gst/id3demux/id3v2frames.c: Extract license/copyright URIs from ID3v2 WCOP frames (Fixes #447000).

Original commit message from CVS:
Based on patch by: Jason Kivlighn  <jkivlighn gmail com>
* gst/id3demux/id3v2frames.c:
Extract license/copyright URIs from ID3v2 WCOP frames
(Fixes #447000).
* tests/check/elements/id3demux.c:
* tests/files/Makefile.am:
* tests/files/id3-447000-wcop.tag:
Add simple unit test.

5 files changed, 99 insertions, 1 deletions

diff --git a/ChangeLog b/ChangeLog
index 0b10b965..b2273257 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,18 @@
 2007-10-11  Tim-Philipp Müller  <tim at centricular dot net>
 
+	Based on patch by: Jason Kivlighn  <jkivlighn gmail com>
+
+	* gst/id3demux/id3v2frames.c:
+	  Extract license/copyright URIs from ID3v2 WCOP frames
+	  (Fixes #447000).
+
+	* tests/check/elements/id3demux.c:
+	* tests/files/Makefile.am:
+	* tests/files/id3-447000-wcop.tag:
+	  Add simple unit test.
+
+2007-10-11  Tim-Philipp Müller  <tim at centricular dot net>
+
 	* ext/taglib/gstid3v2mux.cc:
 	  Add support for license/copyright URI tags (ID3v2 WCOP frame).
 	  Prerequisite for #447000.
diff --git a/gst/id3demux/id3v2frames.c b/gst/id3demux/id3v2frames.c
index 76d4cbc3..efbef079 100644
--- a/gst/id3demux/id3v2frames.c
+++ b/gst/id3demux/id3v2frames.c
@@ -38,6 +38,8 @@ GST_DEBUG_CATEGORY_EXTERN (id3demux_debug);
 #define GST_CAT_DEFAULT (id3demux_debug)
 
 static gboolean parse_comment_frame (ID3TagsWorking * work);
+static gchar *parse_url_link_frame (ID3TagsWorking * work,
+    const gchar ** tag_name);
 static GArray *parse_text_identification_frame (ID3TagsWorking * work);
 static gchar *parse_user_text_identification_frame (ID3TagsWorking * work,
     const gchar ** tag_name);
@@ -162,6 +164,9 @@ id3demux_id3v2_parse_frame (ID3TagsWorking * work)
       /* Text identification frame */
       tag_fields = parse_text_identification_frame (work);
     }
+  } else if (work->frame_id[0] == 'W' && strcmp (work->frame_id, "WXXX") != 0) {
+    /* URL link frame: ISO-8859-1 encoded, one frame per tag */
+    tag_str = parse_url_link_frame (work, &tag_name);
   } else if (!strcmp (work->frame_id, "COMM")) {
     /* Comment */
     result = parse_comment_frame (work);
@@ -297,6 +302,58 @@ parse_text_identification_frame (ID3TagsWorking * work)
   return fields;
 }
 
+static gboolean
+link_is_known_license (const gchar * url)
+{
+  return g_str_has_prefix (url, "http://creativecommons.org/licenses/");
+}
+
+static gchar *
+parse_url_link_frame (ID3TagsWorking * work, const gchar ** tag_name)
+{
+  gsize len;
+  gchar *nul, *data, *link;
+
+  *tag_name = NULL;
+
+  if (work->parse_size == 0)
+    return NULL;
+
+  data = (gchar *) work->parse_data;
+  /* if there's more data then the string is long, we only want to parse the
+   * data up to the terminating zero to g_convert and ignore the rest, as
+   * per spec */
+  nul = memchr (data, '\0', work->parse_size);
+  if (nul != NULL) {
+    len = (gsize) (nul - data);
+  } else {
+    len = work->parse_size;
+  }
+
+  link = g_convert (data, len, "UTF-8", "ISO-8859-1", NULL, NULL, NULL);
+
+  if (link == NULL || !gst_uri_is_valid (link)) {
+    GST_DEBUG ("Invalid URI in %s frame: %s", work->frame_id,
+        GST_STR_NULL (link));
+    g_free (link);
+    return NULL;
+  }
+
+  /* we don't know if it's a link to a page that explains the copyright
+   * situation, or a link that points to/represents a license, the ID3 spec
+   * does not separate those two things; for now only put known license URIs
+   * into GST_TAG_LICENSE_URI and everything else into GST_TAG_COPYRIGHT_URI */
+  if (strcmp (work->frame_id, "WCOP") == 0) {
+    if (link_is_known_license (link))
+      *tag_name = GST_TAG_LICENSE_URI;
+    else
+      *tag_name = GST_TAG_COPYRIGHT_URI;
+  }
+
+  return link;
+}
+
+
 static gchar *
 parse_user_text_identification_frame (ID3TagsWorking * work,
     const gchar ** tag_name)
diff --git a/tests/check/elements/id3demux.c b/tests/check/elements/id3demux.c
index d0f385d9..96aafdcb 100644
--- a/tests/check/elements/id3demux.c
+++ b/tests/check/elements/id3demux.c
@@ -175,6 +175,32 @@ GST_START_TEST (test_tdat_tyer)
 
 GST_END_TEST;
 
+static void
+check_wcop (const GstTagList * tags, const gchar * file)
+{
+  gchar *copyright = NULL;
+  gchar *uri = NULL;
+
+  fail_unless (gst_tag_list_get_string (tags, GST_TAG_LICENSE_URI, &uri));
+  fail_unless (uri != NULL);
+  fail_unless_equals_string (uri,
+      "http://creativecommons.org/licenses/by/3.0/");
+  g_free (uri);
+
+  fail_unless (gst_tag_list_get_string (tags, GST_TAG_COPYRIGHT, &copyright));
+  fail_unless (copyright != NULL);
+  fail_unless_equals_string (copyright,
+      " Steadman. Licensed to the public under http://creativecommons.org/licenses/by/3.0/ verify at http://test.com");
+  g_free (copyright);
+}
+
+GST_START_TEST (test_wcop)
+{
+  run_check_for_file ("id3-447000-wcop.tag", check_wcop);
+}
+
+GST_END_TEST;
+
 static Suite *
 id3demux_suite (void)
 {
@@ -183,6 +209,7 @@ id3demux_suite (void)
 
   suite_add_tcase (s, tc_chain);
   tcase_add_test (tc_chain, test_tdat_tyer);
+  tcase_add_test (tc_chain, test_wcop);
 
   return s;
 }
diff --git a/tests/files/Makefile.am b/tests/files/Makefile.am
index c065d007..6fc51c12 100644
--- a/tests/files/Makefile.am
+++ b/tests/files/Makefile.am
@@ -1,5 +1,6 @@
 
 EXTRA_DIST = \
 	id3-407349-1.tag \
-	id3-407349-2.tag
+	id3-407349-2.tag \
+	id3-447000-wcop.tag
 
diff --git a/tests/files/id3-447000-wcop.tag b/tests/files/id3-447000-wcop.tag
new file mode 100644
index 00000000..0f81c260
--- /dev/null
+++ b/tests/files/id3-447000-wcop.tag