Example #1
0
static void merge_range(const char *range)
{
	int start, end, i, count;
	pdf_graft_map *graft_map;

	count = pdf_count_pages(ctx, doc_src);
	graft_map = pdf_new_graft_map(ctx, doc_des);

	fz_try(ctx)
	{
		while ((range = fz_parse_page_range(ctx, range, &start, &end, count)))
		{
			if (start < end)
				for (i = start; i <= end; ++i)
					page_merge(i, -1, graft_map);
			else
				for (i = start; i >= end; --i)
					page_merge(i, -1, graft_map);
		}
	}
	fz_always(ctx)
	{
		pdf_drop_graft_map(ctx, graft_map);
	}
	fz_catch(ctx)
	{
		fz_rethrow(ctx);
	}
}
Example #2
0
static int _pdf_doc_count_pages(const struct _pdf_doc *self)
{
    if (self->xref)
        return pdf_count_pages(self->xref);

    return 0;
}
Example #3
0
int j_mu_count_pages(void* p_ctx, void* p_doc)
{
	fz_context* ctx = (fz_context*) p_ctx;
	pdf_document* doc = (pdf_document*) p_doc;
	int n = pdf_count_pages(ctx, doc);
	return n;
}
Example #4
0
PDFDocument* PDFDocument::Open(const std::string& path,
                               int page_cache_size) {
  fz_context* context = fz_new_context(nullptr, nullptr, FZ_STORE_DEFAULT);
  pdf_document* raw_pdf_document = nullptr;
  fz_try(context) {
    raw_pdf_document = pdf_open_document(context, path.c_str());
    if ((raw_pdf_document == nullptr) ||
        (!pdf_count_pages(context, raw_pdf_document))) {
      fz_throw(
          context,
          FZ_ERROR_GENERIC,
          const_cast<char*>("Cannot open document \"%s\""),
          path.c_str());
    }
  } fz_catch(context) {
    if (raw_pdf_document != nullptr) {
      pdf_drop_document(context, raw_pdf_document);
    }
    fz_drop_context(context);
    return nullptr;
  }

  PDFDocument* document = new PDFDocument(page_cache_size);
  document->_fz_context = context;
  document->_pdf_document = raw_pdf_document;
  return document;
}
Example #5
0
int pdfsign_main(int argc, char **argv)
{
	fz_context *ctx;
	pdf_document *doc;
	char *password = "";
	int i, n, c;
	pdf_page *page = NULL;

	while ((c = fz_getopt(argc, argv, "p:")) != -1)
	{
		switch (c)
		{
		case 'p': password = fz_optarg; break;
		default: usage(); break;
		}
	}

	if (argc - fz_optind < 1)
		usage();

	filename = argv[fz_optind++];

	ctx = fz_new_context(NULL, NULL, FZ_STORE_UNLIMITED);
	if (!ctx)
	{
		fprintf(stderr, "cannot initialize context\n");
		exit(1);
	}

	fz_var(page);

	doc = pdf_open_document(ctx, filename);
	fz_try(ctx)
	{
		if (pdf_needs_password(ctx, doc))
			if (!pdf_authenticate_password(ctx, doc, password))
				fz_warn(ctx, "cannot authenticate password: %s", filename);

		n = pdf_count_pages(ctx, doc);
		for (i = 0; i < n; ++i)
		{
			page = pdf_load_page(ctx, doc, i);
			verify_page(ctx, doc, i, page);
			fz_drop_page(ctx, (fz_page*)page);
			page = NULL;
		}
	}
	fz_always(ctx)
		pdf_drop_document(ctx, doc);
	fz_catch(ctx)
	{
		fz_drop_page(ctx, (fz_page*)page);
		fprintf(stderr, "error verify signatures: %s\n", fz_caught_message(ctx));
	}

	fz_flush_warnings(ctx);
	fz_drop_context(ctx);
	return 0;
}
Example #6
0
static void
showinfo(char *filename, int show, char *pagelist)
{
	int page, spage, epage;
	char *spec, *dash;
	int allpages;
	int pagecount;

	if (!doc)
		infousage();

	allpages = !strcmp(pagelist, "1-");

	pagecount = pdf_count_pages(doc);
	spec = fz_strsep(&pagelist, ",");
	while (spec && pagecount)
	{
		dash = strchr(spec, '-');

		if (dash == spec)
			spage = epage = pagecount;
		else
			spage = epage = atoi(spec);

		if (dash)
		{
			if (strlen(dash) > 1)
				epage = atoi(dash + 1);
			else
				epage = pagecount;
		}

		if (spage > epage)
			page = spage, spage = epage, epage = page;

		spage = fz_clampi(spage, 1, pagecount);
		epage = fz_clampi(epage, 1, pagecount);

		if (allpages)
			printf("Retrieving info from pages %d-%d...\n", spage, epage);
		for (page = spage; page <= epage; page++)
		{
			gatherpageinfo(page, show);
			if (!allpages)
			{
				printf("Page %d:\n", page);
				printinfo(filename, show, page);
				printf("\n");
			}
		}

		spec = fz_strsep(&pagelist, ",");
	}

	if (allpages)
		printinfo(filename, show, -1);
}
Example #7
0
void
pdf_insert_page(pdf_document *doc, pdf_page *page, int at)
{
	fz_context *ctx = doc->ctx;
	int count = pdf_count_pages(doc);
	pdf_obj *parent, *kids;
	pdf_obj *page_ref;
	int i;

	page_ref = pdf_new_ref(doc, page->me);

	fz_try(ctx)
	{
		if (count == 0)
		{
			/* TODO: create new page tree? */
			fz_throw(ctx, FZ_ERROR_GENERIC, "empty page tree, cannot insert page");
		}
		else if (at >= count)
		{
			if (at > count)
				fz_throw(ctx, FZ_ERROR_GENERIC, "cannot insert page beyond end of page tree");

			/* append after last page */
			pdf_lookup_page_loc(doc, count - 1, &parent, &i);
			kids = pdf_dict_gets(parent, "Kids");
			pdf_array_insert(kids, page_ref, i + 1);
		}
		else
		{
			/* insert before found page */
			pdf_lookup_page_loc(doc, at, &parent, &i);
			kids = pdf_dict_gets(parent, "Kids");
			pdf_array_insert(kids, page_ref, i);
		}

		pdf_dict_puts(page->me, "Parent", parent);

		/* Adjust page counts */
		while (parent)
		{
			int count = pdf_to_int(pdf_dict_gets(parent, "Count"));
			pdf_dict_puts_drop(parent, "Count", pdf_new_int(doc, count + 1));
			parent = pdf_dict_gets(parent, "Parent");
		}

	}
	fz_always(ctx)
	{
		pdf_drop_obj(page_ref);
	}
	fz_catch(ctx)
	{
		fz_rethrow(ctx);
	}
}
Example #8
0
static void wmupdf_convert_pages_to_forms(pdf_document *xref,fz_context *ctx,int *srcpageused)

    {
    int i,pagecount;

    pagecount = pdf_count_pages(xref);
    for (i=0;i<=pagecount;i++)
        if (srcpageused[i])
            wmupdf_convert_single_page_to_form(xref,ctx,i);
    }
Example #9
0
void juggler_page_tree_changed(juggler_t *juggler)
{
	juggler->pagecount = pdf_count_pages(juggler->ctx, juggler->pdf);
	size_t i;
	for(i = 0; i < RENDER_BUFFER_COUNT; i++)
		juggler->rendered[i].samples = NULL;
	juggler->recent_rendered = 0;
	juggler->sizes = NULL;
	get_all_pages_size(juggler);
}
Example #10
0
JNIEXPORT int JNICALL
Java_com_artifex_mupdf_MuPDFCore_openFile(JNIEnv * env, jobject thiz, jstring jfilename)
{
	const char *filename;
	int pages = 0;

	filename = (*env)->GetStringUTFChars(env, jfilename, NULL);
	if (filename == NULL)
	{
		LOGE("Failed to get filename");
		return 0;
	}

	/* 128 MB store for low memory devices. Tweak as necessary. */
	ctx = fz_new_context(NULL, NULL, 128 << 20);
	if (!ctx)
	{
		LOGE("Failed to initialise context");
		return 0;
	}

	xref = NULL;
	fz_try(ctx)
	{
		colorspace = fz_device_rgb;

		LOGE("Opening document...");
		fz_try(ctx)
		{
			xref = pdf_open_document(ctx, filename);
		}
		fz_catch(ctx)
		{
			fz_throw(ctx, "Cannot open document: '%s'\n", filename);
		}
		pages = pdf_count_pages(xref);
		LOGE("Done! %d pages", pages);
	}
	fz_catch(ctx)
	{
		LOGE("Failed: %s", ctx->error->message);
		pdf_close_document(xref);
		xref = NULL;
		fz_free_context(ctx);
		ctx = NULL;
	}

	return pages;
}
Example #11
0
void benchfile(char *pdffilename, int loadonly, int pageNo)
{
	pdf_document *xref = NULL;
	mstimer timer;
	int page_count;
	int curpage;

	fz_context *ctx = fz_new_context(NULL, NULL, FZ_STORE_DEFAULT);
	if (!ctx) {
		logbench("Error: fz_new_context() failed\n");
		return;
	}

	logbench("Starting: %s\n", pdffilename);
	timerstart(&timer);
	fz_var(xref);
	fz_try(ctx) {
		xref = openxref(ctx, pdffilename);
	}
	fz_catch(ctx) {
		goto Exit;
	}
	timerstop(&timer);
	logbench("load: %.2f ms\n", timeinms(&timer));

	page_count = pdf_count_pages(xref);
	logbench("page count: %d\n", page_count);

	if (loadonly)
		goto Exit;
	for (curpage = 1; curpage <= page_count; curpage++) {
		pdf_page *page;
		if ((-1 != pageNo) && (pageNo != curpage))
			continue;
		page = benchloadpage(ctx, xref, curpage);
		if (page) {
			benchrenderpage(ctx, xref, page, curpage);
			pdf_free_page(xref, page);
		}
	}

Exit:
	logbench("Finished: %s\n", pdffilename);
	pdf_close_document(xref);
	fz_free_context(ctx);
}
Example #12
0
static int
showpages(fz_context *ctx, pdf_document *doc, fz_output *out, char *pagelist)
{
	int page, spage, epage;
	char *spec, *dash;
	int pagecount;
	int ret = 0;

	if (!doc)
		infousage();

	pagecount = pdf_count_pages(ctx, doc);
	spec = fz_strsep(&pagelist, ",");
	while (spec && pagecount)
	{
		dash = strchr(spec, '-');

		if (dash == spec)
			spage = epage = pagecount;
		else
			spage = epage = atoi(spec);

		if (dash)
		{
			if (strlen(dash) > 1)
				epage = atoi(dash + 1);
			else
				epage = pagecount;
		}

		if (spage > epage)
			page = spage, spage = epage, epage = page;

		spage = fz_clampi(spage, 1, pagecount);
		epage = fz_clampi(epage, 1, pagecount);

		for (page = spage; page <= epage; page++)
		{
			ret |= showpage(ctx, doc, out, page);
		}

		spec = fz_strsep(&pagelist, ",");
	}

	return ret;
}
Example #13
0
static int
showpages(fz_context *ctx, pdf_document *doc, fz_output *out, const char *pagelist)
{
	int page, spage, epage;
	int pagecount;
	int ret = 0;

	if (!doc)
		infousage();

	pagecount = pdf_count_pages(ctx, doc);
	while ((pagelist = fz_parse_page_range(ctx, pagelist, &spage, &epage, pagecount)))
	{
		if (spage > epage)
			page = spage, spage = epage, epage = page;
		for (page = spage; page <= epage; page++)
			ret |= showpage(ctx, doc, out, page);
	}

	return ret;
}
Example #14
0
pdf_document *openxref(fz_context *ctx, char *filename)
{
	pdf_document *xref = pdf_open_document(ctx, filename);

	fz_try(ctx)
	{
		if (pdf_needs_password(xref))
		{
			logbench("Warning: password protected document\n");
			fz_throw(ctx, "document requires password");
		}

		pdf_count_pages(xref);
	}
	fz_catch(ctx)
	{
		pdf_close_document(xref);
		fz_rethrow(ctx);
	}

	return xref;
}
Example #15
0
int pdfinfo_main(int argc, char **argv)
{
	enum { NO_FILE_OPENED, NO_INFO_GATHERED, INFO_SHOWN } state;
	char *filename = "";
	char *password = "";
	int show = ALL;
	int c;

	while ((c = fz_getopt(argc, argv, "mfispxd:")) != -1)
	{
		switch (c)
		{
		case 'm': if (show == ALL) show = DIMENSIONS; else show |= DIMENSIONS; break;
		case 'f': if (show == ALL) show = FONTS; else show |= FONTS; break;
		case 'i': if (show == ALL) show = IMAGES; else show |= IMAGES; break;
		case 's': if (show == ALL) show = SHADINGS; else show |= SHADINGS; break;
		case 'p': if (show == ALL) show = PATTERNS; else show |= PATTERNS; break;
		case 'x': if (show == ALL) show = XOBJS; else show |= XOBJS; break;
		case 'd': password = fz_optarg; break;
		default:
			infousage();
			break;
		}
	}

	if (fz_optind == argc)
		infousage();

	ctx = fz_new_context(NULL, NULL, FZ_STORE_UNLIMITED);
	if (!ctx)
	{
		fprintf(stderr, "cannot initialise context\n");
		exit(1);
	}

	state = NO_FILE_OPENED;
	while (fz_optind < argc)
	{
		if (state == NO_FILE_OPENED || !arg_is_page_range(argv[fz_optind]))
		{
			if (state == NO_INFO_GATHERED)
			{
				showinfo(filename, show, "1-");
			}

			closexref();

			filename = argv[fz_optind];
			printf("%s:\n", filename);
			doc = pdf_open_document_no_run(ctx, filename);
			if (pdf_needs_password(doc))
				if (!pdf_authenticate_password(doc, password))
					fz_throw(ctx, FZ_ERROR_GENERIC, "cannot authenticate password: %s", filename);
			pagecount = pdf_count_pages(doc);

			showglobalinfo();
			state = NO_INFO_GATHERED;
		}
		else
		{
			showinfo(filename, show, argv[fz_optind]);
			state = INFO_SHOWN;
		}

		fz_optind++;
	}

	if (state == NO_INFO_GATHERED)
		showinfo(filename, show, "1-");

	closexref();
	fz_free_context(ctx);
	return 0;
}
Example #16
0
static void retainpages(fz_context *ctx, globals *glo, int argc, char **argv)
{
	pdf_obj *oldroot, *root, *pages, *kids, *countobj, *parent, *olddests;
	pdf_document *doc = glo->doc;
	int argidx = 0;
	pdf_obj *names_list = NULL;
	pdf_obj *outlines;
	int pagecount;
	int i;
	int *page_object_nums;

	/* Keep only pages/type and (reduced) dest entries to avoid
	 * references to unretained pages */
	oldroot = pdf_dict_get(ctx, pdf_trailer(ctx, doc), PDF_NAME_Root);
	pages = pdf_dict_get(ctx, oldroot, PDF_NAME_Pages);
	olddests = pdf_load_name_tree(ctx, doc, PDF_NAME_Dests);
	outlines = pdf_dict_get(ctx, oldroot, PDF_NAME_Outlines);

	root = pdf_new_dict(ctx, doc, 3);
	pdf_dict_put(ctx, root, PDF_NAME_Type, pdf_dict_get(ctx, oldroot, PDF_NAME_Type));
	pdf_dict_put(ctx, root, PDF_NAME_Pages, pdf_dict_get(ctx, oldroot, PDF_NAME_Pages));
	pdf_dict_put(ctx, root, PDF_NAME_Outlines, outlines);

	pdf_update_object(ctx, doc, pdf_to_num(ctx, oldroot), root);

	/* Create a new kids array with only the pages we want to keep */
	parent = pdf_new_indirect(ctx, doc, pdf_to_num(ctx, pages), pdf_to_gen(ctx, pages));
	kids = pdf_new_array(ctx, doc, 1);

	/* Retain pages specified */
	while (argc - argidx)
	{
		int page, spage, epage;
		char *spec, *dash;
		char *pagelist = argv[argidx];

		pagecount = pdf_count_pages(ctx, doc);
		spec = fz_strsep(&pagelist, ",");
		while (spec)
		{
			dash = strchr(spec, '-');

			if (dash == spec)
				spage = epage = pagecount;
			else
				spage = epage = atoi(spec);

			if (dash)
			{
				if (strlen(dash) > 1)
					epage = atoi(dash + 1);
				else
					epage = pagecount;
			}

			spage = fz_clampi(spage, 1, pagecount);
			epage = fz_clampi(epage, 1, pagecount);

			if (spage < epage)
				for (page = spage; page <= epage; ++page)
					retainpage(ctx, doc, parent, kids, page);
			else
				for (page = spage; page >= epage; --page)
					retainpage(ctx, doc, parent, kids, page);

			spec = fz_strsep(&pagelist, ",");
		}

		argidx++;
	}

	pdf_drop_obj(ctx, parent);

	/* Update page count and kids array */
	countobj = pdf_new_int(ctx, doc, pdf_array_len(ctx, kids));
	pdf_dict_put(ctx, pages, PDF_NAME_Count, countobj);
	pdf_drop_obj(ctx, countobj);
	pdf_dict_put(ctx, pages, PDF_NAME_Kids, kids);
	pdf_drop_obj(ctx, kids);

	/* Force the next call to pdf_count_pages to recount */
	glo->doc->page_count = 0;

	pagecount = pdf_count_pages(ctx, doc);
	page_object_nums = fz_calloc(ctx, pagecount, sizeof(*page_object_nums));
	for (i = 0; i < pagecount; i++)
	{
		pdf_obj *pageref = pdf_lookup_page_obj(ctx, doc, i);
		page_object_nums[i] = pdf_to_num(ctx, pageref);
	}

	/* If we had an old Dests tree (now reformed as an olddests
	 * dictionary), keep any entries in there that point to
	 * valid pages. This may mean we keep more than we need, but
	 * it's safe at least. */
	if (olddests)
	{
		pdf_obj *names = pdf_new_dict(ctx, doc, 1);
		pdf_obj *dests = pdf_new_dict(ctx, doc, 1);
		int len = pdf_dict_len(ctx, olddests);

		names_list = pdf_new_array(ctx, doc, 32);

		for (i = 0; i < len; i++)
		{
			pdf_obj *key = pdf_dict_get_key(ctx, olddests, i);
			pdf_obj *val = pdf_dict_get_val(ctx, olddests, i);
			pdf_obj *dest = pdf_dict_get(ctx, val, PDF_NAME_D);

			dest = pdf_array_get(ctx, dest ? dest : val, 0);
			if (dest_is_valid_page(ctx, dest, page_object_nums, pagecount))
			{
				pdf_obj *key_str = pdf_new_string(ctx, doc, pdf_to_name(ctx, key), strlen(pdf_to_name(ctx, key)));
				pdf_array_push(ctx, names_list, key_str);
				pdf_array_push(ctx, names_list, val);
				pdf_drop_obj(ctx, key_str);
			}
		}

		pdf_dict_put(ctx, dests, PDF_NAME_Names, names_list);
		pdf_dict_put(ctx, names, PDF_NAME_Dests, dests);
		pdf_dict_put(ctx, root, PDF_NAME_Names, names);

		pdf_drop_obj(ctx, names);
		pdf_drop_obj(ctx, dests);
		pdf_drop_obj(ctx, olddests);
	}

	/* Edit each pages /Annot list to remove any links that point to
	 * nowhere. */
	for (i = 0; i < pagecount; i++)
	{
		pdf_obj *pageref = pdf_lookup_page_obj(ctx, doc, i);
		pdf_obj *pageobj = pdf_resolve_indirect(ctx, pageref);

		pdf_obj *annots = pdf_dict_get(ctx, pageobj, PDF_NAME_Annots);

		int len = pdf_array_len(ctx, annots);
		int j;

		for (j = 0; j < len; j++)
		{
			pdf_obj *o = pdf_array_get(ctx, annots, j);

			if (!pdf_name_eq(ctx, pdf_dict_get(ctx, o, PDF_NAME_Subtype), PDF_NAME_Link))
				continue;

			if (!dest_is_valid(ctx, o, pagecount, page_object_nums, names_list))
			{
				/* Remove this annotation */
				pdf_array_delete(ctx, annots, j);
				j--;
			}
		}
	}

	if (strip_outlines(ctx, doc, outlines, pagecount, page_object_nums, names_list) == 0)
	{
		pdf_dict_del(ctx, root, PDF_NAME_Outlines);
	}

	fz_free(ctx, page_object_nums);
	pdf_drop_obj(ctx, names_list);
	pdf_drop_obj(ctx, root);
}
Example #17
0
static void retainpages(int argc, char **argv)
{
	fz_obj *oldroot, *root, *pages, *kids, *countobj, *parent, *olddests;

	/* Keep only pages/type and (reduced) dest entries to avoid
	 * references to unretained pages */
	oldroot = fz_dict_gets(xref->trailer, "Root");
	pages = fz_dict_gets(oldroot, "Pages");
	olddests = pdf_load_name_tree(xref, "Dests");

	root = fz_new_dict(ctx, 2);
	fz_dict_puts(root, "Type", fz_dict_gets(oldroot, "Type"));
	fz_dict_puts(root, "Pages", fz_dict_gets(oldroot, "Pages"));

	pdf_update_object(xref, fz_to_num(oldroot), fz_to_gen(oldroot), root);

	fz_drop_obj(root);

	/* Create a new kids array with only the pages we want to keep */
	parent = fz_new_indirect(ctx, fz_to_num(pages), fz_to_gen(pages), xref);
	kids = fz_new_array(ctx, 1);

	/* Retain pages specified */
	while (argc - fz_optind)
	{
		int page, spage, epage;
		char *spec, *dash;
		char *pagelist = argv[fz_optind];

		spec = fz_strsep(&pagelist, ",");
		while (spec)
		{
			dash = strchr(spec, '-');

			if (dash == spec)
				spage = epage = pdf_count_pages(xref);
			else
				spage = epage = atoi(spec);

			if (dash)
			{
				if (strlen(dash) > 1)
					epage = atoi(dash + 1);
				else
					epage = pdf_count_pages(xref);
			}

			if (spage > epage)
				page = spage, spage = epage, epage = page;

			if (spage < 1)
				spage = 1;
			if (epage > pdf_count_pages(xref))
				epage = pdf_count_pages(xref);

			for (page = spage; page <= epage; page++)
			{
				fz_obj *pageobj = xref->page_objs[page-1];
				fz_obj *pageref = xref->page_refs[page-1];

				fz_dict_puts(pageobj, "Parent", parent);

				/* Store page object in new kids array */
				fz_array_push(kids, pageref);
			}

			spec = fz_strsep(&pagelist, ",");
		}

		fz_optind++;
	}

	fz_drop_obj(parent);

	/* Update page count and kids array */
	countobj = fz_new_int(ctx, fz_array_len(kids));
	fz_dict_puts(pages, "Count", countobj);
	fz_drop_obj(countobj);
	fz_dict_puts(pages, "Kids", kids);
	fz_drop_obj(kids);

	/* Also preserve the (partial) Dests name tree */
	if (olddests)
	{
		int i;
		fz_obj *names = fz_new_dict(ctx, 1);
		fz_obj *dests = fz_new_dict(ctx, 1);
		fz_obj *names_list = fz_new_array(ctx, 32);

		for (i = 0; i < fz_dict_len(olddests); i++)
		{
			fz_obj *key = fz_dict_get_key(olddests, i);
			fz_obj *val = fz_dict_get_val(olddests, i);
			fz_obj *key_str = fz_new_string(ctx, fz_to_name(key), strlen(fz_to_name(key)));
			fz_obj *dest = fz_dict_gets(val, "D");

			dest = fz_array_get(dest ? dest : val, 0);
			if (fz_array_contains(fz_dict_gets(pages, "Kids"), dest))
			{
				fz_array_push(names_list, key_str);
				fz_array_push(names_list, val);
			}
			fz_drop_obj(key_str);
		}

		root = fz_dict_gets(xref->trailer, "Root");
		fz_dict_puts(dests, "Names", names_list);
		fz_dict_puts(names, "Dests", dests);
		fz_dict_puts(root, "Names", names);

		fz_drop_obj(names);
		fz_drop_obj(dests);
		fz_drop_obj(names_list);
		fz_drop_obj(olddests);
	}
}
Example #18
0
void
pdf_insert_page(pdf_document *doc, pdf_page *page, int at)
{
	fz_context *ctx = doc->ctx;
	int count = pdf_count_pages(doc);
	pdf_obj *parent, *kids;
	pdf_obj *page_ref;
	int i;

	page_ref = pdf_new_ref(doc, page->me);

	fz_try(ctx)
	{
		if (count == 0)
		{
			pdf_obj *root = pdf_dict_gets(pdf_trailer(doc), "Root");
			parent = pdf_dict_gets(root, "Pages");
			if (!parent)
				fz_throw(doc->ctx, FZ_ERROR_GENERIC, "cannot find page tree");

			kids = pdf_dict_gets(parent, "Kids");
			if (!kids)
				fz_throw(doc->ctx, FZ_ERROR_GENERIC, "malformed page tree");

			pdf_array_insert(kids, page_ref, 0);
		}
		else if (at >= count)
		{
			if (at == INT_MAX)
				at = count;

			if (at > count)
				fz_throw(ctx, FZ_ERROR_GENERIC, "cannot insert page beyond end of page tree");

			/* append after last page */
			pdf_lookup_page_loc(doc, count - 1, &parent, &i);
			kids = pdf_dict_gets(parent, "Kids");
			pdf_array_insert(kids, page_ref, i + 1);
		}
		else
		{
			/* insert before found page */
			pdf_lookup_page_loc(doc, at, &parent, &i);
			kids = pdf_dict_gets(parent, "Kids");
			pdf_array_insert(kids, page_ref, i);
		}

		pdf_dict_puts(page->me, "Parent", parent);

		/* Adjust page counts */
		while (parent)
		{
			int count = pdf_to_int(pdf_dict_gets(parent, "Count"));
			pdf_dict_puts_drop(parent, "Count", pdf_new_int(doc, count + 1));
			parent = pdf_dict_gets(parent, "Parent");
		}

	}
	fz_always(ctx)
	{
		pdf_drop_obj(page_ref);
	}
	fz_catch(ctx)
	{
		fz_rethrow(ctx);
	}

	doc->page_count = 0; /* invalidate cached value */
}
Example #19
0
static int wmupdf_pdfdoc_newpages(pdf_document *xref,fz_context *ctx,WPDFPAGEINFO *pageinfo,
                                  int use_forms,FILE *out)

    {
    static char *funcname="wmupdf_pdfdoc_newpages";
    pdf_obj *root,*oldroot,*pages,*kids,*countobj,*parent,*olddests;
    pdf_obj *srcpageobj,*srcpagecontents;
    pdf_obj *destpageobj,*destpagecontents,*destpageresources;
    double srcx0,srcy0;
    int qref,i,i0,pagecount,srccount,destpageref,nbb;
    int *srcpageused;
    char *bigbuf;
    double srcpagerot;

    /* Avoid compiler warning */
    destpageref = 0;
    destpageobj = NULL;
    srcx0=srcy0=0.;
    /* Keep only pages/type and (reduced) dest entries to avoid references to unretained pages */
    pagecount = pdf_count_pages(xref);
    if (use_forms)
        {
        willus_mem_alloc_warn((void **)&srcpageused,sizeof(int)*(pagecount+1),funcname,10);
        /* Mark all source pages as "not done" */
        for (i=0;i<=pagecount;i++)
            srcpageused[i]=0;
        nbb=4096;
        willus_mem_alloc_warn((void **)&bigbuf,nbb,funcname,10);
        bigbuf[0]='\0';
        }
    oldroot = pdf_dict_gets(xref->trailer,"Root");
    /*
    ** pages points to /Pages object in PDF file.
    ** Has:  /Type /Pages, /Count <numpages>, /Kids [ obj obj obj obj ]
    */
    pages = pdf_dict_gets(oldroot,"Pages");
    olddests = pdf_load_name_tree(xref,"Dests");

    /*
    ** Create new root object with only /Pages and /Type (and reduced dest entries)
    ** to avoid references to unretained pages.
    */
    root = pdf_new_dict(ctx,4);
    pdf_dict_puts(root,"Type",pdf_dict_gets(oldroot,"Type"));
    pdf_dict_puts(root,"Pages",pages);
    pdf_update_object(xref,pdf_to_num(oldroot),root);
    pdf_drop_obj(root);

    /* Parent indirectly references the /Pages object in the file */
    /* (Each new page we create has to point to this.)            */
    parent = pdf_new_indirect(ctx, pdf_to_num(pages), pdf_to_gen(pages), xref);
    /* Create a new kids array with only the pages we want to keep */
    kids = pdf_new_array(ctx, 1);


    qref=0;
    /* Avoid compiler warnings */
    destpageresources=NULL;
    destpagecontents=NULL;
    srcpagecontents=NULL;
    srcpagerot=0.;
    for (i=0;i<=pageinfo->boxes.n;i++)
        if (pageinfo->boxes.box[i].dstpage>0)
            break;
    if (i>0)
        {
        if (i<pageinfo->boxes.n)
            memmove(&pageinfo->boxes.box[0],&pageinfo->boxes.box[i],sizeof(WPDFBOX)*pageinfo->boxes.n-i);
        pageinfo->boxes.n -= i;
        }
    /* Walk through PFDBOXES array */
    for (i=srccount=i0=0;i<=pageinfo->boxes.n;i++)
        {
        WPDFBOX *box;
        int j,k,newsrc;
        static char buf[512];
        pdf_obj *s1indirect,*qindirect,*rotobj;
        static double cpm[3][3],m[3][3],m1[3][3];
        static double xclip[4],yclip[4];

/*
printf("box[%d]\n",i);
if (i<pageinfo->boxes.n)
{
box=&pageinfo->boxes.box[i];
printf("    srcpage=%d, dstpage=%d\n",box->srcbox.pageno,box->dstpage);
printf("    x0=%g, y0=%g\n",box->x0,box->y0);
printf("    w=%g, h=%g\n",box->w,box->h);
printf("    x1=%g, y1=%g\n",box->x1,box->y1);
printf("    sr=%g, dr=%g\n",box->srcrot_deg,box->dstrot_deg);
printf("    scale=%g\n",box->scale);
}
*/
        /* Check to see if we are done with an output page */
        if (srccount>0 && (i==pageinfo->boxes.n
               || (i>0 && pageinfo->boxes.box[i].dstpage!=pageinfo->boxes.box[i-1].dstpage)))
            {
            pdf_obj *newpageref;
            /*
            ** Store destination page into document structure
            */
/*
printf("    ADDING NEW PAGE. (srccount=%d)\n",srccount);
*/
            if (use_forms)
                {
                pdf_obj *dest_stream;

                /* Create new object in document for destination page stream */
                dest_stream = pdf_new_indirect(ctx,new_stream_object(xref,ctx,bigbuf),
                                               0,(void *)xref);
                /* Store this into the destination page contents array */
                pdf_array_push(destpagecontents,dest_stream);
                pdf_drop_obj(dest_stream);
                }
            newpageref=pdf_new_indirect(ctx,destpageref,0,(void *)xref);
            /* Reference parent list of pages */
            pdf_dict_puts(destpageobj,"Parent",parent);
            pdf_dict_puts(destpageobj,"Contents",destpagecontents);
            pdf_dict_puts(destpageobj,"Resources",destpageresources);
            /* Store page object in document's kids array */
            pdf_array_push(kids,newpageref);
            /* Update document with new page */
            pdf_update_object(xref,destpageref,destpageobj);
            /* Clean up */
            pdf_drop_obj(newpageref);
            pdf_drop_obj(destpageresources);
            pdf_drop_obj(destpagecontents);
            pdf_drop_obj(destpageobj);
            /* Reset source page and index to start of new destination page */
            i0=i;
            srccount=0;
            }
        /* Quit loop if beyond last box */
        if (i>=pageinfo->boxes.n)
            break;
        box=&pageinfo->boxes.box[i];
        if (box->srcbox.pageno<1 || box->srcbox.pageno>pagecount)
            continue;
        /* Is this a source page we haven't processed yet (for this destination page)? */
        for (newsrc=1,j=i0;j<i;j++)
            {
            if (pageinfo->boxes.box[j].srcbox.pageno==box->srcbox.pageno)
                {
                newsrc=0;
                break;
                }
            }
        if (newsrc)
            {
            double v[4];

            srccount++;
            if (use_forms)
                srcpageused[box->srcbox.pageno]=1;
/*
printf("    NEW SOURCE PAGE (srccount=%d)\n",srccount);
*/
            if (srccount==1)
                {
                /*
                ** Start a new destination page.
                **
                ** Each new page object is a dict type with:
                ** /Type /Page
                ** /Contents (array of objects)
                ** /Resources (dict)
                ** /MediaBox [0 0 612 792]
                ** /Parent <PagesObj>
                ** [Can have /Rotate 90, for example.]
                **
                */
/*
printf("        (STARTING NEW DEST. PAGE)\n");
*/
                destpageobj=start_new_destpage(ctx,box->dst_width_pts,box->dst_height_pts);
                destpageresources=pdf_new_dict(ctx,1);
                if (use_forms)
                    pdf_dict_puts(destpageresources,"XObject",pdf_new_dict(ctx,1));
                destpageref=pdf_create_object(xref);
                destpagecontents=pdf_new_array(ctx,1);
                /* Init the destination page stream for forms */
                if (use_forms)
                    bigbuf[0]='\0';
                }
            /* New source page, so get the source page objects */
            srcpageobj = xref->page_objs[box->srcbox.pageno-1];
            wmupdf_page_bbox(srcpageobj,v);
            srcx0=v[0];
            srcy0=v[1];
/*
printf("SRCX0=%g, SRCY0=%g\n",srcx0,srcy0);
*/
            rotobj=pdf_dict_gets(srcpageobj,"Rotate");
            srcpagerot = rotobj!=NULL ? pdf_to_real(rotobj) : 0.;
/*
printf("Page rotation = %g\n",srcpagerot);
*/
            srcpagecontents=pdf_dict_gets(srcpageobj,"Contents");
/*
if (pdf_is_array(srcpagecontents))
{
int k;
printf("    source page contents = array.\n");
for (k=0;k<pdf_array_len(srcpagecontents);k++)
{
pdf_obj *obj;
obj=pdf_array_get(srcpagecontents,k);
if (pdf_is_indirect(obj))
{
printf("    contents[%d] = indirect (%d)\n",k,pdf_to_num(obj));
pdf_resolve_indirect(obj);
}
}
}
*/
            if (use_forms)
                {
                pdf_obj *xobjdict;
                int pageno;

                xobjdict=pdf_dict_gets(destpageresources,"XObject");
                pageno=box->srcbox.pageno;
                pdf_dict_puts(xobjdict,xobject_name(pageno),xref->page_refs[pageno-1]);
                pdf_dict_puts(destpageresources,"XObject",xobjdict);
                }
            else
                {
                pdf_obj *srcpageresources;

                /* Merge source page resources into destination page resources */
                srcpageresources=pdf_dict_gets(srcpageobj,"Resources");
/*
printf("box->dstpage=%d, srcpage=%d (ind.#=%d)\n",box->dstpage,box->srcbox.pageno,pdf_to_num(xref->page_refs[box->srcbox.pageno-1]));
*/
                wmupdf_dict_merge(ctx,"Resources",destpageresources,srcpageresources);
                }
            }
        /*
        ** Process this source box:
        **
        ** Create a tranformation matrix and clipping path to only show the
        ** desired part of the source page at the appropriate place on the
        ** destination page.
        **
        ** How the tranformation matrix works:
        ** - Translations shall be specified as [ 1 0 0 1 tx ty ], where tx and ty
        **   shall be the distances to translate the origin of the coordinate system
        **   in the horizontal and vertical dimensions, respectively.
        **
        ** - Scaling shall be obtained by [ sx 0 0 sy 0 0 ]. This scales the coordinates
        **   so that 1 unit in the horizontal and vertical dimensions of the new coordinate
        **   system is the same size as sx and sy units, respectively, in the previous
        **   coordinate system.
        **
        ** - Rotations shall be produced by [ cos q sin q -sin q cos q 0 0 ], which has the
        **   effect of rotating the coordinate system axes by an angle q counter-clockwise.
        **
        ** - Skew shall be specified by [ 1 tan a tan b 1 0 0 ], which skews the x axis by
        **   an angle a and the y axis by an angle b.
        **
        */
        wpdfbox_determine_original_source_position(box);
/*
printf("Before unrotate.\n");
printf("box->srcrot=%g\n",box->srcrot_deg);
printf("box->x0=%g, box->y0=%g\n",box->x0,box->y0);
printf("box->w=%g, box->h=%g\n",box->w,box->h);
printf("box->pw=%g, box->ph=%g\n",box->src_width_pts,box->src_height_pts);
*/
        if (fabs(srcpagerot) > 1.0e-4)
            wpdfbox_unrotate(box,srcpagerot);
/*
printf("box->srcrot=%g\n",box->srcrot_deg);
printf("box->x0=%g, box->y0=%g\n",box->x0,box->y0);
printf("box->w=%g, box->h=%g\n",box->w,box->h);
printf("box->pw=%g, box->ph=%g\n",box->src_width_pts,box->src_height_pts);
*/
        matrix_unity(m,1.);
/*
printf("xfmatrix = [  %9.6f   %9.6f   %9.6f  ]\n"
       "           [  %9.6f   %9.6f   %9.6f  ]\n"
       "           [  %9.6f   %9.6f   %9.6f  ]\n",
        m[0][0],m[0][1],m[0][2],
        m[1][0],m[1][1],m[1][2],
        m[2][0],m[2][1],m[2][2]);
*/
        matrix_translate(m1,-box->x0-srcx0,-box->y0-srcy0);
        matrix_mul(m,m1);
        matrix_rotate(m1,-box->srcrot_deg+box->dstrot_deg);
        matrix_mul(m,m1);
        matrix_unity(m1,box->scale);
        matrix_mul(m,m1);
        matrix_translate(m1,box->x1,box->y1);
        matrix_mul(m,m1);
        matrix_zero_round(m);
        matrix_rotate(cpm,box->srcrot_deg);
        matrix_translate(m1,box->x0+srcx0,box->y0+srcy0);
        matrix_mul(cpm,m1);
/*
printf("Clip matrix:\n");
printf("xfmatrix = [  %9.6f   %9.6f   %9.6f  ]\n"
       "           [  %9.6f   %9.6f   %9.6f  ]\n"
       "           [  %9.6f   %9.6f   %9.6f  ]\n",
        cpm[0][0],cpm[0][1],cpm[0][2],
        cpm[1][0],cpm[1][1],cpm[1][2],
        cpm[2][0],cpm[2][1],cpm[2][2]);
*/


        set_clip_array(xclip,yclip,box->srcrot_deg,box->w,box->h);
        for (k=0;k<4;k++)
            matrix_xymul(cpm,&xclip[k],&yclip[k]);
/*
printf("Clip path:\n    %7.2f %7.2f\n    %7.2f,%7.2f\n    %7.2f,%7.2f\n"
                   "    %7.2f %7.2f\n    %7.2f,%7.2f\n",
                xclip[0],yclip[0],xclip[1],yclip[1],xclip[2],yclip[2],
                xclip[3],yclip[3],xclip[0],yclip[0]);
*/
        strcpy(buf,"q");
        for (k=0;k<=2;k++)
            {
            cat_pdf_double(buf,m[k][0]);
            cat_pdf_double(buf,m[k][1]);
            }
        strcat(buf," cm");
        for (k=0;k<=4;k++)
            {
            cat_pdf_double(buf,xclip[k&3]);
            cat_pdf_double(buf,yclip[k&3]);
            strcat(buf,k==0 ? " m" : " l");
            }
        strcat(buf," W n");
        if (use_forms)
            {
            /* FORM METHOD */
            sprintf(&buf[strlen(buf)]," /%s Do Q\n",xobject_name(box->srcbox.pageno));
            if (strlen(bigbuf)+strlen(buf) > nbb)
                {
                int newsize;
                newsize=nbb*2;
                willus_mem_realloc_robust_warn((void **)&bigbuf,newsize,nbb,funcname,10);
                nbb=newsize;
                }
            strcat(bigbuf,buf);
            }
        else
            {
            /* NO-FORMS METHOD */
            strcat(buf,"\n");
            /* Create new objects in document for tx matrix and restore matrix */
            s1indirect = pdf_new_indirect(ctx,new_stream_object(xref,ctx,buf),0,(void *)xref);
            if (qref==0)
                qref=new_stream_object(xref,ctx,"Q\n");
            qindirect = pdf_new_indirect(ctx,qref,0,(void *)xref);
            /* Store this region into the destination page contents array */
            pdf_array_push(destpagecontents,s1indirect);
            if (pdf_is_array(srcpagecontents))
                {
                int k;
                for (k=0;k<pdf_array_len(srcpagecontents);k++)
                    pdf_array_push(destpagecontents,pdf_array_get(srcpagecontents,k));
                }
            else
                pdf_array_push(destpagecontents,srcpagecontents);
            pdf_array_push(destpagecontents,qindirect);
            pdf_drop_obj(s1indirect);
            pdf_drop_obj(qindirect);
            }
        }
    pdf_drop_obj(parent);

    /* For forms, convert all original source pages to XObject Forms */
    if (use_forms)
        wmupdf_convert_pages_to_forms(xref,ctx,srcpageused);

    /* Update page count and kids array */
    countobj = pdf_new_int(ctx, pdf_array_len(kids));
    pdf_dict_puts(pages, "Count", countobj);
    pdf_drop_obj(countobj);
    pdf_dict_puts(pages, "Kids", kids);
    pdf_drop_obj(kids);

    /* Also preserve the (partial) Dests name tree */
    if (olddests)
        wmupdf_preserve_old_dests(olddests,ctx,xref,pages);
    if (use_forms)
        {
        /* Free memory */
        willus_mem_free((double **)&bigbuf,funcname);
        willus_mem_free((double **)&srcpageused,funcname);
        }
    return(0);
    }
static void retainpages(fz_context *ctx, globals *glo, int argc, char **argv)
{
	pdf_obj *oldroot, *root, *pages, *kids, *countobj, *parent, *olddests;
	pdf_document *doc = glo->doc;
	int argidx = 0;
	pdf_obj *names_list = NULL;
	int pagecount;
	int i;

	/* Keep only pages/type and (reduced) dest entries to avoid
	 * references to unretained pages */
	oldroot = pdf_dict_get(ctx, pdf_trailer(ctx, doc), PDF_NAME_Root);
	pages = pdf_dict_get(ctx, oldroot, PDF_NAME_Pages);
	olddests = pdf_load_name_tree(ctx, doc, PDF_NAME_Dests);

	root = pdf_new_dict(ctx, doc, 2);
	pdf_dict_put(ctx, root, PDF_NAME_Type, pdf_dict_get(ctx, oldroot, PDF_NAME_Type));
	pdf_dict_put(ctx, root, PDF_NAME_Pages, pdf_dict_get(ctx, oldroot, PDF_NAME_Pages));

	pdf_update_object(ctx, doc, pdf_to_num(ctx, oldroot), root);

	pdf_drop_obj(ctx, root);

	/* Create a new kids array with only the pages we want to keep */
	parent = pdf_new_indirect(ctx, doc, pdf_to_num(ctx, pages), pdf_to_gen(ctx, pages));
	kids = pdf_new_array(ctx, doc, 1);

	/* Retain pages specified */
	while (argc - argidx)
	{
		int page, spage, epage;
		char *spec, *dash;
		char *pagelist = argv[argidx];

		pagecount = pdf_count_pages(ctx, doc);
		spec = fz_strsep(&pagelist, ",");
		while (spec)
		{
			dash = strchr(spec, '-');

			if (dash == spec)
				spage = epage = pagecount;
			else
				spage = epage = atoi(spec);

			if (dash)
			{
				if (strlen(dash) > 1)
					epage = atoi(dash + 1);
				else
					epage = pagecount;
			}

			spage = fz_clampi(spage, 1, pagecount);
			epage = fz_clampi(epage, 1, pagecount);

			if (spage < epage)
				for (page = spage; page <= epage; ++page)
					retainpage(ctx, doc, parent, kids, page);
			else
				for (page = spage; page >= epage; --page)
					retainpage(ctx, doc, parent, kids, page);

			spec = fz_strsep(&pagelist, ",");
		}

		argidx++;
	}

	pdf_drop_obj(ctx, parent);

	/* Update page count and kids array */
	countobj = pdf_new_int(ctx, doc, pdf_array_len(ctx, kids));
	pdf_dict_put(ctx, pages, PDF_NAME_Count, countobj);
	pdf_drop_obj(ctx, countobj);
	pdf_dict_put(ctx, pages, PDF_NAME_Kids, kids);
	pdf_drop_obj(ctx, kids);

	/* Also preserve the (partial) Dests name tree */
	if (olddests)
	{
		pdf_obj *names = pdf_new_dict(ctx, doc, 1);
		pdf_obj *dests = pdf_new_dict(ctx, doc, 1);
		int len = pdf_dict_len(ctx, olddests);

		names_list = pdf_new_array(ctx, doc, 32);

		for (i = 0; i < len; i++)
		{
			pdf_obj *key = pdf_dict_get_key(ctx, olddests, i);
			pdf_obj *val = pdf_dict_get_val(ctx, olddests, i);
			pdf_obj *dest = pdf_dict_get(ctx, val, PDF_NAME_D);

			dest = pdf_array_get(ctx, dest ? dest : val, 0);
			if (pdf_array_contains(ctx, pdf_dict_get(ctx, pages, PDF_NAME_Kids), dest))
			{
				pdf_obj *key_str = pdf_new_string(ctx, doc, pdf_to_name(ctx, key), strlen(pdf_to_name(ctx, key)));
				pdf_array_push(ctx, names_list, key_str);
				pdf_array_push(ctx, names_list, val);
				pdf_drop_obj(ctx, key_str);
			}
		}

		root = pdf_dict_get(ctx, pdf_trailer(ctx, doc), PDF_NAME_Root);
		pdf_dict_put(ctx, dests, PDF_NAME_Names, names_list);
		pdf_dict_put(ctx, names, PDF_NAME_Dests, dests);
		pdf_dict_put(ctx, root, PDF_NAME_Names, names);

		pdf_drop_obj(ctx, names);
		pdf_drop_obj(ctx, dests);
		pdf_drop_obj(ctx, names_list);
		pdf_drop_obj(ctx, olddests);
	}

	/* Force the next call to pdf_count_pages to recount */
	glo->doc->page_count = 0;

	/* Edit each pages /Annot list to remove any links that point to
	 * nowhere. */
	pagecount = pdf_count_pages(ctx, doc);
	for (i = 0; i < pagecount; i++)
	{
		pdf_obj *pageref = pdf_lookup_page_obj(ctx, doc, i);
		pdf_obj *pageobj = pdf_resolve_indirect(ctx, pageref);

		pdf_obj *annots = pdf_dict_get(ctx, pageobj, PDF_NAME_Annots);

		int len = pdf_array_len(ctx, annots);
		int j;

		for (j = 0; j < len; j++)
		{
			pdf_obj *o = pdf_array_get(ctx, annots, j);
			pdf_obj *p;

			if (!pdf_name_eq(ctx, pdf_dict_get(ctx, o, PDF_NAME_Subtype), PDF_NAME_Link))
				continue;

			p = pdf_dict_get(ctx, o, PDF_NAME_A);
			if (!pdf_name_eq(ctx, pdf_dict_get(ctx, p, PDF_NAME_S), PDF_NAME_GoTo))
				continue;

			if (string_in_names_list(ctx, pdf_dict_get(ctx, p, PDF_NAME_D), names_list))
				continue;

			/* FIXME: Should probably look at Next too */

			/* Remove this annotation */
			pdf_array_delete(ctx, annots, j);
			j--;
		}
	}
}
Example #21
0
void pdfapp_open(pdfapp_t *app, char *filename)
{
#if 0
	fz_error error;
	fz_obj *obj;
	char *password = "";
#else
	fz_error error;
	fz_stream *file;
	char *password = "";
	fz_obj *obj;
	fz_obj *info;

	int fd;
	fd = open(filename, O_BINARY | O_RDONLY, 0666);
	if (fd < 0)
		fprintf(stderr, "error, file %s does not exist\n", filename);
#endif

	/*
	 * Open PDF and load xref table
	 */

	app->filename = filename;

#if 0
	app->xref = pdf_newxref();
	error = pdf_loadxref(app->xref, filename);
	if (error)
	{
		fz_catch(error, "trying to air");
		error = pdf_repairxref(app->xref, filename);
		if (error)
			pdfapp_error(app, error);
	}

	error = pdf_decryptxref(app->xref);
	if (error)
		pdfapp_error(app, error);
#else
	file = fz_open_fd(fd);
	error = pdf_open_xref_with_stream(&app->xref, file, NULL);
	if (error)
		pdfapp_error(app, fz_rethrow(error, "cannot open document '%s'", filename));
	fz_close(file);
#endif

	/*
	 * Handle encrypted PDF files
	 */

	if (pdf_needs_password(app->xref))
	{
		int okay = pdf_authenticate_password(app->xref, password);
		while (!okay)
		{	
		        //password = winpassword(app, filename);
			if (!password)
				exit(1);
			okay = pdf_authenticate_password(app->xref, password);
			if (!okay)
				pdfapp_warn(app, "Invalid password.");
		}
	}

	/*
	 * Load meta information
	 * TODO: move this into mupdf library
	 */

#if 0
	obj = fz_dictgets(app->xref->trailer, "Root");
	app->xref->root = fz_resolveindirect(obj);
	if (!app->xref->root)
		pdfapp_error(app, fz_throw("syntaxerror: missing Root object"));
	fz_keepobj(app->xref->root);

	obj = fz_dictgets(app->xref->trailer, "Info");
	app->xref->info = fz_resolveindirect(obj);
	if (!app->xref->info)
		pdfapp_warn(app, "Could not load PDF meta information.");
	if (app->xref->info)
		fz_keepobj(app->xref->info);

	/*app->outline = pdf_loadoutline(app->xref);*/

	app->doctitle = filename;
	if (strrchr(app->doctitle, '\\'))
		app->doctitle = strrchr(app->doctitle, '\\') + 1;
	if (strrchr(app->doctitle, '/'))
		app->doctitle = strrchr(app->doctitle, '/') + 1;
	if (app->xref->info)
	{
		obj = fz_dictgets(app->xref->info, "Title");
		if (obj)
		{
			app->doctitle = pdf_toutf8(obj);
		}
	}
#else
	app->outline = pdf_load_outline(app->xref);

	app->doctitle = filename;
	if (strrchr(app->doctitle, '\\'))
		app->doctitle = strrchr(app->doctitle, '\\') + 1;
	if (strrchr(app->doctitle, '/'))
		app->doctitle = strrchr(app->doctitle, '/') + 1;
	info = fz_dict_gets(app->xref->trailer, "Info");
	if (info)
	{
		obj = fz_dict_gets(info, "Title");
		if (obj)
			app->doctitle = pdf_to_utf8(obj);
	}
#endif

	/*
	 * Start at first page
	 */

#if 0
	app->pagecount = pdf_getpagecount(app->xref);
#else
	error = pdf_load_page_tree(app->xref);
	if (error)
		pdfapp_error(app, fz_rethrow(error, "cannot load page tree"));

	app->pagecount = pdf_count_pages(app->xref);
#endif
	
	app->rotate = 0;
	
}
Example #22
0
static void decimatepages(fz_context *ctx, pdf_document *doc)
{
	pdf_obj *oldroot, *root, *pages, *kids, *parent;
	int num_pages = pdf_count_pages(ctx, doc);
	int page, kidcount;

	oldroot = pdf_dict_get(ctx, pdf_trailer(ctx, doc), PDF_NAME_Root);
	pages = pdf_dict_get(ctx, oldroot, PDF_NAME_Pages);

	root = pdf_new_dict(ctx, doc, 2);
	pdf_dict_put(ctx, root, PDF_NAME_Type, pdf_dict_get(ctx, oldroot, PDF_NAME_Type));
	pdf_dict_put(ctx, root, PDF_NAME_Pages, pdf_dict_get(ctx, oldroot, PDF_NAME_Pages));

	pdf_update_object(ctx, doc, pdf_to_num(ctx, oldroot), root);

	pdf_drop_obj(ctx, root);

	/* Create a new kids array with our new pages in */
	parent = pdf_new_indirect(ctx, doc, pdf_to_num(ctx, pages), pdf_to_gen(ctx, pages));
	kids = pdf_new_array(ctx, doc, 1);

	kidcount = 0;
	for (page=0; page < num_pages; page++)
	{
		pdf_page *page_details = pdf_load_page(ctx, doc, page);
		int xf = x_factor, yf = y_factor;
		int x, y;
		float w = page_details->mediabox.x1 - page_details->mediabox.x0;
		float h = page_details->mediabox.y1 - page_details->mediabox.y0;

		if (xf == 0 && yf == 0)
		{
			/* Nothing specified, so split along the long edge */
			if (w > h)
				xf = 2, yf = 1;
			else
				xf = 1, yf = 2;
		}
		else if (xf == 0)
			xf = 1;
		else if (yf == 0)
			yf = 1;

		for (y = yf-1; y >= 0; y--)
		{
			for (x = 0; x < xf; x++)
			{
				pdf_obj *newpageobj, *newpageref, *newmediabox;
				fz_rect mb;
				int num;

				newpageobj = pdf_copy_dict(ctx, pdf_lookup_page_obj(ctx, doc, page));
				num = pdf_create_object(ctx, doc);
				pdf_update_object(ctx, doc, num, newpageobj);
				newpageref = pdf_new_indirect(ctx, doc, num, 0);

				newmediabox = pdf_new_array(ctx, doc, 4);

				mb.x0 = page_details->mediabox.x0 + (w/xf)*x;
				if (x == xf-1)
					mb.x1 = page_details->mediabox.x1;
				else
					mb.x1 = page_details->mediabox.x0 + (w/xf)*(x+1);
				mb.y0 = page_details->mediabox.y0 + (h/yf)*y;
				if (y == yf-1)
					mb.y1 = page_details->mediabox.y1;
				else
					mb.y1 = page_details->mediabox.y0 + (h/yf)*(y+1);

				pdf_array_push(ctx, newmediabox, pdf_new_real(ctx, doc, mb.x0));
				pdf_array_push(ctx, newmediabox, pdf_new_real(ctx, doc, mb.y0));
				pdf_array_push(ctx, newmediabox, pdf_new_real(ctx, doc, mb.x1));
				pdf_array_push(ctx, newmediabox, pdf_new_real(ctx, doc, mb.y1));

				pdf_dict_put(ctx, newpageobj, PDF_NAME_Parent, parent);
				pdf_dict_put(ctx, newpageobj, PDF_NAME_MediaBox, newmediabox);

				/* Store page object in new kids array */
				pdf_array_push(ctx, kids, newpageref);

				kidcount++;
			}
		}
	}

	pdf_drop_obj(ctx, parent);

	/* Update page count and kids array */
	pdf_dict_put(ctx, pages, PDF_NAME_Count, pdf_new_int(ctx, doc, kidcount));
	pdf_dict_put(ctx, pages, PDF_NAME_Kids, kids);
	pdf_drop_obj(ctx, kids);
}
Example #23
0
static void decimatepages(pdf_document *xref)
{
	pdf_obj *oldroot, *root, *pages, *kids, *parent;
	fz_context *ctx = xref->ctx;
	int num_pages = pdf_count_pages(xref);
	int page, kidcount;

	/* Keep only pages/type and (reduced) dest entries to avoid
	 * references to unretained pages */
	oldroot = pdf_dict_gets(pdf_trailer(xref), "Root");
	pages = pdf_dict_gets(oldroot, "Pages");

	root = pdf_new_dict(ctx, 2);
	pdf_dict_puts(root, "Type", pdf_dict_gets(oldroot, "Type"));
	pdf_dict_puts(root, "Pages", pdf_dict_gets(oldroot, "Pages"));

	pdf_update_object(xref, pdf_to_num(oldroot), root);

	pdf_drop_obj(root);

	/* Create a new kids array with only the pages we want to keep */
	parent = pdf_new_indirect(ctx, pdf_to_num(pages), pdf_to_gen(pages), xref);
	kids = pdf_new_array(ctx, 1);

	kidcount = 0;
	for (page=0; page < num_pages; page++)
	{
		pdf_page *page_details = pdf_load_page(xref, page);
		int xf = x_factor, yf = y_factor;
		int x, y;
		float w = page_details->mediabox.x1 - page_details->mediabox.x0;
		float h = page_details->mediabox.y1 - page_details->mediabox.y0;

		if (xf == 0 && yf == 0)
		{
			/* Nothing specified, so split along the long edge */
			if (w > h)
				xf = 2, yf = 1;
			else
				xf = 1, yf = 2;
		}
		else if (xf == 0)
			xf = 1;
		else if (yf == 0)
			yf = 1;

		for (y = yf-1; y >= 0; y--)
		{
			for (x = 0; x < xf; x++)
			{
				pdf_obj *newpageobj, *newpageref, *newmediabox;
				fz_rect mb;
				int num;

				newpageobj = pdf_copy_dict(ctx, xref->page_objs[page]);
				num = pdf_create_object(xref);
				pdf_update_object(xref, num, newpageobj);
				newpageref = pdf_new_indirect(ctx, num, 0, xref);

				newmediabox = pdf_new_array(ctx, 4);

				mb.x0 = page_details->mediabox.x0 + (w/xf)*x;
				if (x == xf-1)
					mb.x1 = page_details->mediabox.x1;
				else
					mb.x1 = page_details->mediabox.x0 + (w/xf)*(x+1);
				mb.y0 = page_details->mediabox.y0 + (h/yf)*y;
				if (y == yf-1)
					mb.y1 = page_details->mediabox.y1;
				else
					mb.y1 = page_details->mediabox.y0 + (h/yf)*(y+1);

				pdf_array_push(newmediabox, pdf_new_real(ctx, mb.x0));
				pdf_array_push(newmediabox, pdf_new_real(ctx, mb.y0));
				pdf_array_push(newmediabox, pdf_new_real(ctx, mb.x1));
				pdf_array_push(newmediabox, pdf_new_real(ctx, mb.y1));

				pdf_dict_puts(newpageobj, "Parent", parent);
				pdf_dict_puts(newpageobj, "MediaBox", newmediabox);

				/* Store page object in new kids array */
				pdf_array_push(kids, newpageref);

				kidcount++;
			}
		}
	}

	pdf_drop_obj(parent);

	/* Update page count and kids array */
	pdf_dict_puts(pages, "Count", pdf_new_int(ctx, kidcount));
	pdf_dict_puts(pages, "Kids", kids);
	pdf_drop_obj(kids);
}
Example #24
0
	int Document::GetNumPages () const
	{
		return pdf_count_pages (MuDoc_);
	}
Example #25
0
static int _pdf_doc_load(struct _pdf_doc *self, mume_stream_t *stm)
{
    fz_error error;
    fz_stream *fzstm;
    fz_rect *mbox;
    fz_obj *page_obj, *box_obj;
    int i, c;

    _pdf_doc_clear(self);

    fzstm = fz_new_stream(stm, _pdf_stream_read, _pdf_stream_close);
    mume_stream_reference(stm);
    fzstm->seek = _pdf_stream_seek;
    error = pdf_open_xref_with_stream(&self->xref, fzstm, NULL);
    fz_close(fzstm);
    if (error) {
        mume_error(("Read xref failed\n", error));
        return 0;
    }

    assert(!pdf_needs_password(self->xref));

    /* Load meta information. */
    error = pdf_load_page_tree(self->xref);
    if (error) {
        mume_error(("Cannot load page tree\n"));
        return 0;
    }

    c = pdf_count_pages(self->xref);
    self->glyph_cache = fz_new_glyph_cache();
    self->pages = calloc_abort(c, sizeof(pdf_page*));
    self->disps = calloc_abort(c, sizeof(fz_display_list*));
    self->media_boxes = malloc_abort(c * sizeof(fz_rect));
    self->page_rotates = malloc_abort(c * sizeof(int));

    /* Extract each pages' media box and rotation. */
    for (i = 0; i < c; ++i) {
        mbox = self->media_boxes + i;
        page_obj = self->xref->page_objs[i];
        if (!page_obj) {
            *mbox = fz_empty_rect;
            continue;
        }

        box_obj = fz_dict_gets(page_obj, "MediaBox");
        *mbox = pdf_to_rect(box_obj);
        if (fz_is_empty_rect(*mbox)) {
            fz_warn("Cannot find page bounds, guessing page bounds.");
            mbox->x1 = 612;
            mbox->y1 = 792;
        }

        box_obj = fz_dict_gets(page_obj, "CropBox");
        if (fz_is_array(box_obj))
            *mbox = fz_intersect_rect(*mbox, pdf_to_rect(box_obj));

        self->page_rotates[i] = fz_to_int(
            fz_dict_gets(page_obj, "Rotate"));

        if (self->page_rotates[i] % 90)
            self->page_rotates[i] = 0;
    }

    return 1;
}
Example #26
0
int PDFDocument::GetNumPages() {
  return pdf_count_pages(_fz_context, _pdf_document);
}
Example #27
0
int PDFDocument::GetPageCount() {
  return pdf_count_pages(_pdf_document);
}
Example #28
0
char *
pdf_parse_link_action(fz_context *ctx, pdf_document *doc, pdf_obj *action, int pagenum)
{
	pdf_obj *obj, *dest, *file_spec;

	if (!action)
		return NULL;

	obj = pdf_dict_get(ctx, action, PDF_NAME_S);
	if (pdf_name_eq(ctx, PDF_NAME_GoTo, obj))
	{
		dest = pdf_dict_get(ctx, action, PDF_NAME_D);
		return pdf_parse_link_dest(ctx, doc, dest);
	}
	else if (pdf_name_eq(ctx, PDF_NAME_URI, obj))
	{
		/* URI entries are ASCII strings */
		const char *uri = pdf_to_str_buf(ctx, pdf_dict_get(ctx, action, PDF_NAME_URI));
		if (!fz_is_external_link(ctx, uri))
		{
			pdf_obj *uri_base_obj = pdf_dict_getp(ctx, pdf_trailer(ctx, doc), "Root/URI/Base");
			const char *uri_base = uri_base_obj ? pdf_to_str_buf(ctx, uri_base_obj) : "file://";
			char *new_uri = fz_malloc(ctx, strlen(uri_base) + strlen(uri) + 1);
			strcpy(new_uri, uri_base);
			strcat(new_uri, uri);
			return new_uri;
		}
		return fz_strdup(ctx, uri);
	}
	else if (pdf_name_eq(ctx, PDF_NAME_Launch, obj))
	{
		file_spec = pdf_dict_get(ctx, action, PDF_NAME_F);
		return pdf_parse_file_spec(ctx, doc, file_spec, NULL);
	}
	else if (pdf_name_eq(ctx, PDF_NAME_GoToR, obj))
	{
		dest = pdf_dict_get(ctx, action, PDF_NAME_D);
		file_spec = pdf_dict_get(ctx, action, PDF_NAME_F);
		return pdf_parse_file_spec(ctx, doc, file_spec, dest);
	}
	else if (pdf_name_eq(ctx, PDF_NAME_Named, obj))
	{
		dest = pdf_dict_get(ctx, action, PDF_NAME_N);

		if (pdf_name_eq(ctx, PDF_NAME_FirstPage, dest))
			pagenum = 0;
		else if (pdf_name_eq(ctx, PDF_NAME_LastPage, dest))
			pagenum = pdf_count_pages(ctx, doc) - 1;
		else if (pdf_name_eq(ctx, PDF_NAME_PrevPage, dest) && pagenum >= 0)
		{
			if (pagenum > 0)
				pagenum--;
		}
		else if (pdf_name_eq(ctx, PDF_NAME_NextPage, dest) && pagenum >= 0)
		{
			if (pagenum < pdf_count_pages(ctx, doc) - 1)
				pagenum++;
		}
		else
			return NULL;

		return fz_asprintf(ctx, "#%d", pagenum + 1);
	}

	return NULL;
}
Example #29
0
static void pdfapp_open_pdf(pdfapp_t *app, char *filename, int fd)
{
	fz_error error;
	fz_stream *file;
	char *password = "";
	fz_obj *obj;
	fz_obj *info;

	/*
	 * Open PDF and load xref table
	 */

	file = fz_open_fd(fd);
	error = pdf_open_xref_with_stream(&app->xref, file, NULL);
	if (error)
		pdfapp_error(app, fz_rethrow(error, "cannot open document '%s'", filename));
	fz_close(file);

	/*
	 * Handle encrypted PDF files
	 */

	if (pdf_needs_password(app->xref))
	{
		int okay = pdf_authenticate_password(app->xref, password);
		while (!okay)
		{
			password = winpassword(app, filename);
			if (!password)
				exit(1);
			okay = pdf_authenticate_password(app->xref, password);
			if (!okay)
				pdfapp_warn(app, "Invalid password.");
		}
	}

	/*
	 * Load meta information
	 */

	app->outline = pdf_load_outline(app->xref);

	app->doctitle = filename;
	if (strrchr(app->doctitle, '\\'))
		app->doctitle = strrchr(app->doctitle, '\\') + 1;
	if (strrchr(app->doctitle, '/'))
		app->doctitle = strrchr(app->doctitle, '/') + 1;
	info = fz_dict_gets(app->xref->trailer, "Info");
	if (info)
	{
		obj = fz_dict_gets(info, "Title");
		if (obj)
			app->doctitle = pdf_to_utf8(obj);
	}

	/*
	 * Start at first page
	 */

	error = pdf_load_page_tree(app->xref);
	if (error)
		pdfapp_error(app, fz_rethrow(error, "cannot load page tree"));

	app->pagecount = pdf_count_pages(app->xref);
}
Example #30
0
int Pdf::numPages() const
      {
      return pdf_count_pages(xref);
      }