tests/test_string.c

/*
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS HEADER.
 *
 * Copyright 2023 Mike Becker, Olaf Wintermann All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 *   1. Redistributions of source code must retain the above copyright
 *      notice, this list of conditions and the following disclaimer.
 *
 *   2. Redistributions in binary form must reproduce the above copyright
 *      notice, this list of conditions and the following disclaimer in the
 *      documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

#include "cx/test.h"
#include "util_allocator.h"

#include "cx/string.h"
#include "cx/compare.h"

#include <limits.h>
#include <errno.h>

#define ASSERT_ZERO_TERMINATED(str) CX_TEST_ASSERTM((str).ptr[(str).length] == '\0', \
    #str " is not zero terminated")

CX_TEST(test_string_construct) {
    cxstring s1 = CX_STR("1234");
    cxstring s2 = cx_strn("abcd", 2);
    cxmutstr s3 = cx_mutstr((char *) "1234");
    cxmutstr s4 = cx_mutstrn((char *) "abcd", 2);
    CX_TEST_DO {
        CX_TEST_ASSERT(s1.length == 4);
        CX_TEST_ASSERT(strncmp(s1.ptr, "1234", 4) == 0);
        CX_TEST_ASSERT(s2.length == 2);
        CX_TEST_ASSERT(strncmp(s2.ptr, "ab", 2) == 0);
        CX_TEST_ASSERT(s3.length == 4);
        CX_TEST_ASSERT(strncmp(s3.ptr, "1234", 4) == 0);
        CX_TEST_ASSERT(s4.length == 2);
        CX_TEST_ASSERT(strncmp(s4.ptr, "ab", 2) == 0);
    }
}

CX_TEST(test_strfree) {
    CxTestingAllocator talloc;
    cx_testing_allocator_init(&talloc);
    CxAllocator *alloc = &talloc.base;
    CX_TEST_DO {
        char *test = cxMalloc(alloc, 16);
        cxmutstr str = cx_mutstrn(test, 16);
        CX_TEST_ASSERT(str.ptr == test);
        CX_TEST_ASSERT(str.length == 16);
        cx_strfree_a(alloc, &str);
        CX_TEST_ASSERT(str.ptr == NULL);
        CX_TEST_ASSERT(str.length == 0);
        CX_TEST_ASSERT(cx_testing_allocator_verify(&talloc));
    }
    cx_testing_allocator_destroy(&talloc);
}

CX_TEST(test_strdup) {
    cxstring str = CX_STR("test");
    cxmutstr dup = cx_strdup(str);
    CX_TEST_DO {
        CX_TEST_ASSERT(dup.length == str.length);
        CX_TEST_ASSERT(0 == strcmp(dup.ptr, str.ptr));
        ASSERT_ZERO_TERMINATED(dup);
    }
    cx_strfree(&dup);
}

CX_TEST(test_strdup_shortened) {
    cxstring str = CX_STR("test");
    str.length = 2;
    cxmutstr dup = cx_strdup(str);
    CX_TEST_DO {
        CX_TEST_ASSERT(dup.length == str.length);
        CX_TEST_ASSERT(0 == strcmp(dup.ptr, "te"));
        ASSERT_ZERO_TERMINATED(dup);
    }
    cx_strfree(&dup);
}

CX_TEST(test_strlen) {
    cxstring s1 = CX_STR("1234");
    cxstring s2 = CX_STR(".:.:.");
    cxstring s3 = CX_STR("X");
    CX_TEST_DO {
        size_t len0 = cx_strlen(0);
        size_t len1 = cx_strlen(1, s1);
        size_t len2 = cx_strlen(2, s1, s2);
        size_t len3 = cx_strlen(3, s1, s2, s3);

        CX_TEST_ASSERT(len0 == 0);
        CX_TEST_ASSERT(len1 == 4);
        CX_TEST_ASSERT(len2 == 9);
        CX_TEST_ASSERT(len3 == 10);
    }
}

CX_TEST(test_strsubs) {
    cxstring str = CX_STR("A test string");

    CX_TEST_DO {
        cxstring sub = cx_strsubs(str, 0);
        CX_TEST_ASSERT(0 == cx_strcmp(sub, str));

        sub = cx_strsubs(str, 2);
        CX_TEST_ASSERT(0 == cx_strcmp(sub, CX_STR("test string")));

        sub = cx_strsubs(str, 7);
        CX_TEST_ASSERT(0 == cx_strcmp(sub, CX_STR("string")));

        sub = cx_strsubs(str, 15);
        CX_TEST_ASSERT(0 == cx_strcmp(sub, CX_STR("")));

        sub = cx_strsubsl(str, 2, 4);
        CX_TEST_ASSERT(0 == cx_strcmp(sub, CX_STR("test")));

        sub = cx_strsubsl(str, 7, 3);
        CX_TEST_ASSERT(0 == cx_strcmp(sub, CX_STR("str")));

        sub = cx_strsubsl(str, 7, 20);
        CX_TEST_ASSERT(0 == cx_strcmp(sub, CX_STR("string")));

        // just for coverage, call the _m variant
        cxmutstr m = cx_strsubs_m(cx_mutstrn(NULL, 0), 0);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(m), CX_STR("")));
    }
}

CX_TEST(test_strchr) {
    cxstring str = CX_STR("I will find you - and I will kill you");

    CX_TEST_DO {
        cxstring notfound = cx_strchr(str, 'x');
        CX_TEST_ASSERT(notfound.length == 0);

        cxstring result = cx_strchr(str, 'w');
        CX_TEST_ASSERT(result.length == 35);
        CX_TEST_ASSERT(0 == strcmp(result.ptr, "will find you - and I will kill you"));

        // just for coverage, call the _m variant
        cxmutstr m = cx_strchr_m(cx_mutstrn(NULL, 0), 'a');
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(m), CX_STR("")));
    }
}

CX_TEST(test_strrchr) {
    cxstring str = CX_STR("I will find you - and I will kill you");

    CX_TEST_DO {
        cxstring notfound = cx_strrchr(str, 'x');
        CX_TEST_ASSERT(notfound.length == 0);

        cxstring result = cx_strrchr(str, 'w');
        CX_TEST_ASSERT(result.length == 13);
        CX_TEST_ASSERT(0 == strcmp(result.ptr, "will kill you"));

        // just for coverage, call the _m variant
        cxmutstr m = cx_strrchr_m(cx_mutstrn(NULL, 0), 'a');
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(m), CX_STR("")));
    }
}

CX_TEST(test_strstr) {
    cxstring str = CX_STR("find the match in this string");

    const size_t longstrpatternlen = 64 + cx_strstr_sbo_size;
    const size_t longstrlen = 320 + longstrpatternlen + 14;

    // it is more expensive to use calloc here, because we will overwrite
    // the memory anyway in the test preparation - but it is more reliable
    // in case we are doing something horribly wrong
    char *longstrc = calloc(longstrlen+1, 1);
    char *longstrpatternc = calloc(longstrpatternlen+1, 1);

    memcpy(longstrc,
           "abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijkl"
           "mnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwx"
           "yzabcdeababababnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghij"
           "klmnopqrstuvwxyzaababababababababrstuvwxyzabcdefghijklmnopqrstuv"
           "abababababababababababababababababababababababababababababababab",
           320
    );
    memcpy(longstrpatternc,
           "abababababababababababababababababababababababababababababababab",
           64
    );
    char x = 'a', y='b', z;
    for (size_t i = 0; i < cx_strstr_sbo_size ; i++) {
        longstrpatternc[64+i] = x;
        longstrc[320+i] = x;
        z=x; x=y; y=z;
    }
    longstrpatternc[longstrpatternlen] = '\0';
    memcpy(longstrc+longstrlen-14, "wxyz1234567890", 15);

    cxmutstr longstr = cx_mutstrn(longstrc, longstrlen);
    cxstring longstrpattern = cx_strn(longstrpatternc, longstrpatternlen);
    cxmutstr longstrresult = cx_mutstrn(longstrc+256, longstrlen-256);

    CX_TEST_DO {
        cxstring notfound = cx_strstr(str, CX_STR("no match"));
        CX_TEST_ASSERT(notfound.length == 0);

        cxstring result = cx_strstr(str, CX_STR("match"));
        CX_TEST_ASSERT(result.length == 20);
        CX_TEST_ASSERT(0 == strcmp(result.ptr, "match in this string"));

        result = cx_strstr(str, CX_STR(""));
        CX_TEST_ASSERT(result.length == str.length);
        CX_TEST_ASSERT(0 == strcmp(result.ptr, str.ptr));

        cxmutstr resultm = cx_strstr_m(longstr, longstrpattern);
        CX_TEST_ASSERT(resultm.length == longstrresult.length);
        CX_TEST_ASSERT(0 == strcmp(resultm.ptr, longstrresult.ptr));
    }

    free(longstrc);
    free(longstrpatternc);
}

CX_TEST(test_strcmp) {
    cxstring str = CX_STR("compare this");
    CX_TEST_DO {
        CX_TEST_ASSERT(0 == cx_strcmp(CX_STR(""), CX_STR("")));
        CX_TEST_ASSERT(0 < cx_strcmp(str, CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(str, CX_STR("compare this")));
        CX_TEST_ASSERT(0 != cx_strcmp(str, CX_STR("Compare This")));
        CX_TEST_ASSERT(0 > cx_strcmp(str, CX_STR("compare tool")));
        CX_TEST_ASSERT(0 < cx_strcmp(str, CX_STR("compare shit")));
        CX_TEST_ASSERT(0 > cx_strcmp(str, CX_STR("compare this not")));
        CX_TEST_ASSERT(0 < cx_strcmp(str, CX_STR("compare")));

        cxstring str2 = CX_STR("Compare This");
        CX_TEST_ASSERT(0 != cx_strcmp_p(&str, &str2));
        str2 = CX_STR("compare this");
        CX_TEST_ASSERT(0 == cx_strcmp_p(&str, &str2));
    }
}

CX_TEST(test_strcasecmp) {
    cxstring str = CX_STR("compare this");
    CX_TEST_DO {
        CX_TEST_ASSERT(0 == cx_strcasecmp(CX_STR(""), CX_STR("")));
        CX_TEST_ASSERT(0 < cx_strcasecmp(str, CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcasecmp(str, CX_STR("compare this")));
        CX_TEST_ASSERT(0 == cx_strcasecmp(str, CX_STR("Compare This")));
        CX_TEST_ASSERT(0 > cx_strcasecmp(str, CX_STR("compare tool")));
        CX_TEST_ASSERT(0 < cx_strcasecmp(str, CX_STR("compare shit")));
        CX_TEST_ASSERT(0 > cx_strcasecmp(str, CX_STR("compare this not")));
        CX_TEST_ASSERT(0 < cx_strcasecmp(str, CX_STR("compare")));

        cxstring str2 = CX_STR("Compare This");
        CX_TEST_ASSERT(0 == cx_strcasecmp_p(&str, &str2));
        str2 = CX_STR("Compare Tool");
        CX_TEST_ASSERT(0 > cx_strcasecmp_p(&str, &str2));
    }
}

CX_TEST(test_strcat) {
    cxstring s1 = CX_STR("12");
    cxstring s2 = CX_STR("34");
    cxstring s3 = CX_STR("56");
    cxstring sn = {NULL, 0};

    CxTestingAllocator talloc;
    cx_testing_allocator_init(&talloc);
    CxAllocator *alloc = &talloc.base;

    CX_TEST_DO {
        cxmutstr t1 = cx_strcat_a(alloc, 2, s1, s2);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(t1), CX_STR("1234")));
        ASSERT_ZERO_TERMINATED(t1);
        cx_strfree_a(alloc, &t1);

        cxmutstr t2 = cx_strcat_a(alloc, 3, s1, s2, s3);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(t2), CX_STR("123456")));
        ASSERT_ZERO_TERMINATED(t2);
        cx_strfree_a(alloc, &t2);

        cxmutstr t3 = cx_strcat_a(alloc, 6, s1, sn, s2, sn, s3, sn);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(t3), CX_STR("123456")));
        ASSERT_ZERO_TERMINATED(t3);
        cx_strfree_a(alloc, &t3);

        cxmutstr t4 = cx_strcat_a(alloc, 2, sn, sn);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(t4), CX_STR("")));
        ASSERT_ZERO_TERMINATED(t4);
        cx_strfree_a(alloc, &t4);

        CX_TEST_ASSERT(cx_testing_allocator_verify(&talloc));

        // use the macro
        cxmutstr t5 = cx_strcat(3, s3, s1, s2);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(t5), CX_STR("561234")));
        ASSERT_ZERO_TERMINATED(t5);
        cx_strfree(&t5);

        // use an initial string
        cxmutstr t6 = cx_strdup(CX_STR("Hello"));
        t6 = cx_strcat_m(t6, 2, CX_STR(", "), CX_STR("World!"));
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(t6), CX_STR("Hello, World!")));
        ASSERT_ZERO_TERMINATED(t6);
        cx_strfree(&t6);
    }
    cx_testing_allocator_destroy(&talloc);
}

CX_TEST(test_strcat_more_than_eight) {
    cxstring s1 = CX_STR("12");
    cxstring s2 = CX_STR("34");
    cxstring s3 = CX_STR("56");
    cxstring s4 = CX_STR("78");
    cxstring s5 = CX_STR("9a");
    cxstring s6 = CX_STR("bc");
    cxstring s7 = CX_STR("de");
    cxstring s8 = CX_STR("f0");
    cxstring s9 = CX_STR("xy");

    CX_TEST_DO {
        cxmutstr r = cx_strcat(9, s1, s2, s3, s4, s5, s6, s7, s8, s9);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(r), CX_STR("123456789abcdef0xy")));
        ASSERT_ZERO_TERMINATED(r);
        cx_strfree(&r);
    }
}

CX_TEST(test_strsplit) {
    cxstring test = CX_STR("this,is,a,csv,string");
    size_t capa = 8;
    cxstring list[8];
    size_t n;
    CX_TEST_DO {
        // special case: empty string
        n = cx_strsplit(test, CX_STR(""), capa, list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));

        // no delimiter occurrence
        n = cx_strsplit(test, CX_STR("z"), capa, list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));

        // partially matching delimiter
        n = cx_strsplit(test, CX_STR("is,not"), capa, list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));

        // matching single-char delimiter
        n = cx_strsplit(test, CX_STR(","), capa, list);
        CX_TEST_ASSERT(n == 5);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("this")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("is")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("a")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[3], CX_STR("csv")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[4], CX_STR("string")));

        // matching multi-char delimiter
        n = cx_strsplit(test, CX_STR("is"), capa, list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR(",")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR(",a,csv,string")));

        // bounded list using single-char delimiter
        n = cx_strsplit(test, CX_STR(","), 3, list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("this")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("is")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("a,csv,string")));

        // bounded list using multi-char delimiter
        n = cx_strsplit(test, CX_STR("is"), 2, list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR(",is,a,csv,string")));

        // start with delimiter
        n = cx_strsplit(test, CX_STR("this"), capa, list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR(",is,a,csv,string")));

        // end with delimiter
        n = cx_strsplit(test, CX_STR("string"), capa, list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("this,is,a,csv,")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("")));


        // end with delimiter exceed bound
        n = cx_strsplit(CX_STR("a,b,c,"), CX_STR(","), 3, list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("a")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("b")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("c,")));

        // exact match
        n = cx_strsplit(test, CX_STR("this,is,a,csv,string"), capa, list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("")));

        // string to be split is only substring
        n = cx_strsplit(test, CX_STR("this,is,a,csv,string,with,extension"), capa, list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));

        // subsequent encounter of delimiter (the string between is empty)
        n = cx_strsplit(test, CX_STR("is,"), capa, list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("a,csv,string")));

        // call the _m variant just for coverage
        cxmutstr mtest = cx_strdup(test);
        cxmutstr mlist[4];
        n = cx_strsplit_m(mtest, CX_STR("is,"), 4, mlist);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(mlist[0]), CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(mlist[1]), CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(mlist[2]), CX_STR("a,csv,string")));
        cx_strfree(&mtest);
    }
}

CX_TEST(test_strsplit_a) {
    CxTestingAllocator talloc;
    cx_testing_allocator_init(&talloc);
    CxAllocator *alloc = &talloc.base;

    cxstring test = CX_STR("this,is,a,csv,string");
    size_t capa = 8;
    cxstring *list;
    size_t n;
    CX_TEST_DO {
        // special case: empty string
        n = cx_strsplit_a(alloc, test, CX_STR(""), capa, &list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));
        cxFree(alloc, list);

        // no delimiter occurrence
        n = cx_strsplit_a(alloc, test, CX_STR("z"), capa, &list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));
        cxFree(alloc, list);

        // partially matching delimiter
        n = cx_strsplit_a(alloc, test, CX_STR("is,not"), capa, &list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));
        cxFree(alloc, list);

        // matching single-char delimiter
        n = cx_strsplit_a(alloc, test, CX_STR(","), capa, &list);
        CX_TEST_ASSERT(n == 5);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("this")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("is")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("a")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[3], CX_STR("csv")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[4], CX_STR("string")));
        cxFree(alloc, list);

        // matching multi-char delimiter
        n = cx_strsplit_a(alloc, test, CX_STR("is"), capa, &list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR(",")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR(",a,csv,string")));
        cxFree(alloc, list);

        // bounded list using single-char delimiter
        n = cx_strsplit_a(alloc, test, CX_STR(","), 3, &list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("this")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("is")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("a,csv,string")));
        cxFree(alloc, list);

        // bounded list using multi-char delimiter
        n = cx_strsplit_a(alloc, test, CX_STR("is"), 2, &list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR(",is,a,csv,string")));
        cxFree(alloc, list);

        // start with delimiter
        n = cx_strsplit_a(alloc, test, CX_STR("this"), capa, &list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR(",is,a,csv,string")));
        cxFree(alloc, list);

        // end with delimiter
        n = cx_strsplit_a(alloc, test, CX_STR("string"), capa, &list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("this,is,a,csv,")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("")));
        cxFree(alloc, list);

        // end with delimiter exceed bound
        n = cx_strsplit_a(alloc, CX_STR("a,b,c,"), CX_STR(","), 3, &list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("a")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("b")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("c,")));
        cxFree(alloc, list);

        // exact match
        n = cx_strsplit_a(alloc, test, CX_STR("this,is,a,csv,string"), capa, &list);
        CX_TEST_ASSERT(n == 2);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("")));
        cxFree(alloc, list);

        // string to be split is only substring
        n = cx_strsplit_a(alloc, test, CX_STR("this,is,a,csv,string,with,extension"), capa, &list);
        CX_TEST_ASSERT(n == 1);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], test));
        cxFree(alloc, list);

        // subsequent encounter of delimiter (the string between is empty)
        n = cx_strsplit_a(alloc, test, CX_STR("is,"), capa, &list);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(list[0], CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[1], CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(list[2], CX_STR("a,csv,string")));
        cxFree(alloc, list);

        // call the _m variant just for coverage
        cxmutstr mtest = cx_strdup(test);
        cxmutstr *mlist;
        n = cx_strsplit_ma(alloc, mtest, CX_STR("is,"), 4, &mlist);
        CX_TEST_ASSERT(n == 3);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(mlist[0]), CX_STR("th")));
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(mlist[1]), CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(mlist[2]), CX_STR("a,csv,string")));
        cxFree(alloc, mlist);
        cx_strfree(&mtest);

        CX_TEST_ASSERT(cx_testing_allocator_verify(&talloc));
    }
    cx_testing_allocator_destroy(&talloc);
}

CX_TEST(test_strtrim) {
    cxstring t1 = cx_strtrim(CX_STR("  ein test  \t "));
    cxstring t2 = cx_strtrim(CX_STR("abc"));
    cxstring t3 = cx_strtrim(CX_STR(" 123"));
    cxstring t4 = cx_strtrim(CX_STR("xyz "));
    cxstring t5 = cx_strtrim(CX_STR("   "));
    cxstring empty = cx_strtrim(CX_STR(""));

    CX_TEST_DO {
        CX_TEST_ASSERT(0 == cx_strcmp(t1, CX_STR("ein test")));
        CX_TEST_ASSERT(0 == cx_strcmp(t2, CX_STR("abc")));
        CX_TEST_ASSERT(0 == cx_strcmp(t3, CX_STR("123")));
        CX_TEST_ASSERT(0 == cx_strcmp(t4, CX_STR("xyz")));
        CX_TEST_ASSERT(0 == cx_strcmp(t5, CX_STR("")));
        CX_TEST_ASSERT(0 == cx_strcmp(empty, CX_STR("")));

        // call the _m variant just for coverage
        cxmutstr m1 = cx_strtrim_m(cx_mutstr((char *) "  ein test  \t "));
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(m1), CX_STR("ein test")));
    }
}

CX_TEST(test_strprefix) {
    cxstring str = CX_STR("test my prefix and my suffix");
    cxstring empty = CX_STR("");
    CX_TEST_DO {
        CX_TEST_ASSERT(!cx_strprefix(empty, CX_STR("pref")));
        CX_TEST_ASSERT(cx_strprefix(str, empty));
        CX_TEST_ASSERT(cx_strprefix(empty, empty));
        CX_TEST_ASSERT(cx_strprefix(str, CX_STR("test ")));
        CX_TEST_ASSERT(!cx_strprefix(str, CX_STR("8-) fsck ")));
    }
}

CX_TEST(test_strsuffix) {
    cxstring str = CX_STR("test my prefix and my suffix");
    cxstring empty = CX_STR("");
    CX_TEST_DO {
        CX_TEST_ASSERT(!cx_strsuffix(empty, CX_STR("suf")));
        CX_TEST_ASSERT(cx_strsuffix(str, empty));
        CX_TEST_ASSERT(cx_strsuffix(empty, empty));
        CX_TEST_ASSERT(cx_strsuffix(str, CX_STR("fix")));
        CX_TEST_ASSERT(!cx_strsuffix(str, CX_STR("fox")));
    }
}

CX_TEST(test_strcaseprefix) {
    cxstring str = CX_STR("test my prefix and my suffix");
    cxstring empty = CX_STR("");
    CX_TEST_DO {
        CX_TEST_ASSERT(!cx_strcaseprefix(empty, CX_STR("pREf")));
        CX_TEST_ASSERT(cx_strcaseprefix(str, empty));
        CX_TEST_ASSERT(cx_strcaseprefix(empty, empty));
        CX_TEST_ASSERT(cx_strcaseprefix(str, CX_STR("TEST ")));
        CX_TEST_ASSERT(!cx_strcaseprefix(str, CX_STR("8-) fsck ")));
    }
}

CX_TEST(test_strcasesuffix) {
    cxstring str = CX_STR("test my prefix and my suffix");
    cxstring empty = CX_STR("");
    CX_TEST_DO {
        CX_TEST_ASSERT(!cx_strcasesuffix(empty, CX_STR("sUf")));
        CX_TEST_ASSERT(cx_strcasesuffix(str, empty));
        CX_TEST_ASSERT(cx_strcasesuffix(empty, empty));
        CX_TEST_ASSERT(cx_strcasesuffix(str, CX_STR("FIX")));
        CX_TEST_ASSERT(!cx_strcasesuffix(str, CX_STR("fox")));
    }
}

CX_TEST(test_strreplace) {
    CxTestingAllocator talloc;
    cx_testing_allocator_init(&talloc);
    CxAllocator *alloc = &talloc.base;

    cxstring str = CX_STR("test ababab string aba");
    cxstring longstr = CX_STR(
            "xyaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaacd");
    cxstring notrail = CX_STR("test abab");
    cxstring empty = CX_STR("");
    cxstring astr = CX_STR("aaaaaaaaaa");
    cxstring csstr = CX_STR("test AB ab TEST xyz");

    cxmutstr repl = cx_strreplace(str, CX_STR("abab"), CX_STR("muchlonger"));
    const char *expected = "test muchlongerab string aba";

    cxmutstr repln = cx_strreplacen(str, CX_STR("ab"), CX_STR("c"), 2);
    const char *expectedn = "test ccab string aba";

    cxmutstr longrepl = cx_strreplace(longstr, CX_STR("a"), CX_STR("z"));
    const char *longexpect = "xyzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzcd";

    cxmutstr replnotrail = cx_strreplace(notrail, CX_STR("ab"), CX_STR("z"));
    const char *notrailexpect = "test zz";

    cxmutstr repleq = cx_strreplace(str, str, CX_STR("hello"));
    const char *eqexpect = "hello";

    cxmutstr replempty1 = cx_strreplace(empty, CX_STR("ab"), CX_STR("c")); // expect: empty
    cxmutstr replempty2 = cx_strreplace(str, CX_STR("abab"), empty);
    const char *emptyexpect2 = "test ab string aba";

    cxmutstr replpre = cx_strreplace(str, CX_STR("test "), CX_STR("TEST "));
    const char *preexpected = "TEST ababab string aba";

    cxmutstr replan1 = cx_strreplacen(astr, CX_STR("a"), CX_STR("x"), 1);
    const char *an1expected = "xaaaaaaaaa";

    cxmutstr replan4 = cx_strreplacen(astr, CX_STR("a"), CX_STR("x"), 4);
    const char *an4expected = "xxxxaaaaaa";

    cxmutstr replan9 = cx_strreplacen(astr, CX_STR("a"), CX_STR("x"), 9);
    const char *an9expected = "xxxxxxxxxa";

    cxmutstr replan10 = cx_strreplacen(astr, CX_STR("a"), CX_STR("x"), 10);
    const char *an10expected = "xxxxxxxxxx";

    CX_TEST_DO {
        cxmutstr repl1_a = cx_strreplace_a(alloc, csstr, CX_STR("AB"), CX_STR("*"));
        const char *expeced1_a = "test * ab TEST xyz";

        cxmutstr repl2_a = cx_strreplace_a(alloc, csstr, CX_STR("test"), CX_STR("TEST"));
        const char *expected2_a = "TEST AB ab TEST xyz";

        CX_TEST_ASSERT(repl.ptr != str.ptr);
        ASSERT_ZERO_TERMINATED(repl);
        CX_TEST_ASSERT(0 == strcmp(repl.ptr, expected));
        ASSERT_ZERO_TERMINATED(repln);
        CX_TEST_ASSERT(0 == strcmp(repln.ptr, expectedn));
        ASSERT_ZERO_TERMINATED(longrepl);
        CX_TEST_ASSERT(0 == strcmp(longrepl.ptr, longexpect));
        ASSERT_ZERO_TERMINATED(replnotrail);
        CX_TEST_ASSERT(0 == strcmp(replnotrail.ptr, notrailexpect));
        ASSERT_ZERO_TERMINATED(repleq);
        CX_TEST_ASSERT(0 == strcmp(repleq.ptr, eqexpect));
        ASSERT_ZERO_TERMINATED(replempty1);
        CX_TEST_ASSERT(0 == strcmp(replempty1.ptr, ""));
        ASSERT_ZERO_TERMINATED(replempty2);
        CX_TEST_ASSERT(0 == strcmp(replempty2.ptr, emptyexpect2));
        ASSERT_ZERO_TERMINATED(replpre);
        CX_TEST_ASSERT(0 == strcmp(replpre.ptr, preexpected));
        ASSERT_ZERO_TERMINATED(replan1);
        CX_TEST_ASSERT(0 == strcmp(replan1.ptr, an1expected));
        ASSERT_ZERO_TERMINATED(replan4);
        CX_TEST_ASSERT(0 == strcmp(replan4.ptr, an4expected));
        ASSERT_ZERO_TERMINATED(replan9);
        CX_TEST_ASSERT(0 == strcmp(replan9.ptr, an9expected));
        ASSERT_ZERO_TERMINATED(replan10);
        CX_TEST_ASSERT(0 == strcmp(replan10.ptr, an10expected));
        ASSERT_ZERO_TERMINATED(repl1_a);
        CX_TEST_ASSERT(0 == strcmp(repl1_a.ptr, expeced1_a));
        ASSERT_ZERO_TERMINATED(repl2_a);
        CX_TEST_ASSERT(0 == strcmp(repl2_a.ptr, expected2_a));

        cx_strfree_a(alloc, &repl1_a);
        cx_strfree_a(alloc, &repl2_a);
        CX_TEST_ASSERT(cx_testing_allocator_verify(&talloc));
    }

    cx_strfree(&repl);
    cx_strfree(&repln);
    cx_strfree(&longrepl);
    cx_strfree(&replnotrail);
    cx_strfree(&repleq);
    cx_strfree(&replempty1);
    cx_strfree(&replempty2);
    cx_strfree(&replpre);
    cx_strfree(&replan1);
    cx_strfree(&replan4);
    cx_strfree(&replan9);
    cx_strfree(&replan10);
    cx_testing_allocator_destroy(&talloc);
}

CX_TEST(test_strupper) {
    cxmutstr str = cx_strdup(CX_STR("thIs 1s @ Te$t"));
    CX_TEST_DO {
        cx_strupper(str);
        CX_TEST_ASSERT(0 == strcmp(str.ptr, "THIS 1S @ TE$T"));
    }
    cx_strfree(&str);
}

CX_TEST(test_strlower) {
    cxmutstr str = cx_strdup(CX_STR("thIs 1s @ Te$t"));
    CX_TEST_DO {
        cx_strlower(str);
        CX_TEST_ASSERT(0 == strcmp(str.ptr, "this 1s @ te$t"));
    }
    cx_strfree(&str);
}

CX_TEST(test_strtok) {
    cxstring str = CX_STR("a,comma,separated,string");
    cxstring delim = CX_STR(",");
    CX_TEST_DO {
        CxStrtokCtx ctx = cx_strtok(str, delim, 3);
        CX_TEST_ASSERT(ctx.str.ptr == str.ptr);
        CX_TEST_ASSERT(ctx.str.length == str.length);
        CX_TEST_ASSERT(ctx.delim.ptr == delim.ptr);
        CX_TEST_ASSERT(ctx.delim.length == delim.length);
        CX_TEST_ASSERT(ctx.limit == 3);
        CX_TEST_ASSERT(ctx.found == 0);
        CX_TEST_ASSERT(ctx.pos == 0);
        CX_TEST_ASSERT(ctx.next_pos == 0);
        CX_TEST_ASSERT(ctx.delim_more == NULL);
        CX_TEST_ASSERT(ctx.delim_more_count == 0);
    }
}

CX_TEST(test_strtok_m) {
    cxmutstr str = cx_strdup(CX_STR("a,comma,separated,string"));
    cxstring delim = CX_STR(",");
    CX_TEST_DO {
        CxStrtokCtx ctx = cx_strtok_m(str, delim, 3);
        CX_TEST_ASSERT(ctx.str.ptr == str.ptr);
        CX_TEST_ASSERT(ctx.str.length == str.length);
        CX_TEST_ASSERT(ctx.delim.ptr == delim.ptr);
        CX_TEST_ASSERT(ctx.delim.length == delim.length);
        CX_TEST_ASSERT(ctx.limit == 3);
        CX_TEST_ASSERT(ctx.found == 0);
        CX_TEST_ASSERT(ctx.pos == 0);
        CX_TEST_ASSERT(ctx.next_pos == 0);
        CX_TEST_ASSERT(ctx.delim_more == NULL);
        CX_TEST_ASSERT(ctx.delim_more_count == 0);
    }
    cx_strfree(&str);
}

CX_TEST(test_strtok_delim) {
    cxstring str = CX_STR("an,arbitrarily|separated;string");
    cxstring delim = CX_STR(",");
    cxstring delim_more[2] = {CX_STR("|"), CX_STR(";")};
    CX_TEST_DO {
        CxStrtokCtx ctx = cx_strtok(str, delim, 3);
        cx_strtok_delim(&ctx, delim_more, 2);
        CX_TEST_ASSERT(ctx.str.ptr == str.ptr);
        CX_TEST_ASSERT(ctx.str.length == str.length);
        CX_TEST_ASSERT(ctx.delim.ptr == delim.ptr);
        CX_TEST_ASSERT(ctx.delim.length == delim.length);
        CX_TEST_ASSERT(ctx.limit == 3);
        CX_TEST_ASSERT(ctx.found == 0);
        CX_TEST_ASSERT(ctx.pos == 0);
        CX_TEST_ASSERT(ctx.next_pos == 0);
        CX_TEST_ASSERT(ctx.delim_more == delim_more);
        CX_TEST_ASSERT(ctx.delim_more_count == 2);
    }
}

CX_TEST(test_strtok_next_easy) {
    cxstring str = CX_STR("a,comma,separated,string");
    cxstring delim = CX_STR(",");
    CX_TEST_DO {
        CxStrtokCtx ctx = cx_strtok(str, delim, 3);
        bool ret;
        cxstring tok;

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("a")));
        CX_TEST_ASSERT(ctx.pos == 0);
        CX_TEST_ASSERT(ctx.next_pos == 2);
        CX_TEST_ASSERT(ctx.delim_pos == 1);
        CX_TEST_ASSERT(ctx.found == 1);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("comma")));
        CX_TEST_ASSERT(ctx.pos == 2);
        CX_TEST_ASSERT(ctx.next_pos == 8);
        CX_TEST_ASSERT(ctx.delim_pos == 7);
        CX_TEST_ASSERT(ctx.found == 2);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("separated")));
        CX_TEST_ASSERT(ctx.pos == 8);
        CX_TEST_ASSERT(ctx.next_pos == 18);
        CX_TEST_ASSERT(ctx.delim_pos == 17);
        CX_TEST_ASSERT(ctx.found == 3);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(!ret);
        CX_TEST_ASSERT(ctx.pos == 8);
        CX_TEST_ASSERT(ctx.next_pos == 18);
        CX_TEST_ASSERT(ctx.delim_pos == 17);
        CX_TEST_ASSERT(ctx.found == 3);
    }
}

CX_TEST(test_strtok_next_unlimited) {
    cxstring str = CX_STR("some;-;otherwise;-;separated;-;string;-;");
    cxstring delim = CX_STR(";-;");
    CX_TEST_DO {
        CxStrtokCtx ctx = cx_strtok(str, delim, SIZE_MAX);
        bool ret;
        cxstring tok;

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("some")));
        CX_TEST_ASSERT(ctx.pos == 0);
        CX_TEST_ASSERT(ctx.next_pos == 7);
        CX_TEST_ASSERT(ctx.delim_pos == 4);
        CX_TEST_ASSERT(ctx.found == 1);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("otherwise")));
        CX_TEST_ASSERT(ctx.pos == 7);
        CX_TEST_ASSERT(ctx.next_pos == 19);
        CX_TEST_ASSERT(ctx.delim_pos == 16);
        CX_TEST_ASSERT(ctx.found == 2);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("separated")));
        CX_TEST_ASSERT(ctx.pos == 19);
        CX_TEST_ASSERT(ctx.next_pos == 31);
        CX_TEST_ASSERT(ctx.delim_pos == 28);
        CX_TEST_ASSERT(ctx.found == 3);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("string")));
        CX_TEST_ASSERT(ctx.pos == 31);
        CX_TEST_ASSERT(ctx.next_pos == 40);
        CX_TEST_ASSERT(ctx.delim_pos == 37);
        CX_TEST_ASSERT(ctx.found == 4);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(tok, CX_STR("")));
        CX_TEST_ASSERT(ctx.pos == 40);
        CX_TEST_ASSERT(ctx.next_pos == 40);
        CX_TEST_ASSERT(ctx.delim_pos == 40);
        CX_TEST_ASSERT(ctx.found == 5);

        ret = cx_strtok_next(&ctx, &tok);
        CX_TEST_ASSERT(!ret);
        CX_TEST_ASSERT(ctx.pos == 40);
        CX_TEST_ASSERT(ctx.delim_pos == 40);
        CX_TEST_ASSERT(ctx.found == 5);
    }
}

CX_TEST(test_strtok_next_advanced) {
    cxmutstr str = cx_strdup(CX_STR("an,arbitrarily;||separated;string"));
    cxstring delim = CX_STR(",");
    cxstring delim_more[2] = {CX_STR("||"), CX_STR(";")};
    CX_TEST_DO {
        CxStrtokCtx ctx = cx_strtok_m(str, delim, 10);
        cx_strtok_delim(&ctx, delim_more, 2);
        bool ret;
        cxmutstr tok;

        ret = cx_strtok_next_m(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(tok), CX_STR("an")));
        CX_TEST_ASSERT(ctx.pos == 0);
        CX_TEST_ASSERT(ctx.next_pos == 3);
        CX_TEST_ASSERT(ctx.delim_pos == 2);
        CX_TEST_ASSERT(ctx.found == 1);
        cx_strupper(tok);

        ret = cx_strtok_next_m(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(tok), CX_STR("arbitrarily")));
        CX_TEST_ASSERT(ctx.pos == 3);
        CX_TEST_ASSERT(ctx.next_pos == 15);
        CX_TEST_ASSERT(ctx.delim_pos == 14);
        CX_TEST_ASSERT(ctx.found == 2);
        cx_strupper(tok);

        ret = cx_strtok_next_m(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(tok), CX_STR("")));
        CX_TEST_ASSERT(ctx.pos == 15);
        CX_TEST_ASSERT(ctx.next_pos == 17);
        CX_TEST_ASSERT(ctx.delim_pos == 15);
        CX_TEST_ASSERT(ctx.found == 3);
        cx_strupper(tok);

        ret = cx_strtok_next_m(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(tok), CX_STR("separated")));
        CX_TEST_ASSERT(ctx.pos == 17);
        CX_TEST_ASSERT(ctx.next_pos == 27);
        CX_TEST_ASSERT(ctx.delim_pos == 26);
        CX_TEST_ASSERT(ctx.found == 4);
        cx_strupper(tok);

        ret = cx_strtok_next_m(&ctx, &tok);
        CX_TEST_ASSERT(ret);
        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(tok), CX_STR("string")));
        CX_TEST_ASSERT(ctx.pos == 27);
        CX_TEST_ASSERT(ctx.next_pos == 33);
        CX_TEST_ASSERT(ctx.delim_pos == 33);
        CX_TEST_ASSERT(ctx.found == 5);
        cx_strupper(tok);

        ret = cx_strtok_next_m(&ctx, &tok);
        CX_TEST_ASSERT(!ret);
        CX_TEST_ASSERT(ctx.pos == 27);
        CX_TEST_ASSERT(ctx.next_pos == 33);
        CX_TEST_ASSERT(ctx.delim_pos == 33);
        CX_TEST_ASSERT(ctx.found == 5);

        CX_TEST_ASSERT(0 == cx_strcmp(cx_strcast(str), CX_STR("AN,ARBITRARILY;||SEPARATED;STRING")));
    }
    cx_strfree(&str);
}

#define test_strtoint_impl(suffix, num, base, var, min, max) \
    do { \
        errno = 0; \
        int r = cx_strto##var(cx_str( #num), &var, base); \
        if ((min) <= (num##suffix) && (num##suffix) <= (max)) { \
            CX_TEST_ASSERTM(0 == r, "failed for "#num); \
            CX_TEST_ASSERT(0 == errno); \
            CX_TEST_ASSERT((num##suffix) == (0##suffix)+(var)); \
        } else { \
            CX_TEST_ASSERTM(0 != r, "out-of-range not detected for "#num " in variant "#var); \
            CX_TEST_ASSERT(ERANGE == errno); \
        } \
    } while (0)

#define test_strtoint_rollout_signed_impl(num, base) \
    test_strtoint_impl(LL, num, base, s, SHRT_MIN, SHRT_MAX); \
    test_strtoint_impl(LL, num, base, i, INT_MIN, INT_MAX); \
    test_strtoint_impl(LL, num, base, l, LONG_MIN, LONG_MAX); \
    test_strtoint_impl(LL, num, base, ll, LLONG_MIN, LLONG_MAX); \
    test_strtoint_impl(LL, num, base, i8, INT8_MIN, INT8_MAX); \
    test_strtoint_impl(LL, num, base, i16, INT16_MIN, INT16_MAX); \
    test_strtoint_impl(LL, num, base, i32, INT32_MIN, INT32_MAX); \
    test_strtoint_impl(LL, num, base, i64, INT64_MIN, INT64_MAX); \
    test_strtoint_impl(LL, num, base, z, -SSIZE_MAX-1, SSIZE_MAX)

#define test_strtoint_rollout_signed(num, base) \
    test_strtoint_rollout_signed_impl(num, base); \
    test_strtoint_rollout_signed_impl(-num, base)

#define test_strtoint_rollout(num, base) \
    test_strtoint_impl(ULL, num, base, us, 0, USHRT_MAX); \
    test_strtoint_impl(ULL, num, base, u, 0, UINT_MAX); \
    test_strtoint_impl(ULL, num, base, ul, 0, ULONG_MAX); \
    test_strtoint_impl(ULL, num, base, ull, 0, ULLONG_MAX); \
    test_strtoint_impl(ULL, num, base, u8, 0, UINT8_MAX); \
    test_strtoint_impl(ULL, num, base, u16, 0, UINT16_MAX); \
    test_strtoint_impl(ULL, num, base, u32, 0, UINT32_MAX); \
    test_strtoint_impl(ULL, num, base, u64, 0, UINT64_MAX); \
    test_strtoint_impl(ULL, num, base, uz, 0, SIZE_MAX)

CX_TEST(test_string_to_signed_integer) {
    short s;
    int i;
    long l;
    long long ll;
    int8_t i8;
    int16_t i16;
    int32_t i32;
    int64_t i64;
    ssize_t z;
    CX_TEST_DO {
        // do some brute force tests with all ranges
        test_strtoint_rollout_signed(47, 10);
        test_strtoint_rollout_signed(210, 10);
        test_strtoint_rollout_signed(5678, 10);
        test_strtoint_rollout_signed(40678, 10);
        test_strtoint_rollout_signed(1350266537, 10);
        test_strtoint_rollout_signed(3350266537, 10);
        test_strtoint_rollout_signed(473350266537, 10);
        test_strtoint_rollout_signed(057, 8);
        test_strtoint_rollout_signed(0322, 8);
        test_strtoint_rollout_signed(013056, 8);
        test_strtoint_rollout_signed(0117346, 8);
        test_strtoint_rollout_signed(012036667251, 8);
        test_strtoint_rollout_signed(030754201251, 8);
        test_strtoint_rollout_signed(06706567757251, 8);
        test_strtoint_rollout_signed(0767716340165362204025, 8);
        test_strtoint_rollout_signed(0x65, 16);
        test_strtoint_rollout_signed(0xf5, 16);
        test_strtoint_rollout_signed(0xABC5, 16);
        test_strtoint_rollout_signed(0xFBC5, 16);
        test_strtoint_rollout_signed(0x6df9CE03, 16);
        test_strtoint_rollout_signed(0xFdf9CE03, 16);
        test_strtoint_rollout_signed(0x6df9CE03AbC90815, 16);
        // TODO: roll out base 2 tests, but that needs C23

        // do some special case tests
        // --------------------------

        // can fit only in unsigned long long
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtoll(cx_str("0x8df9CE03AbC90815"), &ll, 16));
        CX_TEST_ASSERT(errno == ERANGE);

        // edge case: only the sign bit is set
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtoi16(cx_str("0x8000"), &i16, 16));
        CX_TEST_ASSERT(errno == ERANGE);
        errno = 0;
        CX_TEST_ASSERT(0 == cx_strtoi16(cx_str("-0x8000"), &i16, 16));
        CX_TEST_ASSERT(errno == 0);
        CX_TEST_ASSERT(i16 == INT16_MIN);
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtoi64(cx_str("X8000000000000000"), &i64, 16));
        CX_TEST_ASSERT(errno == ERANGE);
        errno = 0;
        CX_TEST_ASSERT(0 == cx_strtoi64(cx_str("-X8000000000000000"), &i64, 16));
        CX_TEST_ASSERT(errno == 0);
        CX_TEST_ASSERT(i64 == INT64_MIN);

        // group separators
        CX_TEST_ASSERT(0 == cx_strtoi32(cx_str("  -123,456"), &i32, 10));
        CX_TEST_ASSERT(i32 == -123456);
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtoi16_lc(cx_str("  -Xab,cd"), &i16, 16, "'"));
        CX_TEST_ASSERT(errno == EINVAL);
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtoi16_lc(cx_str("  -X'ab'cd"), &i16, 16, "'"));
        CX_TEST_ASSERT(errno == ERANGE);
        errno = 0;
        CX_TEST_ASSERT(0 == cx_strtoi16_lc(cx_str("  -X'67'89"), &i16, 16, "'"));
        CX_TEST_ASSERT(errno == 0);
        CX_TEST_ASSERT(i16 == -0x6789);

        // binary and (unusual notation of) signed binary
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtoi8_lc(cx_str(" -1010 1011"), &i8, 2, " "));
        CX_TEST_ASSERT(errno == ERANGE);
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtoi8_lc(cx_str(" 1010 1011"), &i8, 2, " "));
        CX_TEST_ASSERT(errno == ERANGE);
        errno = 0;
        CX_TEST_ASSERT(0 == cx_strtoi8_lc(cx_str(" -0101 0101"), &i8, 2, " "));
        CX_TEST_ASSERT(errno == 0);
        CX_TEST_ASSERT(i8 == -0x55);
    }
}

CX_TEST(test_string_to_unsigned_integer) {
    unsigned short us;
    unsigned int u;
    unsigned long ul;
    unsigned long long ull;
    uint8_t u8;
    uint16_t u16;
    uint32_t u32;
    uint64_t u64;
    size_t uz;
    CX_TEST_DO {
        // do some brute force tests with all ranges
        test_strtoint_rollout(47, 10);
        test_strtoint_rollout(210, 10);
        test_strtoint_rollout(5678, 10);
        test_strtoint_rollout(40678, 10);
        test_strtoint_rollout(1350266537, 10);
        test_strtoint_rollout(3350266537, 10);
        test_strtoint_rollout(473350266537, 10);
        test_strtoint_rollout(057, 8);
        test_strtoint_rollout(0322, 8);
        test_strtoint_rollout(013056, 8);
        test_strtoint_rollout(0117346, 8);
        test_strtoint_rollout(012036667251, 8);
        test_strtoint_rollout(030754201251, 8);
        test_strtoint_rollout(06706567757251, 8);
        test_strtoint_rollout(01767716340165362204025, 8);
        test_strtoint_rollout(0x65, 16);
        test_strtoint_rollout(0xf5, 16);
        test_strtoint_rollout(0xABC5, 16);
        test_strtoint_rollout(0xFBC5, 16);
        test_strtoint_rollout(0x6df9CE03, 16);
        test_strtoint_rollout(0xFdf9CE03, 16);
        test_strtoint_rollout(0x6df9CE03AbC90815, 16);
        test_strtoint_rollout(0xfdf9CE03AbC90815, 16);
        // TODO: roll out base 2 tests, but that needs C23

        // do some special case tests
        // --------------------------

        // group separators
        CX_TEST_ASSERT(0 == cx_strtou32(cx_str("  123,456"), &u32, 10));
        CX_TEST_ASSERT(u32 == 123456);
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtou16_lc(cx_str("  ab,cd"), &u16, 16, "'"));
        CX_TEST_ASSERT(errno == EINVAL);
        errno = 0;
        CX_TEST_ASSERT(0 == cx_strtou16_lc(cx_str("  ab'cd"), &u16, 16, "'"));
        CX_TEST_ASSERT(errno == 0);
        CX_TEST_ASSERT(u16 == 0xabcd);

        // binary
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtou8_lc(cx_str("1 1010 1011"), &u8, 2, " "));
        CX_TEST_ASSERT(errno == ERANGE);
        errno = 0;
        CX_TEST_ASSERT(0 == cx_strtou8_lc(cx_str(" 1010 1011"), &u8, 2, " "));
        CX_TEST_ASSERT(errno == 0);
        CX_TEST_ASSERT(u8 == 0xAB);
    }
}

CX_TEST(test_string_to_float) {
    float f;
    CX_TEST_DO {
        CX_TEST_ASSERT(0 == cx_strtof(cx_str("11.3"), &f));
        CX_TEST_ASSERT(0 == cx_vcmp_float(11.3f, f));

        CX_TEST_ASSERT(0 == cx_strtof(cx_str("-4.711e+1"), &f));
        CX_TEST_ASSERT(0 == cx_vcmp_float(-47.11f, f));

        CX_TEST_ASSERT(0 == cx_strtof(cx_str("1.67262192595e-27"), &f));
        CX_TEST_ASSERT(0 == cx_vcmp_float(1.67262192595e-27f, f));

        CX_TEST_ASSERT(0 == cx_strtof_lc(cx_str("138,339.4"), &f, '.', ","));
        CX_TEST_ASSERT(0 == cx_vcmp_float(138339.4f, f));

        CX_TEST_ASSERT(0 == cx_strtof_lc(cx_str("138,339.4"), &f, ',', "."));
        CX_TEST_ASSERT(0 == cx_vcmp_float(138.3394f, f));

        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtof(cx_str("15e"), &f));
        CX_TEST_ASSERT(errno == EINVAL);
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtof(cx_str("15e+"), &f));
        CX_TEST_ASSERT(errno == EINVAL);
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtof(cx_str("15e-"), &f));
        CX_TEST_ASSERT(errno == EINVAL);
        CX_TEST_ASSERT(0 == cx_strtof(cx_str("15e-0"), &f));
        CX_TEST_ASSERT(0 == cx_vcmp_float(15.f, f));

        CX_TEST_ASSERT(0 == cx_strtof(cx_str("3e38"), &f));
        CX_TEST_ASSERT(0 == cx_vcmp_float(3e38f, f));
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtof(cx_str("3e39"), &f));
        CX_TEST_ASSERT(errno == ERANGE);
        CX_TEST_ASSERT(0 == cx_strtof(cx_str("-3e38"), &f));
        CX_TEST_ASSERT(0 == cx_vcmp_float(-3e38f, f));
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtof(cx_str("-3e39"), &f));
        CX_TEST_ASSERT(errno == ERANGE);
        CX_TEST_ASSERT(0 == cx_strtof(cx_str("1.18e-38"), &f));
        CX_TEST_ASSERT(0 == cx_vcmp_float(1.18e-38f, f));
        errno = 0;
        CX_TEST_ASSERT(0 != cx_strtof(cx_str("1.17e-38"), &f));
        CX_TEST_ASSERT(errno == ERANGE);
    }
}

CX_TEST(test_string_to_double) {
    double d;
    CX_TEST_DO {
        CX_TEST_ASSERT(0 == cx_strtod(cx_str("11.3"), &d));
        CX_TEST_ASSERT(0 == cx_vcmp_double(11.3, d));

        CX_TEST_ASSERT(0 == cx_strtod(cx_str("-13.37"), &d));
        CX_TEST_ASSERT(0 == cx_vcmp_double(-13.37, d));

        CX_TEST_ASSERT(0 == cx_strtod(cx_str("-4.711e+1"), &d));
        CX_TEST_ASSERT(0 == cx_vcmp_double(-47.11, d));

        CX_TEST_ASSERT(0 == cx_strtod(cx_str("1.67262192595e-27"), &d));
        CX_TEST_ASSERT(0 == cx_vcmp_double(1.67262192595e-27, d));

        CX_TEST_ASSERT(0 == cx_strtod_lc(cx_str("138,339.4"), &d, '.', ","));
        CX_TEST_ASSERT(0 == cx_vcmp_double(138339.4, d));

        CX_TEST_ASSERT(0 == cx_strtod_lc(cx_str("138,339.4"), &d, ',', "."));
        CX_TEST_ASSERT(0 == cx_vcmp_double(138.3394, d));

        // TODO: test and improve support for big numbers, precision, and out-of-range detection
    }
}

CxTestSuite *cx_test_suite_string(void) {
    CxTestSuite *suite = cx_test_suite_new("string");

    cx_test_register(suite, test_string_construct);
    cx_test_register(suite, test_strfree);
    cx_test_register(suite, test_strdup);
    cx_test_register(suite, test_strdup_shortened);
    cx_test_register(suite, test_strlen);
    cx_test_register(suite, test_strsubs);
    cx_test_register(suite, test_strchr);
    cx_test_register(suite, test_strrchr);
    cx_test_register(suite, test_strstr);
    cx_test_register(suite, test_strcmp);
    cx_test_register(suite, test_strcasecmp);
    cx_test_register(suite, test_strcat);
    cx_test_register(suite, test_strcat_more_than_eight);
    cx_test_register(suite, test_strsplit);
    cx_test_register(suite, test_strsplit_a);
    cx_test_register(suite, test_strtrim);
    cx_test_register(suite, test_strprefix);
    cx_test_register(suite, test_strsuffix);
    cx_test_register(suite, test_strcaseprefix);
    cx_test_register(suite, test_strcasesuffix);
    cx_test_register(suite, test_strreplace);
    cx_test_register(suite, test_strupper);
    cx_test_register(suite, test_strlower);
    cx_test_register(suite, test_strtok);
    cx_test_register(suite, test_strtok_m);
    cx_test_register(suite, test_strtok_delim);
    cx_test_register(suite, test_strtok_next_easy);
    cx_test_register(suite, test_strtok_next_unlimited);
    cx_test_register(suite, test_strtok_next_advanced);

    return suite;
}

CxTestSuite *cx_test_suite_string_to_number(void) {
    CxTestSuite *suite = cx_test_suite_new("string to number");

    cx_test_register(suite, test_string_to_signed_integer);
    cx_test_register(suite, test_string_to_unsigned_integer);
    cx_test_register(suite, test_string_to_float);
    cx_test_register(suite, test_string_to_double);

    return suite;
}