From 908fc8266ef81f80964be710dd5a15dbdb86500f Mon Sep 17 00:00:00 2001
From: "Rafael G. Martins" <rafael@rafaelmartins.eng.br>
Date: Thu, 22 Oct 2015 03:06:13 -0200
Subject: content-parser: handle \r\n line endings properly

---
 man/blogc-source.7.ronn      |   4 -
 src/content-parser.c         | 110 +++++++---
 tests/check_content_parser.c | 502 ++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 579 insertions(+), 37 deletions(-)

diff --git a/man/blogc-source.7.ronn b/man/blogc-source.7.ronn
index 83baf8c..8f43413 100644
--- a/man/blogc-source.7.ronn
+++ b/man/blogc-source.7.ronn
@@ -225,10 +225,6 @@ The source content is handled by handwritten parsers, that even being well
 tested, may be subject of parsing bugs. Please report any issues to:
 <https://github.com/blogc/blogc>
 
-At least one bug is known at this point: ``\r\n`` character sequences are
-handled like 2 line breaks. The parsers won't work properly with files edited
-on Windows editors like Notepad.
-
 ## AUTHOR
 
 Rafael G. Martins &lt;<rafael@rafaelmartins.eng.br>&gt;
diff --git a/src/content-parser.c b/src/content-parser.c
index 3b661a8..e8d3143 100644
--- a/src/content-parser.c
+++ b/src/content-parser.c
@@ -305,10 +305,10 @@ blogc_content_parse_inline(const char *src)
             case '\r':
                 if (state == LINK_CLOSED) {
                     if (spaces >= 2) {
-                        b_string_append(rv, "<br />\n");
+                        b_string_append(rv, "<br />");
                         spaces = 0;
                     }
-                    else if (c == '\n' || c == '\r')
+                    if (c == '\n' || c == '\r')
                         b_string_append_c(rv, c);
                 }
                 break;
@@ -403,6 +403,7 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
     size_t start2 = 0;
     size_t end = 0;
     size_t eend = 0;
+    size_t real_end = 0;
 
     unsigned int header_level = 0;
     char *prefix = NULL;
@@ -412,6 +413,12 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
     char *parsed = NULL;
     char *slug = NULL;
 
+    // this isn't empty because we need some reasonable default value in the
+    // unlikely case when we need to print some line ending before evaluating
+    // the "real" value.
+    char line_ending[3] = "\n";
+    char l[2] = {0, 0};
+
     char d = '\0';
 
     b_slist_t *lines = NULL;
@@ -426,6 +433,30 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
         char c = src[current];
         bool is_last = current == src_len - 1;
 
+        if (c == '\n' || c == '\r') {
+            if ((current + 1) < src_len) {
+                if ((c == '\n' && src[current + 1] == '\r') ||
+                    (c == '\r' && src[current + 1] == '\n'))
+                {
+                    if (l[0] == 0) {
+                        l[0] = c;
+                        l[1] = src[current + 1];
+                        line_ending[0] = c;
+                        line_ending[1] = src[current + 1];
+                        line_ending[2] = '\0';
+                    }
+                    real_end = current;
+                    c = src[++current];
+                    is_last = current == src_len - 1;
+                }
+            }
+            if (l[0] == 0) {
+                l[0] = c;
+                line_ending[0] = c;
+                line_ending[1] = '\0';
+            }
+        }
+
         switch (state) {
 
             case CONTENT_START_LINE:
@@ -520,16 +551,18 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
 
             case CONTENT_HEADER_TITLE:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len :
+                        (real_end != 0 ? real_end : current);
                     tmp = b_strndup(src + start, end - start);
                     parsed = blogc_content_parse_inline(tmp);
                     slug = blogc_slugify(tmp);
                     if (slug == NULL)
-                        b_string_append_printf(rv, "<h%d>%s</h%d>\n",
-                            header_level, parsed, header_level);
+                        b_string_append_printf(rv, "<h%d>%s</h%d>%s",
+                            header_level, parsed, header_level, line_ending);
                     else
-                        b_string_append_printf(rv, "<h%d id=\"%s\">%s</h%d>\n",
-                            header_level, slug, parsed, header_level);
+                        b_string_append_printf(rv, "<h%d id=\"%s\">%s</h%d>%s",
+                            header_level, slug, parsed, header_level,
+                            line_ending);
                     free(slug);
                     free(parsed);
                     parsed = NULL;
@@ -543,7 +576,8 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
             case CONTENT_HTML:
                 if (c == '\n' || c == '\r' || is_last) {
                     state = CONTENT_HTML_END;
-                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len :
+                        (real_end != 0 ? real_end : current);
                 }
                 if (!is_last)
                     break;
@@ -551,7 +585,7 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
             case CONTENT_HTML_END:
                 if (c == '\n' || c == '\r' || is_last) {
                     tmp = b_strndup(src + start, end - start);
-                    b_string_append_printf(rv, "%s\n", tmp);
+                    b_string_append_printf(rv, "%s%s", tmp, line_ending);
                     free(tmp);
                     tmp = NULL;
                     state = CONTENT_START_LINE;
@@ -570,7 +604,8 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
 
             case CONTENT_BLOCKQUOTE_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len :
+                        (real_end != 0 ? real_end : current);
                     tmp = b_strndup(src + start2, end - start2);
                     if (b_str_starts_with(tmp, prefix)) {
                         lines = b_slist_append(lines, b_strdup(tmp + strlen(prefix)));
@@ -596,11 +631,12 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
                         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_string_append_printf(tmp_str, "%s%s", l->data,
+                                line_ending);
                     }
                     tmp = blogc_content_parse(tmp_str->str, NULL);
-                    b_string_append_printf(rv, "<blockquote>%s</blockquote>\n",
-                        tmp);
+                    b_string_append_printf(rv, "<blockquote>%s</blockquote>%s",
+                        tmp, line_ending);
                     free(tmp);
                     tmp = NULL;
                     b_string_free(tmp_str, true);
@@ -627,7 +663,8 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
 
             case CONTENT_CODE_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len :
+                        (real_end != 0 ? real_end : current);
                     tmp = b_strndup(src + start2, end - start2);
                     if (b_str_starts_with(tmp, prefix)) {
                         lines = b_slist_append(lines, b_strdup(tmp + strlen(prefix)));
@@ -656,9 +693,10 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
                         if (l->next == NULL)
                             b_string_append_printf(rv, "%s", l->data);
                         else
-                            b_string_append_printf(rv, "%s\n", l->data);
+                            b_string_append_printf(rv, "%s%s", l->data,
+                                line_ending);
                     }
-                    b_string_append(rv, "</code></pre>\n");
+                    b_string_append_printf(rv, "</code></pre>%s", line_ending);
                     b_slist_free_full(lines, free);
                     lines = NULL;
                     free(prefix);
@@ -691,7 +729,7 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
                 }
 hr:
                 if (c == '\n' || c == '\r' || is_last) {
-                    b_string_append(rv, "<hr />\n");
+                    b_string_append_printf(rv, "<hr />%s", line_ending);
                     state = CONTENT_START_LINE;
                     start = current;
                     d = '\0';
@@ -702,7 +740,8 @@ hr:
 
             case CONTENT_UNORDERED_LIST_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len :
+                        (real_end != 0 ? real_end : current);
                     tmp = b_strndup(src + start2, end - start2);
                     tmp2 = b_strdup_printf("%-*s", strlen(prefix), "");
                     if (b_str_starts_with(tmp, prefix)) {
@@ -712,7 +751,8 @@ hr:
                                 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_string_append_printf(tmp_str, "%s%s", l->data,
+                                        line_ending);
                             }
                             b_slist_free_full(lines2, free);
                             lines2 = NULL;
@@ -760,7 +800,8 @@ hr:
                             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_string_append_printf(tmp_str, "%s%s", l->data,
+                                    line_ending);
                         }
                         b_slist_free_full(lines2, free);
                         lines2 = NULL;
@@ -770,10 +811,11 @@ hr:
                         free(parsed);
                         parsed = NULL;
                     }
-                    b_string_append(rv, "<ul>\n");
+                    b_string_append_printf(rv, "<ul>%s", line_ending);
                     for (b_slist_t *l = lines; l != NULL; l = l->next)
-                        b_string_append_printf(rv, "<li>%s</li>\n", l->data);
-                    b_string_append(rv, "</ul>\n");
+                        b_string_append_printf(rv, "<li>%s</li>%s", l->data,
+                            line_ending);
+                    b_string_append_printf(rv, "</ul>%s", line_ending);
                     b_slist_free_full(lines, free);
                     lines = NULL;
                     free(prefix);
@@ -807,7 +849,8 @@ hr:
 
             case CONTENT_ORDERED_LIST_START:
                 if (c == '\n' || c == '\r' || is_last) {
-                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len :
+                        (real_end != 0 ? real_end : current);
                     tmp = b_strndup(src + start2, end - start2);
                     tmp2 = b_strdup_printf("%-*s", prefix_len, "");
                     if (blogc_is_ordered_list_item(tmp, prefix_len)) {
@@ -817,7 +860,8 @@ hr:
                                 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_string_append_printf(tmp_str, "%s%s", l->data,
+                                        line_ending);
                             }
                             b_slist_free_full(lines2, free);
                             lines2 = NULL;
@@ -865,7 +909,8 @@ hr:
                             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_string_append_printf(tmp_str, "%s%s", l->data,
+                                    line_ending);
                         }
                         b_slist_free_full(lines2, free);
                         lines2 = NULL;
@@ -875,10 +920,11 @@ hr:
                         free(parsed);
                         parsed = NULL;
                     }
-                    b_string_append(rv, "<ol>\n");
+                    b_string_append_printf(rv, "<ol>%s", line_ending);
                     for (b_slist_t *l = lines; l != NULL; l = l->next)
-                        b_string_append_printf(rv, "<li>%s</li>\n", l->data);
-                    b_string_append(rv, "</ol>\n");
+                        b_string_append_printf(rv, "<li>%s</li>%s", l->data,
+                            line_ending);
+                    b_string_append_printf(rv, "</ol>%s", line_ending);
                     b_slist_free_full(lines, free);
                     lines = NULL;
                     free(prefix);
@@ -895,7 +941,8 @@ hr:
             case CONTENT_PARAGRAPH:
                 if (c == '\n' || c == '\r' || is_last) {
                     state = CONTENT_PARAGRAPH_END;
-                    end = is_last && c != '\n' && c != '\r' ? src_len : current;
+                    end = is_last && c != '\n' && c != '\r' ? src_len :
+                        (real_end != 0 ? real_end : current);
                 }
                 if (!is_last)
                     break;
@@ -905,7 +952,8 @@ para:
                 if (c == '\n' || c == '\r' || is_last) {
                     tmp = b_strndup(src + start, end - start);
                     parsed = blogc_content_parse_inline(tmp);
-                    b_string_append_printf(rv, "<p>%s</p>\n", parsed);
+                    b_string_append_printf(rv, "<p>%s</p>%s", parsed,
+                        line_ending);
                     free(parsed);
                     parsed = NULL;
                     free(tmp);
diff --git a/tests/check_content_parser.c b/tests/check_content_parser.c
index 37c1761..994e900 100644
--- a/tests/check_content_parser.c
+++ b/tests/check_content_parser.c
@@ -144,6 +144,84 @@ test_content_parse(void **state)
 }
 
 
+static void
+test_content_parse_crlf(void **state)
+{
+    size_t l = 0;
+    char *html = blogc_content_parse(
+        "# um\r\n"
+        "## dois\r\n"
+        "### tres\r\n"
+        "#### quatro\r\n"
+        "##### cinco\r\n"
+        "###### seis\r\n"
+        "\r\n"
+        "bola\r\n"
+        "chunda\r\n"
+        "\r\n"
+        ">  bola  \r\n"
+        ">  guda\r\n"
+        ">  buga\r\n"
+        ">  \r\n"
+        ">    asd\r\n"
+        "\r\n"
+        "    bola\r\n"
+        "     asd\r\n"
+        "    qwewer\r\n"
+        "\r\n"
+        "+++\r\n"
+        "1. chunda\r\n"
+        "3. fuuuu\r\n"
+        "\r\n"
+        "+  chunda2\r\n"
+        "+  fuuuu2\r\n"
+        "\r\n"
+        "<style>\r\n"
+        "   chunda\r\n"
+        "</style>\r\n"
+        "\r\n"
+        "guda\r\n"
+        "yay\r\n"
+        "\r\n"
+        "**bola**\r\n", &l);
+    assert_non_null(html);
+    assert_int_equal(l, 0);
+    assert_string_equal(html,
+        "<h1 id=\"um\">um</h1>\r\n"
+        "<h2 id=\"dois\">dois</h2>\r\n"
+        "<h3 id=\"tres\">tres</h3>\r\n"
+        "<h4 id=\"quatro\">quatro</h4>\r\n"
+        "<h5 id=\"cinco\">cinco</h5>\r\n"
+        "<h6 id=\"seis\">seis</h6>\r\n"
+        "<p>bola\r\n"
+        "chunda</p>\r\n"
+        "<blockquote><p>bola  <br />\r\n"
+        "guda\r\n"
+        "buga</p>\r\n"
+        "<pre><code>asd</code></pre>\r\n"
+        "</blockquote>\r\n"
+        "<pre><code>bola\r\n"
+        " asd\r\n"
+        "qwewer</code></pre>\r\n"
+        "<hr />\r\n"
+        "<ol>\r\n"
+        "<li>chunda</li>\r\n"
+        "<li>fuuuu</li>\r\n"
+        "</ol>\r\n"
+        "<ul>\r\n"
+        "<li>chunda2</li>\r\n"
+        "<li>fuuuu2</li>\r\n"
+        "</ul>\r\n"
+        "<style>\r\n"
+        "   chunda\r\n"
+        "</style>\r\n"
+        "<p>guda\r\n"
+        "yay</p>\r\n"
+        "<p><strong>bola</strong></p>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_with_excerpt(void **state)
 {
@@ -186,6 +264,48 @@ test_content_parse_with_excerpt(void **state)
 }
 
 
+static void
+test_content_parse_with_excerpt_crlf(void **state)
+{
+    size_t l = 0;
+    char *html = blogc_content_parse(
+        "# test\r\n"
+        "\r\n"
+        "chunda\r\n"
+        "\r\n"
+        "..\r\n"
+        "\r\n"
+        "guda\r\n"
+        "lol", &l);
+    assert_non_null(html);
+    assert_int_equal(l, 40);
+    assert_string_equal(html,
+        "<h1 id=\"test\">test</h1>\r\n"
+        "<p>chunda</p>\r\n"
+        "<p>guda\r\n"
+        "lol</p>\r\n");
+    free(html);
+    l = 0;
+    html = blogc_content_parse(
+        "# test\r\n"
+        "\r\n"
+        "chunda\r\n"
+        "\r\n"
+        "...\r\n"
+        "\r\n"
+        "guda\r\n"
+        "lol", &l);
+    assert_non_null(html);
+    assert_int_equal(l, 40);
+    assert_string_equal(html,
+        "<h1 id=\"test\">test</h1>\r\n"
+        "<p>chunda</p>\r\n"
+        "<p>guda\r\n"
+        "lol</p>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_header(void **state)
 {
@@ -212,6 +332,32 @@ test_content_parse_header(void **state)
 }
 
 
+static void
+test_content_parse_header_crlf(void **state)
+{
+    char *html = blogc_content_parse("## bola", NULL);
+    assert_non_null(html);
+    assert_string_equal(html, "<h2 id=\"bola\">bola</h2>\n");
+    free(html);
+    html = blogc_content_parse("## bola\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html, "<h2 id=\"bola\">bola</h2>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "bola\r\n"
+        "\r\n"
+        "## bola\r\n"
+        "\r\n"
+        "guda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\r\n"
+        "<h2 id=\"bola\">bola</h2>\r\n"
+        "<p>guda</p>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_html(void **state)
 {
@@ -239,6 +385,33 @@ test_content_parse_html(void **state)
 }
 
 
+static void
+test_content_parse_html_crlf(void **state)
+{
+    char *html = blogc_content_parse("<div>\r\n</div>", NULL);
+    assert_non_null(html);
+    assert_string_equal(html, "<div>\r\n</div>\r\n");
+    free(html);
+    html = blogc_content_parse("<div>\r\n</div>\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html, "<div>\r\n</div>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "bola\r\n"
+        "\r\n"
+        "<div>\r\n"
+        "</div>\r\n"
+        "\r\n"
+        "chunda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\r\n"
+        "<div>\r\n</div>\r\n"
+        "<p>chunda</p>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_blockquote(void **state)
 {
@@ -274,6 +447,41 @@ test_content_parse_blockquote(void **state)
 }
 
 
+static void
+test_content_parse_blockquote_crlf(void **state)
+{
+    char *html = blogc_content_parse(">  bola\r\n>  guda", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<blockquote><p>bola\r\n"
+        "guda</p>\r\n"
+        "</blockquote>\r\n");
+    free(html);
+    html = blogc_content_parse(">  bola\r\n>  guda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<blockquote><p>bola\r\n"
+        "guda</p>\r\n"
+        "</blockquote>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "bola\r\n"
+        "\r\n"
+        ">   bola\r\n"
+        ">   guda\r\n"
+        "\r\n"
+        "chunda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\r\n"
+        "<blockquote><p>bola\r\n"
+        "guda</p>\r\n"
+        "</blockquote>\r\n"
+        "<p>chunda</p>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_code(void **state)
 {
@@ -306,6 +514,38 @@ test_content_parse_code(void **state)
 }
 
 
+static void
+test_content_parse_code_crlf(void **state)
+{
+    char *html = blogc_content_parse("  bola\r\n  guda", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<pre><code>bola\r\n"
+        "guda</code></pre>\r\n");
+    free(html);
+    html = blogc_content_parse("  bola\r\n  guda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<pre><code>bola\r\n"
+        "guda</code></pre>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "bola\r\n"
+        "\r\n"
+        "   bola\r\n"
+        "   guda\r\n"
+        "\r\n"
+        "chunda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\r\n"
+        "<pre><code>bola\r\n"
+        "guda</code></pre>\r\n"
+        "<p>chunda</p>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_horizontal_rule(void **state)
 {
@@ -364,6 +604,64 @@ test_content_parse_horizontal_rule(void **state)
 }
 
 
+static void
+test_content_parse_horizontal_rule_crlf(void **state)
+{
+    char *html = blogc_content_parse("bola\r\nguda\r\n\r\n**", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola\r\n"
+        "guda</p>\r\n"
+        "<hr />\r\n");
+    free(html);
+    html = blogc_content_parse("bola\r\nguda\r\n\r\n++++", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola\r\n"
+        "guda</p>\r\n"
+        "<hr />\r\n");
+    free(html);
+    html = blogc_content_parse("bola\r\nguda\r\n\r\n--\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola\r\n"
+        "guda</p>\r\n"
+        "<hr />\r\n");
+    free(html);
+    html = blogc_content_parse("bola\r\nguda\r\n\r\n****\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola\r\n"
+        "guda</p>\r\n"
+        "<hr />\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "bola\r\n"
+        "\r\n"
+        "**\r\n"
+        "\r\n"
+        "chunda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\r\n"
+        "<hr />\r\n"
+        "<p>chunda</p>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "bola\r\n"
+        "\r\n"
+        "----\r\n"
+        "\r\n"
+        "chunda\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>bola</p>\r\n"
+        "<hr />\r\n"
+        "<p>chunda</p>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_unordered_list(void **state)
 {
@@ -452,6 +750,94 @@ test_content_parse_unordered_list(void **state)
 }
 
 
+static void
+test_content_parse_unordered_list_crlf(void **state)
+{
+    char *html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "*  asd\r\n"
+        "*  qwe\r\n"
+        "*  zxc", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ul>\r\n"
+        "<li>asd</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc</li>\r\n"
+        "</ul>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "*  asd\r\n"
+        "*  qwe\r\n"
+        "*  zxc\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ul>\r\n"
+        "<li>asd</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc</li>\r\n"
+        "</ul>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "*  asd\r\n"
+        "*  qwe\r\n"
+        "*  zxc\r\n"
+        "\r\n"
+        "fuuuu\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ul>\r\n"
+        "<li>asd</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc</li>\r\n"
+        "</ul>\r\n"
+        "<p>fuuuu</p>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "*  asd\r\n"
+        "   cvb\r\n"
+        "*  qwe\r\n"
+        "*  zxc\r\n"
+        "   1234\r\n"
+        "\r\n"
+        "fuuuu\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ul>\r\n"
+        "<li>asd\r\n"
+        "cvb</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc\r\n"
+        "1234</li>\r\n"
+        "</ul>\r\n"
+        "<p>fuuuu</p>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "*  asd\r\n"
+        "*   qwe\r\n"
+        "*    zxc", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<ul>\r\n"
+        "<li>asd</li>\r\n"
+        "<li> qwe</li>\r\n"
+        "<li>  zxc</li>\r\n"
+        "</ul>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_ordered_list(void **state)
 {
@@ -540,6 +926,94 @@ test_content_parse_ordered_list(void **state)
 }
 
 
+static void
+test_content_parse_ordered_list_crlf(void **state)
+{
+    char *html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "1.  asd\r\n"
+        "2.  qwe\r\n"
+        "3.  zxc", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ol>\r\n"
+        "<li>asd</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc</li>\r\n"
+        "</ol>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "1.  asd\r\n"
+        "2.  qwe\r\n"
+        "3.  zxc\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ol>\r\n"
+        "<li>asd</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc</li>\r\n"
+        "</ol>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "1.  asd\r\n"
+        "2.  qwe\r\n"
+        "3.  zxc\r\n"
+        "\r\n"
+        "fuuuu\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ol>\r\n"
+        "<li>asd</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc</li>\r\n"
+        "</ol>\r\n"
+        "<p>fuuuu</p>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "lol\r\n"
+        "\r\n"
+        "1.  asd\r\n"
+        "    cvb\r\n"
+        "2.  qwe\r\n"
+        "3.  zxc\r\n"
+        "    1234\r\n"
+        "\r\n"
+        "fuuuu\r\n", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<p>lol</p>\r\n"
+        "<ol>\r\n"
+        "<li>asd\r\n"
+        "cvb</li>\r\n"
+        "<li>qwe</li>\r\n"
+        "<li>zxc\r\n"
+        "1234</li>\r\n"
+        "</ol>\r\n"
+        "<p>fuuuu</p>\r\n");
+    free(html);
+    html = blogc_content_parse(
+        "1.  asd\r\n"
+        "2.   qwe\r\n"
+        "3.    zxc", NULL);
+    assert_non_null(html);
+    assert_string_equal(html,
+        "<ol>\r\n"
+        "<li>asd</li>\r\n"
+        "<li> qwe</li>\r\n"
+        "<li>  zxc</li>\r\n"
+        "</ol>\r\n");
+    free(html);
+}
+
+
 static void
 test_content_parse_invalid_excerpt(void **state)
 {
@@ -1126,11 +1600,11 @@ test_content_parse_inline_line_break(void **state)
     free(html);
     html = blogc_content_parse_inline("asd  ");
     assert_non_null(html);
-    assert_string_equal(html, "asd  <br />\n");
+    assert_string_equal(html, "asd  <br />");
     free(html);
     html = blogc_content_parse_inline("asd   ");
     assert_non_null(html);
-    assert_string_equal(html, "asd   <br />\n");
+    assert_string_equal(html, "asd   <br />");
     free(html);
     // invalid
     html = blogc_content_parse_inline("asd ");
@@ -1144,6 +1618,20 @@ test_content_parse_inline_line_break(void **state)
 }
 
 
+static void
+test_content_parse_inline_line_break_crlf(void **state)
+{
+    char *html = blogc_content_parse_inline("asd  \r\n");
+    assert_non_null(html);
+    assert_string_equal(html, "asd  <br />\r\n");
+    free(html);
+    html = blogc_content_parse_inline("asd \r\n");
+    assert_non_null(html);
+    assert_string_equal(html, "asd \r\n");
+    free(html);
+}
+
+
 int
 main(void)
 {
@@ -1151,14 +1639,23 @@ main(void)
         unit_test(test_slugify),
         unit_test(test_is_ordered_list_item),
         unit_test(test_content_parse),
+        unit_test(test_content_parse_crlf),
         unit_test(test_content_parse_with_excerpt),
+        unit_test(test_content_parse_with_excerpt_crlf),
         unit_test(test_content_parse_header),
+        unit_test(test_content_parse_header_crlf),
         unit_test(test_content_parse_html),
+        unit_test(test_content_parse_html_crlf),
         unit_test(test_content_parse_blockquote),
+        unit_test(test_content_parse_blockquote_crlf),
         unit_test(test_content_parse_code),
+        unit_test(test_content_parse_code_crlf),
         unit_test(test_content_parse_horizontal_rule),
+        unit_test(test_content_parse_horizontal_rule_crlf),
         unit_test(test_content_parse_unordered_list),
+        unit_test(test_content_parse_unordered_list_crlf),
         unit_test(test_content_parse_ordered_list),
+        unit_test(test_content_parse_ordered_list_crlf),
         unit_test(test_content_parse_invalid_excerpt),
         unit_test(test_content_parse_invalid_header),
         unit_test(test_content_parse_invalid_header_empty),
@@ -1175,6 +1672,7 @@ main(void)
         unit_test(test_content_parse_inline_link_auto),
         unit_test(test_content_parse_inline_image),
         unit_test(test_content_parse_inline_line_break),
+        unit_test(test_content_parse_inline_line_break_crlf),
     };
     return run_tests(tests);
 }
-- 
cgit v1.2.3-18-g5258