add splitter tests

meson.build

@@ -2,7 +2,7 @@ project('ssp', 'cpp',
   default_options : 
   ['warning_level=3', 
     'cpp_std=c++17', 
-    'buildtype=debug'])
+    'buildtype=debugoptimized'])
 
 includes = include_directories('include')
 subdir('test')

test/test_converter.cpp

@@ -1,26 +1,7 @@
+#include "test_helpers.hpp"
 #include <algorithm>
 #include <ss/converter.hpp>
 
-#ifdef CMAKE_GITHUB_CI
-#include <doctest/doctest.h>
-#else
-#include <doctest.h>
-#endif
-
-class buffer {
-    constexpr static auto buff_size = 1024;
-    char data_[buff_size];
-
-public:
-    char* operator()(const char* data) {
-        memset(data_, '\0', buff_size);
-        strcpy(data_, data);
-        return data_;
-    }
-};
-
-static buffer buff;
-
 TEST_CASE("testing split") {
     ss::converter c;
     for (const auto& [s, expected, delim] :

test/test_extractions.cpp

@@ -1,11 +1,6 @@
-#include <ss/extract.hpp>
+#include "test_helpers.hpp"
 #include <algorithm>
-
+#include <ss/extract.hpp>
-#ifdef CMAKE_GITHUB_CI
-#include <doctest/doctest.h>
-#else
-#include <doctest.h>
-#endif
 
 constexpr auto eps = 0.000001;
 using ld = long double;

test/test_helpers.hpp

@@ -0,0 +1,22 @@
+#pragma once
+#include <cstring>
+
+#ifdef CMAKE_GITHUB_CI
+#include <doctest/doctest.h>
+#else
+#include <doctest.h>
+#endif
+
+class buffer {
+    constexpr static auto buff_size = 1024;
+    char data_[buff_size];
+
+public:
+    char* operator()(const char* data) {
+        memset(data_, '\0', buff_size);
+        strcpy(data_, data);
+        return data_;
+    }
+};
+
+[[maybe_unused]] inline buffer buff;

test/test_parser.cpp

@@ -1,20 +1,18 @@
+#include "test_helpers.hpp"
 #include <algorithm>
 #include <filesystem>
 #include <fstream>
 #include <ss/parser.hpp>
 
-#ifdef CMAKE_GITHUB_CI
-#include <doctest/doctest.h>
-#else
-#include <doctest.h>
-#endif
-
 struct unique_file_name {
     const std::string name;
 
-    unique_file_name() : name{std::tmpnam(nullptr)} {}
+    unique_file_name() : name{std::tmpnam(nullptr)} {
+    }
 
-    ~unique_file_name() { std::filesystem::remove(name); }
+    ~unique_file_name() {
+        std::filesystem::remove(name);
+    }
 };
 
 struct X {
@@ -30,7 +28,9 @@ struct X {
             .append(delim)
             .append(s);
     }
-    auto tied() const { return std::tie(i, d, s); }
+    auto tied() const {
+        return std::tie(i, d, s);
+    }
 };
 
 template <typename T>
@@ -167,10 +167,13 @@ struct test_struct {
     int i;
     double d;
     char c;
-    auto tied() { return std::tie(i, d, c); }
+    auto tied() {
+        return std::tie(i, d, c);
+    }
 };
 
-void expect_test_struct(const test_struct&) {}
+void expect_test_struct(const test_struct&) {
+}
 
 // various scenarios
 TEST_CASE("testing composite conversion") {
@@ -392,7 +395,9 @@ struct my_string {
 
     my_string() = default;
 
-    ~my_string() { delete[] data; }
+    ~my_string() {
+        delete[] data;
+    }
 
     // make sure no object is copied
     my_string(const my_string&) = delete;
@@ -423,7 +428,9 @@ struct xyz {
     my_string x;
     my_string y;
     my_string z;
-    auto tied() { return std::tie(x, y, z); }
+    auto tied() {
+        return std::tie(x, y, z);
+    }
 };
 
 TEST_CASE("testing the moving of parsed values") {

test/test_splitter.cpp

@@ -1,48 +1,369 @@
+#include "test_helpers.hpp"
 #include <algorithm>
 #include <cstring>
 #include <iostream>
 #include <ss/splitter.hpp>
 
-#ifdef CMAKE_GITHUB_CI
+// TODO make ss::quote accept only one argument
-#include <doctest/doctest.h>
-#else
-#include <doctest.h>
-#endif
 
-TEST_CASE("testing splitter with escaping") {
+namespace {
-    std::vector<std::string> values{"10",   "he\\\"llo",
+constexpr static auto combinations_size_default = 4;
-                                    "\\\"", "\\\"a\\,a\\\"",
+size_t combinations_size = combinations_size_default;
-                                    "3.33", "a\\\""};
 
-    char buff[128];
+struct set_combinations_size {
-    // with quote
+    set_combinations_size(size_t size) {
-    ss::splitter<ss::quote<'"'>, ss::escape<'\\'>> s;
+        combinations_size = size;
-    std::string delim = ",";
-    for (size_t i = 0; i < values.size() * values.size(); ++i) {
-        std::string input1;
-        std::string input2;
-        for (size_t j = 0; j < values.size(); ++j) {
-            if (i & (1 << j) && j != 2 && j != 3) {
-                input1.append(values[j]);
-                input2.append(values.at(values.size() - 1 - j));
-            } else {
-                input1.append("\"" + values[j] + "\"");
-                input2.append("\"" + values.at(values.size() - 1 - j) + "\"");
     }
-            input1.append(delim);
+    ~set_combinations_size() {
-            input2.append(delim);
+        combinations_size = combinations_size_default;
     }
-        input1.pop_back();
+};
-        input2.pop_back();
-        input1.append("\0\"");
-        input2.append("\0\"");
 
-        memcpy(buff, input1.c_str(), input1.size() + 1);
+std::vector<std::string> words(const ss::split_input& input) {
-        auto tup1 = s.split(buff, delim);
+    std::vector<std::string> ret;
-        CHECK(tup1.size() == 6);
+    for (const auto& [begin, end] : input) {
+        ret.emplace_back(begin, end);
+    }
+    return ret;
+}
 
-        memcpy(buff, input2.c_str(), input2.size() + 1);
+[[maybe_unused]] std::string concat(const std::vector<std::string>& v) {
-        auto tup2 = s.split(buff, delim);
+    std::string ret = "{";
-        CHECK(tup2.size() == 6);
+    for (const auto& i : v) {
+        ret.append(i).append(" ");
+    }
+    ret.back() = ('}');
+    return ret;
+}
+
+std::vector<std::string> combinations(const std::vector<std::string>& v,
+                                      const std::string& delim, size_t n) {
+    if (n <= 1) {
+        return v;
+    }
+    std::vector<std::string> ret;
+    auto inner_combinations = combinations(v, delim, n - 1);
+    for (auto& i : v) {
+        for (auto& j : inner_combinations) {
+            ret.push_back(i + delim + j);
+        }
+    }
+    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 (auto& i : v) {
+            ret.push_back({i});
+        }
+        return ret;
+    }
+
+    auto inner_combinations = vector_combinations(v, n - 1);
+    for (auto& i : v) {
+        for (auto j : inner_combinations) {
+            j.insert(j.begin(), i);
+            ret.push_back(std::move(j));
+        }
+    }
+    return ret;
+}
+
+using case_type = std::vector<std::string>;
+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) {
+        auto l = combinations(input, delim, i);
+        lines.reserve(lines.size() + l.size());
+        lines.insert(lines.end(), l.begin(), l.end());
+
+        auto e = vector_combinations(output, i);
+        expectations.reserve(expectations.size() + e.size());
+        expectations.insert(expectations.end(), e.begin(), e.end());
+    }
+
+    return {std::move(lines), std::move(expectations)};
+}
+
+auto spaced(const case_type& input, std::string s) {
+    case_type ret = input;
+    for (auto& i : input) {
+        ret.push_back(s + i + s);
+        ret.push_back(i + s);
+        ret.push_back(s + i);
+        ret.push_back(s + s + i);
+        ret.push_back(s + s + i + s);
+        ret.push_back(s + s + i + s + s);
+        ret.push_back(s + i + s + s);
+        ret.push_back(i + s + s);
+    }
+
+    return ret;
+}
+
+auto spaced(const case_type& input, std::string s1, std::string s2) {
+    case_type ret = input;
+    for (auto& i : input) {
+        ret.push_back(s1 + i + s2);
+        ret.push_back(s2 + i + s1);
+        ret.push_back(s2 + s2 + s1 + s1 + i);
+        ret.push_back(i + s1 + s2 + s1 + s2);
+        ret.push_back(s1 + s1 + s1 + i + s2 + s2 + s2);
+        ret.push_back(s2 + s2 + s2 + i + s1 + s1 + s1);
+    }
+
+    return ret;
+}
+} /* namespace */
+
+/* ********************************** */
+/* ********************************** */
+
+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) {
+
+    ss::splitter<Matchers...> s;
+    std::vector<std::string> inputs;
+    std::vector<std::string> outputs;
+    for (const auto& [cases, e] : matches) {
+        for (const auto& c : cases) {
+            inputs.emplace_back(c);
+            outputs.emplace_back(e);
+        }
+    }
+
+    for (const auto& delim : delims) {
+        auto [lines, expectations] = make_combinations(inputs, outputs, delim);
+
+        REQUIRE(lines.size() == expectations.size());
+
+        for (size_t i = 0; i < lines.size(); ++i) {
+            auto vec = s.split(buff(lines[i].c_str()), delim);
+            CHECK(s.valid());
+            CHECK(words(vec) == expectations[i]);
+        }
+    }
+}
+
+TEST_CASE("testing splitter no setup") {
+    {
+        matches_type p{{{"x"}, "x"},        {{"\""}, "\""},
+                       {{""}, ""},          {{"\n"}, "\n"},
+                       {{"\"\""}, "\"\""},  {{"\" \\ \""}, "\" \\ \""},
+                       {{"     "}, "     "}};
+        test_combinations(p, {",", ";", "\t", "::"});
+    }
+}
+
+TEST_CASE("testing splitter quote") {
+    case_type case1 = {R"("""")"};
+    case_type case2 = {R"("x""x")", R"(x"x)"};
+    case_type case3 = {R"("")", R"()"};
+    case_type case4 = {R"("x")", R"(x)"};
+    case_type case5 = {R"("""""")"};
+    case_type case6 = {R"("\")", R"(\)"};
+    case_type case7 = {R"("xxxxxxxxxx")", R"(xxxxxxxxxx)"};
+
+    std::vector<std::string> delims = {",", "::", " ", "\t", "\n"};
+
+    {
+        matches_type p{{case1, "\""},        {case2, "x\"x"}, {case3, ""},
+                       {case4, "x"},         {case5, "\"\""}, {case6, "\\"},
+                       {case7, "xxxxxxxxxx"}};
+        test_combinations<ss::quote<'"'>>(p, delims);
+    }
+
+    case_type case8 = {R"(",")"};
+    case_type case9 = {R"("x,")"};
+    case_type case10 = {R"(",x")"};
+    case_type case11 = {R"("x,x")"};
+    case_type case12 = {R"(",,")"};
+    {
+        matches_type p{{case1, "\""}, {case3, ""},    {case8, ","},
+                       {case9, "x,"}, {case10, ",x"}, {case11, "x,x"},
+                       {case12, ",,"}};
+        test_combinations<ss::quote<'"'>>(p, {","});
+    }
+
+    case_type case13 = {R"("::")"};
+    case_type case14 = {R"("x::")"};
+    case_type case15 = {R"("::x")"};
+    case_type case16 = {R"("x::x")"};
+    case_type case17 = {R"("::::")"};
+
+    {
+        matches_type p{{case1, "\""},   {case3, ""},     {case13, "::"},
+                       {case14, "x::"}, {case15, "::x"}, {case16, "x::x"},
+                       {case17, "::::"}};
+        test_combinations<ss::quote<'"'>>(p, {"::"});
+    }
+}
+
+TEST_CASE("testing splitter trim") {
+    auto guard = set_combinations_size(3);
+    case_type case1 = spaced({R"(x)"}, " ");
+    case_type case2 = spaced({R"(yy)"}, " ");
+    case_type case3 = spaced({R"(y y)"}, " ");
+    case_type case4 = spaced({R"()"}, " ");
+
+    std::vector<std::string> delims = {",", "::", "\t", "\n"};
+
+    {
+        matches_type p{{case1, "x"},
+                       {case2, "yy"},
+                       {case3, "y y"},
+                       {case4, ""}};
+        test_combinations<ss::trim<' '>>(p, delims);
+    }
+
+    case_type case5 = spaced({"z"}, "\t");
+    case_type case6 = spaced({"ab"}, " ", "\t");
+    case_type case7 = spaced({"a\tb"}, " ", "\t");
+    case_type case8 = spaced({"a \t b"}, " ", "\t");
+
+    {
+        matches_type p{{case1, "x"},    {case2, "yy"},    {case3, "y y"},
+                       {case4, ""},     {case5, "z"},     {case6, "ab"},
+                       {case7, "a\tb"}, {case8, "a \t b"}};
+        test_combinations<ss::trim<' ', '\t'>>(p, {",", "::", "\n"});
+    }
+}
+
+TEST_CASE("testing splitter escape") {
+    case_type case1 = {R"(x)", R"(\x)"};
+    case_type case2 = {R"(xx)", R"(\xx)", R"(x\x)", R"(\x\x)"};
+    case_type case3 = {R"(\\)"};
+
+    std::vector<std::string> delims = {",", "::", " ", "\t", "\n"};
+
+    {
+        matches_type p{{case1, "x"}, {case2, "xx"}, {case3, "\\"}};
+        test_combinations<ss::escape<'\\'>>(p, delims);
+    }
+
+    case_type case4 = {R"(\,)"};
+    case_type case5 = {R"(x#,)"};
+    case_type case6 = {R"(#,x)"};
+    case_type case7 = {R"(x\,x)"};
+
+    {
+        matches_type p{{case1, "x"},  {case2, "xx"}, {case3, "\\"},
+                       {case4, ","},  {case5, "x,"}, {case6, ",x"},
+                       {case7, "x,x"}};
+        test_combinations<ss::escape<'\\', '#'>>(p, {","});
+    }
+
+    case_type case8 = {R"(\:\:)"};
+    case_type case9 = {R"(x\::x)"};
+
+    {
+        matches_type p{{case1, "x"},
+                       {case2, "xx"},
+                       {case3, "\\"},
+                       {case8, "::"},
+                       {case9, "x::x"}};
+        test_combinations<ss::escape<'\\'>>(p, {"::"});
+    }
+}
+
+TEST_CASE("testing splitter quote and trim") {
+    auto guard = set_combinations_size(3);
+    case_type case1 = spaced({R"("""")"}, " ");
+    case_type case2 = spaced({R"("x""x")", R"(x"x)"}, " ");
+    case_type case3 = spaced({R"("")", R"()"}, " ");
+    case_type case4 = spaced({R"("x")", R"(x)"}, " ");
+    case_type case5 = spaced({R"("""""")"}, " ");
+    case_type case6 = spaced({R"("\")", R"(\)"}, " ");
+    case_type case7 = spaced({R"("xxxxxxxxxx")", R"(xxxxxxxxxx)"}, " ");
+
+    std::vector<std::string> delims = {",", "::", "\t", "\n"};
+
+    {
+        matches_type p{{case1, "\""},        {case2, "x\"x"}, {case3, ""},
+                       {case4, "x"},         {case5, "\"\""}, {case6, "\\"},
+                       {case7, "xxxxxxxxxx"}};
+        test_combinations<ss::quote<'"'>, ss::trim<' '>>(p, delims);
+    }
+
+    return;
+    case_type case8 = {R"(",")"};
+    case_type case9 = {R"("x,")"};
+    case_type case10 = {R"(",x")"};
+    case_type case11 = {R"("x,x")"};
+    case_type case12 = {R"(",,")"};
+    {
+        matches_type p{{case1, "\""}, {case3, ""},    {case8, ","},
+                       {case9, "x,"}, {case10, ",x"}, {case11, "x,x"},
+                       {case12, ",,"}};
+        test_combinations<ss::quote<'"'>>(p, {","});
+    }
+
+    case_type case13 = {R"("::")"};
+    case_type case14 = {R"("x::")"};
+    case_type case15 = {R"("::x")"};
+    case_type case16 = {R"("x::x")"};
+    case_type case17 = {R"("::::")"};
+
+    {
+        matches_type p{{case1, "\""},   {case3, ""},     {case13, "::"},
+                       {case14, "x::"}, {case15, "::x"}, {case16, "x::x"},
+                       {case17, "::::"}};
+        test_combinations<ss::quote<'"'>>(p, {"::"});
+    }
+}
+
+TEST_CASE("testing splitter quote and escape") {
+    case_type case1 = {R"("\"")", R"(\")", R"("""")"};
+    case_type case2 = {R"("x\"x")", R"(x\"x)", R"(x"x)", R"("x""x")"};
+    case_type case3 = {R"("")", R"()"};
+    case_type case4 = {R"("x")", R"(x)"};
+    case_type case5 = {R"("\"\"")", R"("""""")", R"("\"""")", R"("""\"")"};
+    case_type case6 = {R"("\\")", R"(\\)"};
+    case_type case7 = {R"("xxxxxxxxxx")", R"(xxxxxxxxxx)"};
+
+    std::vector<std::string> delims = {",", "::", " ", "\t", "\n"};
+
+    ss::splitter<ss::quote<'"'>, ss::escape<'\\'>> s;
+
+    {
+        matches_type p{{case1, "\""},        {case2, "x\"x"}, {case3, ""},
+                       {case4, "x"},         {case5, "\"\""}, {case6, "\\"},
+                       {case7, "xxxxxxxxxx"}};
+        test_combinations<ss::quote<'"'>, ss::escape<'\\'>>(p, delims);
+    }
+
+    case_type case8 = {R"('xxxxxxxxxx')", R"(xxxxxxxxxx)"};
+    case_type case9 = {R"('')", R"()"};
+    case_type case10 = {R"('#\')", R"(#\)"};
+    case_type case11 = {R"('#'')", R"(#')", R"('''')"};
+    case_type case12 = {R"('##')", R"(##)"};
+    {
+        matches_type p{{case8, "xxxxxxxxxx"},
+                       {case9, ""},
+                       {case10, "\\"},
+                       {case11, "'"},
+                       {case12, "#"}};
+        test_combinations<ss::quote<'\''>, ss::escape<'#'>>(p, delims);
+    }
+
+    case_type case13 = {R"("x,x")",  R"(x\,x)",   R"(x#,x)",
+                        R"("x\,x")", R"("x#,x")", R"("x#,x")"};
+    case_type case14 = {R"("#\\#")", R"(#\\#)", R"(\\##)", R"("\\##")"};
+
+    {
+        matches_type p{{case1, "\""},
+                       {case2, "x\"x"},
+                       {case3, ""},
+                       {case13, "x,x"},
+                       {case14, "\\#"}};
+        test_combinations<ss::quote<'"'>, ss::escape<'\\', '#'>>(p, {","});
     }
 }