content-parser: implemented multiline support for lists

3 files changed, 217 insertions, 51 deletions

diff --git a/src/content-parser.c b/src/content-parser.c
index e5bb16c..9052aec 100644
--- a/src/content-parser.c
+++ b/src/content-parser.c
@@ -17,7 +17,7 @@
 #include "content-parser.h"
 
 // this is a half ass implementation of a markdown-like syntax. bugs are
-// expected. feel free to improve the parser and and new features.
+// expected. feel free to improve the parser and add new features.
 
 
 typedef enum {
@@ -356,6 +356,33 @@ blogc_content_parse_inline(const char *src)
 }
 
 
+bool
+blogc_is_ordered_list_item(const char *str, size_t prefix_len)
+{
+    if (str == NULL)
+        return false;
+
+    if (strlen(str) < 2)
+        return false;
+
+    size_t i;
+
+    for (i = 0; str[i] >= '0' && str[i] <= '9'; i++);
+
+    if (i == 0)
+        return false;
+    if (str[i] != '.')
+        return false;
+
+    for (i++; i < prefix_len && (str[i] == ' ' || str[i] == '\t'); i++);
+
+    if (str[i] == '\0')
+        return false;
+
+    return i == prefix_len;
+}
+
+
 char*
 blogc_content_parse(const char *src, size_t *end_excerpt)
 {
@@ -373,13 +400,12 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
     size_t prefix_len = 0;
     char *tmp = NULL;
     char *tmp2 = NULL;
-    char *tmp3 = NULL;
     char *parsed = NULL;
-    char **tmpv = NULL;
 
     char d = '\0';
 
     b_slist_t *lines = NULL;
+    b_slist_t *lines2 = NULL;
 
     b_string_t *rv = b_string_new();
     b_string_t *tmp_str = NULL;
@@ -664,22 +690,39 @@ hr:
                 if (c == '\n' || c == '\r' || is_last) {
                     end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start2, end - start2);
+                    tmp2 = b_strdup_printf("%-*s", strlen(prefix), "");
                     if (b_str_starts_with(tmp, prefix)) {
-                        tmp3 = b_strdup(tmp + strlen(prefix));
-                        parsed = blogc_content_parse_inline(tmp3);
-                        free(tmp3);
-                        tmp3 = NULL;
-                        lines = b_slist_append(lines, b_strdup(parsed));
-                        free(parsed);
-                        parsed = NULL;
+                        if (lines2 != NULL) {
+                            tmp_str = b_string_new();
+                            for (b_slist_t *l = lines2; l != NULL; l = l->next) {
+                                if (l->next == NULL)
+                                    b_string_append_printf(tmp_str, "%s", l->data);
+                                else
+                                    b_string_append_printf(tmp_str, "%s\n", l->data);
+                            }
+                            b_slist_free_full(lines2, free);
+                            lines2 = NULL;
+                            parsed = blogc_content_parse_inline(tmp_str->str);
+                            b_string_free(tmp_str, true);
+                            lines = b_slist_append(lines, b_strdup(parsed));
+                            free(parsed);
+                            parsed = NULL;
+                        }
+                        lines2 = b_slist_append(lines2, b_strdup(tmp + strlen(prefix)));
+                    }
+                    else if (b_str_starts_with(tmp, tmp2)) {
+                        lines2 = b_slist_append(lines2, b_strdup(tmp + strlen(prefix)));
                     }
                     else {
                         state = CONTENT_PARAGRAPH_END;
                         free(tmp);
                         tmp = NULL;
+                        free(tmp2);
+                        tmp2 = NULL;
                         free(prefix);
                         prefix = NULL;
                         b_slist_free_full(lines, free);
+                        b_slist_free_full(lines2, free);
                         lines = NULL;
                         if (is_last)
                             goto para;
@@ -687,6 +730,8 @@ hr:
                     }
                     free(tmp);
                     tmp = NULL;
+                    free(tmp2);
+                    tmp2 = NULL;
                     state = CONTENT_UNORDERED_LIST_END;
                 }
                 if (!is_last)
@@ -694,6 +739,23 @@ hr:
 
             case CONTENT_UNORDERED_LIST_END:
                 if (c == '\n' || c == '\r' || is_last) {
+                    if (lines2 != NULL) {
+                        // FIXME: avoid repeting the code below
+                        tmp_str = b_string_new();
+                        for (b_slist_t *l = lines2; l != NULL; l = l->next) {
+                            if (l->next == NULL)
+                                b_string_append_printf(tmp_str, "%s", l->data);
+                            else
+                                b_string_append_printf(tmp_str, "%s\n", l->data);
+                        }
+                        b_slist_free_full(lines2, free);
+                        lines2 = NULL;
+                        parsed = blogc_content_parse_inline(tmp_str->str);
+                        b_string_free(tmp_str, true);
+                        lines = b_slist_append(lines, b_strdup(parsed));
+                        free(parsed);
+                        parsed = NULL;
+                    }
                     b_string_append(rv, "<ul>\n");
                     for (b_slist_t *l = lines; l != NULL; l = l->next)
                         b_string_append_printf(rv, "<li>%s</li>\n", l->data);
@@ -733,52 +795,72 @@ hr:
                 if (c == '\n' || c == '\r' || is_last) {
                     end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start2, end - start2);
-                    if (strlen(tmp) >= prefix_len) {
-                        tmp2 = b_strndup(tmp, prefix_len);
-                        tmpv = b_str_split(tmp2, '.', 2);
-                        free(tmp2);
-                        tmp2 = NULL;
-                        if (b_strv_length(tmpv) != 2) {
-                            state = CONTENT_PARAGRAPH_END;
-                            b_strv_free(tmpv);
-                            tmpv = NULL;
-                            free(tmp);
-                            tmp = NULL;
-                            b_slist_free_full(lines, free);
-                            lines = NULL;
-                            goto para;
-                        }
-                        for (unsigned int i = 0; tmpv[0][i] != '\0'; i++) {
-                            if (!(tmpv[0][i] >= '0' && tmpv[0][i] <= '9')) {
-                                state = CONTENT_PARAGRAPH_END;
-                                b_strv_free(tmpv);
-                                tmpv = NULL;
-                                free(tmp);
-                                tmp = NULL;
-                                b_slist_free_full(lines, free);
-                                lines = NULL;
-                                goto para;
+                    tmp2 = b_strdup_printf("%-*s", prefix_len, "");
+                    if (blogc_is_ordered_list_item(tmp, prefix_len)) {
+                        if (lines2 != NULL) {
+                            tmp_str = b_string_new();
+                            for (b_slist_t *l = lines2; l != NULL; l = l->next) {
+                                if (l->next == NULL)
+                                    b_string_append_printf(tmp_str, "%s", l->data);
+                                else
+                                    b_string_append_printf(tmp_str, "%s\n", l->data);
                             }
+                            b_slist_free_full(lines2, free);
+                            lines2 = NULL;
+                            parsed = blogc_content_parse_inline(tmp_str->str);
+                            b_string_free(tmp_str, true);
+                            lines = b_slist_append(lines, b_strdup(parsed));
+                            free(parsed);
+                            parsed = NULL;
                         }
-                        tmp3 = b_strdup(tmp + prefix_len);
-                        parsed = blogc_content_parse_inline(tmp3);
-                        free(tmp3);
-                        tmp3 = NULL;
-                        lines = b_slist_append(lines, b_strdup(parsed));
-                        state = CONTENT_ORDERED_LIST_END;
+                        lines2 = b_slist_append(lines2, b_strdup(tmp + prefix_len));
+                    }
+                    else if (b_str_starts_with(tmp, tmp2)) {
+                        lines2 = b_slist_append(lines2, b_strdup(tmp + prefix_len));
+                    }
+                    else {
+                        state = CONTENT_PARAGRAPH_END;
+                        free(tmp);
+                        tmp = NULL;
+                        free(tmp2);
+                        tmp2 = NULL;
                         free(parsed);
                         parsed = NULL;
-                        b_strv_free(tmpv);
-                        tmpv = NULL;
+                        b_slist_free_full(lines, free);
+                        b_slist_free_full(lines2, free);
+                        lines = NULL;
+                        if (is_last)
+                            goto para;
+                        break;
                     }
                     free(tmp);
                     tmp = NULL;
+                    free(tmp2);
+                    tmp2 = NULL;
+                    state = CONTENT_ORDERED_LIST_END;
                 }
-                if (state == CONTENT_PARAGRAPH || !is_last)
+                if (!is_last)
                     break;
 
             case CONTENT_ORDERED_LIST_END:
                 if (c == '\n' || c == '\r' || is_last) {
+                    if (lines2 != NULL) {
+                        // FIXME: avoid repeting the code below
+                        tmp_str = b_string_new();
+                        for (b_slist_t *l = lines2; l != NULL; l = l->next) {
+                            if (l->next == NULL)
+                                b_string_append_printf(tmp_str, "%s", l->data);
+                            else
+                                b_string_append_printf(tmp_str, "%s\n", l->data);
+                        }
+                        b_slist_free_full(lines2, free);
+                        lines2 = NULL;
+                        parsed = blogc_content_parse_inline(tmp_str->str);
+                        b_string_free(tmp_str, true);
+                        lines = b_slist_append(lines, b_strdup(parsed));
+                        free(parsed);
+                        parsed = NULL;
+                    }
                     b_string_append(rv, "<ol>\n");
                     for (b_slist_t *l = lines; l != NULL; l = l->next)
                         b_string_append_printf(rv, "<li>%s</li>\n", l->data);
diff --git a/src/content-parser.h b/src/content-parser.h
index b98769d..4579719 100644
--- a/src/content-parser.h
+++ b/src/content-parser.h
@@ -10,8 +10,10 @@
 #define _CONTENT_PARSER_H
 
 #include <stdlib.h>
+#include <stdbool.h>
 
 char* blogc_content_parse_inline(const char *src);
+bool blogc_is_ordered_list_item(const char *str, size_t prefix_len);
 char* blogc_content_parse(const char *src, size_t *end_excerpt);
 
 #endif /* _CONTENT_PARSER_H */
diff --git a/tests/check_content_parser.c b/tests/check_content_parser.c
index 1eb2aa5..b9f83b0 100644
--- a/tests/check_content_parser.c
+++ b/tests/check_content_parser.c
@@ -20,6 +20,24 @@
 
 
 static void
+test_is_ordered_list_item(void **state)
+{
+    assert_true(blogc_is_ordered_list_item("1.bola", 2));
+    assert_true(blogc_is_ordered_list_item("1. bola", 3));
+    assert_true(blogc_is_ordered_list_item("12. bola", 4));
+    assert_true(blogc_is_ordered_list_item("123. bola", 5));
+    assert_true(blogc_is_ordered_list_item("1.    bola", 6));
+    assert_true(blogc_is_ordered_list_item("1.    bola", 5));
+    assert_false(blogc_is_ordered_list_item("1bola", 1));
+    assert_false(blogc_is_ordered_list_item("12bola", 2));
+    assert_false(blogc_is_ordered_list_item("1 .   bola", 6));
+    assert_false(blogc_is_ordered_list_item("1.   bola", 6));
+    assert_false(blogc_is_ordered_list_item("1.", 2));
+    assert_false(blogc_is_ordered_list_item(NULL, 2));
+}
+
+
+static void
 test_content_parse(void **state)
 {
     size_t l = 0;
@@ -368,6 +386,40 @@ test_content_parse_unordered_list(void **state)
         "</ul>\n"
         "<p>fuuuu</p>\n");
     free(html);
+    html = blogc_content_parse(
+        "lol\n"
+        "\n"
+        "*  asd\n"
+        "   cvb\n"
+        "*  qwe\n"
+        "*  zxc\n"
+        "   1234\n"
+        "\n"
+        "fuuuu\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\n"
+        "<ul>\n"
+        "<li>asd\n"
+        "cvb</li>\n"
+        "<li>qwe</li>\n"
+        "<li>zxc\n"
+        "1234</li>\n"
+        "</ul>\n"
+        "<p>fuuuu</p>\n");
+    free(html);
+    html = blogc_content_parse(
+        "*  asd\n"
+        "*   qwe\n"
+        "*    zxc", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<ul>\n"
+        "<li>asd</li>\n"
+        "<li> qwe</li>\n"
+        "<li>  zxc</li>\n"
+        "</ul>\n");
+    free(html);
 }
 
 
@@ -423,21 +475,37 @@ test_content_parse_ordered_list(void **state)
         "<p>fuuuu</p>\n");
     free(html);
     html = blogc_content_parse(
-        "1.\nasd\n"
-        "2. qwe\n", NULL);
+        "lol\n"
+        "\n"
+        "1.  asd\n"
+        "    cvb\n"
+        "2.  qwe\n"
+        "3.  zxc\n"
+        "    1234\n"
+        "\n"
+        "fuuuu\n", NULL);
     assert_non_null(html);
     assert_string_equal(html,
-        "<p>1.\n"
-        "asd</p>\n"
+        "<p>lol</p>\n"
         "<ol>\n"
+        "<li>asd\n"
+        "cvb</li>\n"
         "<li>qwe</li>\n"
-        "</ol>\n");
+        "<li>zxc\n"
+        "1234</li>\n"
+        "</ol>\n"
+        "<p>fuuuu</p>\n");
     free(html);
-    html = blogc_content_parse("1.\n", NULL);
+    html = blogc_content_parse(
+        "1.  asd\n"
+        "2.   qwe\n"
+        "3.    zxc", NULL);
     assert_non_null(html);
     assert_string_equal(html,
         "<ol>\n"
-        "<li></li>\n"
+        "<li>asd</li>\n"
+        "<li> qwe</li>\n"
+        "<li>  zxc</li>\n"
         "</ol>\n");
     free(html);
 }
@@ -719,6 +787,19 @@ test_content_parse_invalid_ordered_list(void **state)
         "<p>a. asd\n"
         "2. qwe</p>\n");
     free(html);
+    html = blogc_content_parse(
+        "1.\nasd\n"
+        "2. qwe\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>1.\n"
+        "asd\n"
+        "2. qwe</p>\n");
+    free(html);
+    html = blogc_content_parse("1.\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html, "<p>1.</p>\n");
+    free(html);
 }
 
 
@@ -1038,6 +1119,7 @@ int
 main(void)
 {
     const UnitTest tests[] = {
+        unit_test(test_is_ordered_list_item),
         unit_test(test_content_parse),
         unit_test(test_content_parse_with_excerpt),
         unit_test(test_content_parse_header),