G_MODULE_EXPORT gboolean
tracker_extract_get_metadata (TrackerExtractInfo *info)
{
TrackerResource *resource;
GXPSDocument *document;
GXPSFile *xps_file;
GFile *file;
gchar *filename;
GError *error = NULL;
file = tracker_extract_info_get_file (info);
xps_file = gxps_file_new (file, &error);
filename = g_file_get_path (file);
if (error != NULL) {
g_warning ("Unable to open '%s': %s", filename, error->message);
g_error_free (error);
g_free (filename);
return FALSE;
}
document = gxps_file_get_document (xps_file, 0, &error);
g_object_unref (xps_file);
if (error != NULL) {
g_warning ("Unable to read '%s': %s", filename, error->message);
g_error_free (error);
g_free (filename);
return FALSE;
}
resource = tracker_resource_new (NULL);
tracker_resource_add_uri (resource, "rdf:type", "nfo:PaginatedTextDocument");
tracker_resource_set_int64 (resource, "nfo:pageCount", gxps_document_get_n_pages (document));
g_object_unref (document);
g_free (filename);
tracker_extract_info_set_resource (info, resource);
g_object_unref (resource);
return TRUE;
}
G_MODULE_EXPORT gboolean
tracker_extract_get_metadata (TrackerExtractInfo *info)
{
TrackerResource *resource;
GFile *file;
gchar *filename;
DviContext *context;
file = tracker_extract_info_get_file (info);
filename = g_file_get_path (file);
context = mdvi_init_context (filename);
if (context == NULL) {
g_warning ("Could not open dvi file '%s'\n", filename);
g_free (filename);
return FALSE;
}
resource = tracker_resource_new (NULL);
tracker_resource_add_uri (resource, "rdf:type", "nfo:PaginatedTextDocument");
tracker_resource_set_int64 (resource, "nfo:pageCount", context->npages);
if (context->fileid) {
tracker_resource_set_string (resource, "nie:comment", context->fileid);
}
mdvi_destroy_context (context);
tracker_extract_info_set_resource (info, resource);
g_object_unref (resource);
return TRUE;
}
G_MODULE_EXPORT gboolean
tracker_extract_get_metadata (TrackerExtractInfo *info)
{
TrackerResource *metadata;
guint max_width;
guint max_height;
GFile *file;
gchar *uri;
file = tracker_extract_info_get_file (info);
uri = g_file_get_uri (file);
metadata = tracker_resource_new (NULL);
/* The Windows Icon file format may contain the same icon with different
* sizes inside, so there's no clear way of setting single width and
* height values. Thus, we set maximum sizes found. */
tracker_resource_add_uri (metadata, "rdf:type", "nfo:Image");
tracker_resource_add_uri (metadata, "rdf:type", "nfo:Icon");
if (find_max_width_and_height (uri, &max_width, &max_height)) {
if (max_width > 0) {
tracker_resource_set_int64 (metadata, "nfo:width", (gint64)max_width);
}
if (max_height > 0) {
tracker_resource_set_int64 (metadata, "nfo:height", (gint64)max_height);
}
}
g_free (uri);
tracker_extract_info_set_resource (info, metadata);
g_object_unref (metadata);
return TRUE;
}
G_MODULE_EXPORT gboolean
tracker_extract_get_metadata (TrackerExtractInfo *info)
{
TrackerResource *metadata;
GFile *file;
TrackerConfig *config;
htmlDocPtr doc;
parser_data pd;
gchar *filename;
xmlSAXHandler handler = {
NULL, /* internalSubset */
NULL, /* isStandalone */
NULL, /* hasInternalSubset */
NULL, /* hasExternalSubset */
NULL, /* resolveEntity */
NULL, /* getEntity */
NULL, /* entityDecl */
NULL, /* notationDecl */
NULL, /* attributeDecl */
NULL, /* elementDecl */
NULL, /* unparsedEntityDecl */
NULL, /* setDocumentLocator */
NULL, /* startDocument */
NULL, /* endDocument */
parser_start_element, /* startElement */
parser_end_element, /* endElement */
NULL, /* reference */
parser_characters, /* characters */
NULL, /* ignorableWhitespace */
NULL, /* processingInstruction */
NULL, /* comment */
NULL, /* xmlParserWarning */
NULL, /* xmlParserError */
NULL, /* xmlParserError */
NULL, /* getParameterEntity */
NULL, /* cdataBlock */
NULL, /* externalSubset */
1, /* initialized */
NULL, /* private */
NULL, /* startElementNsSAX2Func */
NULL, /* endElementNsSAX2Func */
NULL /* xmlStructuredErrorFunc */
};
file = tracker_extract_info_get_file (info);
metadata = tracker_resource_new (NULL);
tracker_resource_add_uri (metadata, "rdf:type", "nfo:HtmlDocument");
pd.metadata = metadata;
pd.current = -1;
pd.in_body = FALSE;
pd.plain_text = g_string_new (NULL);
pd.title = g_string_new (NULL);
config = tracker_main_get_config ();
pd.n_bytes_remaining = tracker_config_get_max_bytes (config);
filename = g_file_get_path (file);
doc = htmlSAXParseFile (filename, NULL, &handler, &pd);
g_free (filename);
if (doc) {
xmlFreeDoc (doc);
}
g_strstrip (pd.plain_text->str);
g_strstrip (pd.title->str);
if (pd.title->str &&
*pd.title->str != '\0') {
tracker_resource_set_string (metadata, "nie:title", pd.title->str);
}
if (pd.plain_text->str &&
*pd.plain_text->str != '\0') {
tracker_resource_set_string (metadata, "nie:plainTextContent", pd.plain_text->str);
}
g_string_free (pd.plain_text, TRUE);
g_string_free (pd.title, TRUE);
tracker_extract_info_set_resource (info, metadata);
g_object_unref (metadata);
return TRUE;
}
G_MODULE_EXPORT gboolean
tracker_extract_get_metadata (TrackerExtractInfo *extract_info)
{
TrackerResource *metadata;
TrackerConfig *config;
ODTMetadataParseInfo info = { 0 };
ODTFileType file_type;
GFile *file;
gchar *uri;
const gchar *mime_used;
GMarkupParseContext *context;
GMarkupParser parser = {
xml_start_element_handler_metadata,
xml_end_element_handler_metadata,
xml_text_handler_metadata,
NULL,
NULL
};
if (G_UNLIKELY (maximum_size_error_quark == 0)) {
maximum_size_error_quark = g_quark_from_static_string ("maximum_size_error");
}
metadata = tracker_resource_new (NULL);
mime_used = tracker_extract_info_get_mimetype (extract_info);
file = tracker_extract_info_get_file (extract_info);
uri = g_file_get_uri (file);
/* Setup conf */
config = tracker_main_get_config ();
g_debug ("Extracting OASIS metadata and contents from '%s'", uri);
/* First, parse metadata */
tracker_resource_add_uri (metadata, "rdf:type", "nfo:PaginatedTextDocument");
/* Create parse info */
info.metadata = metadata;
info.current = ODT_TAG_TYPE_UNKNOWN;
info.uri = uri;
/* Create parsing context */
context = g_markup_parse_context_new (&parser, 0, &info, NULL);
/* Load the internal XML file from the Zip archive, and parse it
* using the given context */
tracker_gsf_parse_xml_in_zip (uri, "meta.xml", context, NULL);
g_markup_parse_context_free (context);
if (g_ascii_strcasecmp (mime_used, "application/vnd.oasis.opendocument.text") == 0) {
file_type = FILE_TYPE_ODT;
} else if (g_ascii_strcasecmp (mime_used, "application/vnd.oasis.opendocument.presentation") == 0) {
file_type = FILE_TYPE_ODP;
} else if (g_ascii_strcasecmp (mime_used, "application/vnd.oasis.opendocument.spreadsheet") == 0) {
file_type = FILE_TYPE_ODS;
} else if (g_ascii_strcasecmp (mime_used, "application/vnd.oasis.opendocument.graphics") == 0) {
file_type = FILE_TYPE_ODG;
} else {
g_message ("Mime type was not recognised:'%s'", mime_used);
file_type = FILE_TYPE_INVALID;
}
/* Extract content with the given limitations */
extract_oasis_content (uri,
tracker_config_get_max_bytes (config),
file_type,
metadata);
g_free (uri);
tracker_extract_info_set_resource (extract_info, metadata);
g_object_unref (metadata);
return TRUE;
}
G_MODULE_EXPORT gboolean
tracker_extract_get_metadata (TrackerExtractInfo *info)
{
TrackerConfig *config;
GTime creation_date;
GError *error = NULL;
TrackerResource *metadata;
TrackerXmpData *xd = NULL;
PDFData pd = { 0 }; /* actual data */
PDFData md = { 0 }; /* for merging */
PopplerDocument *document;
gchar *xml = NULL;
gchar *content, *uri;
guint n_bytes;
GPtrArray *keywords;
guint i;
GFile *file;
gchar *filename;
int fd;
gchar *contents = NULL;
gsize len;
struct stat st;
file = tracker_extract_info_get_file (info);
filename = g_file_get_path (file);
fd = tracker_file_open_fd (filename);
if (fd == -1) {
g_warning ("Could not open pdf file '%s': %s\n",
filename,
g_strerror (errno));
g_free (filename);
return FALSE;
}
if (fstat (fd, &st) == -1) {
g_warning ("Could not fstat pdf file '%s': %s\n",
filename,
g_strerror (errno));
close (fd);
g_free (filename);
return FALSE;
}
if (st.st_size == 0) {
contents = NULL;
len = 0;
} else {
contents = (gchar *) mmap (NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
if (contents == NULL || contents == MAP_FAILED) {
g_warning ("Could not mmap pdf file '%s': %s\n",
filename,
g_strerror (errno));
close (fd);
g_free (filename);
return FALSE;
}
len = st.st_size;
}
g_free (filename);
uri = g_file_get_uri (file);
document = poppler_document_new_from_data (contents, len, NULL, &error);
if (error) {
if (error->code == POPPLER_ERROR_ENCRYPTED) {
metadata = tracker_resource_new (NULL);
tracker_resource_add_uri (metadata, "rdf:type", "nfo:PaginatedTextDocument");
tracker_resource_set_boolean (metadata, "nfo:isContentEncrypted", TRUE);
tracker_extract_info_set_resource (info, metadata);
g_object_unref (metadata);
g_error_free (error);
g_free (uri);
close (fd);
return TRUE;
} else {
g_warning ("Couldn't create PopplerDocument from uri:'%s', %s",
uri,
error->message ? error->message : "no error given");
g_error_free (error);
g_free (uri);
close (fd);
return FALSE;
}
}
if (!document) {
g_warning ("Could not create PopplerDocument from uri:'%s', "
"NULL returned without an error",
uri);
g_free (uri);
close (fd);
return FALSE;
}
metadata = tracker_resource_new (NULL);
tracker_resource_add_uri (metadata, "rdf:type", "nfo:PaginatedTextDocument");
g_object_get (document,
"title", &pd.title,
"author", &pd.author,
"subject", &pd.subject,
"keywords", &pd.keywords,
"creation-date", &creation_date,
"metadata", &xml,
NULL);
if (creation_date > 0) {
pd.creation_date = tracker_date_to_string ((time_t) creation_date);
}
keywords = g_ptr_array_new_with_free_func ((GDestroyNotify) g_free);
if (xml && *xml &&
(xd = tracker_xmp_new (xml, strlen (xml), uri)) != NULL) {
/* The casts here are well understood and known */
md.title = (gchar *) tracker_coalesce_strip (4, pd.title, xd->title, xd->title2, xd->pdf_title);
md.subject = (gchar *) tracker_coalesce_strip (2, pd.subject, xd->subject);
md.date = (gchar *) tracker_coalesce_strip (3, pd.creation_date, xd->date, xd->time_original);
md.author = (gchar *) tracker_coalesce_strip (2, pd.author, xd->creator);
write_pdf_data (md, metadata, keywords);
if (xd->keywords) {
tracker_keywords_parse (keywords, xd->keywords);
}
if (xd->pdf_keywords) {
tracker_keywords_parse (keywords, xd->pdf_keywords);
}
if (xd->publisher) {
TrackerResource *publisher = tracker_extract_new_contact (xd->publisher);
tracker_resource_set_relation (metadata, "nco:publisher", publisher);
g_object_unref (publisher);
}
if (xd->type) {
tracker_resource_set_string (metadata, "dc:type", xd->type);
}
if (xd->format) {
tracker_resource_set_string (metadata, "dc:format", xd->format);
}
if (xd->identifier) {
tracker_resource_set_string (metadata, "dc:identifier", xd->identifier);
}
if (xd->source) {
tracker_resource_set_string (metadata, "dc:source", xd->source);
}
if (xd->language) {
tracker_resource_set_string (metadata, "dc:language", xd->language);
}
if (xd->relation) {
tracker_resource_set_string (metadata, "dc:relation", xd->relation);
}
if (xd->coverage) {
tracker_resource_set_string (metadata, "dc:coverage", xd->coverage);
}
if (xd->license) {
tracker_resource_set_string (metadata, "nie:license", xd->license);
}
if (xd->make || xd->model) {
TrackerResource *equipment = tracker_extract_new_equipment (xd->make, xd->model);
tracker_resource_set_relation (metadata, "nfo:equipment", equipment);
g_object_unref (equipment);
}
if (xd->orientation) {
tracker_resource_set_string (metadata, "nfo:orientation", xd->orientation);
}
if (xd->rights) {
tracker_resource_set_string (metadata, "nie:copyright", xd->rights);
}
if (xd->white_balance) {
tracker_resource_set_string (metadata, "nmm:whiteBalance", xd->white_balance);
}
if (xd->fnumber) {
gdouble value;
value = g_strtod (xd->fnumber, NULL);
tracker_resource_set_double (metadata, "nmm:fnumber", value);
}
if (xd->flash) {
tracker_resource_set_string (metadata, "nmm:flash", xd->flash);
}
if (xd->focal_length) {
gdouble value;
value = g_strtod (xd->focal_length, NULL);
tracker_resource_set_double (metadata, "nmm:focalLength", value);
}
/* Question: Shouldn't xd->Artist be merged with md.author instead? */
if (xd->artist || xd->contributor) {
TrackerResource *artist;
const gchar *artist_name;
artist_name = tracker_coalesce_strip (2, xd->artist, xd->contributor);
artist = tracker_extract_new_contact (artist_name);
tracker_resource_set_relation (metadata, "nco:contributor", artist);
g_object_unref (artist);
}
if (xd->exposure_time) {
gdouble value;
value = g_strtod (xd->exposure_time, NULL);
tracker_resource_set_double (metadata, "nmm:exposureTime", value);
}
if (xd->iso_speed_ratings) {
gdouble value;
value = g_strtod (xd->iso_speed_ratings, NULL);
tracker_resource_set_double (metadata, "nmm:isoSpeed", value);
}
if (xd->description) {
tracker_resource_set_string (metadata, "nie:description", xd->description);
}
if (xd->metering_mode) {
tracker_resource_set_string (metadata, "nmm:meteringMode", xd->metering_mode);
}
if (xd->address || xd->state || xd->country || xd->city ||
xd->gps_altitude || xd->gps_latitude || xd-> gps_longitude) {
TrackerResource *location = tracker_extract_new_location (xd->address,
xd->state, xd->city, xd->country, xd->gps_altitude,
xd->gps_latitude, xd->gps_longitude);
tracker_resource_set_relation (metadata, "slo:location", location);
g_object_unref (location);
}
if (xd->regions) {
tracker_xmp_apply_regions_to_resource (metadata, xd);
}
tracker_xmp_free (xd);
} else {
/* So if we are here we have NO XMP data and we just
* write what we know from Poppler.
*/
write_pdf_data (pd, metadata, keywords);
}
for (i = 0; i < keywords->len; i++) {
TrackerResource *tag;
const gchar *p;
p = g_ptr_array_index (keywords, i);
tag = tracker_extract_new_tag (p);
tracker_resource_add_relation (metadata, "nao:hasTag", tag);
g_object_unref (tag);
}
g_ptr_array_free (keywords, TRUE);
tracker_resource_set_int64 (metadata, "nfo:pageCount", poppler_document_get_n_pages(document));
config = tracker_main_get_config ();
n_bytes = tracker_config_get_max_bytes (config);
content = extract_content_text (document, n_bytes);
if (content) {
tracker_resource_set_string (metadata, "nie:plainTextContent", content);
g_free (content);
}
read_outline (document, metadata);
g_free (xml);
g_free (pd.keywords);
g_free (pd.title);
g_free (pd.subject);
g_free (pd.creation_date);
g_free (pd.author);
g_free (pd.date);
g_free (uri);
g_object_unref (document);
if (contents) {
munmap (contents, len);
}
close (fd);
tracker_extract_info_set_resource (info, metadata);
g_object_unref (metadata);
return TRUE;
}
