Merge pull request #1445 from radarhere/pdf

Added PDF multipage saving

PIL/PdfImagePlugin.py

@@ -51,17 +51,21 @@ def _endobj(fp):
     fp.write("endobj\n")
 
 
+def _save_all(im, fp, filename):
+    _save(im, fp, filename, save_all=True)
+
+
 ##
 # (Internal) Image save plugin for the PDF format.
 
-def _save(im, fp, filename):
+def _save(im, fp, filename, save_all=False):
     resolution = im.encoderinfo.get("resolution", 72.0)
 
     #
     # make sure image data is available
     im.load()
 
-    xref = [0]*(5+1)  # placeholders
+    xref = [0]
 
     class TextWriter(object):
         def __init__(self, fp):
@@ -78,11 +82,6 @@ def _save(im, fp, filename):
     fp.write("%PDF-1.2\n")
     fp.write("% created by PIL PDF driver " + __version__ + "\n")
 
-    #
-    # Get image characteristics
-
-    width, height = im.size
-
     # FIXME: Should replace ASCIIHexDecode with RunLengthDecode (packbits)
     # or LZWDecode (tiff/lzw compression).  Note that PDF 1.2 also supports
     # Flatedecode (zip compression).
@@ -125,7 +124,7 @@ def _save(im, fp, filename):
     #
     # catalogue
 
-    xref[1] = fp.tell()
+    xref.append(fp.tell())
     _obj(
         fp, 1,
         Type="/Catalog",
@@ -134,89 +133,108 @@ def _save(im, fp, filename):
 
     #
     # pages
+    numberOfPages = 1
+    if save_all:
+        try:
+            numberOfPages = im.n_frames
+        except AttributeError:
+            # Image format does not have n_frames. It is a single frame image
+            pass
+    pages = [str(pageNumber*3+4)+" 0 R"
+             for pageNumber in range(0, numberOfPages)]
 
-    xref[2] = fp.tell()
+    xref.append(fp.tell())
     _obj(
         fp, 2,
         Type="/Pages",
-        Count=1,
+        Count=len(pages),
-        Kids="[4 0 R]")
+        Kids="["+"\n".join(pages)+"]")
     _endobj(fp)
 
-    #
+    for pageNumber in range(0, numberOfPages):
-    # image
+        im.seek(pageNumber)
 
-    op = io.BytesIO()
+        #
+        # image
 
-    if filter == "/ASCIIHexDecode":
+        op = io.BytesIO()
-        if bits == 1:
-            # FIXME: the hex encoder doesn't support packed 1-bit
-            # images; do things the hard way...
-            data = im.tobytes("raw", "1")
-            im = Image.new("L", (len(data), 1), None)
-            im.putdata(data)
-        ImageFile._save(im, op, [("hex", (0, 0)+im.size, 0, im.mode)])
-    elif filter == "/DCTDecode":
-        Image.SAVE["JPEG"](im, op, filename)
-    elif filter == "/FlateDecode":
-        ImageFile._save(im, op, [("zip", (0, 0)+im.size, 0, im.mode)])
-    elif filter == "/RunLengthDecode":
-        ImageFile._save(im, op, [("packbits", (0, 0)+im.size, 0, im.mode)])
-    else:
-        raise ValueError("unsupported PDF filter (%s)" % filter)
 
-    xref[3] = fp.tell()
+        if filter == "/ASCIIHexDecode":
-    _obj(
+            if bits == 1:
-        fp, 3,
+                # FIXME: the hex encoder doesn't support packed 1-bit
-        Type="/XObject",
+                # images; do things the hard way...
-        Subtype="/Image",
+                data = im.tobytes("raw", "1")
-        Width=width,  # * 72.0 / resolution,
+                im = Image.new("L", (len(data), 1), None)
-        Height=height,  # * 72.0 / resolution,
+                im.putdata(data)
-        Length=len(op.getvalue()),
+            ImageFile._save(im, op, [("hex", (0, 0)+im.size, 0, im.mode)])
-        Filter=filter,
+        elif filter == "/DCTDecode":
-        BitsPerComponent=bits,
+            Image.SAVE["JPEG"](im, op, filename)
-        DecodeParams=params,
+        elif filter == "/FlateDecode":
-        ColorSpace=colorspace)
+            ImageFile._save(im, op, [("zip", (0, 0)+im.size, 0, im.mode)])
+        elif filter == "/RunLengthDecode":
+            ImageFile._save(im, op, [("packbits", (0, 0)+im.size, 0, im.mode)])
+        else:
+            raise ValueError("unsupported PDF filter (%s)" % filter)
 
-    fp.write("stream\n")
+        #
-    fp.fp.write(op.getvalue())
+        # Get image characteristics
-    fp.write("\nendstream\n")
 
-    _endobj(fp)
+        width, height = im.size
 
-    #
+        xref.append(fp.tell())
-    # page
+        _obj(
+            fp, pageNumber*3+3,
+            Type="/XObject",
+            Subtype="/Image",
+            Width=width,  # * 72.0 / resolution,
+            Height=height,  # * 72.0 / resolution,
+            Length=len(op.getvalue()),
+            Filter=filter,
+            BitsPerComponent=bits,
+            DecodeParams=params,
+            ColorSpace=colorspace)
 
-    xref[4] = fp.tell()
+        fp.write("stream\n")
-    _obj(fp, 4)
+        fp.fp.write(op.getvalue())
-    fp.write(
+        fp.write("\nendstream\n")
-        "<<\n/Type /Page\n/Parent 2 0 R\n"
-        "/Resources <<\n/ProcSet [ /PDF %s ]\n"
-        "/XObject << /image 3 0 R >>\n>>\n"
-        "/MediaBox [ 0 0 %d %d ]\n/Contents 5 0 R\n>>\n" % (
-            procset,
-            int(width * 72.0 / resolution),
-            int(height * 72.0 / resolution)))
-    _endobj(fp)
 
-    #
+        _endobj(fp)
-    # page contents
 
-    op = TextWriter(io.BytesIO())
+        #
+        # page
 
-    op.write(
+        xref.append(fp.tell())
-        "q %d 0 0 %d 0 0 cm /image Do Q\n" % (
+        _obj(fp, pageNumber*3+4)
-            int(width * 72.0 / resolution),
+        fp.write(
-            int(height * 72.0 / resolution)))
+            "<<\n/Type /Page\n/Parent 2 0 R\n"
+            "/Resources <<\n/ProcSet [ /PDF %s ]\n"
+            "/XObject << /image %d 0 R >>\n>>\n"
+            "/MediaBox [ 0 0 %d %d ]\n/Contents %d 0 R\n>>\n" % (
+                procset,
+                pageNumber*3+3,
+                int(width * 72.0 / resolution),
+                int(height * 72.0 / resolution),
+                pageNumber*3+5))
+        _endobj(fp)
 
-    xref[5] = fp.tell()
+        #
-    _obj(fp, 5, Length=len(op.fp.getvalue()))
+        # page contents
 
-    fp.write("stream\n")
+        op = TextWriter(io.BytesIO())
-    fp.fp.write(op.fp.getvalue())
-    fp.write("\nendstream\n")
 
-    _endobj(fp)
+        op.write(
+            "q %d 0 0 %d 0 0 cm /image Do Q\n" % (
+                int(width * 72.0 / resolution),
+                int(height * 72.0 / resolution)))
+
+        xref.append(fp.tell())
+        _obj(fp, pageNumber*3+5, Length=len(op.fp.getvalue()))
+
+        fp.write("stream\n")
+        fp.fp.write(op.fp.getvalue())
+        fp.write("\nendstream\n")
+
+        _endobj(fp)
 
     #
     # trailer
@@ -232,6 +250,7 @@ def _save(im, fp, filename):
 # --------------------------------------------------------------------
 
 Image.register_save("PDF", _save)
+Image.register_save_all("PDF", _save_all)
 
 Image.register_extension("PDF", ".pdf")
 

Tests/test_file_pdf.py

@@ -1,17 +1,20 @@
 from helper import unittest, PillowTestCase, hopper
-
+from PIL import Image
 import os.path
 
 
 class TestFilePdf(PillowTestCase):
 
-    def helper_save_as_pdf(self, mode):
+    def helper_save_as_pdf(self, mode, save_all=False):
         # Arrange
         im = hopper(mode)
         outfile = self.tempfile("temp_" + mode + ".pdf")
 
         # Act
-        im.save(outfile)
+        if save_all:
+            im.save(outfile, save_all=True)
+        else:
+            im.save(outfile)
 
         # Assert
         self.assertTrue(os.path.isfile(outfile))
@@ -58,6 +61,19 @@ class TestFilePdf(PillowTestCase):
 
         self.assertRaises(ValueError, lambda: im.save(outfile))
 
+    def test_save_all(self):
+        # Single frame image
+        self.helper_save_as_pdf("RGB", save_all=True)
+
+        # Multiframe image
+        im = Image.open("Tests/images/dispose_bgnd.gif")
+
+        outfile = self.tempfile('temp.pdf')
+        im.save(outfile, save_all=True)
+
+        self.assertTrue(os.path.isfile(outfile))
+        self.assertGreater(os.path.getsize(outfile), 0)
+
 
 if __name__ == '__main__':
     unittest.main()

docs/handbook/image-file-formats.rst

@@ -786,6 +786,11 @@ PIL can write PDF (Acrobat) images. Such images are written as binary PDF 1.1
 files, using either JPEG or HEX encoding depending on the image mode (and
 whether JPEG support is available or not).
 
+When calling :py:meth:`~PIL.Image.Image.save`, if a multiframe image is used,
+by default, only the first image will be saved. To save all frames, each frame
+to a separate page of the PDF, the ``save_all`` parameter must be present and
+set to ``True``.
+
 PIXAR (read only)
 ^^^^^^^^^^^^^^^^^