131 Commits

Author SHA1 Message Date
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
red0124
8b928de086 Merge pull request #22 from red0124/bugfix/multiline_on_custom_delimiter
Fix bug where the default delimiter would be used for multiline data
2023-07-25 00:46:43 +02:00
ado
6edce88d79 Fix bug where the default delimiter would be used for multiline data 2023-07-25 00:36:53 +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
red0124
6196f7796b Merge pull request #21 from red0124/improvement/markdown_labels
Update README links to sections
2023-07-20 00:21:38 +02:00
ado
5672aa635e Update README links to sections 2023-07-20 00:19:31 +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
ado
3eefac93b1 Fix README issues 2023-05-25 01:10:37 +02:00
ado
774f452689 fix typo 2023-03-19 19:48:22 +01:00
ado
448066b173 Update version 2023-02-18 14:20:32 +01:00
red0124
a7eca6064a Merge pull request #20 from red0124/improvement/conditional_t
Replace ss::ternary_t with std::conditional_t
2023-02-12 12:57:36 +01:00
ado
a4ecbd4dc8 Replace ss::ternary_t with std::conditional_t 2023-02-12 12:45:49 +01:00
red0124
a9e9783e6a Merge pull request #19 from red0124/improvement/doctest_update
Make project used forked doctest
2023-02-12 12:38:34 +01:00
ado
04edf1e532 Make project used forked doctest 2023-02-12 12:23:25 +01:00
red0124
8dcb69aa2c Merge pull request #18 from red0124/improvement/cmake_update
Update CMake files
2023-02-09 20:20:27 +01:00
ado
db2a72c18b Update CMake files 2023-02-08 21:54:44 +01:00
red0124
0e28a06799 Merge pull request #17 from red0124/improvement/meson_update
Improvement/meson update
2023-02-07 23:29:29 +01:00
ado
2218b01b81 Merge remote-tracking branch 'origin/master' into improvement/meson_update 2023-02-07 23:22:36 +01:00
ado
f777b04eb8 Fix failing mingw unit test 2023-02-01 22:45:50 +01:00
ado
7831bbd735 Add forcefallback wrap_mode option in meson.build 2023-02-01 01:22:43 +01:00
ado
6efb39b2db Add version number to meson.build 2023-02-01 00:57:32 +01:00
ado
f2b49e6d6c Update meson dependency usage 2023-02-01 00:52:23 +01:00
ado
5cb3c65b24 Update doctest revision 2023-01-31 21:23:04 +01:00
red0124
d58644fd67 Merge pull request #16 from red0124/feature/single_header
Feature/single header
2022-03-30 20:26:51 +02:00
ado
56447eb1d6 update README 2022-03-30 20:21:23 +02:00
ado
86d732e743 update README 2022-03-30 20:20:13 +02:00
ado
031ab5f7fc update README 2022-03-30 20:18:18 +02:00
ado
420625b25c update README 2022-03-30 20:15:14 +02:00
ado
9fa9edb24d add spp.hpp 2022-03-30 20:11:55 +02:00
ado
a2d72cdef3 remove unused script 2022-03-30 19:56:57 +02:00
ado
c214975ca0 add unit tests for conversion without the fast float library 2022-03-30 19:54:50 +02:00
ado
d328f7d59d add test_single_header.sh 2022-03-30 18:35:10 +02:00
ado
62055f03c7 add script to generate single header 2022-03-30 18:14:30 +02:00
ado
999992e579 fix README typos 2022-03-29 12:31:59 +02:00
31 changed files with 6594 additions and 829 deletions

63
.github/workflows/win-msvc.yml vendored Normal file
View File

@@ -0,0 +1,63 @@
name: win-msvc-ci
on:
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
runs-on: ${{ matrix.config.os }}
strategy:
fail-fast: false
matrix:
config:
- os: windows-2019
vs: "Visual Studio 16 2019"
- os: windows-latest
vs: "Visual Studio 17 2022"
build: [Debug, Release]
platform: [Win32, x64]
name: "${{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 -j ${{steps.cores.outputs.count}}

2
.gitignore vendored
View File

@@ -6,3 +6,5 @@ build/
hbuild/
subprojects/*
!subprojects/*.wrap
ssp.cpp
ssp.bin

View File

@@ -2,8 +2,8 @@ cmake_minimum_required(VERSION 3.14)
project(
ssp
VERSION 0.0.1
DESCRIPTION "Static split parser"
VERSION 1.4.0
DESCRIPTION "csv parser"
HOMEPAGE_URL "https://github.com/red0124/ssp"
LANGUAGES CXX
)
@@ -11,22 +11,23 @@ project(
# ---- Warning guard ----
# Protect dependents from this project's warnings if the guard isn't disabled
set(ssp_warning_guard SYSTEM)
if(ssp_INCLUDE_WITHOUT_SYSTEM)
set(ssp_warning_guard "")
set(SSP_WARNING_GUARD SYSTEM)
if(SSP_INCLUDE_WITHOUT_SYSTEM)
set(SSP_WARNING_GUARD "")
endif()
# ---- Dependencies ----
include(FetchContent)
FetchContent_Declare(
fast_float
GIT_REPOSITORY https://github.com/red0124/fast_float.git
GIT_TAG origin/meson
GIT_SHALLOW TRUE)
fetchcontent_declare(
fast_float
GIT_REPOSITORY https://github.com/red0124/fast_float.git
GIT_TAG origin/meson
GIT_SHALLOW TRUE
)
FetchContent_MakeAvailable(fast_float)
set(fast_float_source_dir "${FETCHCONTENT_BASE_DIR}/fast_float-src")
fetchcontent_makeavailable(fast_float)
set(FAST_FLOAT_SOURCE_DIR "${FETCHCONTENT_BASE_DIR}/fast_float-src")
# ---- Declare library ----
@@ -35,10 +36,10 @@ add_library(ssp::ssp ALIAS ssp)
target_include_directories(
ssp
${ssp_warning_guard}
${SSP_WARNING_GUARD}
INTERFACE
"$<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/include>"
"$<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/fast_float>"
"$<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/include>"
"$<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/fast_float>"
)
target_compile_features(ssp INTERFACE cxx_std_17)
@@ -46,8 +47,8 @@ target_compile_features(ssp INTERFACE cxx_std_17)
target_link_libraries(
ssp
INTERFACE
"$<$<AND:$<CXX_COMPILER_ID:AppleClang,Clang>,$<VERSION_LESS:$<CXX_COMPILER_VERSION>,9.0>>:c++fs>"
"$<$<AND:$<CXX_COMPILER_ID:GNU>,$<VERSION_LESS:$<CXX_COMPILER_VERSION>,9.1>>:stdc++fs>"
"$<$<AND:$<CXX_COMPILER_ID:AppleClang,Clang>,$<VERSION_LESS:$<CXX_COMPILER_VERSION>,9.0>>:c++fs>"
"$<$<AND:$<CXX_COMPILER_ID:GNU>,$<VERSION_LESS:$<CXX_COMPILER_VERSION>,9.1>>:stdc++fs>"
)
# ---- Install ----
@@ -55,19 +56,21 @@ target_link_libraries(
include(CMakePackageConfigHelpers)
include(GNUInstallDirs)
set(ssp_directory "ssp-${PROJECT_VERSION}")
set(ssp_include_directory "${CMAKE_INSTALL_INCLUDEDIR}")
set(SSP_DIRECTORY "ssp-${PROJECT_VERSION}")
set(SSP_INCLUDE_DIRECTORY "${CMAKE_INSTALL_INCLUDEDIR}")
install(
DIRECTORY "${PROJECT_SOURCE_DIR}/include/" "${fast_float_source_dir}/include/"
DESTINATION "${ssp_include_directory}"
DIRECTORY
"${PROJECT_SOURCE_DIR}/include/"
"${FAST_FLOAT_SOURCE_DIR}/include/"
DESTINATION "${SSP_INCLUDE_DIRECTORY}"
COMPONENT ssp_Development
)
install(
TARGETS ssp
EXPORT sspTargets
INCLUDES DESTINATION "${ssp_include_directory}"
INCLUDES DESTINATION "${SSP_INCLUDE_DIRECTORY}"
)
write_basic_package_version_file(
@@ -76,11 +79,11 @@ write_basic_package_version_file(
ARCH_INDEPENDENT
)
set(ssp_install_cmakedir "${CMAKE_INSTALL_LIBDIR}/cmake/${ssp_directory}")
set(SSP_INSTALL_CMAKEDIR "${CMAKE_INSTALL_LIBDIR}/cmake/${SSP_DIRECTORY}")
install(
FILES "${PROJECT_BINARY_DIR}/ssp-config-version.cmake"
DESTINATION "${ssp_install_cmakedir}"
DESTINATION "${SSP_INSTALL_CMAKEDIR}"
COMPONENT ssp_Development
)
@@ -88,10 +91,10 @@ install(
EXPORT sspTargets
FILE ssp-config.cmake
NAMESPACE ssp::
DESTINATION "${ssp_install_cmakedir}"
DESTINATION "${SSP_INSTALL_CMAKEDIR}"
COMPONENT ssp_Development
)
if(CMAKE_SOURCE_DIR STREQUAL CMAKE_CURRENT_SOURCE_DIR)
include(CPack)
include(CPack)
endif()

254
README.md
View File

@@ -8,25 +8,25 @@
```
[![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)
[![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)
[![win-msvc-ci](https://github.com/red0124/ssp/workflows/win-msvc-ci/badge.svg)](https://github.com/red0124/ssp/actions/workflows/win-msvc.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)
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 +34,10 @@ 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;
@@ -53,22 +51,26 @@ Brian S. Wolfe 40 1.9
Bill (Heath) Gates 65 3.3
```
# Features
* [Works on any type](#Custom-conversions)
* [Works on any type](#custom-conversions)
* Easy to use
* No exceptions
* [Works with headers](#Headers)
* [Works with quotes, escapes and spacings](#Setup)
* [Works with values containing new lines](#Multiline)
* [Columns and rows can be ignored](#Special-types)
* Works with any type of delimiter
* Can work without exceptions
* [Works with headers](#headers)
* [Works with quotes, escapes and spacings](#setup)
* [Works with values containing new lines](#multiline)
* [Columns and rows can be ignored](#special-types)
* [Works with any type of delimiter](#delimiter)
* Can return whole objects composed of converted values
* [Descriptive error handling can be enabled](#Error-handling)
* [Restrictions can be added for each column](#Restrictions)
* [Works with `std::optional` and `std::variant`](#Special-types)
* [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`**
* [Conversions can be chained if invalid](#Substitute-conversions)
* [Conversions can be chained if invalid](#substitute-conversions)
* Fast
# 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.
# Installation
```shell
@@ -78,8 +80,8 @@ $ cmake --configure .
$ sudo make install
```
*Note, this will also install the fast_float library*
The library supports [CMake](#Cmake) and [meson](#Meson) build systems
*Note, this will also install the fast_float library.*\
The library supports [CMake](#Cmake) and [meson](#Meson) build systems
# Usage
@@ -88,20 +90,20 @@ The library supports [CMake](#Cmake) and [meson](#Meson) build systems
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.
```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;
}
// ...
```
```shell
$ ./a.out
@@ -113,20 +115,19 @@ The header can be ignored using the **`ss::ignore_header`** [setup](#Setup) opti
```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 paser can also check if a field is present within the header by using the **`has_field`** method.
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.
```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");
const auto& [name, grade] = p.get_next<std::string, float>();
std::cout << name << ' ' << grade << std::endl;
const auto& [id, grade] = p.get_next<std::string, float>();
std::cout << id << ' ' << 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;
}
}
// ...
@@ -134,21 +135,26 @@ The fields with which the parser works with can be modified at any given time. T
```shell
$ ./a.out
James Bailey 2.5
1.9 Brian S. Wolfe 40
3.3 Bill (Heath) Gates 65
40 Brian S. Wolfe
65 Bill (Heath) Gates
```
## Conversions
An alternate loop to the example above would look like:
```cpp
while(!p.eof()) {
auto [name, age, grade] = p.get_next<std::string, int, float>();
// ...
ss::parser p{"students.csv"};
while (!p.eof()) {
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.
@@ -157,14 +163,14 @@ If **`get_next`** is called with a **`tuple`** as template parameter it would be
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;
};
@@ -175,24 +181,23 @@ 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.
@@ -220,7 +225,14 @@ 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 slightliy faster when using single character delimiters.*
### Empty lines
Empty lines can be ignored by defining **`ss::ignore_empty`** within the setup parameters:
@@ -302,14 +314,11 @@ 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:
```
@@ -330,22 +339,22 @@ 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`**:
@@ -356,30 +365,30 @@ 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>();
if(grade) {
// do something with grade
auto [id, age, grade] = p.get_next<std::string, int, std::optional<float>>();
if (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>();
if(std::holds_alternative<float>(grade)) {
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)) {
} else if (std::holds_alternative<char>(grade)) {
// grade set as char
}
```
## Restrictions
Custom **`restrictions`** can be used to narrow down the conversions of unwanted values. **`ss::ir`** (in range) and **`ss::ne`** (none empty) are one of those:
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*):
@@ -406,9 +415,11 @@ 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
@@ -435,21 +446,34 @@ 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.
```cpp
const std::string& parser::error_msg();
bool parser::valid();
bool parser::eof();
size_t parser::line();
// ...
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
@@ -468,7 +492,6 @@ The delimiter is " ", and the number of columns varies depending on which shape
```cpp
ss::parser p{"shapes.txt", " "};
if (!p.valid()) {
std::cout << p.error_msg() << std::endl;
exit(EXIT_FAILURE);
}
@@ -479,11 +502,16 @@ while (!p.eof()) {
using udbl = ss::gte<double, 0>;
auto [circle_or_square, rectangle, triangle] =
p.try_next<ss::nx<shape, shape::circle, shape::square>, udbl>()
p.try_next<ss::nx<shape, shape::circle, shape::square>, udbl>()
.or_else<ss::nx<shape, shape::rectangle>, udbl, udbl>()
.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;
@@ -510,7 +538,7 @@ while (!p.eof()) {
```
It is quite hard to make an error this way since most things will be checked at compile time.
The **`try_next`** method works in a similar way as **`get_next`** but returns a **`composit`** which holds a **`tuple`** with an **`optional`** to the **`tuple`** returned by **`get_next`**. This **`composite`** has an **`or_else`** method (looks a bit like **`tl::expected`**) which is able to try additional conversions if the previous failed. **`or_else`** also returns a **`composite`**, but in its tuple is the **`optional`** to the **`tuple`** of the previous conversions and an **`optional`** to the **`tuple`** of the new conversion. (sounds more complicated than it is.
The **`try_next`** method works in a similar way as **`get_next`** but returns a **`composite`** which holds a **`tuple`** with an **`optional`** to the **`tuple`** returned by **`get_next`**. This **`composite`** has an **`or_else`** method (looks a bit like **`tl::expected`**) which is able to try additional conversions if the previous failed. **`or_else`** also returns a **`composite`**, but in its tuple is the **`optional`** to the **`tuple`** of the previous conversions and an **`optional`** to the **`tuple`** of the new conversion. (sounds more complicated than it is.
To fetch the **`tuple`** from the **`composite`** the **`values`** method is used. The value of the above used conversion would look something like this:
```cpp
@@ -527,31 +555,34 @@ Each of those **`composite`** conversions can accept a lambda (or anything calla
// non negative double
using udbl = ss::gte<double, 0>;
p.try_next<ss::nx<shape, shape::circle, shape::square>, udbl>(
[&](const auto& data) {
const auto& [s, x] = data;
double area = (s == shape::circle) ? x * x * M_PI : x * x;
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);
})
.or_else<ss::nx<shape, shape::triangle>, udbl, udbl, udbl>(
[&](auto&& s, auto& a, const double& b, double& 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);
}
while (!p.eof()) {
p.try_next<ss::nx<shape, shape::circle, shape::square>, udbl>(
[&](const auto& data) {
const auto& [s, x] = data;
double area = (s == shape::circle) ? x * x * M_PI : x * x;
shapes.emplace_back(s, area);
})
.or_else<ss::nx<shape, shape::rectangle>, udbl, udbl>(
[&](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, 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.*
@@ -580,7 +611,7 @@ 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", " ");
@@ -628,6 +659,5 @@ revision = origin/master
```
Then simply fetch the dependency and it is ready to be used:
```meson
ssp_sub = subproject('ssp')
ssp_dep = ssp_sub.get_variable('ssp_dep')
ssp_dep = dependency('ssp')
```

View File

@@ -19,6 +19,12 @@ inline void assert_string_error_defined() {
"'string_error' needs to be enabled to use 'error_msg'");
}
template <bool ThrowOnError>
inline void assert_throw_on_error_not_defined() {
static_assert(!ThrowOnError, "cannot handle errors manually if "
"'throw_on_error' is enabled");
}
#if __unix__
inline ssize_t get_line(char** lineptr, size_t* n, FILE* stream) {
return getline(lineptr, n, stream);

View File

@@ -1,4 +1,5 @@
#pragma once
#include "exception.hpp"
#include "extract.hpp"
#include "function_traits.hpp"
#include "restrictions.hpp"
@@ -95,14 +96,15 @@ 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 = ss::ternary_t<string_error, std::string, bool>;
using error_type = std::conditional_t<string_error, std::string, bool>;
public:
// parses line with given delimiter, returns a 'T' object created with
@@ -119,7 +121,12 @@ public:
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_.split_data_);
} else {
handle_error_bad_split();
return {};
}
}
// parses already split line, returns 'T' object with extracted values
@@ -162,6 +169,8 @@ public:
bool valid() const {
if constexpr (string_error) {
return error_.empty();
} else if constexpr (throw_on_error) {
return true;
} else {
return !error_;
}
@@ -194,7 +203,7 @@ private:
////////////////
const split_data& resplit(line_ptr_type new_line, ssize_t new_size,
const std::string& delim = default_delimiter) {
const std::string& delim) {
return splitter_.resplit(new_line, new_size, delim);
}
@@ -225,98 +234,111 @@ 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";
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,
size_t mapping_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;
}
@@ -329,29 +351,28 @@ private:
template <typename... Ts>
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),
column_mappings_.size());
return return_type{};
handle_error_incompatible_mapping(sizeof...(Ts),
column_mappings_.size());
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 {};
}
}
@@ -380,19 +401,9 @@ private:
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;
number_of_columns_ = number_of_columns;
}
@@ -419,16 +430,17 @@ private:
}
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;
}
@@ -472,7 +484,7 @@ private:
////////////////
error_type error_{};
splitter<Matchers...> splitter_;
splitter<Options...> splitter_;
template <typename...>
friend class parser;

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(const std::string& msg): msg_{msg} {
}
virtual char const* what() const noexcept {
return msg_.c_str();
}
};
} /* ss */

View File

@@ -2,21 +2,29 @@
#include "type_traits.hpp"
#include <cstring>
#include <fast_float/fast_float.h>
#include <functional>
#include <limits>
#include <optional>
#include <stdexcept>
#include <string>
#include <string_view>
#include <variant>
#ifndef SSP_DISABLE_FAST_FLOAT
#include <fast_float/fast_float.h>
#else
#include <algorithm>
#include <cstdlib>
#endif
// TODO try from_chars for integer conversions
namespace ss {
////////////////
// number converters
////////////////
#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) {
@@ -29,6 +37,42 @@ std::enable_if_t<std::is_floating_point_v<T>, std::optional<T>> to_num(
return ret;
}
#else
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) {
constexpr static auto buff_max = 64;
char buff[buff_max];
size_t string_range = std::distance(begin, end);
if (string_range > buff_max) {
return std::nullopt;
}
std::copy_n(begin, string_range, buff);
buff[string_range] = '\0';
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);
} else if constexpr (std::is_same_v<T, long double>) {
ret = std::strtold(buff, &parse_end);
}
if (parse_end != buff + string_range) {
return std::nullopt;
}
return ret;
}
#endif
inline std::optional<short> from_char(char c) {
if (c >= '0' && c <= '9') {
return c - '0';
@@ -145,6 +189,7 @@ inline bool sub_overflow(long long& result, long long operand) {
return __builtin_ssubll_overflow(result, operand, &result);
}
// Note: sub_overflow function should be unreachable for unsigned values
template <>
inline bool sub_overflow(unsigned int& result, unsigned int operand) {
return __builtin_usub_overflow(result, operand, &result);
@@ -162,8 +207,8 @@ inline bool sub_overflow(unsigned long long& 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)) {
bool shift_and_add_overflow(T& value, T digit, F add_last_digit_overflow) {
if (mul_overflow<T>(value, 10) || add_last_digit_overflow(value, digit)) {
return true;
}
return false;
@@ -201,17 +246,17 @@ std::enable_if_t<std::is_integral_v<T>, std::optional<T>> to_num(
#if (defined(__clang__) || defined(__GNUC__) || defined(__GUNG__)) && \
!defined(MINGW32_CLANG)
auto add_last_digit_owerflow =
auto add_last_digit_overflow =
(is_negative) ? sub_overflow<T> : add_overflow<T>;
#else
auto add_last_digit_owerflow = is_negative;
auto add_last_digit_overflow = 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)) {
add_last_digit_overflow)) {
return std::nullopt;
}
}

View File

@@ -1,7 +1,9 @@
#pragma once
// TODO add single header tests
#include "common.hpp"
#include "converter.hpp"
#include "exception.hpp"
#include "extract.hpp"
#include "restrictions.hpp"
#include <cstdlib>
@@ -12,22 +14,23 @@
namespace ss {
template <typename... Matchers>
template <typename... Options>
class parser {
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;
using multiline = typename setup<Matchers...>::multiline;
using error_type = ss::ternary_t<string_error, std::string, bool>;
using multiline = typename setup<Options...>::multiline;
using error_type = std::conditional_t<string_error, std::string, bool>;
constexpr static bool escaped_multiline_enabled =
multiline::enabled && setup<Matchers...>::escape::enabled;
multiline::enabled && setup<Options...>::escape::enabled;
constexpr static bool quoted_multiline_enabled =
multiline::enabled && setup<Matchers...>::quote::enabled;
multiline::enabled && setup<Options...>::quote::enabled;
constexpr static bool ignore_header = setup<Matchers...>::ignore_header;
constexpr static bool ignore_header = setup<Options...>::ignore_header;
constexpr static bool ignore_empty = setup<Matchers...>::ignore_empty;
constexpr static bool ignore_empty = setup<Options...>::ignore_empty;
public:
parser(const std::string& file_name,
@@ -38,10 +41,10 @@ public:
if constexpr (ignore_header) {
ignore_next();
} else {
header_ = reader_.get_next_row();
raw_header_ = reader_.get_buffer();
}
} else {
set_error_file_not_open();
handle_error_file_not_open();
eof_ = true;
}
}
@@ -56,6 +59,8 @@ public:
bool valid() const {
if constexpr (string_error) {
return error_.empty();
} else if constexpr (throw_on_error) {
return true;
} else {
return !error_;
}
@@ -80,22 +85,56 @@ public:
}
size_t line() const {
return valid() ? reader_.line_number_ - 1 : 0;
return reader_.line_number_ > 1 ? reader_.line_number_ - 1
: reader_.line_number_;
}
template <typename T, typename... Ts>
no_void_validator_tup_t<T, Ts...> get_next() {
std::optional<std::string> error;
if (!eof_) {
if constexpr (throw_on_error) {
try {
reader_.parse();
} catch (const ss::exception& e) {
read_line();
decorate_rethrow(e);
}
} else {
reader_.parse();
}
}
reader_.update();
clear_error();
if (eof_) {
set_error_eof_reached();
if (!reader_.converter_.valid()) {
handle_error_invalid_conversion();
read_line();
return {};
}
clear_error();
if (eof_) {
handle_error_eof_reached();
return {};
}
if constexpr (throw_on_error) {
try {
auto value = reader_.converter_.template convert<T, Ts...>();
read_line();
return value;
} catch (const ss::exception& e) {
read_line();
decorate_rethrow(e);
}
}
auto value = reader_.converter_.template convert<T, Ts...>();
if (!reader_.converter_.valid()) {
set_error_invalid_conversion();
handle_error_invalid_conversion();
}
read_line();
@@ -103,33 +142,47 @@ public:
}
bool field_exists(const std::string& field) {
if (header_.empty()) {
split_header_data();
}
return header_index(field).has_value();
}
template <typename... Ts>
void use_fields(const Ts&... fields_args) {
if constexpr (ignore_header) {
set_error_header_ignored();
handle_error_header_ignored();
return;
}
if (header_.empty()) {
split_header_data();
}
if (!valid()) {
return;
}
auto fields = std::vector<std::string>{fields_args...};
if (fields.empty()) {
handle_error_empty_mapping();
return;
}
std::vector<size_t> column_mappings;
for (const auto& field : fields) {
if (std::count(fields.begin(), fields.end(), field) != 1) {
set_error_field_used_multiple_times(field);
handle_error_field_used_multiple_times(field);
return;
}
auto index = header_index(field);
if (!index) {
set_error_invalid_field(field);
handle_error_invalid_field(field);
return;
}
@@ -139,7 +192,8 @@ public:
reader_.converter_.set_column_mapping(column_mappings, header_.size());
reader_.next_line_converter_.set_column_mapping(column_mappings,
header_.size());
if (line() == 0) {
if (line() == 1) {
ignore_next();
}
}
@@ -151,23 +205,28 @@ public:
template <bool get_object, typename T, typename... Ts>
struct iterable {
struct iterator {
using value =
ss::ternary_t<get_object, T, no_void_validator_tup_t<T, Ts...>>;
using value = std::conditional_t<get_object, T,
no_void_validator_tup_t<T, Ts...>>;
iterator() : parser_{nullptr} {
iterator() : parser_{nullptr}, value_{} {
}
iterator(parser<Matchers...>* parser) : parser_{parser} {
iterator(parser<Options...>* parser) : parser_{parser}, value_{} {
}
iterator(const iterator& other) = default;
iterator(iterator&& other) = default;
value& operator*() {
return value_;
}
value* operator->() {
return &value_;
}
iterator& operator++() {
if (parser_->eof()) {
if (!parser_ || parser_->eof()) {
parser_ = nullptr;
} else {
if constexpr (get_object) {
@@ -196,22 +255,23 @@ public:
}
private:
parser<Options...>* parser_;
value value_;
parser<Matchers...>* parser_;
};
iterable(parser<Matchers...>* parser) : parser_{parser} {
iterable(parser<Options...>* parser) : parser_{parser} {
}
iterator begin() {
return ++iterator{parser_};
}
iterator end() {
return iterator{};
}
private:
parser<Matchers...>* parser_;
parser<Options...>* parser_;
};
template <typename... Ts>
@@ -264,6 +324,7 @@ public:
template <typename Fun>
auto on_error(Fun&& fun) {
assert_throw_on_error_not_defined<throw_on_error>();
if (!parser_.valid()) {
if constexpr (std::is_invocable_v<Fun>) {
fun();
@@ -313,7 +374,7 @@ public:
auto value =
parser_.reader_.converter_.template convert<U, Us...>();
if (!parser_.reader_.converter_.valid()) {
parser_.set_error_invalid_conversion();
parser_.handle_error_invalid_conversion();
}
return value;
}
@@ -331,6 +392,7 @@ public:
template <typename... Ts, typename Fun = none>
composite<std::optional<no_void_validator_tup_t<Ts...>>> try_next(
Fun&& fun = none{}) {
assert_throw_on_error_not_defined<throw_on_error>();
using Ret = no_void_validator_tup_t<Ts...>;
return try_invoke_and_make_composite<
std::optional<Ret>>(get_next<Ts...>(), std::forward<Fun>(fun));
@@ -340,6 +402,7 @@ public:
// tuple
template <typename T, typename... Ts, typename Fun = none>
composite<std::optional<T>> try_object(Fun&& fun = none{}) {
assert_throw_on_error_not_defined<throw_on_error>();
return try_invoke_and_make_composite<
std::optional<T>>(get_object<T, Ts...>(), std::forward<Fun>(fun));
}
@@ -358,7 +421,7 @@ private:
constexpr bool returns_void = std::is_same_v<Ret, void>;
if constexpr (!returns_void) {
if (!try_invoke_impl(arg, std::forward<Fun>(fun))) {
set_error_failed_check();
handle_error_failed_check();
}
} else {
try_invoke_impl(arg, std::forward<Fun>(fun));
@@ -399,6 +462,22 @@ private:
// header
////////////////
void split_header_data() {
ss::splitter<Options...> splitter;
std::string raw_header_copy = raw_header_;
splitter.split(raw_header_copy.data(), reader_.delim_);
for (const auto& [begin, end] : splitter.split_data_) {
std::string field{begin, end};
if (std::find(header_.begin(), header_.end(), field) !=
header_.end()) {
handle_error_invalid_header(field);
header_.clear();
return;
}
header_.push_back(std::move(field));
}
}
std::optional<size_t> header_index(const std::string& field) {
auto it = std::find(header_.begin(), header_.end(), field);
@@ -421,77 +500,135 @@ private:
}
}
void set_error_failed_check() {
void handle_error_failed_check() {
constexpr static auto error_msg = " failed check";
if constexpr (string_error) {
error_.append(file_name_).append(" failed check.");
error_.clear();
error_.append(file_name_).append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{file_name_ + error_msg};
} else {
error_ = true;
}
}
void set_error_file_not_open() {
void handle_error_file_not_open() {
constexpr static auto error_msg = " could not be opened";
if constexpr (string_error) {
error_.append(file_name_).append(" could not be opened.");
error_.clear();
error_.append(file_name_).append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{file_name_ + error_msg};
} else {
error_ = true;
}
}
void set_error_eof_reached() {
void handle_error_eof_reached() {
constexpr static auto error_msg = " read on end of file";
if constexpr (string_error) {
error_.append(file_name_).append(" reached end of file.");
error_.clear();
error_.append(file_name_).append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{file_name_ + error_msg};
} else {
error_ = true;
}
}
void set_error_invalid_conversion() {
void handle_error_invalid_conversion() {
if constexpr (string_error) {
error_.clear();
error_.append(file_name_)
.append(" ")
.append(std::to_string(reader_.line_number_))
.append(": ")
.append(reader_.converter_.error_msg())
.append(": \"")
.append(reader_.buffer_)
.append("\"");
.append(reader_.converter_.error_msg());
} else if constexpr (!throw_on_error) {
error_ = true;
}
}
void handle_error_header_ignored() {
constexpr static auto error_msg =
": the header row is ignored within the setup it cannot be used";
if constexpr (string_error) {
error_.clear();
error_.append(file_name_).append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{file_name_ + error_msg};
} else {
error_ = true;
}
}
void set_error_header_ignored() {
void handle_error_invalid_field(const std::string& field) {
constexpr static auto error_msg =
": header does not contain given field: ";
if constexpr (string_error) {
error_.append(file_name_)
.append(": \"")
.append("the header row is ignored within the setup, it cannot "
"be used")
.append("\"");
error_.clear();
error_.append(file_name_).append(error_msg).append(field);
} else if constexpr (throw_on_error) {
throw ss::exception{file_name_ + error_msg + field};
} else {
error_ = true;
}
}
void set_error_invalid_field(const std::string& field) {
void handle_error_field_used_multiple_times(const std::string& field) {
constexpr static auto error_msg = ": given field used multiple times: ";
if constexpr (string_error) {
error_.append(file_name_)
.append(": header does not contain given field: ")
.append(field);
error_.clear();
error_.append(file_name_).append(error_msg).append(field);
} else if constexpr (throw_on_error) {
throw ss::exception{file_name_ + error_msg + field};
} else {
error_ = true;
}
}
void set_error_field_used_multiple_times(const std::string& field) {
void handle_error_empty_mapping() {
constexpr static auto error_msg = "received empty mapping";
if constexpr (string_error) {
error_.append(file_name_)
.append(": given field used multiple times: ")
.append(field);
error_.clear();
error_.append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg};
} else {
error_ = true;
}
}
void handle_error_invalid_header(const std::string& field) {
constexpr static auto error_msg = "header contains duplicates: ";
if constexpr (string_error) {
error_.clear();
error_.append(error_msg).append(error_msg);
} else if constexpr (throw_on_error) {
throw ss::exception{error_msg + field};
} else {
error_ = true;
}
}
void decorate_rethrow(const ss::exception& e) const {
static_assert(throw_on_error,
"throw_on_error needs to be enabled to use this method");
throw ss::exception{std::string{file_name_}
.append(" ")
.append(std::to_string(line()))
.append(": ")
.append(e.what())};
}
////////////////
// line reading
////////////////
@@ -511,10 +648,12 @@ private:
helper_buffer_{other.helper_buffer_}, converter_{std::move(
other.converter_)},
next_line_converter_{std::move(other.next_line_converter_)},
size_{other.size_}, next_line_size_{other.next_line_size_},
buffer_size_{other.buffer_size_},
next_line_buffer_size_{other.next_line_buffer_size_},
helper_size_{other.helper_size_}, delim_{std::move(other.delim_)},
file_{other.file_}, crlf_{other.crlf_}, line_number_{
other.line_number_} {
file_{other.file_}, crlf_{other.crlf_},
line_number_{other.line_number_}, next_line_size_{
other.next_line_size_} {
other.buffer_ = nullptr;
other.next_line_buffer_ = nullptr;
other.helper_buffer_ = nullptr;
@@ -528,13 +667,14 @@ private:
helper_buffer_ = other.helper_buffer_;
converter_ = std::move(other.converter_);
next_line_converter_ = std::move(other.next_line_converter_);
size_ = other.size_;
next_line_size_ = other.next_line_size_;
buffer_size_ = other.buffer_size_;
next_line_buffer_size_ = other.next_line_buffer_size_;
helper_size_ = other.helper_size_;
delim_ = std::move(other.delim_);
file_ = other.file_;
crlf_ = other.crlf_;
line_number_ = other.line_number_;
next_line_size_ = other.next_line_size_;
other.buffer_ = nullptr;
other.next_line_buffer_ = nullptr;
@@ -559,16 +699,18 @@ private:
reader(const reader& other) = delete;
reader& operator=(const reader& other) = delete;
// read next line each time in order to set eof_
bool read_next() {
ssize_t ssize;
next_line_converter_.clear_error();
ssize_t ssize = 0;
size_t size = 0;
while (size == 0) {
++line_number_;
if (next_line_size_ > 0) {
if (next_line_buffer_size_ > 0) {
next_line_buffer_[0] = '\0';
}
ssize = get_line(&next_line_buffer_, &next_line_size_, file_);
ssize = get_line(&next_line_buffer_, &next_line_buffer_size_,
file_);
if (ssize == -1) {
return false;
@@ -581,17 +723,23 @@ private:
}
}
next_line_size_ = size;
return true;
}
void parse() {
size_t limit = 0;
if constexpr (escaped_multiline_enabled) {
while (escaped_eol(size)) {
while (escaped_eol(next_line_size_)) {
if (multiline_limit_reached(limit)) {
return true;
return;
}
if (!append_next_line_to_buffer(next_line_buffer_, size)) {
remove_eol(next_line_buffer_, ssize);
next_line_converter_.set_error_unterminated_escape();
return true;
if (!append_next_line_to_buffer(next_line_buffer_,
next_line_size_)) {
next_line_converter_.handle_error_unterminated_escape();
return;
}
}
}
@@ -600,46 +748,49 @@ private:
if constexpr (quoted_multiline_enabled) {
while (unterminated_quote()) {
next_line_size_ -= next_line_converter_.size_shifted();
if (multiline_limit_reached(limit)) {
return true;
return;
}
if (!append_next_line_to_buffer(next_line_buffer_, size)) {
remove_eol(next_line_buffer_, ssize);
return true;
if (!append_next_line_to_buffer(next_line_buffer_,
next_line_size_)) {
next_line_converter_.handle_error_unterminated_quote();
return;
}
if constexpr (escaped_multiline_enabled) {
while (escaped_eol(size)) {
while (escaped_eol(next_line_size_)) {
if (multiline_limit_reached(limit)) {
return true;
return;
}
if (!append_next_line_to_buffer(next_line_buffer_,
size)) {
remove_eol(next_line_buffer_, ssize);
next_line_size_)) {
next_line_converter_
.set_error_unterminated_escape();
return true;
.handle_error_unterminated_escape();
return;
}
}
}
next_line_converter_.resplit(next_line_buffer_, size);
next_line_converter_.resplit(next_line_buffer_,
next_line_size_, delim_);
}
}
return true;
}
void update() {
std::swap(buffer_, next_line_buffer_);
std::swap(size_, next_line_size_);
std::swap(buffer_size_, next_line_buffer_size_);
std::swap(converter_, next_line_converter_);
}
bool multiline_limit_reached(size_t& limit) {
if constexpr (multiline::size > 0) {
if (limit++ >= multiline::size) {
next_line_converter_.set_error_multiline_limit_reached();
next_line_converter_.handle_error_multiline_limit_reached();
return true;
}
}
@@ -650,23 +801,17 @@ private:
const char* curr;
for (curr = next_line_buffer_ + size - 1;
curr >= next_line_buffer_ &&
setup<Matchers...>::escape::match(*curr);
setup<Options...>::escape::match(*curr);
--curr) {
}
return (next_line_buffer_ - curr + size) % 2 == 0;
}
bool unterminated_quote() {
if (next_line_converter_.unterminated_quote()) {
return true;
}
return false;
return next_line_converter_.unterminated_quote();
}
void undo_remove_eol(char* buffer, size_t& string_end) {
if (next_line_converter_.unterminated_quote()) {
string_end -= next_line_converter_.size_shifted();
}
if (crlf_) {
std::copy_n("\r\n\0", 3, buffer + string_end);
string_end += 2;
@@ -676,24 +821,30 @@ private:
}
}
size_t remove_eol(char*& buffer, size_t size) {
size_t new_size = size - 1;
if (size >= 2 && buffer[size - 2] == '\r') {
size_t remove_eol(char*& buffer, size_t ssize) {
size_t size = ssize - 1;
if (ssize >= 2 && buffer[ssize - 2] == '\r') {
crlf_ = true;
new_size--;
size--;
} else {
crlf_ = false;
}
buffer[new_size] = '\0';
return new_size;
buffer[size] = '\0';
return size;
}
void realloc_concat(char*& first, size_t& first_size,
const char* const second, size_t second_size) {
next_line_size_ = first_size + second_size + 3;
first = static_cast<char*>(
realloc(static_cast<void*>(first), next_line_size_));
// TODO make buffer_size an argument
next_line_buffer_size_ = first_size + second_size + 3;
auto new_first = static_cast<char*>(
realloc(static_cast<void*>(first), next_line_buffer_size_));
if (!first) {
throw std::bad_alloc{};
}
first = new_first;
std::copy_n(second, second_size + 1, first + first_size);
first_size += second_size;
}
@@ -713,13 +864,8 @@ private:
return true;
}
std::vector<std::string> get_next_row() const {
std::vector<std::string> next_row;
auto& next_row_raw = next_line_converter_.splitter_.split_data_;
for (const auto& [begin, end] : next_row_raw) {
next_row.emplace_back(begin, end);
}
return next_row;
std::string get_buffer() {
return std::string{next_line_buffer_, next_line_buffer_size_};
}
////////////////
@@ -729,18 +875,20 @@ private:
char* next_line_buffer_{nullptr};
char* helper_buffer_{nullptr};
converter<Matchers...> converter_;
converter<Matchers...> next_line_converter_;
converter<Options...> converter_;
converter<Options...> next_line_converter_;
size_t size_{0};
size_t next_line_size_{0};
size_t buffer_size_{0};
size_t next_line_buffer_size_{0};
size_t helper_size_{0};
std::string delim_;
FILE* file_{nullptr};
bool crlf_;
bool crlf_{false};
size_t line_number_{0};
size_t next_line_size_{0};
};
////////////////
@@ -751,6 +899,7 @@ private:
error_type error_{};
reader reader_;
std::vector<std::string> header_;
std::string raw_header_;
bool eof_{false};
};

View File

@@ -109,10 +109,10 @@ 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 = ternary_t<is_matcher<T>::value, T,
typename get_matcher<Matcher, Ts...>::type>;
using type = std::conditional_t<is_matcher<T>::value, T,
typename get_matcher<Matcher, Ts...>::type>;
};
template <template <char...> class Matcher>
@@ -149,8 +149,8 @@ struct get_multiline;
template <typename T, typename... Ts>
struct get_multiline<T, Ts...> {
using type = ternary_t<is_instance_of_multiline<T>::value, T,
typename get_multiline<Ts...>::type>;
using type = std::conditional_t<is_instance_of_multiline<T>::value, T,
typename get_multiline<Ts...>::type>;
};
template <>
@@ -179,20 +179,26 @@ class ignore_header;
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> {};
@@ -203,37 +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 = ternary_t<trim_all::enabled, trim_all, trim_left_only>;
using trim_right = ternary_t<trim_all::enabled, trim_all, trim_right_only>;
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"
@@ -252,21 +269,28 @@ 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 */

View File

@@ -1,5 +1,6 @@
#pragma once
#include "common.hpp"
#include "exception.hpp"
#include "setup.hpp"
#include "type_traits.hpp"
#include <algorithm>
@@ -11,26 +12,29 @@
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 = ss::ternary_t<string_error, std::string, bool>;
using error_type = std::conditional_t<string_error, std::string, bool>;
public:
using line_ptr_type = ternary_t<is_const_line, const char*, char*>;
using line_ptr_type = std::conditional_t<is_const_line, const char*, char*>;
bool valid() const {
if constexpr (string_error) {
return error_.empty();
} else if constexpr (throw_on_error) {
return true;
} else {
return !error_;
}
@@ -77,7 +81,7 @@ 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_;
}
@@ -86,7 +90,7 @@ private:
// 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 +116,75 @@ 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;
}
@@ -235,7 +259,9 @@ 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 +308,7 @@ 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]);
@@ -362,7 +388,9 @@ 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 +407,10 @@ 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 +449,7 @@ 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;

View File

@@ -357,26 +357,6 @@ struct is_instance_of<Template, Template<Ts...>> {
template <template <typename...> class Template, typename... Ts>
constexpr bool is_instance_of_v = is_instance_of<Template, Ts...>::value;
////////////////
// ternary
////////////////
template <bool B, typename T, typename U>
struct ternary;
template <typename T, typename U>
struct ternary<true, T, U> {
using type = T;
};
template <typename T, typename U>
struct ternary<false, T, U> {
using type = U;
};
template <bool B, typename T, typename U>
using ternary_t = typename ternary<B, T, U>::type;
////////////////
// tuple to struct
////////////////

View File

@@ -1,17 +1,24 @@
project('ssp', 'cpp',
project(
'ssp',
['cpp'],
default_options :
['warning_level=3',
'cpp_std=c++17',
'buildtype=debugoptimized'])
['warning_level=3',
'cpp_std=c++17',
'buildtype=debugoptimized',
'wrap_mode=forcefallback'],
version: '1.4.0',
meson_version:'>=0.54.0')
fast_float_sub = subproject('fast_float')
fast_float_dep = fast_float_sub.get_variable('fast_float_dep')
fast_float_dep = dependency('fast_float')
ssp_dep = declare_dependency(
include_directories: include_directories('include'),
dependencies: fast_float_dep
)
meson.override_dependency('ssp', ssp_dep)
if not meson.is_subproject()
subdir('test')
endif

View File

@@ -5,7 +5,7 @@ BUILD_TYPE=Debug
set -eux
git clone https://github.com/onqtam/doctest -b 2.4.4 --depth 1
git clone https://github.com/red0124/doctest -b master --depth 1
cmake -S doctest -B doctest/build \
-D CMAKE_BUILD_TYPE=${BUILD_TYPE} \

View File

@@ -0,0 +1,35 @@
#!/bin/python3
headers_dir = 'include/ss/'
headers = ['type_traits.hpp',
'exception.hpp',
'function_traits.hpp',
'restrictions.hpp',
'common.hpp',
'setup.hpp',
'splitter.hpp',
'extract.hpp',
'converter.hpp',
'parser.hpp']
combined_file = []
includes = []
for header in headers:
with open(headers_dir + header) as f:
for line in f.read().splitlines():
if '#include "' in line or '#include <fast_float' in line:
continue
if '#include <' in line:
includes.append(line)
continue
if '#pragma once' not in line:
combined_file.append(line)
includes = sorted(set(includes))
print('\n'.join(includes))
print('#define SSP_DISABLE_FAST_FLOAT')
print('\n'.join(combined_file))

3208
ssp.hpp Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -1,3 +1,3 @@
[wrap-git]
url = https://github.com/onqtam/doctest
revision = 2.4.4
url = https://github.com/red0124/doctest
revision = master

View File

@@ -4,34 +4,41 @@ project(ssp_tests CXX)
# ---- Dependencies ----
set(
ssp_INCLUDE_WITHOUT_SYSTEM
YES
CACHE
INTERNAL
"Turn the warning guard off to have errors appear in test builds"
)
include(FetchContent)
FetchContent_Declare(ssp SOURCE_DIR "${PROJECT_SOURCE_DIR}/..")
FetchContent_MakeAvailable(ssp)
if(CMAKE_CXX_COMPILER_ID MATCHES "GNU|Clang")
target_compile_options(ssp INTERFACE -Wall -Wextra)
target_compile_options(ssp INTERFACE -Wall -Wextra)
endif()
find_package(doctest 2.4.4 CONFIG REQUIRED)
# for doctest_discover_tests
include(doctest)
if (MSVC)
add_compile_options(/bigobj)
elseif (MINGW)
add_compile_options(-Wa,-mbig-obj)
endif ()
include(FetchContent)
FetchContent_Declare(
DOCTEST
GIT_REPOSITORY https://github.com/red0124/doctest
GIT_TAG origin/master
GIT_SHALLOW TRUE)
FetchContent_MakeAvailable(DOCTEST)
set(DOCTEST "${FETCHCONTENT_BASE_DIR}/doctest-src")
# ---- Test ----
enable_testing()
foreach(name IN ITEMS test_splitter test_parser test_converter test_extractions)
add_executable("${name}" "${name}.cpp")
target_link_libraries("${name}" PRIVATE ssp::ssp fast_float doctest::doctest)
target_compile_definitions("${name}" PRIVATE
DOCTEST_CONFIG_IMPLEMENT_WITH_MAIN CMAKE_GITHUB_CI)
doctest_discover_tests("${name}")
foreach(name IN ITEMS test_splitter test_parser test_converter test_extractions
test_parser2_1 test_parser2_2 test_parser2_3
test_parser2_4 test_extractions_without_fast_float)
add_executable("${name}" "${name}.cpp")
target_link_libraries("${name}" PRIVATE ssp::ssp fast_float
doctest::doctest)
target_compile_definitions(
"${name}" PRIVATE DOCTEST_CONFIG_IMPLEMENT_WITH_MAIN CMAKE_GITHUB_CI)
add_test(NAME "${name}" COMMAND "${name}")
endforeach()

View File

@@ -1,19 +1,27 @@
test_sources = files([
'test_main.cpp',
'test_splitter.cpp',
'test_converter.cpp',
'test_parser.cpp',
'test_extractions.cpp',
])
doctest_dep = dependency('doctest')
add_project_arguments('-DDOCTEST_CONFIG_IMPLEMENT_WITH_MAIN', language: 'cpp')
doctest_proj = subproject('doctest')
doctest_dep = doctest_proj.get_variable('doctest_dep')
tests = [
'parser',
'splitter',
'converter',
'extractions',
'parser2_1',
'parser2_2',
'parser2_3',
'parser2_4',
'extractions_without_fast_float',
]
test_exe = executable(
'test_ssp',
sources: test_sources,
dependencies: [doctest_dep, ssp_dep],
cpp_args: '-lstdc++fs'
)
foreach name : tests
test_name = 'test_' + name
exe = executable(
test_name,
test_name + '.cpp',
dependencies: [doctest_dep, ssp_dep]
)
test(test_name, exe, timeout: 60)
endforeach
test('test_ssp', test_exe)

View File

@@ -22,6 +22,30 @@ TEST_CASE("converter test split") {
}
}
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"}, ","),
{"", {}, " "},
{" 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]);
}
}
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
}
TEST_CASE("converter test valid conversions") {
ss::converter c;
@@ -116,12 +140,153 @@ TEST_CASE("converter test valid conversions") {
}
}
TEST_CASE("converter test valid conversions with exceptions") {
ss::converter<ss::throw_on_error> c;
try {
auto tup = c.convert<int>("5");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<int, 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, int>("junk,5");
REQUIRE(c.valid());
CHECK_EQ(tup, 5);
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup = c.convert<int, 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, int, 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, int>("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<int>>("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<int, 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<int, 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, int, 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<int>, 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<int>, 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));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup =
c.convert<void, std::variant<int, 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));
} catch (ss::exception& e) {
FAIL(std::string{e.what()});
}
try {
auto tup =
c.convert<void, std::variant<int, 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));
} 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("converter test invalid conversions") {
ss::converter c;
c.convert<int>("");
REQUIRE_FALSE(c.valid());
c.convert<int>("1", "");
REQUIRE_FALSE(c.valid());
c.convert<int>("10", "");
REQUIRE_FALSE(c.valid());
@@ -150,6 +315,23 @@ TEST_CASE("converter test invalid conversions") {
REQUIRE_FALSE(c.valid());
}
TEST_CASE("converter test invalid conversions with exceptions") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(c.convert<int>(""));
REQUIRE_EXCEPTION(c.convert<int>("1", ""));
REQUIRE_EXCEPTION(c.convert<int>("10", ""));
REQUIRE_EXCEPTION(c.convert<int, void>(""));
REQUIRE_EXCEPTION(c.convert<int, void>(",junk"));
REQUIRE_EXCEPTION(c.convert<void, int>("junk,"));
REQUIRE_EXCEPTION(c.convert<int>("x"));
REQUIRE_EXCEPTION(c.convert<int, void>("x"));
REQUIRE_EXCEPTION(c.convert<int, void>("x,junk"));
REQUIRE_EXCEPTION(c.convert<void, int>("junk,x"));
REQUIRE_EXCEPTION(
c.convert<void, std::variant<int, double>, double>("junk;.5.5;6", ";"));
}
TEST_CASE("converter test ss:ax restriction (all except)") {
ss::converter c;
@@ -181,6 +363,32 @@ TEST_CASE("converter test ss:ax restriction (all except)") {
}
}
TEST_CASE("converter test ss:ax restriction (all except) with exceptions") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(c.convert<ss::ax<int, 0>>("0"));
REQUIRE_EXCEPTION(c.convert<ss::ax<int, 0, 1, 2>>("1"));
REQUIRE_EXCEPTION(c.convert<void, char, ss::ax<int, 0, 1, 2>>("junk,c,1"));
REQUIRE_EXCEPTION(c.convert<ss::ax<int, 1>, char>("1,c"));
try {
{
int tup = c.convert<ss::ax<int, 1>>("3");
CHECK_EQ(tup, 3);
}
{
std::tuple<char, int> tup = c.convert<char, ss::ax<int, 1>>("c,3");
CHECK_EQ(tup, std::make_tuple('c', 3));
}
{
std::tuple<int, char> tup = c.convert<ss::ax<int, 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;
@@ -215,6 +423,39 @@ TEST_CASE("converter test ss:nx restriction (none except)") {
}
}
TEST_CASE("converter test ss:nx restriction (none except) with exceptions") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(c.convert<ss::nx<int, 1>>("3"));
REQUIRE_EXCEPTION(c.convert<char, ss::nx<int, 1, 2, 69>>("c,3"));
REQUIRE_EXCEPTION(c.convert<ss::nx<int, 1>, char>("3,c"));
try {
{
auto tup = c.convert<ss::nx<int, 3>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
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::nx<int, 0, 1, 2>>("c,junk,1");
REQUIRE(c.valid());
CHECK_EQ(tup, std::make_tuple('c', 1));
}
{
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("converter test ss:ir restriction (in range)") {
ss::converter c;
@@ -249,6 +490,39 @@ TEST_CASE("converter test ss:ir restriction (in range)") {
}
}
TEST_CASE("converter test ss:ir restriction (in range) with exceptions") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(c.convert<ss::ir<int, 0, 2>>("3"));
REQUIRE_EXCEPTION(c.convert<char, ss::ir<int, 4, 69>>("c,3"));
REQUIRE_EXCEPTION(c.convert<ss::ir<int, 1, 2>, char>("3,c"));
try {
{
auto tup = c.convert<ss::ir<int, 1, 5>>("3");
REQUIRE(c.valid());
CHECK_EQ(tup, 3);
}
{
auto tup = c.convert<ss::ir<int, 0, 2>>("2");
REQUIRE(c.valid());
CHECK_EQ(tup, 2);
}
{
auto tup = c.convert<char, void, ss::ir<int, 0, 1>>("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");
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;
@@ -283,7 +557,38 @@ 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(c.convert<ss::oor<int, 1, 5>>("3"));
REQUIRE_EXCEPTION(c.convert<ss::oor<int, 0, 2>>("2"));
REQUIRE_EXCEPTION(c.convert<char, ss::oor<int, 0, 1>, void>("c,1,junk"));
REQUIRE_EXCEPTION(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 <>
@@ -331,6 +636,37 @@ 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(c.convert<ss::ne<std::string>>(""));
REQUIRE_EXCEPTION(c.convert<int, ss::ne<std::string>>("3,"));
REQUIRE_EXCEPTION(c.convert<ss::ne<std::string>, int>(",3"));
REQUIRE_EXCEPTION(c.convert<void, ss::ne<std::string>, int>("junk,,3"));
REQUIRE_EXCEPTION(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;
@@ -390,6 +726,58 @@ 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(c.convert<ss::lt<int, 3>>("3"));
REQUIRE_EXCEPTION(c.convert<ss::lt<int, 2>>("3"));
REQUIRE_EXCEPTION(c.convert<ss::gt<int, 3>>("3"));
REQUIRE_EXCEPTION(c.convert<ss::gt<int, 4>>("3"));
REQUIRE_EXCEPTION(c.convert<ss::lte<int, 2>>("3"));
REQUIRE_EXCEPTION(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");
@@ -397,6 +785,11 @@ TEST_CASE("converter test error mode") {
CHECK_FALSE(c.error_msg().empty());
}
TEST_CASE("converter test throw on error mode") {
ss::converter<ss::throw_on_error> c;
REQUIRE_EXCEPTION(c.convert<int>("junk"));
}
TEST_CASE("converter test converter with quotes spacing and escaping") {
{
ss::converter c;
@@ -444,6 +837,66 @@ 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 +904,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// mismatched quote
auto tup = c.convert<std::string, std::string, double, char>(
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 +913,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated quote
auto tup = c.convert<std::string, std::string, double, std::string>(
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 +922,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated escape
auto tup = c.convert<std::string, std::string, double, std::string>(
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 +931,7 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated escape while quoting
auto tup = c.convert<std::string, std::string, double, std::string>(
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 +940,42 @@ TEST_CASE("converter test invalid split conversions") {
{
// unterminated escaped quote
auto tup = c.convert<std::string, std::string, double, std::string>(
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(c.convert<std::string, std::string, double, char>(
buff(R"( "just , some , "12.3","a" )")));
CHECK_FALSE(c.unterminated_quote());
// unterminated quote
REQUIRE_EXCEPTION(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(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(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(c.convert<std::string, std::string, double, std::string>(
buff(R"(just,some,2,"strings\")")));
CHECK(c.unterminated_quote());
}

View File

@@ -2,32 +2,15 @@
#include <algorithm>
#include <ss/extract.hpp>
#define CHECK_FLOATING_CONVERSION(input, type) \
{ \
auto eps = std::numeric_limits<type>::min(); \
std::string s = #input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
REQUIRE(t.has_value()); \
CHECK_LT(std::abs(t.value() - type(input)), eps); \
} \
{ \
/* check negative too */ \
auto eps = std::numeric_limits<type>::min(); \
auto s = std::string("-") + #input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
REQUIRE(t.has_value()); \
CHECK_LT(std::abs(t.value() - type(-input)), eps); \
}
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);
@@ -70,13 +53,6 @@ TEST_CASE("extract test functions for decimal values") {
CHECK_DECIMAL_CONVERSION(1234567891011, ull);
}
#define CHECK_INVALID_CONVERSION(input, type) \
{ \
std::string s = input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
CHECK_FALSE(t.has_value()); \
}
TEST_CASE("extract test functions for numbers with invalid inputs") {
// negative unsigned value
CHECK_INVALID_CONVERSION("-1234", ul);
@@ -200,15 +176,6 @@ TEST_CASE("extract test functions for std::optional") {
}
}
#define REQUIRE_VARIANT(var, el, type) \
{ \
auto ptr = std::get_if<type>(&var); \
REQUIRE(ptr); \
REQUIRE_EQ(el, *ptr); \
}
#define CHECK_NOT_VARIANT(var, type) CHECK(!std::holds_alternative<type>(var));
TEST_CASE("extract test functions for std::variant") {
{
std::string s = "22";

View File

@@ -0,0 +1,132 @@
#include "test_helpers.hpp"
#include <algorithm>
#define SSP_DISABLE_FAST_FLOAT
#include <ss/extract.hpp>
TEST_CASE(
"testing extract functions for floating point values without fast float") {
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(420., float);
CHECK_FLOATING_CONVERSION(420., double);
CHECK_FLOATING_CONVERSION(0.123, float);
CHECK_FLOATING_CONVERSION(0.123, double);
CHECK_FLOATING_CONVERSION(123e4, float);
CHECK_FLOATING_CONVERSION(123e4, double);
}
TEST_CASE("extract test functions for numbers with invalid inputs without fast "
"float") {
// floating pint for int
CHECK_INVALID_CONVERSION("123.4", int);
// random input for float
CHECK_INVALID_CONVERSION("xxx1", float);
}
TEST_CASE("extract test functions for std::variant without fast float") {
{
std::string s = "22";
{
std::variant<int, 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);
}
{
std::variant<double, int, std::string> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, std::string);
REQUIRE_VARIANT(var, 22, double);
}
{
std::variant<std::string, double, int> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "22", std::string);
}
{
std::variant<int> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, 22, int);
}
}
{
std::string s = "22.2";
{
std::variant<int, 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, std::string);
REQUIRE_VARIANT(var, 22.2, double);
}
{
std::variant<double, int, std::string> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, std::string);
REQUIRE_VARIANT(var, 22.2, double);
}
{
std::variant<std::string, double, int> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "22.2", std::string);
}
}
{
std::string s = "2.2.2";
{
std::variant<int, 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, double);
REQUIRE_VARIANT(var, "2.2.2", std::string);
}
{
std::variant<double, std::string, int> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "2.2.2", std::string);
}
{
std::variant<std::string, double, int> var;
REQUIRE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
CHECK_NOT_VARIANT(var, int);
CHECK_NOT_VARIANT(var, double);
REQUIRE_VARIANT(var, "2.2.2", std::string);
}
{
std::variant<int, double> var;
REQUIRE_FALSE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, int{}, int);
CHECK_NOT_VARIANT(var, double);
}
{
std::variant<double, int> var;
REQUIRE_FALSE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, double{}, double);
CHECK_NOT_VARIANT(var, int);
}
{
std::variant<int> var;
REQUIRE_FALSE(ss::extract(s.c_str(), s.c_str() + s.size(), var));
REQUIRE_VARIANT(var, int{}, int);
}
}
}

View File

@@ -1,6 +1,10 @@
#pragma once
#include <cstdlib>
#include <cstring>
#include <ctime>
#include <iomanip>
#include <sstream>
#include <string>
#include <vector>
#include <filesystem>
#ifdef CMAKE_GITHUB_CI
#include <doctest/doctest.h>
@@ -8,41 +12,111 @@
#include <doctest.h>
#endif
namespace {
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;
std::string time_now_rand() {
std::stringstream ss;
auto t = std::time(nullptr);
auto tm = *std::localtime(&t);
ss << std::put_time(&tm, "%d%m%Y%H%M%S");
srand(time(nullptr));
return ss.str() + std::to_string(rand());
}
struct unique_file_name {
static inline int i = 0;
const std::string name;
unique_file_name(const std::string& test)
: name{"random_" + test + "_" + std::to_string(i++) + "_" +
time_now_rand() + "_file.csv"} {
}
~unique_file_name() {
std::filesystem::remove(name);
}
};
#define CHECK_FLOATING_CONVERSION(input, type) \
{ \
auto eps = std::numeric_limits<type>::min(); \
std::string s = #input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
REQUIRE(t.has_value()); \
CHECK_LT(std::abs(t.value() - type(input)), eps); \
} \
{ \
/* check negative too */ \
auto eps = std::numeric_limits<type>::min(); \
auto s = std::string("-") + #input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
REQUIRE(t.has_value()); \
CHECK_LT(std::abs(t.value() - type(-input)), eps); \
}
#define CHECK_INVALID_CONVERSION(input, type) \
{ \
std::string s = input; \
auto t = ss::to_num<type>(s.c_str(), s.c_str() + s.size()); \
CHECK_FALSE(t.has_value()); \
}
#define REQUIRE_VARIANT(var, el, type) \
{ \
auto ptr = std::get_if<type>(&var); \
REQUIRE(ptr); \
REQUIRE_EQ(el, *ptr); \
}
#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()); \
}
template <typename T>
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(move(j));
}
}
return ret;
}
} /* namespace */

File diff suppressed because it is too large Load Diff

653
test/test_parser2.hpp Normal file
View File

@@ -0,0 +1,653 @@
#include "test_helpers.hpp"
#include <algorithm>
#include <filesystem>
#include <fstream>
#include <iomanip>
#include <iostream>
#include <regex>
#include <ss/parser.hpp>
#include <sstream>
#include <unordered_map>
#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>
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>
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) {
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 << 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 <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{"test_parser2" + 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});
}
if (include_header) {
auto header_data = generate_csv_data<Ts...>(field_header, delim);
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);
write_to_file(data, delim, f.name);
/*
std::cout << "[.";
for (const auto& el : data) {
std::cout << el << '.';
}
std::cout << "]" << std::endl;
*/
}
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) {
ss::parser<setup> p{f.name, delim};
if (include_header && !setup::ignore_header) {
std::vector<std::string> fields;
for (const auto& index : layout) {
fields.push_back(header[index]);
}
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] {
CHECK(p.valid());
if (!p.valid()) {
if constexpr (setup::string_error) {
std::cout << p.error_msg() << std::endl;
}
}
};
int num_columns = layout.size();
for (size_t i = 0; i < n + 1; ++i) {
try {
switch (num_columns) {
case 1: {
auto s0 = p.template get_next<std::string>();
if (i < n) {
check_error();
// std::cout << s0 << std::endl;
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();
// std::cout << s0 << ' ' << s1 << std::endl;
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();
// std::cout << s0 << ' ' << s1 << ' ' << s2 <<
// std::endl;
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();
/*
std::cout << s0 << ' ' << s1 << ' ' << s2 << ' ' << s3
<< std::endl;
*/
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();
// std::cout << s0 << ' ' << s1 << ' ' << s2 << ' ' <<
// s3
// << ' ' << s4 << std::endl;
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<Ts...>(columns, delimiter, false);
test_data_combinations<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>();
}
} /* namespace */
#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<' '>;
// segment 1
test_option_combinations3<>();
test_option_combinations3<escape>();
test_option_combinations3<quote>();
// segment 2
test_option_combinations3<escape, quote>();
test_option_combinations3<escape, multiline>();
test_option_combinations3<quote, multiline>();
// segment 3
test_option_combinations3<escape, quote, multiline>();
test_option_combinations3<escape, quote, multiline_r>();
// segment 4
test_option_combinations<escape, quote, multiline, triml>();
test_option_combinations<escape, quote, multiline, trimr>();
#else
test_option_combinations3<escape, quote, multiline>();
#endif
}
#endif

14
test/test_parser2_1.cpp Normal file
View File

@@ -0,0 +1,14 @@
#define SEGMENT_NAME "segment1"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 1") {
#ifdef CMAKE_GITHUB_CI
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
test_option_combinations3<>();
test_option_combinations3<escape>();
test_option_combinations3<quote>();
#endif
}

15
test/test_parser2_2.cpp Normal file
View File

@@ -0,0 +1,15 @@
#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<'\\'>;
using multiline = ss::multiline;
test_option_combinations3<escape, quote>();
test_option_combinations3<escape, multiline>();
test_option_combinations3<quote, multiline>();
#endif
}

15
test/test_parser2_3.cpp Normal file
View File

@@ -0,0 +1,15 @@
#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;
using multiline_r = ss::multiline_restricted<10>;
test_option_combinations3<escape, quote, multiline>();
test_option_combinations3<escape, quote, multiline_r>();
#endif
}

19
test/test_parser2_4.cpp Normal file
View File

@@ -0,0 +1,19 @@
#define SEGMENT_NAME "segment4"
#include "test_parser2.hpp"
TEST_CASE("parser test various cases version 2 segment 4") {
using quote = ss::quote<'"'>;
using escape = ss::escape<'\\'>;
using multiline = ss::multiline;
#ifdef CMAKE_GITHUB_CI
using trimr = ss::trim_right<' '>;
using triml = ss::trim_left<' '>;
test_option_combinations<escape, quote, multiline, triml>();
test_option_combinations<escape, quote, multiline, trimr>();
#else
test_option_combinations3<escape, quote, multiline>();
#endif
}

14
test/test_single_header.sh Executable file
View File

@@ -0,0 +1,14 @@
#!/bin/bash
set -x
set -e
python3 script/single_header_generator.py > ssp.cpp
echo 'int main(){ ss::parser p{""}; p.get_next<int, float>(); return 0; }' \
>> ssp.cpp
g++ -std=c++17 ssp.cpp -o ssp.bin -Wall -Wextra
./ssp.bin
rm ssp.cpp ssp.bin

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());
@@ -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) {
auto vec = s.split(buff(line.c_str()));
CHECK_FALSE(s.valid());
CHECK(s.unterminated_quote());
return vec;
static inline auto expect_unterminated_quote(Splitter& s,
const std::string& line) {
try {
auto vec = s.split(buff(line.c_str()));
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 {
@@ -550,7 +555,9 @@ 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);
}
@@ -770,6 +777,269 @@ TEST_CASE("splitter test resplit unterminated quote") {
}
}
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()});
}
}
TEST_CASE("splitter test invalid splits") {
ss::converter<ss::string_error, ss::quote<'"'>, ss::trim<' '>,
ss::escape<'\\'>>
@@ -808,14 +1078,44 @@ TEST_CASE("splitter test invalid splits") {
// 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_FALSE(s.unterminated_quote());
CHECK_FALSE(s.error_msg().empty());
}
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());
// 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 +1145,7 @@ 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 +1176,7 @@ 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 +1194,7 @@ 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")"}, "_");