Attachment 86849 Details for Bug 69965 – glib: Add poppler_documents_save_pages()

[patch] glib: Add poppler_documents_save_pages()

0001-glib-Add-poppler_documents_save_pages.patch (text/plain), 7.21 KB, created by Marek Kasik on 2013-09-30 13:53:22 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Marek Kasik

Created: 2013-09-30 13:53:22 UTC

Size: 7.21 KB

patch

obsolete

>From 0deb05560241d23ca08f420258b84b71b949e106 Mon Sep 17 00:00:00 2001
>From: Marek Kasik <mkasik@redhat.com>
>Date: Thu, 26 Sep 2013 16:20:18 +0200
>Subject: [PATCH] glib: Add poppler_documents_save_pages()
>
>Function poppler_documents_save_pages() saves pages from given documents
>to the specified uri. It saves only pages specified in "pages" parameter
>which is a list of lists of indices terminated by "-1". Number of the lists
>is the same as number of given documents.
>This function allows to merge / split PDF documents. It is based on the code
>of pdfunite utility.
>---
> glib/poppler-document.cc | 170 +++++++++++++++++++++++++++++++++++++++++++++++
> glib/poppler-document.h  |   6 ++
> 2 files changed, 176 insertions(+)
>
>diff --git a/glib/poppler-document.cc b/glib/poppler-document.cc
>index 61d92e8..f5f7304 100644
>--- a/glib/poppler-document.cc
>+++ b/glib/poppler-document.cc
>@@ -18,6 +18,7 @@
> 
> #include "config.h"
> #include <string.h>
>+#include <errno.h>
> 
> #ifndef __GI_SCANNER__
> #include <goo/GooList.h>
>@@ -1527,6 +1528,175 @@ poppler_document_init (PopplerDocument *document)
> {
> }
> 
>+/**
>+ * poppler_documents_save_pages:
>+ * @documents: array of #PopplerDocument
>+ * @pages: list of lists of number of pages to save, each terminated by -1
>+ * @n_documents: number of documents
>+ * @uri: uri of file to save
>+ * @error: (allow-none): return location for an error, or %NULL
>+ *
>+ * Saves specified pages from given documents.
>+ * If @error is set, %FALSE will be returned. Possible errors
>+ * include those in the #G_FILE_ERROR domain.
>+ * 
>+ * Return value: %TRUE, if the document was successfully saved
>+ **/
>+gboolean
>+poppler_documents_save_pages (PopplerDocument **documents,
>+			      int             **pages,
>+			      int               n_documents,
>+			      const char       *uri,
>+			      GError          **error)
>+{
>+  int objectsCount = 0;
>+  Guint numOffset = 0;
>+  std::vector<Object> new_pages;
>+  std::vector<Guint> offsets;
>+  XRef *yRef, *countRef;
>+  FILE *f;
>+  OutStream *outStr;
>+  int i, j, rootNum;
>+  std::vector<PDFDoc *>docs;
>+  int majorVersion = 0;
>+  int minorVersion = 0;
>+  Object page;
>+  Object value;
>+  PDFDoc *doc;
>+  Ref *refPage;
>+  Dict *pageDict;
>+  Goffset uxrefOffset;
>+  Ref ref;
>+  Dict *trailerDict;
>+  const char *key;
>+
>+  if (documents == NULL || pages == NULL) {
>+    g_set_error (error, POPPLER_ERROR,
>+                 POPPLER_ERROR_INVALID,
>+                 "No document neither pages given");
>+    return FALSE;
>+  }
>+
>+  for (i = 0; i < n_documents; i++) {
>+    g_return_val_if_fail (POPPLER_IS_DOCUMENT (documents[i]), FALSE);
>+  }
>+
>+  if (!globalParams) {
>+    globalParams = new GlobalParams ();
>+  }
>+
>+  for (i = 0; i < n_documents; i++) {
>+    doc = documents[i]->doc;
>+    if (doc->isOk () && !doc->isEncrypted ()) {
>+      docs.push_back (doc);
>+      if (doc->getPDFMajorVersion () > majorVersion) {
>+        majorVersion = doc->getPDFMajorVersion ();
>+        minorVersion = doc->getPDFMinorVersion ();
>+      } else if (doc->getPDFMajorVersion () == majorVersion) {
>+        if (doc->getPDFMinorVersion () > minorVersion) {
>+          minorVersion = doc->getPDFMinorVersion ();
>+        }
>+      }
>+    } else if (doc->isOk ()) {
>+      g_set_error (error, POPPLER_ERROR,
>+                   POPPLER_ERROR_ENCRYPTED,
>+                   "Document is encrypted");
>+      return FALSE;
>+    } else {
>+      g_set_error (error, POPPLER_ERROR,
>+                   POPPLER_ERROR_DAMAGED,
>+                   "PDF document is damaged");
>+      return FALSE;
>+    }
>+  }
>+
>+  if (!(f = fopen (uri, "wb"))) {
>+    g_set_error (error, G_FILE_ERROR,
>+                 g_file_error_from_errno (errno),
>+                 "%s", g_strerror (errno));
>+    return FALSE;
>+  }
>+
>+  outStr = new FileOutStream (f, 0);
>+  yRef = new XRef ();
>+  countRef = new XRef ();
>+
>+  yRef->add (0, 65535, 0, gFalse);
>+  PDFDoc::writeHeader (outStr, majorVersion, minorVersion);
>+
>+  for (i = 0; i < (int) docs.size (); i++) {
>+    if (pages[i] != NULL) {
>+      for (j = 0; pages[i][j] != -1; j++) {
>+        refPage = docs[i]->getCatalog ()->getPageRef (pages[i][j]);
>+        docs[i]->getXRef ()->fetch (refPage->num, refPage->gen, &page);
>+        new_pages.push_back (page);
>+        offsets.push_back (numOffset);
>+        pageDict = page.getDict ();
>+        docs[i]->markPageObjects (pageDict, yRef, countRef, numOffset);
>+      }
>+
>+      objectsCount += docs[i]->writePageObjects (outStr, yRef, numOffset, gTrue);
>+      numOffset = yRef->getNumObjects () + 1;
>+    }
>+  }
>+
>+  rootNum = yRef->getNumObjects () + 1;
>+  yRef->add (rootNum, 0, outStr->getPos (), gTrue);
>+  outStr->printf ("%d 0 obj\n", rootNum);
>+  outStr->printf ("<< /Type /Catalog /Pages %d 0 R", rootNum + 1);
>+  outStr->printf (">>\nendobj\n");
>+  objectsCount++;
>+
>+  yRef->add (rootNum + 1, 0, outStr->getPos (), gTrue);
>+  outStr->printf ("%d 0 obj\n", rootNum + 1);
>+  outStr->printf ("<< /Type /Pages /Kids [");
>+  for (j = 0; j < (int) new_pages.size (); j++)
>+    outStr->printf (" %d 0 R", rootNum + j + 2);
>+  outStr->printf (" ] /Count %zd >>\nendobj\n", new_pages.size ());
>+  objectsCount++;
>+
>+  for (i = 0; i < (int) new_pages.size (); i++) {
>+    yRef->add (rootNum + i + 2, 0, outStr->getPos (), gTrue);
>+    outStr->printf ("%d 0 obj\n", rootNum + i + 2);
>+    outStr->printf ("<< ");
>+    pageDict = new_pages[i].getDict ();
>+    for (j = 0; j < pageDict->getLength (); j++) {
>+      if (j > 0)
>+        outStr->printf (" ");
>+      key = pageDict->getKey (j);
>+      pageDict->getValNF (j, &value);
>+      if (strcmp (key, "Parent") == 0) {
>+        outStr->printf ("/Parent %d 0 R", rootNum + 1);
>+      } else {
>+        outStr->printf ("/%s ", key);
>+        PDFDoc::writeObject (&value, outStr, yRef, offsets[i], NULL, cryptRC4, 0, 0, 0);
>+      }
>+      value.free ();
>+    }
>+    outStr->printf (" >>\nendobj\n");
>+    objectsCount++;
>+    new_pages[i].free ();
>+  }
>+
>+  ref.num = rootNum;
>+  ref.gen = 0;
>+  uxrefOffset = outStr->getPos ();
>+  trailerDict = PDFDoc::createTrailerDict (objectsCount, gFalse, 0, &ref, yRef,
>+                                           uri, outStr->getPos ());
>+
>+  PDFDoc::writeXRefTableTrailer (trailerDict, yRef, gFalse /* do not write unnecessary entries */,
>+                                 uxrefOffset, outStr, yRef);
>+
>+  delete trailerDict;
>+  delete yRef;
>+  delete countRef;
>+
>+  outStr->close ();
>+  fclose (f);
>+
>+  return TRUE;
>+}
>+
> /* PopplerIndexIter: For determining the index of a tree */
> struct _PopplerIndexIter
> {
>diff --git a/glib/poppler-document.h b/glib/poppler-document.h
>index a34e88c..314c581 100644
>--- a/glib/poppler-document.h
>+++ b/glib/poppler-document.h
>@@ -227,6 +227,12 @@ PopplerDest       *poppler_document_find_dest              (PopplerDocument  *do
> PopplerFormField  *poppler_document_get_form_field         (PopplerDocument  *document,
> 							    gint              id);
> 
>+gboolean           poppler_documents_save_pages            (PopplerDocument **documents,
>+							    int             **pages,
>+							    int               n_documents,
>+							    const char       *uri,
>+							    GError          **error);
>+
> /* Interface for getting the Index of a poppler_document */
> #define POPPLER_TYPE_INDEX_ITER                 (poppler_index_iter_get_type ())
> GType             poppler_index_iter_get_type   (void) G_GNUC_CONST;
>-- 
>1.8.3.1
>

Actions: View | Splinter Review

Attachments on bug 69965: 86849 | 87030

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.