tests/tests_HTMLparser_htmlValidateUtf8.c

#include "unity/unity.h"
#include <libxml/HTMLparser.h>

#include <libxml/parserInternals.h> /* for xmlNewInputStream, xmlPushInput */
#include <libxml/parser.h>          /* for xmlInitParser, xmlCleanupParser */
#include <stdlib.h>
#include <string.h>

/* Wrapper provided in the module to access the static function */
int test_htmlValidateUtf8(xmlParserCtxtPtr ctxt, const xmlChar *str, size_t len, int partial);

/* Helpers to create and free a clean parser context with a valid input */
static xmlParserCtxtPtr make_ctxt(void) {
    htmlParserCtxtPtr hctxt = htmlNewParserCtxt();
    TEST_ASSERT_NOT_NULL(hctxt);
    xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr)hctxt;

    xmlParserInputPtr input = xmlNewInputStream(ctxt);
    TEST_ASSERT_NOT_NULL(input);
    xmlPushInput(ctxt, input);

    /* Ensure flags start at 0 for predictable testing */
    if (ctxt->input)
        ctxt->input->flags = 0;

    TEST_ASSERT_NOT_NULL(ctxt->input);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    return ctxt;
}

static void free_ctxt(xmlParserCtxtPtr ctxt) {
    if (ctxt != NULL) {
        htmlFreeParserCtxt((htmlParserCtxtPtr)ctxt);
    }
}

void setUp(void) {
    /* Nothing required per-test */
}

void tearDown(void) {
    /* Nothing required per-test */
}

/* ---------- Tests ---------- */

void test_htmlValidateUtf8_valid_2byte_returns_size_and_no_error_flag(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char seq[] = { 0xC2, 0xA2 }; /* U+00A2 */
    int ret = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq, sizeof(seq), 0);

    TEST_ASSERT_EQUAL_INT(2, ret);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    free_ctxt(ctxt);
}

void test_htmlValidateUtf8_invalid_start_byte_below_C2_sets_flag_and_returns_minus1(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char seq[] = { 0xC1, 0x80 }; /* Overlong lead byte -> invalid */
    unsigned int before = ctxt->input->flags;
    int ret = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq, sizeof(seq), 0);

    TEST_ASSERT_EQUAL_INT(-1, ret);
    TEST_ASSERT_TRUE(ctxt->input->flags != before);

    free_ctxt(ctxt);
}

void test_htmlValidateUtf8_invalid_2byte_continuation_sets_flag(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char seq[] = { 0xC2, 0x20 }; /* Bad continuation (not 10xxxxxx) */
    int ret = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq, sizeof(seq), 0);

    TEST_ASSERT_EQUAL_INT(-1, ret);
    TEST_ASSERT_NOT_EQUAL(0u, ctxt->input->flags);

    free_ctxt(ctxt);
}

void test_htmlValidateUtf8_incomplete_2byte_partial_returns_zero_no_flag(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char seq[] = { 0xC2 }; /* Incomplete */
    int ret = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq, sizeof(seq), 1);

    TEST_ASSERT_EQUAL_INT(0, ret);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    free_ctxt(ctxt);
}

void test_htmlValidateUtf8_incomplete_2byte_nonpartial_sets_flag_and_minus1(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char seq[] = { 0xC2 }; /* Incomplete */
    int ret = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq, sizeof(seq), 0);

    TEST_ASSERT_EQUAL_INT(-1, ret);
    TEST_ASSERT_NOT_EQUAL(0u, ctxt->input->flags);

    free_ctxt(ctxt);
}

void test_htmlValidateUtf8_valid_3byte_general_and_minimum(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char euro[] = { 0xE2, 0x82, 0xAC }; /* U+20AC */
    int r1 = test_htmlValidateUtf8(ctxt, (const xmlChar *)euro, sizeof(euro), 0);
    TEST_ASSERT_EQUAL_INT(3, r1);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    const unsigned char min3[] = { 0xE0, 0xA0, 0x80 }; /* U+0800 minimal 3-byte */
    int r2 = test_htmlValidateUtf8(ctxt, (const xmlChar *)min3, sizeof(min3), 0);
    TEST_ASSERT_EQUAL_INT(3, r2);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    /* 0xED with first cont < 0xA0 should be valid (below surrogates) */
    const unsigned char ed_valid[] = { 0xED, 0x9F, 0xBF }; /* U+D7FF */
    int r3 = test_htmlValidateUtf8(ctxt, (const xmlChar *)ed_valid, sizeof(ed_valid), 0);
    TEST_ASSERT_EQUAL_INT(3, r3);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    free_ctxt(ctxt);
}

void test_htmlValidateUtf8_invalid_3byte_overlong_and_surrogate(void) {
    /* Overlong: E0 80 80 */
    {
        xmlParserCtxtPtr ctxt = make_ctxt();
        const unsigned char overlong[] = { 0xE0, 0x80, 0x80 };
        int r = test_htmlValidateUtf8(ctxt, (const xmlChar *)overlong, sizeof(overlong), 0);
        TEST_ASSERT_EQUAL_INT(-1, r);
        TEST_ASSERT_NOT_EQUAL(0u, ctxt->input->flags);
        free_ctxt(ctxt);
    }
    /* Surrogate range: ED A0 80 */
    {
        xmlParserCtxtPtr ctxt = make_ctxt();
        const unsigned char surrogate[] = { 0xED, 0xA0, 0x80 }; /* U+D800 -> invalid */
        int r = test_htmlValidateUtf8(ctxt, (const xmlChar *)surrogate, sizeof(surrogate), 0);
        TEST_ASSERT_EQUAL_INT(-1, r);
        TEST_ASSERT_NOT_EQUAL(0u, ctxt->input->flags);
        free_ctxt(ctxt);
    }
}

void test_htmlValidateUtf8_valid_4byte_boundaries(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char min4[] = { 0xF0, 0x90, 0x80, 0x80 }; /* U+10000 */
    int r1 = test_htmlValidateUtf8(ctxt, (const xmlChar *)min4, sizeof(min4), 0);
    TEST_ASSERT_EQUAL_INT(4, r1);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    const unsigned char max4[] = { 0xF4, 0x8F, 0xBF, 0xBF }; /* U+10FFFF */
    int r2 = test_htmlValidateUtf8(ctxt, (const xmlChar *)max4, sizeof(max4), 0);
    TEST_ASSERT_EQUAL_INT(4, r2);
    TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);

    free_ctxt(ctxt);
}

void test_htmlValidateUtf8_invalid_4byte_overlong_and_above_max(void) {
    /* Overlong for F0: second byte too low */
    {
        xmlParserCtxtPtr ctxt = make_ctxt();
        const unsigned char overlong4[] = { 0xF0, 0x80, 0x80, 0x80 };
        int r = test_htmlValidateUtf8(ctxt, (const xmlChar *)overlong4, sizeof(overlong4), 0);
        TEST_ASSERT_EQUAL_INT(-1, r);
        TEST_ASSERT_NOT_EQUAL(0u, ctxt->input->flags);
        free_ctxt(ctxt);
    }
    /* Above Unicode max: F4 90 80 80 */
    {
        xmlParserCtxtPtr ctxt = make_ctxt();
        const unsigned char above_max[] = { 0xF4, 0x90, 0x80, 0x80 };
        int r = test_htmlValidateUtf8(ctxt, (const xmlChar *)above_max, sizeof(above_max), 0);
        TEST_ASSERT_EQUAL_INT(-1, r);
        TEST_ASSERT_NOT_EQUAL(0u, ctxt->input->flags);
        free_ctxt(ctxt);
    }
}

void test_htmlValidateUtf8_incomplete_3_and_4_byte_partial_behavior(void) {
    /* Incomplete 3-byte, partial=1 -> 0, no flag */
    {
        xmlParserCtxtPtr ctxt = make_ctxt();
        const unsigned char seq3[] = { 0xE2, 0x82 }; /* Missing 3rd byte */
        int r = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq3, sizeof(seq3), 1);
        TEST_ASSERT_EQUAL_INT(0, r);
        TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);
        free_ctxt(ctxt);
    }
    /* Incomplete 4-byte, partial=1 -> 0, no flag */
    {
        xmlParserCtxtPtr ctxt = make_ctxt();
        const unsigned char seq4[] = { 0xF0, 0x90, 0x80 }; /* Missing 4th byte */
        int r = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq4, sizeof(seq4), 1);
        TEST_ASSERT_EQUAL_INT(0, r);
        TEST_ASSERT_EQUAL_UINT(0u, ctxt->input->flags);
        free_ctxt(ctxt);
    }
    /* Incomplete 4-byte, partial=0 -> -1, sets flag */
    {
        xmlParserCtxtPtr ctxt = make_ctxt();
        const unsigned char seq4[] = { 0xF0, 0x90, 0x80 }; /* Missing 4th byte */
        int r = test_htmlValidateUtf8(ctxt, (const xmlChar *)seq4, sizeof(seq4), 0);
        TEST_ASSERT_EQUAL_INT(-1, r);
        TEST_ASSERT_NOT_EQUAL(0u, ctxt->input->flags);
        free_ctxt(ctxt);
    }
}

void test_htmlValidateUtf8_error_flag_only_set_once_per_context(void) {
    xmlParserCtxtPtr ctxt = make_ctxt();

    const unsigned char bad1[] = { 0xC1, 0x80 };      /* invalid */
    const unsigned char bad2[] = { 0xE0, 0x80, 0x80 };/* invalid */

    int r1 = test_htmlValidateUtf8(ctxt, (const xmlChar *)bad1, sizeof(bad1), 0);
    TEST_ASSERT_EQUAL_INT(-1, r1);
    unsigned int flags_after_first = ctxt->input->flags;
    TEST_ASSERT_NOT_EQUAL(0u, flags_after_first);

    int r2 = test_htmlValidateUtf8(ctxt, (const xmlChar *)bad2, sizeof(bad2), 0);
    TEST_ASSERT_EQUAL_INT(-1, r2);
    unsigned int flags_after_second = ctxt->input->flags;

    /* Should not clear or change the encoding error flag after the first error */
    TEST_ASSERT_EQUAL_UINT(flags_after_first, flags_after_second);

    free_ctxt(ctxt);
}

int main(void) {
    xmlInitParser();

    UNITY_BEGIN();
    RUN_TEST(test_htmlValidateUtf8_valid_2byte_returns_size_and_no_error_flag);
    RUN_TEST(test_htmlValidateUtf8_invalid_start_byte_below_C2_sets_flag_and_returns_minus1);
    RUN_TEST(test_htmlValidateUtf8_invalid_2byte_continuation_sets_flag);
    RUN_TEST(test_htmlValidateUtf8_incomplete_2byte_partial_returns_zero_no_flag);
    RUN_TEST(test_htmlValidateUtf8_incomplete_2byte_nonpartial_sets_flag_and_minus1);
    RUN_TEST(test_htmlValidateUtf8_valid_3byte_general_and_minimum);
    RUN_TEST(test_htmlValidateUtf8_invalid_3byte_overlong_and_surrogate);
    RUN_TEST(test_htmlValidateUtf8_valid_4byte_boundaries);
    RUN_TEST(test_htmlValidateUtf8_invalid_4byte_overlong_and_above_max);
    RUN_TEST(test_htmlValidateUtf8_incomplete_3_and_4_byte_partial_behavior);
    RUN_TEST(test_htmlValidateUtf8_error_flag_only_set_once_per_context);
    int rc = UNITY_END();

    xmlCleanupParser();
    return rc;
}