Commit 8551c49f authored by Stefan Kost's avatar Stefan Kost
Browse files

tags: add basic xmp metadata support

XMP metadata can be embedded in many media container formats. Implement own
parser and formatter that can be used to convert between an xpacket and a
GstTagList. Add unit tests.
parent 91fbd420
......@@ -186,6 +186,7 @@
<xi:include href="xml/gsttag.xml" />
<xi:include href="xml/gsttagvorbis.xml" />
<xi:include href="xml/gsttagid3.xml" />
<xi:include href="xml/gsttagxmp.xml" />
<xi:include href="xml/gsttagdemux.xml" />
<xi:include href="xml/gsttaglanguagecodes.xml" />
</chapter>
......
......@@ -1663,6 +1663,14 @@ gst_tag_to_id3_tag
gst_tag_list_add_id3_image
</SECTION>
<SECTION>
<FILE>gsttagxmp</FILE>
<INCLUDE>gst/tag/tag.h</INCLUDE>
gst_tag_list_from_xmp_buffer
gst_tag_list_to_xmp_buffer
<SUBSECTION Standard>
</SECTION>
<SECTION>
<FILE>gsttagdemux</FILE>
<INCLUDE>gst/tag/gsttagdemux.h</INCLUDE>
......
......@@ -6,7 +6,9 @@ libgsttaginclude_HEADERS = \
lib_LTLIBRARIES = libgsttag-@GST_MAJORMINOR@.la
libgsttag_@GST_MAJORMINOR@_la_SOURCES = gstvorbistag.c gstid3tag.c lang.c tags.c gsttagdemux.c
libgsttag_@GST_MAJORMINOR@_la_SOURCES = \
gstvorbistag.c gstid3tag.c gstxmptag.c \
lang.c tags.c gsttagdemux.c
libgsttag_@GST_MAJORMINOR@_la_CFLAGS = $(GST_PLUGINS_BASE_CFLAGS) $(GST_BASE_CFLAGS) $(GST_CFLAGS)
libgsttag_@GST_MAJORMINOR@_la_LIBADD = $(GST_BASE_LIBS) $(GST_LIBS)
libgsttag_@GST_MAJORMINOR@_la_LDFLAGS = $(GST_LIB_LDFLAGS) $(GST_ALL_LDFLAGS) $(GST_LT_LDFLAGS)
......
/* GStreamer
* Copyright (C) 2010 Stefan Kost <stefan.kost@nokia.com>
*
* gstxmptag.c: library for reading / modifying xmp tags
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Library General Public License for more details.
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
* Boston, MA 02111-1307, USA.
*/
/**
* SECTION:gsttagxmp
* @short_description: tag mappings and support functions for plugins
* dealing with xmp packets
* @see_also: #GstTagList
*
* Contains various utility functions for plugins to parse or create
* xmp packets and map them to and from #GstTagList<!-- -->s.
*
* Please note that the xmp parser is very lightweight and not strict at all.
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include <gst/gsttagsetter.h>
#include "gsttageditingprivate.h"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <time.h>
/* look at this page for addtional schemas
* http://www.sno.phy.queensu.ca/~phil/exiftool/TagNames/XMP.html
*/
static const GstTagEntryMatch tag_matches[] = {
/* dublic code metadata
* http://dublincore.org/documents/dces/
*/
{GST_TAG_ARTIST, "dc:creator"},
{GST_TAG_COPYRIGHT, "dc:rights"},
{GST_TAG_DATE, "dc:date"},
{GST_TAG_DATE, "exif:DateTimeOriginal"},
{GST_TAG_DESCRIPTION, "dc:description"},
{GST_TAG_KEYWORDS, "dc:subject"},
{GST_TAG_TITLE, "dc:title"},
/* FIXME: we probably want GST_TAG_{,AUDIO_,VIDEO_}MIME_TYPE */
{GST_TAG_VIDEO_CODEC, "dc:format"},
/* */
{NULL, NULL}
};
typedef struct _GstXmpNamespaceMatch GstXmpNamespaceMatch;
struct _GstXmpNamespaceMatch
{
const gchar *ns_prefix;
const gchar *ns_uri;
};
static const GstXmpNamespaceMatch ns_match[] = {
{"dc", "http://purl.org/dc/elements/1.1/"},
{"exif", "http://ns.adobe.com/exif/1.0/"},
{"tiff", "http://ns.adobe.com/tiff/1.0/"},
{"xap", "http://ns.adobe.com/xap/1.0/"},
{NULL, NULL}
};
typedef struct _GstXmpNamespaceMap GstXmpNamespaceMap;
struct _GstXmpNamespaceMap
{
const gchar *original_ns;
gchar *gstreamer_ns;
};
static GstXmpNamespaceMap ns_map[] = {
{"dc", NULL},
{"exif", NULL},
{"tiff", NULL},
{"xap", NULL},
{NULL, NULL}
};
/* parsing */
static void
read_one_tag (GstTagList * list, const gchar * tag, const gchar * v)
{
GType tag_type = gst_tag_get_type (tag);
/* add gstreamer tag depending on type */
switch (tag_type) {
case G_TYPE_STRING:{
gst_tag_list_add (list, GST_TAG_MERGE_REPLACE, tag, v, NULL);
break;
}
default:
if (tag_type == GST_TYPE_DATE) {
GDate *date;
gint d, m, y;
/* this is ISO 8601 Date and Time Format
* %F Equivalent to %Y-%m-%d (the ISO 8601 date format). (C99)
* %T The time in 24-hour notation (%H:%M:%S). (SU)
* e.g. 2009-05-30T18:26:14+03:00 */
/* FIXME: this would be the proper way, but needs
#define _XOPEN_SOURCE before #include <time.h>
date = g_date_new ();
struct tm tm={0,};
strptime (dts, "%FT%TZ", &tm);
g_date_set_time_t (date, mktime(&tm));
*/
/* FIXME: this cannot parse the date
date = g_date_new ();
g_date_set_parse (date, v);
if (g_date_valid (date)) {
gst_tag_list_add (list, GST_TAG_MERGE_REPLACE, tag,
date, NULL);
} else {
GST_WARNING ("unparsable date: '%s'", v);
}
*/
/* poor mans straw */
sscanf (v, "%04d-%02d-%02dT", &y, &m, &d);
date = g_date_new_dmy (d, m, y);
gst_tag_list_add (list, GST_TAG_MERGE_REPLACE, tag, date, NULL);
g_date_free (date);
} else {
GST_WARNING ("unhandled type for %s from xmp", tag);
}
break;
}
}
/**
* gst_tag_list_from_xmp_buffer:
* @list: buffer
*
* Parse a xmp packet into a taglist.
*
* Returns: new taglist or %NULL, free the list when done
*
* Since: 0.10.29
*/
GstTagList *
gst_tag_list_from_xmp_buffer (const GstBuffer * buffer)
{
GstTagList *list = NULL;
const gchar *xps, *xp1, *xp2, *xpe, *ns, *ne;
guint len, max_ft_len;
gboolean in_tag;
gchar *part, *pp;
guint i;
const gchar *last_tag = NULL;
g_return_val_if_fail (GST_IS_BUFFER (buffer), NULL);
g_return_val_if_fail (GST_BUFFER_SIZE (buffer) > 0, NULL);
xps = (const gchar *) GST_BUFFER_DATA (buffer);
len = GST_BUFFER_SIZE (buffer);
xpe = &xps[len + 1];
/* check header and footer */
xp1 = g_strstr_len (xps, len, "<?xpacket begin");
if (!xp1)
goto missing_header;
xp1 = &xp1[strlen ("<?xpacket begin")];
while (*xp1 != '>' && *xp1 != '<' && xp1 < xpe)
xp1++;
if (*xp1 != '>')
goto missing_header;
max_ft_len = 1 + strlen ("<?xpacket end=\".\"?>\n");
if (len < max_ft_len)
goto missing_footer;
GST_DEBUG ("checking footer: [%s]", &xps[len - max_ft_len]);
xp2 = g_strstr_len (&xps[len - max_ft_len], max_ft_len, "<?xpacket ");
if (!xp2)
goto missing_footer;
GST_INFO ("xmp header okay");
/* skip > and text until first xml-node */
xp1++;
while (*xp1 != '<' && xp1 < xpe)
xp1++;
/* no tag can be longer that the whole buffer */
part = g_malloc (xp2 - xp1);
list = gst_tag_list_new ();
/* parse data into a list of nodes */
/* data is between xp1..xp2 */
in_tag = TRUE;
ns = ne = xp1;
pp = part;
while (ne < xp2) {
if (in_tag) {
ne++;
while (ne < xp2 && *ne != '>' && *ne != '<') {
if (*ne == '\n' || *ne == '\t' || *ne == ' ') {
while (ne < xp2 && (*ne == '\n' || *ne == '\t' || *ne == ' '))
ne++;
*pp++ = ' ';
} else {
*pp++ = *ne++;
}
}
*pp = '\0';
if (*ne != '>')
goto broken_xml;
/* create node */
/* {XML, ns, ne-ns} */
if (ns[0] != '/') {
gchar *as = strchr (part, ' ');
/* only log start nodes */
GST_INFO ("xml: %s", part);
if (as) {
gchar *ae, *d;
/* skip ' ' and scan the attributes */
as++;
d = ae = as;
/* split attr=value pairs */
while (*ae != '\0') {
if (*ae == '=') {
/* attr/value delimmiter */
d = ae;
} else if (*ae == '"') {
/* scan values */
gchar *v;
ae++;
while (*ae != '\0' && *ae != '"')
ae++;
*d = *ae = '\0';
v = &d[2];
GST_INFO (" : [%s][%s]", as, v);
if (!strncmp (as, "xmlns:", 6)) {
i = 0;
/* we need to rewrite known namespaces to what we use in
* tag_matches */
while (ns_match[i].ns_prefix) {
if (!strcmp (ns_match[i].ns_uri, v))
break;
i++;
}
if (ns_match[i].ns_prefix) {
if (strcmp (ns_map[i].original_ns, &as[6])) {
ns_map[i].gstreamer_ns = g_strdup (&as[6]);
}
}
} else {
/* FIXME: eventualy rewrite ns
* find ':'
* check if ns before ':' is in ns_map and ns_map[i].gstreamer_ns!=NULL
* do 2 stage filter in tag_matches
*/
i = 0;
while (tag_matches[i].gstreamer_tag) {
if (!strcmp (tag_matches[i].original_tag, as))
break;
i++;
}
if (tag_matches[i].gstreamer_tag) {
read_one_tag (list, tag_matches[i].gstreamer_tag, v);
}
}
/* restore chars overwritten by '\0' */
*d = '=';
*ae = '"';
} else if (*ae == '\0' || *ae == ' ') {
/* end of attr/value pair */
as = &ae[1];
}
/* to next char if not eos */
if (*ae != '\0')
ae++;
}
} else {
/*
<dc:type><rdf:Bag><rdf:li>Image</rdf:li></rdf:Bag></dc:type>
<dc:creator><rdf:Seq><rdf:li/></rdf:Seq></dc:creator>
*/
/* FIXME: eventualy rewrite ns */
/* skip rdf tags for now */
if (strncmp (part, "rdf:", 4)) {
i = 0;
while (tag_matches[i].gstreamer_tag) {
if (!strcmp (tag_matches[i].original_tag, part))
break;
i++;
}
if (tag_matches[i].gstreamer_tag) {
last_tag = tag_matches[i].gstreamer_tag;
}
}
}
}
/* next cycle */
ne++;
if (ne < xp2) {
if (*ne != '<')
in_tag = FALSE;
ns = ne;
pp = part;
}
} else {
while (ne < xp2 && *ne != '<') {
*pp++ = *ne;
ne++;
}
*pp = '\0';
/* create node */
/* {TXT, ns, (ne-ns)-1} */
if (ns[0] != '\n' && &ns[1] < ne) {
/* only log non-newline nodes, we still have to parse them */
GST_INFO ("txt: %s", part);
if (last_tag) {
read_one_tag (list, last_tag, part);
}
}
/* next cycle */
in_tag = TRUE;
ns = ne;
pp = part;
}
}
GST_INFO ("xmp packet parsed, %d entries",
gst_structure_n_fields ((GstStructure *) list));
/* free resources */
i = 0;
while (ns_map[i].original_ns) {
g_free (ns_map[i].gstreamer_ns);
i++;
}
g_free (part);
return list;
/* Errors */
missing_header:
GST_WARNING ("malformed xmp packet header");
return NULL;
missing_footer:
GST_WARNING ("malformed xmp packet footer");
return NULL;
broken_xml:
GST_WARNING ("malformed xml tag: %s", part);
return NULL;
}
/* formatting */
static void
write_one_tag (const GstTagList * list, const gchar * tag, gpointer user_data)
{
guint i = 0, ct = gst_tag_list_get_tag_size (list, tag);
GType tag_type = gst_tag_get_type (tag);
GString *data = user_data;
const gchar *xmp_tag = NULL, *fmt;
/* map gst-tag to xmp tag */
while (tag_matches[i].gstreamer_tag != NULL) {
if (strcmp (tag, tag_matches[i].gstreamer_tag) == 0) {
xmp_tag = tag_matches[i].original_tag;
break;
}
i++;
}
if (!xmp_tag) {
GST_WARNING ("no mapping for %s to xmp", tag);
return;
}
/* write single or multi valued field */
if (ct > 1) {
g_string_append_printf (data, "<%s><rdf:Bag>", xmp_tag);
fmt = "<rdf:li>%s</rdf:li>";
} else {
g_string_append_printf (data, "<%s>", xmp_tag);
fmt = "%s";
}
for (i = 0; i < ct; i++) {
GST_DEBUG ("mapping %s[%u/%u] to xmp", tag, i, ct);
switch (tag_type) {
case G_TYPE_STRING:{
gchar *str;
if (!gst_tag_list_get_string_index (list, tag, i, &str))
g_return_if_reached ();
g_string_append_printf (data, fmt, str);
g_free (str);
break;
}
default:
if (tag_type == GST_TYPE_DATE) {
GDate *date;
gchar *str;
if (!gst_tag_list_get_date_index (list, tag, i, &date))
g_return_if_reached ();
str = g_strdup_printf ("%04d-%02d-%02d",
(gint) g_date_get_year (date), (gint) g_date_get_month (date),
(gint) g_date_get_day (date));
g_string_append_printf (data, fmt, str);
g_free (str);
g_date_free (date);
} else {
GST_WARNING ("unhandled type for %s to xmp", tag);
}
break;
}
}
if (ct > 1) {
g_string_append_printf (data, "</rdf:Bag></%s>\n", xmp_tag);
} else {
g_string_append_printf (data, "</%s>\n", xmp_tag);
}
}
/**
* gst_tag_list_to_xmp_buffer:
* @list: tags
* @read_only: does the container forbid inplace editing
*
* Formats a taglist as a xmp packet.
*
* Returns: new buffer or %NULL, unref the buffer when done
*
* Since: 0.10.29
*/
GstBuffer *
gst_tag_list_to_xmp_buffer (const GstTagList * list, gboolean read_only)
{
GstBuffer *buffer = NULL;
GString *data = g_string_sized_new (4096);
guint i;
g_return_val_if_fail (GST_IS_TAG_LIST (list), NULL);
/* xmp header */
g_string_append (data,
"<?xpacket begin=\"\xEF\xBB\xBF\" id=\"W5M0MpCehiHzreSzNTczkc9d\"?>\n");
g_string_append (data,
"<x:xmpmeta xmlns:x=\"adobe:ns:meta/\" x:xmptk=\"GStreamer\">\n");
g_string_append (data,
"<rdf:RDF xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\"");
i = 0;
while (ns_match[i].ns_prefix) {
g_string_append_printf (data, " xmlns:%s=\"%s\"", ns_match[i].ns_prefix,
ns_match[i].ns_uri);
i++;
}
g_string_append (data, ">\n");
g_string_append (data, "<rdf:Description rdf:about=\"\">\n");
/* iterate the taglist */
gst_tag_list_foreach (list, write_one_tag, data);
/* xmp footer */
g_string_append (data, "</rdf:Description>\n");
g_string_append (data, "</rdf:RDF>\n");
g_string_append (data, "</x:xmpmeta>\n");
if (!read_only) {
/* the xmp spec recommand to add 2-4KB padding for in-place editable xmp */
guint i;
for (i = 0; i < 32; i++) {
g_string_append (data, " " " "
" " " " "\n");
}
}
g_string_append_printf (data, "<?xpacket end=\"%c\"?>\n",
(read_only ? 'r' : 'w'));
buffer = gst_buffer_new ();
GST_BUFFER_SIZE (buffer) = data->len + 1;
GST_BUFFER_DATA (buffer) = (guint8 *) g_string_free (data, FALSE);
GST_BUFFER_MALLOCDATA (buffer) = GST_BUFFER_DATA (buffer);
return buffer;
}
......@@ -219,6 +219,11 @@ gboolean gst_tag_list_add_id3_image (GstTagList * tag_list,
guint image_data_len,
guint id3_picture_type);
/* functions to convert GstBuffers with xmp packets contents to GstTagLists and back */
GstTagList * gst_tag_list_from_xmp_buffer (const GstBuffer * buffer);
GstBuffer * gst_tag_list_to_xmp_buffer (const GstTagList * list,
gboolean read_only);
/* other tag-related functions */
gboolean gst_tag_parse_extended_comment (const gchar * ext_comment,
......
......@@ -749,6 +749,110 @@ GST_START_TEST (test_language_utils)
GST_END_TEST;
GST_START_TEST (test_xmp_formatting)
{
GstTagList *list;
GstBuffer *buf;
const gchar *text;
guint len;
/* test data */
list = gst_tag_list_new_full (GST_TAG_TITLE, "test title",
GST_TAG_DESCRIPTION, "test decription",
GST_TAG_KEYWORDS, "keyword1", GST_TAG_KEYWORDS, "keyword2", NULL);
buf = gst_tag_list_to_xmp_buffer (list, FALSE);
fail_unless (buf != NULL);
text = (const gchar *) GST_BUFFER_DATA (buf);
len = GST_BUFFER_SIZE (buf);
/* check the content */
fail_unless (g_strrstr_len (text, len, "<?xpacket begin") == text);
fail_unless (g_strrstr_len (text, len, ">test title<") != NULL);
fail_unless (g_strrstr_len (text, len, ">test decription<") != NULL);
fail_unless (g_strrstr_len (text, len, ">keyword1<") != NULL);
fail_unless (g_strrstr_len (text, len, ">keyword2<") != NULL);
fail_unless (g_strrstr_len (text, len, "<?xpacket end") != NULL);
gst_buffer_unref (buf);
gst_tag_list_free (list);
}
GST_END_TEST;
GST_START_TEST (test_xmp_parsing)
{
GstTagList *list;
GstBuffer *buf;
guint i, result_size;
gchar *text;
const gchar *xmp_header =
"<?xpacket begin=\"\xEF\xBB\xBF\" id=\"W5M0MpCehiHzreSzNTczkc9d\"?>"
"<x:xmpmeta xmlns:x=\"adobe:ns:meta/\" x:xmptk=\"GStreamer\">"
"<rdf:RDF xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\" xmlns:dc=\"http://purl.org/dc/elements/1.1/\">";
const gchar *xmp_footer =
"</rdf:RDF>" "</x:xmpmeta>" "<?xpacket end=\"r\"?>\n";
struct
{
const gchar *xmp_data;
gint result_size;
gint result_test;
} test_data[] = {
{
"", -1, -1}, {
"<rdf:Description rdf:about=\"\" />", 0, -1}, {
"<rdf:Description rdf:about=\"\"></rdf:Description>", 0, -1}, {
"<rdf:Description rdf:about=\"\" ></rdf:Description>", 0, -1}, {
"<rdf:Description rdf:about=\"\"><dc:description>test</dc:description></rdf:Description>",
1, 0}, {
"<rdf:Description rdf:about=\"\" dc:description=\"test\"></rdf:Description>",
1, 0}, {