test_invalid_fields<T>({"Int"}, {"Int"});
test_invalid_fields<T>({"Int", "1"}, {"Int"});
test_invalid_fields<T>({"Int", "1", "2"}, {"Int"});
test_invalid_fields<T>({"Int,String"}, {"Int", "String"});
test_invalid_fields<T>({"Int,String", "1,hi"}, {"Int", "String"});
test_invalid_fields<T>({"Int,String", "2,hello"}, {"Int", "String"});
test_invalid_fields<T>({"Int,String,Double"}, {"Int", "String", "Double"});
test_invalid_fields<T>({"Int,String,Double", "1,hi,2.34"},
{"Int", "String", "Double"});
test_invalid_fields<T>({"Int,String,Double", "1,hi,2.34", "2,hello,3.45"},
{"Int", "String", "Double"});
test_invalid_fields<T>({"Int,Int,Int"}, {"Int", "Int", "Int"});
test_invalid_fields<T>({"Int,Int,Int", "1,2,3"}, {"Int", "Int", "Int"});
test_invalid_fields<T>({"Int,String,Int"}, {"Int", "String", "Int"});
test_invalid_fields<T>({"Int,String,Int", "1,hi,3"},
{"Int", "String", "Int"});
}
TEST_CASE_TEMPLATE("test invalid rows with header", T,
ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"invalid_rows_with_header"};
{
std::ofstream out{f.name};
out << "Int,String,Double" << std::endl;
out << "1,line1,2.34" << std::endl;
out << "2,line2" << std::endl;
out << "3,line3,67.8" << std::endl;
out << "4,line4,67.8,9" << std::endl;
out << "5,line5,9.10" << std::endl;
out << "six,line6,10.11" << std::endl;
}
std::vector<std::string> header = {"Int", "String", "Double"};
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header()), p.raw_header());
p.use_fields("Int", "String", "Double");
using data = std::tuple<int, std::string, double>;
std::vector<data> i;
CHECK(p.valid());
while (!p.eof()) {
try {
const auto& t = p.template get_next<data>();
if (p.valid()) {
i.push_back(t);
}
} catch (const ss::exception&) {
continue;
}
}
std::vector<data> expected = {{1, "line1", 2.34},
{3, "line3", 67.8},
{5, "line5", 9.10}};
CHECK_EQ(i, expected);
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header()), p.raw_header());
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header()), p.raw_header());
p.use_fields("Double", "Int");
using data = std::tuple<double, int>;
std::vector<data> i;
CHECK(p.valid());
while (!p.eof()) {
try {
const auto& t = p.template get_next<data>();
if (p.valid()) {
i.push_back(t);
}
} catch (const ss::exception&) {
continue;
}
}
std::vector<data> expected = {{2.34, 1}, {67.8, 3}, {9.10, 5}};
CHECK_EQ(i, expected);
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header()), p.raw_header());
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header()), p.raw_header());
p.use_fields("String", "Double");
using data = std::tuple<std::string, double>;
std::vector<data> i;
CHECK(p.valid());
while (!p.eof()) {
try {
const auto& t = p.template get_next<data>();
if (p.valid()) {
i.push_back(t);
}
} catch (const ss::exception&) {
continue;
}
}
std::vector<data> expected = {{"line1", 2.34},
{"line3", 67.8},
{"line5", 9.10},
{"line6", 10.11}};
CHECK_EQ(i, expected);
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(merge_header(p.header()), p.raw_header());
}
}

301
test/test_parser1_5.cpp Normal file
View File

@ -0,0 +1,301 @@
#include "test_parser1.hpp"
TEST_CASE_TEMPLATE("test empty fields header", T, ParserOptionCombinations) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"empty_fields_header"};
// Empty header
{
std::ofstream out{f.name};
out << "" << std::endl;
out << "1" << std::endl;
}
{
std::vector<std::string> expected_header = {""};
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(expected_header, p.header());
CHECK_EQ("", p.raw_header());
CHECK(p.valid());
}
// All empty header fields
{
std::ofstream out{f.name};
out << ",," << std::endl;
out << "1,2,3" << std::endl;
}
{
std::vector<std::string> expected_header = {"", "", ""};
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(expected_header, p.header());
CHECK_EQ(",,", p.raw_header());
CHECK(p.valid());
auto command1 = [&p = p] { std::ignore = p.field_exists("Int"); };
expect_error_on_command(p, command1);
auto command2 = [&p = p] { p.use_fields("Int"); };
expect_error_on_command(p, command2);
}
// One empty field
const std::vector<std::string> valid_fields = {"Int0", "Int1", ""};
using svec = std::vector<std::string>;
const std::vector<std::vector<std::string>> valid_field_combinations =
{svec{"Int0"},
svec{"Int1"},
svec{""},
svec{"", "Int0"},
svec{"Int0", "Int1"},
svec{"Int1", ""},
svec{"Int0", "", "Int1"},
svec{"", "Int1", "Int0"}};
// Last header field empty
{
std::ofstream out{f.name};
out << "Int0,Int1," << std::endl;
out << "1,2,3" << std::endl;
}
{
std::vector<std::string> expected_header = {"Int0", "Int1", ""};
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(expected_header, p.header());
CHECK_EQ("Int0,Int1,", p.raw_header());
CHECK(p.valid());
for (const auto& field : valid_fields) {
CHECK(p.field_exists(field));
CHECK(p.valid());
}
for (const auto& fields : valid_field_combinations) {
p.use_fields(fields);
CHECK(p.valid());
}
}
// First header field empty
{
std::ofstream out{f.name};
out << ",Int0,Int1" << std::endl;
out << "1,2,3" << std::endl;
}
{
std::vector<std::string> expected_header = {"", "Int0", "Int1"};
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(expected_header, p.header());
CHECK_EQ(",Int0,Int1", p.raw_header());
CHECK(p.valid());
for (const auto& field : valid_fields) {
CHECK(p.field_exists(field));
CHECK(p.valid());
}
for (const auto& fields : valid_field_combinations) {
p.use_fields(fields);
CHECK(p.valid());
}
}
// Middle header field empty
{
std::ofstream out{f.name};
out << "Int0,,Int1" << std::endl;
out << "1,2,3" << std::endl;
}
{
std::vector<std::string> expected_header = {"Int0", "", "Int1"};
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name);
CHECK_EQ_ARRAY(expected_header, p.header());
CHECK_EQ("Int0,,Int1", p.raw_header());
CHECK(p.valid());
for (const auto& field : valid_fields) {
CHECK(p.field_exists(field));
CHECK(p.valid());
}
for (const auto& fields : valid_field_combinations) {
p.use_fields(fields);
CHECK(p.valid());
}
}
}
template <typename T, typename... Ts>
void test_unterminated_quote_header() {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"unterminated_quote_header"};
{
std::ofstream out{f.name};
out << "\"Int" << std::endl;
out << "1" << std::endl;
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode, Ts...>(f.name);
auto command0 = [&p = p] { std::ignore = p.header(); };
expect_error_on_command(p, command0);
CHECK_EQ(p.raw_header(), "\"Int");
auto command1 = [&p = p] { std::ignore = p.field_exists("Int"); };
expect_error_on_command(p, command1);
auto command2 = [&p = p] { p.use_fields("Int"); };
expect_error_on_command(p, command2);
}
}
TEST_CASE_TEMPLATE("test unterminated quote header", T,
ParserOptionCombinations) {
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
test_unterminated_quote_header<T, quote>();
test_unterminated_quote_header<T, quote, ss::multiline>();
test_unterminated_quote_header<T, quote, escape>();
test_unterminated_quote_header<T, quote, escape, ss::multiline>();
}
template <typename T, typename... Ts>
void test_unterminated_escape_header() {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"unterminated_escape_header"};
// Unterminated escape in header
{
std::ofstream out{f.name};
out << "Int\\" << std::endl;
out << "1" << std::endl;
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode, Ts...>(f.name);
auto command0 = [&p = p] { std::ignore = p.header(); };
expect_error_on_command(p, command0);
CHECK_EQ(p.raw_header(), "Int\\");
auto command1 = [&p = p] { std::ignore = p.field_exists("Int"); };
expect_error_on_command(p, command1);
auto command2 = [&p = p] { p.use_fields("Int"); };
expect_error_on_command(p, command2);
}
}
TEST_CASE_TEMPLATE("test unterminated escape header", T,
ParserOptionCombinations) {
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
test_unterminated_escape_header<T, escape>();
test_unterminated_escape_header<T, escape, ss::multiline>();
test_unterminated_escape_header<T, escape, quote>();
test_unterminated_escape_header<T, escape, quote, ss::multiline>();
}
template <typename T>
void test_ignore_empty(const std::vector<X>& data) {
constexpr auto buffer_mode = T::BufferMode::value;
using ErrorMode = typename T::ErrorMode;
unique_file_name f{"ignore_empty"};
make_and_write(f.name, data);
std::vector<X> expected;
for (const auto& d : data) {
if (d.s != X::empty) {
expected.push_back(d);
}
}
{
auto [p, _] =
make_parser<buffer_mode, ErrorMode, ss::ignore_empty>(f.name, ",");
std::vector<X> i;
for (const auto& a : p.template iterate<X>()) {
i.push_back(a);
}
CHECK_EQ(i, expected);
}
{
auto [p, _] = make_parser<buffer_mode, ErrorMode>(f.name, ",");
std::vector<X> i;
size_t n = 0;
while (!p.eof()) {
try {
++n;
const auto& a = p.template get_next<X>();
if (data.at(n - 1).s == X::empty) {
CHECK_FALSE(p.valid());
continue;
}
i.push_back(a);
} catch (...) {
CHECK_EQ(data.at(n - 1).s, X::empty);
}
}
CHECK_EQ(i, expected);
}
}
TEST_CASE_TEMPLATE("test various cases with empty lines", T,
ParserOptionCombinations) {
test_ignore_empty<T>(
{{1, 2, "x"}, {3, 4, "y"}, {9, 10, "v"}, {11, 12, "w"}});
test_ignore_empty<T>(
{{1, 2, X::empty}, {3, 4, "y"}, {9, 10, "v"}, {11, 12, "w"}});
test_ignore_empty<T>(
{{1, 2, "x"}, {3, 4, "y"}, {9, 10, "v"}, {11, 12, X::empty}});
test_ignore_empty<T>(
{{1, 2, "x"}, {5, 6, X::empty}, {9, 10, "v"}, {11, 12, "w"}});
test_ignore_empty<T>(
{{1, 2, X::empty}, {5, 6, X::empty}, {9, 10, "v"}, {11, 12, "w"}});
test_ignore_empty<T>(
{{1, 2, X::empty}, {3, 4, "y"}, {9, 10, "v"}, {11, 12, X::empty}});
test_ignore_empty<T>(
{{1, 2, "x"}, {3, 4, "y"}, {9, 10, X::empty}, {11, 12, X::empty}});
test_ignore_empty<T>(
{{1, 2, X::empty}, {3, 4, "y"}, {9, 10, X::empty}, {11, 12, X::empty}});
test_ignore_empty<T>({{1, 2, X::empty},
{3, 4, X::empty},
{9, 10, X::empty},
{11, 12, X::empty}});
test_ignore_empty<T>(
{{1, 2, "x"}, {3, 4, X::empty}, {9, 10, X::empty}, {11, 12, X::empty}});
test_ignore_empty<T>(
{{1, 2, X::empty}, {3, 4, X::empty}, {9, 10, X::empty}, {11, 12, "w"}});
test_ignore_empty<T>({{11, 12, X::empty}});
test_ignore_empty<T>({});
}

673
test/test_parser2.hpp Normal file
View File

@ -0,0 +1,673 @@
#include "test_helpers.hpp"
#include <fstream>
#include <iostream>
#include <ss/parser.hpp>
#include <unordered_set>
#ifndef SEGMENT_NAME
#error "SEGMENT_NAME must be defined"
#endif
// parser tests v2
namespace {
struct random_number_generator {
size_t z1 = 12341;
size_t z2 = 12342;
size_t z3 = 12343;
size_t z4 = 12344;
size_t rand() {
uint32_t b;
b = ((z1 << 6) ^ z1) >> 13;
z1 = ((z1 & 4294967294U) << 18) ^ b;
b = ((z2 << 2) ^ z2) >> 27;
z2 = ((z2 & 4294967288U) << 2) ^ b;
b = ((z3 << 13) ^ z3) >> 21;
z3 = ((z3 & 4294967280U) << 7) ^ b;
b = ((z4 << 3) ^ z4) >> 12;
z4 = ((z4 & 4294967168U) << 13) ^ b;
return (z1 ^ z2 ^ z3 ^ z4);
}
template <typename T>
size_t rand_index(const T& s) {
REQUIRE(!s.empty());
return rand() % s.size();
}
bool rand_bool() {
return (rand() % 100) > 50;
}
template <typename T>
void rand_insert(std::string& dst, const T& src) {
dst.insert(rand_index(dst), std::string{src});
}
template <typename T>
void rand_insert_n(std::string& dst, const T& src, size_t n_max) {
size_t n = rand() % n_max;
for (size_t i = 0; i < n; ++i) {
rand_insert(dst, src);
}
}
} rng;
struct field {
std::string value;
bool is_string = false;
bool has_spaces_left = false;
bool has_spaces_right = false;
bool has_new_line = false;
field(const std::string& input) {
value = input;
is_string = true;
has_spaces_left = !input.empty() && input.front() == ' ';
has_spaces_right = !input.empty() && input.back() == ' ';
has_new_line = input.find_first_of('\n') != std::string::npos;
}
field(int input) {
value = std::to_string(input);
}
field(double input) {
value = std::to_string(input);
}
};
struct column {
std::string header;
std::vector<field> fields;
};
template <typename... Ts>
[[nodiscard]] column make_column(const std::string& input_header,
const std::vector<field>& input_fields) {
using setup = ss::setup<Ts...>;
std::vector<field> filtered_fields;
for (const auto& el : input_fields) {
if (!setup::multiline::enabled && el.has_new_line) {
continue;
}
if (!setup::escape::enabled && !setup::quote::enabled) {
if (setup::trim_left::enabled && el.has_spaces_left) {
continue;
}
if (setup::trim_right::enabled && el.has_spaces_right) {
continue;
}
}
filtered_fields.push_back(el);
}
column c;
c.header = input_header;
c.fields = filtered_fields;
return c;
}
[[maybe_unused]] void replace_all2(std::string& s, const std::string& old_value,
const std::string& new_value) {
for (size_t i = 0; i < 999; ++i) {
size_t pos = s.find(old_value);
if (pos == std::string::npos) {
return;
}
s.replace(pos, old_value.size(), new_value);
}
FAIL("bad replace");
}
template <typename... Ts>
[[nodiscard]] std::vector<std::string> generate_csv_data(
const std::vector<field>& data, const std::string& delim) {
(void)delim;
using setup = ss::setup<Ts...>;
constexpr static auto escape = '\\';
constexpr static auto quote = '"';
constexpr static auto space = ' ';
constexpr static auto new_line = '\n';
constexpr static auto helper0 = '#';
constexpr static auto helper1 = '$';
// constexpr static auto helper3 = '&';
std::vector<std::string> output;
if (setup::escape::enabled && setup::quote::enabled) {
for (const auto& el : data) {
auto value = el.value;
replace_all2(value, {escape, quote}, {helper1});
bool quote_newline = rng.rand_bool();
bool quote_spacings = rng.rand_bool();
bool has_spaces = el.has_spaces_right || el.has_spaces_left;
// handle escape
replace_all2(value, {escape}, {helper0});
rng.rand_insert_n(value, escape, 2);
if (!quote_newline) {
replace_all2(value, {new_line}, {helper1});
replace_all2(value, {helper1}, {escape, new_line});
}
replace_all2(value, {escape, escape}, {escape});
replace_all2(value, {escape, helper0}, {helper0});
replace_all2(value, {helper0, escape}, {helper0});
replace_all2(value, {helper0}, {escape, escape});
replace_all2(value, {helper1}, {escape, quote});
replace_all2(value, {escape, quote}, {helper1});
if (rng.rand_bool() || quote_newline ||
(quote_spacings && has_spaces)) {
replace_all2(value, {quote}, {helper0});
if (rng.rand_bool()) {
replace_all2(value, {helper0}, {escape, quote});
} else {
replace_all2(value, {helper0}, {quote, quote});
}
value = std::string{quote} + value + std::string{quote};
}
replace_all2(value, {helper1}, {escape, quote});
if (!quote_spacings && has_spaces) {
replace_all2(value, {escape, space}, {helper0});
replace_all2(value, {space}, {helper0});
replace_all2(value, {helper0}, {escape, space});
}
output.push_back(value);
}
} else if (setup::escape::enabled) {
for (const auto& el : data) {
auto value = el.value;
replace_all2(value, {escape}, {helper0});
rng.rand_insert_n(value, escape, 3);
replace_all2(value, {new_line}, {helper1});
replace_all2(value, {helper1}, {escape, new_line});
replace_all2(value, {escape, escape}, {escape});
replace_all2(value, {escape, helper0}, {helper0});
replace_all2(value, {helper0, escape}, {helper0});
replace_all2(value, {helper0}, {escape, escape});
if (setup::trim_right::enabled || setup::trim_left::enabled) {
// escape space
replace_all2(value, {escape, space}, {helper0});
replace_all2(value, {space}, {helper0});
replace_all2(value, {helper0}, {escape, space});
}
output.push_back(value);
}
} else if (setup::quote::enabled) {
for (const auto& el : data) {
auto value = el.value;
if (rng.rand_bool() || el.has_new_line || el.has_spaces_left ||
el.has_spaces_right) {
replace_all2(value, {quote}, {helper0});
replace_all2(value, {helper0}, {quote, quote});
value = std::string{quote} + value + std::string{quote};
}
output.push_back(value);
}
} else {
for (const auto& el : data) {
output.push_back(el.value);
}
}
if (setup::trim_right::enabled) {
for (auto& el : output) {
size_t n = rng.rand();
for (size_t i = 0; i < n % 3; ++i) {
el = el + " ";
}
}
}
if (setup::trim_left::enabled) {
for (auto& el : output) {
size_t n = rng.rand();
for (size_t i = 0; i < n % 3; ++i) {
el = " " + el;
}
}
}
return output;
}
[[maybe_unused]] void write_to_file(const std::vector<std::string>& data,
const std::string& delim,
const std::string& file_name,
bool add_new_line = true) {
std::ofstream out{file_name, std::ios_base::app};
std::string line;
for (size_t i = 0; i < data.size(); ++i) {
line += data[i];
if (i != data.size() - 1) {
line += delim;
}
}
out << line;
if (add_new_line) {
out << std::endl;
}
}
#define CHECK_EQ_CRLF(V1, V2) \
if (V1 != V2) { \
auto tmp1 = V1; \
auto tmp2 = V2; \
replace_all2(tmp1, "\r\n", "\n"); \
replace_all2(tmp2, "\r\n", "\n"); \
\
CHECK(tmp1 == tmp2); \
\
if (tmp1 != tmp2) { \
replace_all2(tmp1, "\r", "(r)"); \
replace_all2(tmp2, "\r", "(r)"); \
\
replace_all2(tmp1, "\n", "(n)"); \
replace_all2(tmp2, "\n", "(n)"); \
\
replace_all2(tmp1, " ", "_"); \
replace_all2(tmp2, " ", "_"); \
\
std::cout << "<" << tmp1 << ">" << std::endl; \
std::cout << "<" << tmp2 << ">" << std::endl; \
std::cout << "file: " << f.name << std::endl; \
std::cout << "----------------" << std::endl; \
} \
\
} else { \
CHECK(V1 == V2); \
}
template <bool buffer_mode, typename... Ts>
void test_data_combinations(const std::vector<column>& input_data,
const std::string& delim, bool include_header) {
using setup = ss::setup<Ts...>;
if (setup::ignore_header && !include_header) {
return;
}
unique_file_name f{"parser_data_combinations" + std::string{SEGMENT_NAME}};
std::vector<std::vector<field>> expected_data;
std::vector<std::string> header;
std::vector<field> field_header;
auto add_blank_if_ignore_empty = [&] {
if constexpr (setup::ignore_empty) {
size_t n = rng.rand() % 3;
for (size_t i = 0; i < n; ++i) {
write_to_file({}, delim, f.name);
}
}
};
for (const auto& el : input_data) {
header.push_back(el.header);
field_header.push_back(field{el.header});
}
std::string header_line;
if (include_header) {
auto header_data = generate_csv_data<Ts...>(field_header, delim);
header_line = merge_header(header_data, delim);
if (input_data.size() == 0 && rand() % 10 == 0) {
write_to_file(header_data, delim, f.name, false);
} else {
write_to_file(header_data, delim, f.name);
}
}
std::vector<int> layout;
size_t n = 1 + rng.rand() % 5;
for (size_t i = 0; i < input_data.size(); ++i) {
layout.push_back(i);
}
for (size_t i = 0; i < n; ++i) {
std::vector<field> raw_data;
for (const auto& el : input_data) {
const auto& fields = el.fields;
if (fields.empty()) {
continue;
}
raw_data.push_back(fields[rng.rand_index(fields)]);
}
add_blank_if_ignore_empty();
expected_data.push_back(raw_data);
auto data = generate_csv_data<Ts...>(raw_data, delim);
if (i + 1 == n && rand() % 10 == 0) {
write_to_file(data, delim, f.name, false);
} else {
write_to_file(data, delim, f.name);
}
}
auto layout_combinations = include_header && !setup::ignore_header
? vector_combinations(layout, layout.size())
: std::vector<std::vector<int>>{layout};
auto remove_duplicates = [](const auto& vec) {
std::vector<int> unique_vec;
std::unordered_set<int> vec_set;
for (const auto& el : vec) {
if (vec_set.find(el) == vec_set.end()) {
vec_set.insert(el);
unique_vec.push_back(el);
}
}
return unique_vec;
};
std::vector<std::vector<int>> unique_layout_combinations;
for (const auto& layout : layout_combinations) {
unique_layout_combinations.push_back(remove_duplicates(layout));
}
for (const auto& layout : unique_layout_combinations) {
auto [p, _] = make_parser<buffer_mode, setup>(f.name, delim);
if (include_header && !setup::ignore_header) {
std::vector<std::string> fields;
for (const auto& index : layout) {
fields.push_back(header[index]);
}
if constexpr (!setup::ignore_header) {
p.use_fields(fields);
}
if (!p.valid()) {
if constexpr (setup::string_error) {
std::cout << p.error_msg() << std::endl;
} else {
std::cout << "use_fields failed" << std::endl;
}
}
REQUIRE(p.valid());
}
auto check_error = [&p = p] {
CHECK(p.valid());
if (!p.valid()) {
if constexpr (setup::string_error) {
std::cout << p.error_msg() << std::endl;
}
}
};
auto check_header = [&p = p, &header = header, include_header,
header_line] {
if (include_header) {
if constexpr (!setup::ignore_header) {
CHECK_EQ_ARRAY(header, p.header());
CHECK_EQ(header_line, p.raw_header());
}
}
};
int num_columns = layout.size();
for (size_t i = 0; i < n + 1; ++i) {
check_header();
try {
switch (num_columns) {
case 1: {
auto s0 = p.template get_next<std::string>();
if (i < n) {
check_error();
CHECK_EQ_CRLF(s0, expected_data[i][layout[0]].value);
} else {
CHECK(p.eof());
CHECK(!p.valid());
}
break;
}
case 2: {
auto [s0, s1] =
p.template get_next<std::string, std::string>();
if (i < n) {
check_error();
CHECK_EQ_CRLF(s0, expected_data[i][layout[0]].value);
CHECK_EQ_CRLF(s1, expected_data[i][layout[1]].value);
} else {
CHECK(p.eof());
CHECK(!p.valid());
}
break;
}
case 3: {
auto [s0, s1, s2] =
p.template get_next<std::string, std::string,
std::string>();
if (i < n) {
check_error();
CHECK_EQ_CRLF(s0, expected_data[i][layout[0]].value);
CHECK_EQ_CRLF(s1, expected_data[i][layout[1]].value);
CHECK_EQ_CRLF(s2, expected_data[i][layout[2]].value);
} else {
CHECK(p.eof());
CHECK(!p.valid());
}
break;
}
case 4: {
auto [s0, s1, s2, s3] =
p.template get_next<std::string, std::string,
std::string, std::string>();
if (i < n) {
check_error();
CHECK_EQ_CRLF(s0, expected_data[i][layout[0]].value);
CHECK_EQ_CRLF(s1, expected_data[i][layout[1]].value);
CHECK_EQ_CRLF(s2, expected_data[i][layout[2]].value);
CHECK_EQ_CRLF(s3, expected_data[i][layout[3]].value);
} else {
CHECK(p.eof());
CHECK(!p.valid());
}
break;
}
case 5: {
auto [s0, s1, s2, s3, s4] =
p.template get_next<std::string, std::string,
std::string, std::string,
std::string>();
if (i < n) {
check_error();
CHECK_EQ_CRLF(s0, expected_data[i][layout[0]].value);
CHECK_EQ_CRLF(s1, expected_data[i][layout[1]].value);
CHECK_EQ_CRLF(s2, expected_data[i][layout[2]].value);
CHECK_EQ_CRLF(s3, expected_data[i][layout[3]].value);
CHECK_EQ_CRLF(s4, expected_data[i][layout[4]].value);
} else {
CHECK(p.eof());
CHECK(!p.valid());
}
break;
}
default:
FAIL(("Invalid number of columns: " +
std::to_string(num_columns)));
break;
}
} catch (const std::exception& e) {
if (i < n) {
throw;
}
}
}
}
}
template <typename... Ts>
void test_option_combinations() {
column ints0 =
make_column<Ts...>("ints0", {field{123}, field{45}, field{6}});
column ints1 =
make_column<Ts...>("ints1", {field{123}, field{45}, field{6}});
column ints2 =
make_column<Ts...>("ints2", {field{123}, field{45}, field{6}});
column floats0 =
make_column<Ts...>("floats0", {field{1.23}, field{456.7}, field{0.8},
field{910}, field{123456789.987654321}});
column floats1 =
make_column<Ts...>("floats1", {field{1.23}, field{456.7}, field{0.8},
field{910}, field{123456789.987654321}});
column floats2 =
make_column<Ts...>("floats2", {field{1.23}, field{456.7}, field{0.8},
field{910}, field{123456789.987654321}});
column strings0 =
make_column<Ts...>("strings0", {field{"just"}, field{"some"},
field{"random"}, field{"string"}});
column strings1 =
make_column<Ts...>("strings1", {field{"st\"rings"}, field{"w\"\"ith"},
field{"qu\"otes\\"}, field{"\\a\\n\\d"},
field{"escapes\""}});
column strings2 =
make_column<Ts...>("strings2",
{field{" with "}, field{" spaces"},
field{"and "}, field{"\nnew"}, field{" \nlines"},
field{" a\n\nn\n\nd "}, field{" \nso\n "},
field{"on"}});
auto columns0 = std::vector{ints0, strings0, floats0, strings1, strings2};
auto columns1 = std::vector{strings2, strings1, floats0, strings0, ints0};
auto columns2 = std::vector{floats0, strings1, ints0, strings2, strings0};
auto columns3 = std::vector{ints0, ints1, ints2};
auto columns4 = std::vector{floats0, floats1, floats2};
auto columns5 = std::vector{strings1, strings2};
auto columns6 = std::vector{strings1};
auto columns7 = std::vector{strings2};
for (size_t i = 0; i < 3; ++i) {
for (const auto& delimiter : {",", "-", "--"}) {
for (const auto& columns :
{columns0, columns1, columns2, columns3, columns4, columns5,
columns6, columns7}) {
try {
test_data_combinations<false, Ts...>(columns, delimiter,
false);
test_data_combinations<false, Ts...>(columns, delimiter,
true);
test_data_combinations<true, Ts...>(columns, delimiter,
false);
test_data_combinations<true, Ts...>(columns, delimiter,
true);
} catch (std::exception& e) {
std::cout << typeid(ss::parser<Ts...>).name() << std::endl;
FAIL_CHECK(std::string{e.what()});
}
}
}
}
}
template <typename... Ts>
void test_option_combinations0() {
test_option_combinations<Ts...>();
#ifdef CMAKE_GITHUB_CI
test_option_combinations<Ts..., ss::ignore_empty>();
#endif
}
template <typename... Ts>
void test_option_combinations1() {
test_option_combinations0<Ts...>();
#ifdef CMAKE_GITHUB_CI
test_option_combinations0<Ts..., ss::ignore_header>();
#endif
}
template <typename... Ts>
void test_option_combinations2() {
test_option_combinations1<Ts...>();
#ifdef CMAKE_GITHUB_CI
test_option_combinations1<Ts..., ss::string_error>();
test_option_combinations1<Ts..., ss::throw_on_error>();
#endif
}
template <typename... Ts>
void test_option_combinations3() {
using trim = ss::trim<' '>;
test_option_combinations2<Ts...>();
test_option_combinations2<Ts..., trim>();
}
} /* anonymous namespace */
// Tests split into multiple compilation units
#if 0
TEST_CASE("parser test various cases version 2 segment 1") {
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
using multiline = ss::multiline;
#ifdef CMAKE_GITHUB_CI
using multiline_r = ss::multiline_restricted<10>;
using trimr = ss::trim_right<' '>;
using triml = ss::trim_left<' '>;
using trim = ss::trim<' '>;
// segment 1
test_option_combinations3<>();
test_option_combinations3<escape>();
// segment 2
test_option_combinations3<quote>();
test_option_combinations3<escape, quote>();
// segment 3
test_option_combinations3<escape, multiline>();
test_option_combinations3<quote, multiline>();
// segment 4
test_option_combinations3<escape, quote, multiline>();
test_option_combinations3<escape, quote, multiline_r>();
// segment 5
test_option_combinations<escape, quote, multiline, triml>();
test_option_combinations<escape, quote, multiline, trimr>();
// segment 6
test_option_combinations3<escape, quote, multiline>();
test_option_combinations3<escape, quote, multiline, trim>();
#else
test_option_combinations3<escape, quote, multiline>();
#endif
}
#endif

11
test/test_parser2_1.cpp Normal file
View File

@ -0,0 +1,11 @@
#define SEGMENT_NAME "segment1"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 1") {
#ifdef CMAKE_GITHUB_CI
using escape = ss::escape<'\\'>;
test_option_combinations3<>();
test_option_combinations3<escape>();
#endif
}

12
test/test_parser2_2.cpp Normal file
View File

@ -0,0 +1,12 @@
#define SEGMENT_NAME "segment2"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 2") {
#ifdef CMAKE_GITHUB_CI
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
test_option_combinations3<quote>();
test_option_combinations3<escape, quote>();
#endif
}

13
test/test_parser2_3.cpp Normal file
View File

@ -0,0 +1,13 @@
#define SEGMENT_NAME "segment3"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 3") {
#ifdef CMAKE_GITHUB_CI
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
using multiline = ss::multiline;
test_option_combinations3<escape, multiline>();
test_option_combinations3<quote, multiline>();
#endif
}

14
test/test_parser2_4.cpp Normal file
View File

@ -0,0 +1,14 @@
#define SEGMENT_NAME "segment4"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 4") {
#ifdef CMAKE_GITHUB_CI
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
using multiline = ss::multiline;
using multiline_r = ss::multiline_restricted<10>;
test_option_combinations3<escape, quote, multiline>();
test_option_combinations3<escape, quote, multiline_r>();
#endif
}

15
test/test_parser2_5.cpp Normal file
View File

@ -0,0 +1,15 @@
#define SEGMENT_NAME "segment5"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 5") {
#ifdef CMAKE_GITHUB_CI
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
using multiline = ss::multiline;
using trimr = ss::trim_right<' '>;
using triml = ss::trim_left<' '>;
test_option_combinations<escape, quote, multiline, triml>();
test_option_combinations<escape, quote, multiline, trimr>();
#endif
}

10
test/test_parser2_6.cpp Normal file
View File

@ -0,0 +1,10 @@
#define SEGMENT_NAME "segment6"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 6") {
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
using multiline = ss::multiline;
test_option_combinations3<escape, quote, multiline>();
}

View File

@ -3,12 +3,14 @@
set -x
set -e
python3 script/single_header_generator.py > ssp.cpp
TMP_HDR=test_single_header.hpp
TMP_SRC=test_single_header.cpp
TMP_BIN=test_single_header
echo 'int main(){ ss::parser p{""}; p.get_next<int, float>(); return 0; }' \
>> ssp.cpp
python3 script/single_header_generator.py > ${TMP_HDR}
cat ${TMP_HDR} test/test_single_header_main.txt > ${TMP_SRC}
g++ -std=c++17 ssp.cpp -o ssp.bin -Wall -Wextra
./ssp.bin
g++ -std=c++17 ${TMP_SRC} -o ${TMP_BIN} -Wall -Wextra
./${TMP_BIN}
rm ssp.cpp ssp.bin
rm ${TMP_HDR} ${TMP_SRC} ${TMP_BIN}

View File

@ -0,0 +1,12 @@
int main() {
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
using trim = ss::trim<' '>;
std::string data = "1,string,2.34,c";
ss::parser<quote, escape, trim, ss::multiline> p{data.c_str(), data.size()};
auto tup = p.get_next<int, std::string, float, std::optional<char>>();
return 0;
}

View File

@ -5,15 +5,15 @@
#include <ss/splitter.hpp>
namespace {
constexpr static auto combinations_size_default = 4;
size_t combinations_size = combinations_size_default;
constexpr static auto num_combinations_default = 4;
size_t num_combinations = num_combinations_default;
struct set_combinations_size {
set_combinations_size(size_t size) {
combinations_size = size;
struct set_num_combinations {
set_num_combinations(size_t size) {
num_combinations = size;
}
~set_combinations_size() {
combinations_size = combinations_size_default;
~set_num_combinations() {
num_combinations = num_combinations_default;
}
};
@ -127,33 +127,13 @@ std::vector<std::string> combinations(const std::vector<std::string>& v,
return ret;
}
std::vector<std::vector<std::string>> vector_combinations(
const std::vector<std::string>& v, size_t n) {
std::vector<std::vector<std::string>> ret;
if (n <= 1) {
for (const auto& i : v) {
ret.push_back({i});
}
return ret;
}
auto inner_combinations = vector_combinations(v, n - 1);
for (const auto& i : v) {
for (auto j : inner_combinations) {
j.insert(j.begin(), i);
ret.push_back(move(j));
}
}
return ret;
}
std::pair<std::vector<std::string>, std::vector<std::vector<std::string>>>
make_combinations(const std::vector<std::string>& input,
const std::vector<std::string>& output,
const std::string& delim) {
std::vector<std::string> lines;
std::vector<std::vector<std::string>> expectations;
for (size_t i = 0; i < combinations_size; ++i) {
for (size_t i = 0; i < num_combinations; ++i) {
auto l = combinations(input, delim, i);
lines.reserve(lines.size() + l.size());
lines.insert(lines.end(), l.begin(), l.end());
@ -165,7 +145,7 @@ make_combinations(const std::vector<std::string>& input,
return {std::move(lines), std::move(expectations)};
}
} /* namespace */
} /* anonymous namespace */
/* ********************************** */
/* ********************************** */
@ -173,9 +153,12 @@ make_combinations(const std::vector<std::string>& input,
using matches_type = std::vector<std::pair<case_type, std::string>>;
template <typename... Matchers>
void test_combinations(matches_type& matches, std::vector<std::string> delims) {
static inline void test_combinations(matches_type& matches,
std::vector<std::string> delims) {
ss::splitter<Matchers...> s;
ss::splitter<Matchers..., ss::throw_on_error> st;
std::vector<std::string> inputs;
std::vector<std::string> outputs;
for (const auto& [cases, e] : matches) {
@ -194,6 +177,13 @@ void test_combinations(matches_type& matches, std::vector<std::string> delims) {
auto vec = s.split(buff(lines[i].c_str()), delim);
CHECK(s.valid());
CHECK_EQ(words(vec), expectations[i]);
try {
auto vec = st.split(buff(lines[i].c_str()), delim);
CHECK_EQ(words(vec), expectations[i]);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
}
}
@ -253,7 +243,7 @@ TEST_CASE("splitter test with quote") {
}
TEST_CASE("splitter test with trim") {
auto guard = set_combinations_size(3);
auto guard = set_num_combinations(3);
case_type case1 = spaced({R"(x)"}, " ");
case_type case2 = spaced({R"(yy)"}, " ");
case_type case3 = spaced({R"(y y)"}, " ");
@ -320,7 +310,7 @@ TEST_CASE("splitter test with escape") {
}
TEST_CASE("splitter test with quote and trim") {
auto guard = set_combinations_size(3);
auto guard = set_num_combinations(3);
case_type case1 = spaced({R"("""")"}, " ");
case_type case2 = spaced({R"("x""x")", R"(x"x)"}, " ");
case_type case3 = spaced({R"("")", R"()"}, " ");
@ -438,7 +428,7 @@ TEST_CASE("splitter test with escape and trim") {
}
TEST_CASE("splitter test with quote and escape and trim") {
auto guard = set_combinations_size(3);
auto guard = set_num_combinations(3);
case_type case1 = spaced({R"("\"")", R"(\")", R"("""")"}, " ");
case_type case2 =
spaced({R"("x\"x")", R"(x\"x)", R"(x"x)", R"("x""x")"}, " ");
@ -509,6 +499,15 @@ TEST_CASE("splitter test error mode") {
CHECK_FALSE(s.valid());
CHECK_FALSE(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
try {
ss::splitter<ss::throw_on_error> s;
s.split(buff("just,some,strings"), "");
FAIL("expected exception");
} catch (ss::exception& e) {
CHECK_FALSE(std::string{e.what()}.empty());
CHECK_FALSE(s.unterminated_quote());
}
}
{
@ -522,11 +521,17 @@ TEST_CASE("splitter test error mode") {
}
template <typename Splitter>
auto expect_unterminated_quote(Splitter& s, const std::string& line) {
static inline auto expect_unterminated_quote(Splitter& s,
const std::string& line) {
try {
auto vec = s.split(buff(line.c_str()));
CHECK_FALSE(s.valid());
CHECK(s.valid());
CHECK(s.unterminated_quote());
return vec;
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
return decltype(s.split(buff(line.c_str()))){};
}
}
namespace ss {
@ -543,14 +548,16 @@ public:
return splitter.size_shifted();
}
};
} /* ss */
} /* namespace ss */
TEST_CASE("splitter test resplit unterminated quote") {
{
ss::converter<ss::quote<'"'>, ss::multiline, ss::escape<'\\'>> c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("x)");
CHECK_EQ(vec.size(), 1);
REQUIRE(s.unterminated_quote());
@ -631,7 +638,7 @@ TEST_CASE("splitter test resplit unterminated quote") {
{
auto new_line = buff.append(R"(,dom)");
vec = c.resplit(new_line, strlen(new_line));
CHECK_FALSE(s.valid());
CHECK(s.valid());
CHECK(s.unterminated_quote());
CHECK_EQ(words(vec), expected);
}
@ -768,54 +775,374 @@ TEST_CASE("splitter test resplit unterminated quote") {
CHECK_EQ(words(vec), expected);
}
}
{
ss::converter<ss::quote<'"'>, ss::escape<'\\'>, ss::multiline> c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("just\"some","ra)");
std::vector<std::string> expected{"just\"some"};
auto w = words(vec);
w.pop_back();
CHECK_EQ(w, expected);
REQUIRE(s.unterminated_quote());
{
auto new_line = buff.append(R"(n,dom",str\"ings)");
// invalid resplit size
vec = c.resplit(new_line, 4);
CHECK(!s.valid());
}
}
}
TEST_CASE("splitter test invalid splits") {
ss::converter<ss::string_error, ss::quote<'"'>, ss::trim<' '>,
ss::escape<'\\'>>
TEST_CASE("splitter test resplit unterminated quote with exceptions") {
try {
ss::converter<ss::quote<'"'>, ss::multiline, ss::escape<'\\'>,
ss::throw_on_error>
c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("x)");
CHECK_EQ(vec.size(), 1);
REQUIRE(s.unterminated_quote());
{
auto new_linet =
buff.append_overwrite_last(R"(a\x)", c.size_shifted());
vec = c.resplit(new_linet, strlen(new_linet));
CHECK(s.unterminated_quote());
CHECK_EQ(vec.size(), 1);
}
{
auto new_linet =
buff.append_overwrite_last(R"(")", c.size_shifted());
vec = c.resplit(new_linet, strlen(new_linet));
REQUIRE(s.valid());
CHECK_FALSE(s.unterminated_quote());
REQUIRE_EQ(vec.size(), 1);
CHECK_EQ(words(vec)[0], "xax");
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::multiline, ss::throw_on_error> c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, "\"just");
CHECK_EQ(vec.size(), 1);
auto new_line = buff.append(R"(",strings)");
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
std::vector<std::string> expected{"just", "strings"};
CHECK_EQ(words(vec), expected);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::multiline, ss::throw_on_error> c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, "just,some,\"random");
std::vector<std::string> expected{"just", "some", "just,some,\""};
CHECK_EQ(words(vec), expected);
auto new_line = buff.append(R"(",strings)");
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"just", "some", "random", "strings"};
CHECK_EQ(words(vec), expected);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::multiline, ss::throw_on_error> c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("just","some","ran"")");
std::vector<std::string> expected{"just", "some", R"("just","some",")"};
CHECK_EQ(words(vec), expected);
auto new_line =
buff.append_overwrite_last(R"(,dom","strings")", c.size_shifted());
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"just", "some", "ran\",dom", "strings"};
CHECK_EQ(words(vec), expected);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::multiline, ss::throw_on_error> c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("just","some","ran)");
std::vector<std::string> expected{"just", "some", R"("just","some",")"};
CHECK_EQ(words(vec), expected);
REQUIRE(s.unterminated_quote());
{
auto new_line = buff.append(R"(,dom)");
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK(s.unterminated_quote());
CHECK_EQ(words(vec), expected);
}
{
auto new_line = buff.append(R"(",strings)");
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"just", "some", "ran,dom", "strings"};
CHECK_EQ(words(vec), expected);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::escape<'\\'>, ss::multiline,
ss::throw_on_error>
c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("just\"some","ra)");
std::vector<std::string> expected{"just\"some"};
auto w = words(vec);
w.pop_back();
CHECK_EQ(w, expected);
REQUIRE(s.unterminated_quote());
{
auto new_line = buff.append(R"(n,dom",str\"ings)");
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"just\"some", "ran,dom", "str\"ings"};
CHECK_EQ(words(vec), expected);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::escape<'\\'>, ss::multiline,
ss::throw_on_error>
c;
auto& s = c.splitter;
auto vec =
expect_unterminated_quote(s, "3,4,"
"\"just0\\\n1\\\n22\\\n33333x\\\n4");
std::vector<std::string> expected{"3", "4"};
auto w = words(vec);
w.pop_back();
CHECK_EQ(w, expected);
REQUIRE(s.unterminated_quote());
{
auto new_line =
buff.append_overwrite_last("\nx5strings\"", c.size_shifted());
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"3", "4", "just0\n1\n22\n33333x\n4\nx5strings"};
CHECK_EQ(words(vec), expected);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::escape<'\\'>, ss::multiline,
ss::throw_on_error>
c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("just\"some","ra"")");
std::vector<std::string> expected{"just\"some"};
auto w = words(vec);
w.pop_back();
CHECK_EQ(w, expected);
REQUIRE(s.unterminated_quote());
{
auto new_line = buff.append_overwrite_last(R"(n,dom",str\"ings)",
c.size_shifted());
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"just\"some", "ra\"n,dom", "str\"ings"};
CHECK_EQ(words(vec), expected);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::escape<'\\'>, ss::multiline,
ss::throw_on_error>
c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"("just\"some","r\a\a\\\a\")");
std::vector<std::string> expected{"just\"some"};
auto w = words(vec);
w.pop_back();
CHECK_EQ(w, expected);
REQUIRE(s.unterminated_quote());
{
auto new_line = buff.append_overwrite_last(R"(n,dom",str\"ings)",
c.size_shifted());
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"just\"some", "raa\\a\"n,dom", "str\"ings"};
CHECK_EQ(words(vec), expected);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::trim<' '>, ss::multiline,
ss::throw_on_error>
c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"( "just" ,some, "ra )");
std::vector<std::string> expected{"just", "some"};
auto w = words(vec);
w.pop_back();
CHECK_EQ(w, expected);
REQUIRE(s.unterminated_quote());
{
auto new_line = buff.append(R"( n,dom" , strings )");
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"just", "some", "ra n,dom", "strings"};
CHECK_EQ(words(vec), expected);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
ss::converter<ss::quote<'"'>, ss::trim<' '>, ss::escape<'\\'>,
ss::multiline>
c;
auto& s = c.splitter;
auto vec = expect_unterminated_quote(s, R"( "ju\"st" ,some, "ra \")");
std::vector<std::string> expected{"ju\"st", "some"};
auto w = words(vec);
w.pop_back();
CHECK_EQ(w, expected);
REQUIRE(s.unterminated_quote());
{
auto new_line =
buff.append_overwrite_last(R"( n,dom" , strings )",
c.size_shifted());
vec = c.resplit(new_line, strlen(new_line));
CHECK(s.valid());
CHECK_FALSE(s.unterminated_quote());
expected = {"ju\"st", "some", "ra \" n,dom", "strings"};
CHECK_EQ(words(vec), expected);
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
template <typename... Ts>
void test_invalid_splits() {
ss::converter<ss::quote<'"'>, ss::trim<' '>, ss::escape<'\\'>, Ts...> c;
auto& s = c.splitter;
auto check_error_msg = [&] {
if constexpr (ss::setup<Ts...>::string_error) {
CHECK_FALSE(s.error_msg().empty());
}
};
// empty delimiter
s.split(buff("some,random,strings"), "");
CHECK_FALSE(s.valid());
CHECK_FALSE(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
check_error_msg();
// mismatched delimiter
s.split(buff(R"(some,"random,"strings")"));
CHECK_FALSE(s.valid());
CHECK_FALSE(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
check_error_msg();
// unterminated escape
s.split(buff(R"(some,random,strings\)"));
CHECK_FALSE(s.valid());
CHECK_FALSE(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
check_error_msg();
// unterminated escape
s.split(buff(R"(some,random,"strings\)"));
CHECK_FALSE(s.valid());
CHECK_FALSE(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
check_error_msg();
// unterminated quote
s.split(buff("some,random,\"strings"));
CHECK_FALSE(s.valid());
CHECK(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
check_error_msg();
// invalid resplit
char new_line[] = "some";
auto a = c.resplit(new_line, strlen(new_line));
c.resplit(new_line, strlen(new_line));
CHECK_FALSE(s.valid());
check_error_msg();
}
TEST_CASE("splitter test invalid splits") {
test_invalid_splits();
test_invalid_splits<ss::string_error>();
}
TEST_CASE("splitter test invalid splits with exceptions") {
ss::converter<ss::throw_on_error, ss::quote<'"'>, ss::trim<' '>,
ss::escape<'\\'>>
c;
auto& s = c.splitter;
// empty delimiter
REQUIRE_EXCEPTION(s.split(buff("some,random,strings"), ""));
CHECK_FALSE(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
// mismatched delimiter
REQUIRE_EXCEPTION(s.split(buff(R"(some,"random,"strings")")));
CHECK_FALSE(s.unterminated_quote());
// unterminated escape
REQUIRE_EXCEPTION(s.split(buff(R"(some,random,strings\)")));
CHECK_FALSE(s.unterminated_quote());
// unterminated escape
REQUIRE_EXCEPTION(s.split(buff(R"(some,random,"strings\)")));
CHECK_FALSE(s.unterminated_quote());
// unterminated quote
REQUIRE_EXCEPTION(s.split(buff("some,random,\"strings")));
CHECK(s.unterminated_quote());
// invalid resplit
char new_line[] = "some";
REQUIRE_EXCEPTION(c.resplit(new_line, strlen(new_line)));
}
TEST_CASE("splitter test with trim_left") {
auto guard = set_combinations_size(3);
auto guard = set_num_combinations(3);
case_type case1 = spaced_left({R"(x )"}, " ");
case_type case2 = spaced_left({R"(yy )"}, " ");
case_type case3 = spaced_left({R"(y y )"}, " ");
@ -845,7 +1172,7 @@ TEST_CASE("splitter test with trim_left") {
}
TEST_CASE("splitter test with trim_right") {
auto guard = set_combinations_size(3);
auto guard = set_num_combinations(3);
case_type case1 = spaced_right({R"( x)"}, " ");
case_type case2 = spaced_right({R"( yy)"}, " ");
case_type case3 = spaced_right({R"( y y)"}, " ");
@ -876,7 +1203,7 @@ TEST_CASE("splitter test with trim_right") {
}
TEST_CASE("splitter test with trim_right and trim_left") {
auto guard = set_combinations_size(3);
auto guard = set_num_combinations(3);
case_type case1 = spaced_right({R"(-x)"}, "-");
case_type case2 = spaced_left({R"(yy_)"}, "_");
case_type case3 = spaced_right({R"(-y y)"}, "-");
@ -894,7 +1221,7 @@ TEST_CASE("splitter test with trim_right and trim_left") {
}
TEST_CASE("splitter test with quote and escape, trim_left and trim_right") {
auto guard = set_combinations_size(3);
auto guard = set_num_combinations(3);
case_type case1 = spaced_left({R"("\"")", R"(\")", R"("""")"}, "_");
case_type case2 =
spaced_left({R"("x\"x")", R"(x\"x)", R"(x"x)", R"("x""x")"}, "_");