static int
concat_stream (pdf_obj *dst, pdf_obj *src)
{
const char *stream_data;
long stream_length;
pdf_obj *stream_dict;
pdf_obj *filter;
if (!PDF_OBJ_STREAMTYPE(dst) || !PDF_OBJ_STREAMTYPE(src))
ERROR("Invalid type.");
stream_data = pdf_stream_dataptr(src);
stream_length = pdf_stream_length (src);
stream_dict = pdf_stream_dict (src);
if (pdf_lookup_dict(stream_dict, "DecodeParms")) {
WARN("DecodeParams not supported.");
return -1;
}
filter = pdf_lookup_dict(stream_dict, "Filter");
if (!filter) {
pdf_add_stream(dst, stream_data, stream_length);
return 0;
#if HAVE_ZLIB
} else {
char *filter_name;
if (PDF_OBJ_NAMETYPE(filter)) {
filter_name = pdf_name_value(filter);
if (filter_name && !strcmp(filter_name, "FlateDecode"))
return add_stream_flate(dst, stream_data, stream_length);
else {
WARN("DecodeFilter \"%s\" not supported.", filter_name);
return -1;
}
} else if (PDF_OBJ_ARRAYTYPE(filter)) {
if (pdf_array_length(filter) > 1) {
WARN("Multiple DecodeFilter not supported.");
return -1;
} else {
filter_name = pdf_name_value(pdf_get_array(filter, 0));
if (filter_name && !strcmp(filter_name, "FlateDecode"))
return add_stream_flate(dst, stream_data, stream_length);
else {
WARN("DecodeFilter \"%s\" not supported.", filter_name);
return -1;
}
}
} else
ERROR("Broken PDF file?");
#endif /* HAVE_ZLIB */
}
return -1;
}
/* ximage here is the result. DONT USE IT FOR PASSING OPTIONS! */
int
pdf_include_page (pdf_ximage *ximage,
FILE *image_file,
const char *ident,
load_options options)
{
pdf_file *pf;
xform_info info;
pdf_obj *contents = NULL, *catalog;
pdf_obj *page = NULL, *resources = NULL, *markinfo = NULL;
pf = pdf_open(ident, image_file);
if (!pf)
return -1;
if (pdf_file_get_version(pf) > pdf_get_version()) {
WARN("Trying to include PDF file which has newer version number " \
"than output PDF: 1.%d.", pdf_get_version());
}
pdf_ximage_init_form_info(&info);
if (options.page_no == 0)
options.page_no = 1;
page = pdf_doc_get_page(pf,
options.page_no, options.bbox_type,
&info.bbox, &resources);
if(!page)
goto error_silent;
catalog = pdf_file_get_catalog(pf);
markinfo = pdf_deref_obj(pdf_lookup_dict(catalog, "MarkInfo"));
if (markinfo) {
pdf_obj *tmp = pdf_deref_obj(pdf_lookup_dict(markinfo, "Marked"));
pdf_release_obj(markinfo);
if (!PDF_OBJ_BOOLEANTYPE(tmp)) {
if (tmp)
pdf_release_obj(tmp);
goto error;
} else if (pdf_boolean_value(tmp)) {
WARN("PDF file is tagged... Ignoring tags.");
}
pdf_release_obj(tmp);
}
contents = pdf_deref_obj(pdf_lookup_dict(page, "Contents"));
pdf_release_obj(page);
page = NULL;
/*
* Handle page content stream.
*/
{
pdf_obj *content_new;
if (!contents) {
/*
* Empty page
*/
content_new = pdf_new_stream(0);
/* TODO: better don't include anything if the page is empty */
} else if (PDF_OBJ_STREAMTYPE(contents)) {
/*
* We must import the stream because its dictionary
* may contain indirect references.
*/
content_new = pdf_import_object(contents);
} else if (PDF_OBJ_ARRAYTYPE(contents)) {
/*
* Concatenate all content streams.
*/
int idx, len = pdf_array_length(contents);
content_new = pdf_new_stream(STREAM_COMPRESS);
for (idx = 0; idx < len; idx++) {
pdf_obj *content_seg = pdf_deref_obj(pdf_get_array(contents, idx));
if (!PDF_OBJ_STREAMTYPE(content_seg) ||
pdf_concat_stream(content_new, content_seg) < 0) {
pdf_release_obj(content_seg);
pdf_release_obj(content_new);
goto error;
}
pdf_release_obj(content_seg);
}
} else {
goto error;
}
if (contents)
pdf_release_obj(contents);
contents = content_new;
}
/*
* Add entries to contents stream dictionary.
*/
{
pdf_obj *contents_dict, *bbox, *matrix;
contents_dict = pdf_stream_dict(contents);
pdf_add_dict(contents_dict,
pdf_new_name("Type"), pdf_new_name("XObject"));
pdf_add_dict(contents_dict,
pdf_new_name("Subtype"), pdf_new_name("Form"));
pdf_add_dict(contents_dict,
pdf_new_name("FormType"), pdf_new_number(1.0));
bbox = pdf_new_array();
pdf_add_array(bbox, pdf_new_number(info.bbox.llx));
pdf_add_array(bbox, pdf_new_number(info.bbox.lly));
pdf_add_array(bbox, pdf_new_number(info.bbox.urx));
pdf_add_array(bbox, pdf_new_number(info.bbox.ury));
pdf_add_dict(contents_dict, pdf_new_name("BBox"), bbox);
matrix = pdf_new_array();
pdf_add_array(matrix, pdf_new_number(1.0));
pdf_add_array(matrix, pdf_new_number(0.0));
pdf_add_array(matrix, pdf_new_number(0.0));
pdf_add_array(matrix, pdf_new_number(1.0));
pdf_add_array(matrix, pdf_new_number(0.0));
pdf_add_array(matrix, pdf_new_number(0.0));
pdf_add_dict(contents_dict, pdf_new_name("Matrix"), matrix);
pdf_add_dict(contents_dict, pdf_new_name("Resources"),
pdf_import_object(resources));
pdf_release_obj(resources);
}
pdf_close(pf);
pdf_ximage_set_form(ximage, &info, contents);
return 0;
error:
WARN("Cannot parse document. Broken PDF file?");
error_silent:
if (resources)
pdf_release_obj(resources);
if (markinfo)
pdf_release_obj(markinfo);
if (page)
pdf_release_obj(page);
if (contents)
pdf_release_obj(contents);
pdf_close(pf);
return -1;
}
static pdf_obj*
pdf_get_page_content (pdf_obj* page)
{
pdf_obj *contents, *content_new;
contents = pdf_deref_obj(pdf_lookup_dict(page, "Contents"));
if (!contents)
return NULL;
if (pdf_obj_typeof(contents) == PDF_NULL) {
/* empty page */
pdf_release_obj(contents);
/* TODO: better don't include anything if the page is empty */
contents = pdf_new_stream(0);
} else if (PDF_OBJ_ARRAYTYPE(contents)) {
/*
* Concatenate all content streams.
*/
pdf_obj *content_seg;
int idx = 0;
content_new = pdf_new_stream(STREAM_COMPRESS);
for (;;) {
content_seg = pdf_deref_obj(pdf_get_array(contents, idx));
if (!content_seg)
break;
else if (PDF_OBJ_NULLTYPE(content_seg)) {
/* Silently ignore. */
} else if (!PDF_OBJ_STREAMTYPE(content_seg)) {
WARN("Page content not a stream object. Broken PDF file?");
pdf_release_obj(content_seg);
pdf_release_obj(content_new);
pdf_release_obj(contents);
return NULL;
} else if (pdf_concat_stream(content_new, content_seg) < 0) {
WARN("Could not handle content stream with multiple segments.");
pdf_release_obj(content_seg);
pdf_release_obj(content_new);
pdf_release_obj(contents);
return NULL;
}
pdf_release_obj(content_seg);
idx++;
}
pdf_release_obj(contents);
contents = content_new;
} else {
if (!PDF_OBJ_STREAMTYPE(contents)) {
WARN("Page content not a stream object. Broken PDF file?");
pdf_release_obj(contents);
return NULL;
}
/* Flate the contents if necessary. */
content_new = pdf_new_stream(STREAM_COMPRESS);
if (pdf_concat_stream(content_new, contents) < 0) {
WARN("Could not handle a content stream.");
pdf_release_obj(contents);
pdf_release_obj(content_new);
return NULL;
}
pdf_release_obj(contents);
contents = content_new;
}
return contents;
}
