content-parser: added id attributes to headers

the id is a slugified version of the (unparsed) header content.

4 files changed, 82 insertions, 25 deletions

diff --git a/src/content-parser.c b/src/content-parser.c
index 511d09e..3b661a8 100644
--- a/src/content-parser.c
+++ b/src/content-parser.c
@@ -20,6 +20,27 @@
 // expected. feel free to improve the parser and add new features.
 
 
+char*
+blogc_slugify(const char *str)
+{
+    if (str == NULL)
+        return NULL;
+    char *new_str = b_strdup(str);
+    int diff = 'a' - 'A';  // just to avoid magic numbers
+    for (unsigned int i = 0; new_str[i] != '\0'; i++) {
+        if (new_str[i] >= 'a' && new_str[i] <= 'z')
+            continue;
+        if (new_str[i] >= '0' && new_str[i] <= '9')
+            continue;
+        if (new_str[i] >= 'A' && new_str[i] <= 'Z')
+            new_str[i] += diff;
+        else
+            new_str[i] = '-';
+    }
+    return new_str;
+}
+
+
 typedef enum {
     CONTENT_START_LINE = 1,
     CONTENT_EXCERPT,
@@ -389,6 +410,7 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
     char *tmp = NULL;
     char *tmp2 = NULL;
     char *parsed = NULL;
+    char *slug = NULL;
 
     char d = '\0';
 
@@ -501,8 +523,14 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
                     end = is_last && c != '\n' && c != '\r' ? src_len : current;
                     tmp = b_strndup(src + start, end - start);
                     parsed = blogc_content_parse_inline(tmp);
-                    b_string_append_printf(rv, "<h%d>%s</h%d>\n", header_level,
-                        parsed, header_level);
+                    slug = blogc_slugify(tmp);
+                    if (slug == NULL)
+                        b_string_append_printf(rv, "<h%d>%s</h%d>\n",
+                            header_level, parsed, header_level);
+                    else
+                        b_string_append_printf(rv, "<h%d id=\"%s\">%s</h%d>\n",
+                            header_level, slug, parsed, header_level);
+                    free(slug);
                     free(parsed);
                     parsed = NULL;
                     free(tmp);
diff --git a/src/content-parser.h b/src/content-parser.h
index 4579719..2f6b8b9 100644
--- a/src/content-parser.h
+++ b/src/content-parser.h
@@ -12,6 +12,7 @@
 #include <stdlib.h>
 #include <stdbool.h>
 
+char* blogc_slugify(const char *str);
 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);
diff --git a/tests/check_content_parser.c b/tests/check_content_parser.c
index b9f83b0..f1430ac 100644
--- a/tests/check_content_parser.c
+++ b/tests/check_content_parser.c
@@ -20,6 +20,33 @@
 
 
 static void
+test_slugify(void **state)
+{
+    char *s = blogc_slugify("bola");
+    assert_string_equal(s, "bola");
+    free(s);
+    s = blogc_slugify("bola o");
+    assert_string_equal(s, "bola-o");
+    free(s);
+    s = blogc_slugify("Bola O");
+    assert_string_equal(s, "bola-o");
+    free(s);
+    s = blogc_slugify("bola 0");
+    assert_string_equal(s, "bola-0");
+    free(s);
+    s = blogc_slugify("bola  () o");
+    assert_string_equal(s, "bola-----o");
+    free(s);
+    s = blogc_slugify("Bola O");
+    assert_string_equal(s, "bola-o");
+    free(s);
+    s = blogc_slugify("bolaço");
+    assert_string_equal(s, "bola--o");
+    free(s);
+}
+
+
+static void
 test_is_ordered_list_item(void **state)
 {
     assert_true(blogc_is_ordered_list_item("1.bola", 2));
@@ -80,12 +107,12 @@ test_content_parse(void **state)
     assert_non_null(html);
     assert_int_equal(l, 0);
     assert_string_equal(html,
-        "<h1>um</h1>\n"
-        "<h2>dois</h2>\n"
-        "<h3>tres</h3>\n"
-        "<h4>quatro</h4>\n"
-        "<h5>cinco</h5>\n"
-        "<h6>seis</h6>\n"
+        "<h1 id=\"um\">um</h1>\n"
+        "<h2 id=\"dois\">dois</h2>\n"
+        "<h3 id=\"tres\">tres</h3>\n"
+        "<h4 id=\"quatro\">quatro</h4>\n"
+        "<h5 id=\"cinco\">cinco</h5>\n"
+        "<h6 id=\"seis\">seis</h6>\n"
         "<p>bola\n"
         "chunda</p>\n"
         "<blockquote><p>bola  <br />\n"
@@ -129,9 +156,9 @@ test_content_parse_with_excerpt(void **state)
         "guda\n"
         "lol", &l);
     assert_non_null(html);
-    assert_int_equal(l, 28);
+    assert_int_equal(l, 38);
     assert_string_equal(html,
-        "<h1>test</h1>\n"
+        "<h1 id=\"test\">test</h1>\n"
         "<p>chunda</p>\n"
         "<p>guda\n"
         "lol</p>\n");
@@ -147,9 +174,9 @@ test_content_parse_with_excerpt(void **state)
         "guda\n"
         "lol", &l);
     assert_non_null(html);
-    assert_int_equal(l, 28);
+    assert_int_equal(l, 38);
     assert_string_equal(html,
-        "<h1>test</h1>\n"
+        "<h1 id=\"test\">test</h1>\n"
         "<p>chunda</p>\n"
         "<p>guda\n"
         "lol</p>\n");
@@ -162,11 +189,11 @@ test_content_parse_header(void **state)
 {
     char *html = blogc_content_parse("## bola", NULL);
     assert_non_null(html);
-    assert_string_equal(html, "<h2>bola</h2>\n");
+    assert_string_equal(html, "<h2 id=\"bola\">bola</h2>\n");
     free(html);
     html = blogc_content_parse("## bola\n", NULL);
     assert_non_null(html);
-    assert_string_equal(html, "<h2>bola</h2>\n");
+    assert_string_equal(html, "<h2 id=\"bola\">bola</h2>\n");
     free(html);
     html = blogc_content_parse(
         "bola\n"
@@ -177,7 +204,7 @@ test_content_parse_header(void **state)
     assert_non_null(html);
     assert_string_equal(html,
         "<p>bola</p>\n"
-        "<h2>bola</h2>\n"
+        "<h2 id=\"bola\">bola</h2>\n"
         "<p>guda</p>\n");
     free(html);
 }
@@ -526,7 +553,7 @@ test_content_parse_invalid_excerpt(void **state)
     assert_non_null(html);
     assert_int_equal(l, 0);
     assert_string_equal(html,
-        "<h1>test</h1>\n"
+        "<h1 id=\"test\">test</h1>\n"
         "<p>chunda\n"
         "..</p>\n"
         "<p>guda\n"
@@ -544,7 +571,7 @@ test_content_parse_invalid_excerpt(void **state)
     assert_non_null(html);
     assert_int_equal(l, 0);
     assert_string_equal(html,
-        "<h1>test</h1>\n"
+        "<h1 id=\"test\">test</h1>\n"
         "<p>chunda</p>\n"
         "<p>...\n"
         "guda\n"
@@ -561,7 +588,7 @@ test_content_parse_invalid_excerpt(void **state)
     assert_non_null(html);
     assert_int_equal(l, 0);
     assert_string_equal(html,
-        "<h1>test</h1>\n"
+        "<h1 id=\"test\">test</h1>\n"
         "<p>chunda..</p>\n"
         "<p>guda\n"
         "lol</p>\n");
@@ -577,7 +604,7 @@ test_content_parse_invalid_excerpt(void **state)
     assert_non_null(html);
     assert_int_equal(l, 0);
     assert_string_equal(html,
-        "<h1>test</h1>\n"
+        "<h1 id=\"test\">test</h1>\n"
         "<p>chunda</p>\n"
         "<p>...guda\n"
         "lol</p>\n");
@@ -1119,6 +1146,7 @@ int
 main(void)
 {
     const UnitTest tests[] = {
+        unit_test(test_slugify),
         unit_test(test_is_ordered_list_item),
         unit_test(test_content_parse),
         unit_test(test_content_parse_with_excerpt),
diff --git a/tests/check_source_parser.c b/tests/check_source_parser.c
index 94f75c9..145ba30 100644
--- a/tests/check_source_parser.c
+++ b/tests/check_source_parser.c
@@ -38,10 +38,10 @@ test_source_parse(void **state)
     assert_string_equal(b_trie_lookup(source, "VAR1"), "asd asd");
     assert_string_equal(b_trie_lookup(source, "VAR2"), "123chunda");
     assert_string_equal(b_trie_lookup(source, "EXCERPT"),
-        "<h1>This is a test</h1>\n"
+        "<h1 id=\"this-is-a-test\">This is a test</h1>\n"
         "<p>bola</p>\n");
     assert_string_equal(b_trie_lookup(source, "CONTENT"),
-        "<h1>This is a test</h1>\n"
+        "<h1 id=\"this-is-a-test\">This is a test</h1>\n"
         "<p>bola</p>\n");
     assert_string_equal(b_trie_lookup(source, "RAW_CONTENT"),
         "# This is a test\n"
@@ -71,10 +71,10 @@ test_source_parse_with_spaces(void **state)
     assert_string_equal(b_trie_lookup(source, "VAR1"), "chunda");
     assert_string_equal(b_trie_lookup(source, "BOLA"), "guda");
     assert_string_equal(b_trie_lookup(source, "EXCERPT"),
-        "<h1>This is a test</h1>\n"
+        "<h1 id=\"this-is-a-test\">This is a test</h1>\n"
         "<p>bola</p>\n");
     assert_string_equal(b_trie_lookup(source, "CONTENT"),
-        "<h1>This is a test</h1>\n"
+        "<h1 id=\"this-is-a-test\">This is a test</h1>\n"
         "<p>bola</p>\n");
     assert_string_equal(b_trie_lookup(source, "RAW_CONTENT"),
         "# This is a test\n"
@@ -107,10 +107,10 @@ test_source_parse_with_excerpt(void **state)
     assert_string_equal(b_trie_lookup(source, "VAR1"), "asd asd");
     assert_string_equal(b_trie_lookup(source, "VAR2"), "123chunda");
     assert_string_equal(b_trie_lookup(source, "EXCERPT"),
-        "<h1>This is a test</h1>\n"
+        "<h1 id=\"this-is-a-test\">This is a test</h1>\n"
         "<p>bola</p>\n");
     assert_string_equal(b_trie_lookup(source, "CONTENT"),
-        "<h1>This is a test</h1>\n"
+        "<h1 id=\"this-is-a-test\">This is a test</h1>\n"
         "<p>bola</p>\n"
         "<p>guda\n"
         "yay</p>\n");