content-parser: minor fixes. added tests

2 files changed, 252 insertions, 14 deletions

diff --git a/src/content-parser.c b/src/content-parser.c
index 75c0f34..8249322 100644
--- a/src/content-parser.c
+++ b/src/content-parser.c
@@ -87,7 +87,7 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
         switch (state) {
 
             case CONTENT_START_LINE:
-                if (c == '\n' || c == '\r')
+                if (c == '\n' || c == '\r' || is_last)
                     break;
                 if (c == '#') {
                     header_level = 1;
@@ -140,18 +140,15 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
             case CONTENT_HEADER_TITLE_START:
                 if (c == ' ' || c == '\t')
                     break;
-                if (c != '\n' || c != '\r') {
-                    start = current;
+                start = current;
+                if (c != '\n' && c != '\r') {
                     state = CONTENT_HEADER_TITLE;
                     break;
                 }
-                *err = blogc_error_parser(BLOGC_ERROR_CONTENT_PARSER, src, src_len,
-                    current, "Empty header");
-                break;
 
             case CONTENT_HEADER_TITLE:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start, end - start);
                     b_string_append_printf(rv, "<h%d>%s</h%d>\n", header_level,
                         tmp, header_level);
@@ -159,14 +156,13 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
                     tmp = NULL;
                     state = CONTENT_START_LINE;
                     start = current;
-                    break;
                 }
                 break;
 
             case CONTENT_HTML:
                 if (c == '\n' || c == '\r' || is_last) {
                     state = CONTENT_HTML_END;
-                    end = is_last ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
                 }
                 if (!is_last)
                     break;
@@ -189,10 +185,11 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
                     break;
                 prefix = b_strndup(src + start, current - start);
                 state = CONTENT_BLOCKQUOTE_START;
+                break;
 
             case CONTENT_BLOCKQUOTE_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start, end - start);
                     if (b_str_starts_with(tmp, prefix)) {
                         lines = b_slist_append(lines, b_strdup(tmp + strlen(prefix)));
@@ -202,6 +199,10 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
                         *err = blogc_error_parser(BLOGC_ERROR_CONTENT_PARSER, src, src_len,
                             current, "Malformed blockquote, must use same prefix "
                             "as previous line(s): %s", prefix);
+                        free(prefix);
+                        prefix = NULL;
+                        b_slist_free_full(lines, free);
+                        lines = NULL;
                     }
                     free(tmp);
                     tmp = NULL;
@@ -245,10 +246,11 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
                     break;
                 prefix = b_strndup(src + start, current - start);
                 state = CONTENT_CODE_START;
+                break;
 
             case CONTENT_CODE_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start, end - start);
                     if (b_str_starts_with(tmp, prefix)) {
                         lines = b_slist_append(lines, b_strdup(tmp + strlen(prefix)));
@@ -258,6 +260,10 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
                         *err = blogc_error_parser(BLOGC_ERROR_CONTENT_PARSER, src, src_len,
                             current, "Malformed code block, must use same prefix "
                             "as previous line(s): '%s'", prefix);
+                        free(prefix);
+                        prefix = NULL;
+                        b_slist_free_full(lines, free);
+                        lines = NULL;
                     }
                     free(tmp);
                     tmp = NULL;
@@ -316,7 +322,7 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
 
             case CONTENT_UNORDERED_LIST_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start, end - start);
                     if (b_str_starts_with(tmp, prefix)) {
                         lines = b_slist_append(lines, b_strdup(tmp + strlen(prefix)));
@@ -367,10 +373,11 @@ blogc_content_parse(const char *src, size_t src_len, blogc_error_t **err)
                     break;
                 prefix_len = current - start;
                 state = CONTENT_ORDERED_LIST_START;
+                break;
 
             case CONTENT_ORDERED_LIST_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start, end - start);
                     if (strlen(tmp) >= prefix_len) {
                         tmp2 = b_strndup(tmp, prefix_len);
@@ -436,7 +443,7 @@ err_li:
             case CONTENT_PARAGRAPH:
                 if (c == '\n' || c == '\r' || is_last) {
                     state = CONTENT_PARAGRAPH_END;
-                    end = is_last ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
                 }
                 if (!is_last)
                     break;
diff --git a/tests/check_content_parser.c b/tests/check_content_parser.c
index df14df3..b3e7092 100644
--- a/tests/check_content_parser.c
+++ b/tests/check_content_parser.c
@@ -96,11 +96,242 @@ test_content_parse(void **state)
 }
 
 
+void
+test_content_parse_header(void **state)
+{
+    const char *a = "## bola";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html, "<h2>bola</h2>\n");
+    free(html);
+    a = "## bola\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html, "<h2>bola</h2>\n");
+    free(html);
+    a =
+        "bola\n"
+        "\n"
+        "## bola\n"
+        "\n"
+        "guda\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\n"
+        "<h2>bola</h2>\n"
+        "<p>guda</p>\n");
+    free(html);
+}
+
+
+void
+test_content_parse_html(void **state)
+{
+    const char *a = "<div>\n</div>";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html, "<div>\n</div>\n");
+    free(html);
+    a = "<div>\n</div>\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html, "<div>\n</div>\n");
+    free(html);
+    a =
+        "bola\n"
+        "\n"
+        "<div>\n"
+        "</div>\n"
+        "\n"
+        "chunda\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\n"
+        "<div>\n</div>\n"
+        "<p>chunda</p>\n");
+    free(html);
+}
+
+
+void
+test_content_parse_blockquote(void **state)
+{
+    const char *a = ">  bola\n>  guda";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<blockquote><p>bola\n"
+        "guda</p>\n"
+        "</blockquote>\n");
+    free(html);
+    a = ">  bola\n>  guda\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<blockquote><p>bola\n"
+        "guda</p>\n"
+        "</blockquote>\n");
+    free(html);
+    a =
+        "bola\n"
+        "\n"
+        ">   bola\n"
+        ">   guda\n"
+        "\n"
+        "chunda\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\n"
+        "<blockquote><p>bola\n"
+        "guda</p>\n"
+        "</blockquote>\n"
+        "<p>chunda</p>\n");
+    free(html);
+}
+
+
+void
+test_content_parse_code(void **state)
+{
+    const char *a = "  bola\n  guda";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<pre><code>bola\n"
+        "guda</code></pre>\n");
+    free(html);
+    a = "  bola\n  guda\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<pre><code>bola\n"
+        "guda</code></pre>\n");
+    free(html);
+    a =
+        "bola\n"
+        "\n"
+        "   bola\n"
+        "   guda\n"
+        "\n"
+        "chunda\n";
+    html = blogc_content_parse(a, strlen(a), &err);
+    assert_null(err);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\n"
+        "<pre><code>bola\n"
+        "guda</code></pre>\n"
+        "<p>chunda</p>\n");
+    free(html);
+}
+
+
+void
+test_content_parse_invalid_header(void **state)
+{
+    const char *a =
+        "asd\n"
+        "\n"
+        "##bola\n";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_non_null(err);
+    assert_null(html);
+    assert_int_equal(err->type, BLOGC_ERROR_CONTENT_PARSER);
+    assert_string_equal(err->msg,
+        "Malformed header, no space or tab after '#'\n"
+        "Error occurred near to 'bola'");
+    blogc_error_free(err);
+}
+
+
+void
+test_content_parse_invalid_header_empty(void **state)
+{
+    const char *a =
+        "asd\n"
+        "\n"
+        "##\n"
+        "\n"
+        "qwe\n";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_non_null(err);
+    assert_null(html);
+    assert_int_equal(err->type, BLOGC_ERROR_CONTENT_PARSER);
+    assert_string_equal(err->msg,
+        "Malformed header, no space or tab after '#'");
+    blogc_error_free(err);
+}
+
+
+void
+test_content_parse_invalid_blockquote(void **state)
+{
+    const char *a =
+        ">   asd\n"
+        "> bola\n"
+        ">   foo\n";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_non_null(err);
+    assert_null(html);
+    assert_int_equal(err->type, BLOGC_ERROR_CONTENT_PARSER);
+    assert_string_equal(err->msg,
+        "Malformed blockquote, must use same prefix as previous line(s): >   ");
+    blogc_error_free(err);
+}
+
+
+void
+test_content_parse_invalid_code(void **state)
+{
+    const char *a =
+        "    asd\n"
+        "  bola\n"
+        "    foo\n";
+    blogc_error_t *err = NULL;
+    char *html = blogc_content_parse(a, strlen(a), &err);
+    assert_non_null(err);
+    assert_null(html);
+    assert_int_equal(err->type, BLOGC_ERROR_CONTENT_PARSER);
+    assert_string_equal(err->msg,
+        "Malformed code block, must use same prefix as previous line(s): '    '");
+    blogc_error_free(err);
+}
+
+
 int
 main(void)
 {
     const UnitTest tests[] = {
         unit_test(test_content_parse),
+        unit_test(test_content_parse_header),
+        unit_test(test_content_parse_html),
+        unit_test(test_content_parse_blockquote),
+        unit_test(test_content_parse_code),
+        unit_test(test_content_parse_invalid_header),
+        unit_test(test_content_parse_invalid_header_empty),
+        unit_test(test_content_parse_invalid_blockquote),
+        unit_test(test_content_parse_invalid_code),
     };
     return run_tests(tests);
 }