Initial import

diff --git a/README b/README
new file mode 100644 (file)
index 0000000..c114880
--- /dev/null
+++ b/README
@@ -0,0 +1,49 @@
+pdfstrip.py - strip objects from PDF files by specifying objects IDs
+
+Example of use:
+
+  a. Uncompress the pdf file with PDFtk, e.g.:
+
+      $ pdftk input.pdf output uncompressed.pdf uncompress
+
+  b. Find the objects IDs you want to strip; for example for images, the
+     program pdfimages from poppler-utils can be used like this:
+
+      1. List all the images:
+
+          $ pdfimages -list uncompressed.pdf > pdfimages.txt
+
+      2. Extract all the images:
+
+          $ mkdir images
+          $ pdfimages -p -all uncompressed.pdf images/image
+
+      3. Isolate unique images, for an easier analysis:
+
+          $ mkdir uniq-images
+          $ md5sum images/image-* | sort | uniq -w 32 | tr -s ' ' | cut -d ' ' -f 2 | while read file; do cp "$file" uniq-images/; done
+
+      4. Compare the file names with the content of pdfimages.txt from 1. and
+         find the objects IDs, the result is a list of objects IDs, like this:
+
+          53,52,51,50,49,48,66,65,64,63,62,68,103,102,101,111,110,109,108,107,106
+
+  c. Pass the list of objects to pdfstrip.py (here the list is shown sorted,
+     just for readability):
+
+      $ ./pdfstrip.py uncompressed.pdf stripped.pdf 48,49,50,51,52,53,62,63,64,65,66,68,101,102,103,106,107,108,109,110,111
+
+  d. Re-compress the file:
+
+      $ pdftk stripped.pdf output final.pdf compress
+
+
+Limitations
+
+Sometimes pdfimages misses images, or report them as inlined even when they
+are not, so you may need to look at the PDF source to spot the missing IDs.
+
+Inline images cannot be stripped by pdfstrip, but they are easy to spot in the
+PDf source, they are delimited by markers "BI" and "EI" and there is always an
+"ID" string between the two; removing the source code usually works but this
+is a brute force approach.

diff --git a/pdfstrip.py b/pdfstrip.py
new file mode 100755 (executable)
index 0000000..1b287f8
--- /dev/null
+++ b/pdfstrip.py
@@ -0,0 +1,109 @@
+#!/usr/bin/env python3
+#
+# pdfstrip.py - strip objects from PDF files by specifying objects IDs
+#
+# Copyright (C) 2016  Antonio Ospite <ao2@ao2.it>
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+import argparse
+import logging
+import sys
+
+from pdfrw import PdfReader, PdfWriter
+from pdfrw.objects.pdfindirect import PdfIndirect
+
+
+# pylint: disable=invalid-name
+logger = logging.getLogger(__name__)
+console_handler = logging.StreamHandler()
+formatter = logging.Formatter('[%(levelname)s] %(funcName)s:%(lineno)d %(message)s')
+console_handler.setFormatter(formatter)
+logger.addHandler(console_handler)
+logger.propagate = False
+
+
+def resolve_objects_names(pdf, objects_ids):
+    logger.debug("objects_ids %s", objects_ids)
+    objects_names = []
+    for objnum in objects_ids:
+        indirect_object = pdf.findindirect(objnum, 0)
+        if isinstance(indirect_object, PdfIndirect):
+            try:
+                real_object = indirect_object.real_value()
+                if real_object.Name:
+                    objects_names.append(real_object.Name)
+                else:
+                    logger.warning("Object %d has an empty 'Name' attribute", objnum)
+            except AttributeError:
+                logger.warning("Object %d has no 'Name' attribute", objnum)
+        else:
+            logger.warning("Object %d is not a PdfIndirect but a %s",
+                           objnum, type(indirect_object))
+
+    logger.debug("objects_names %s\n", objects_names)
+    return objects_names
+
+
+def strip_objects(pdf, objects_names):
+    for i, page in enumerate(pdf.pages):
+        logger.debug("Page %d", i + 1)
+        logger.debug("Before %s", page.Resources.XObject.keys())
+        for obj in objects_names:
+            if obj in page.Resources.XObject:
+                del page.Resources.XObject[obj]
+
+        logger.debug("After  %s\n", page.Resources.XObject.keys())
+
+    return pdf
+
+
+def validate_objects_ids(objects_ids_string):
+    try:
+        objects_ids = [int(obj) for obj in objects_ids_string.split(',')]
+    except (IndexError, ValueError):
+        raise argparse.ArgumentTypeError("%s contains an invalid value" % objects_ids_string)
+
+    return objects_ids
+
+
+def main():
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument("input_filename",
+                        help="the input PDF file (better if uncompressed)")
+    parser.add_argument("output_filename",
+                        help="the output PDF file")
+    parser.add_argument("objects_ids",
+                        type=validate_objects_ids,
+                        help="a comma-separated list of objects IDs")
+    parser.add_argument("-d", "--debug", action="store_true",
+                        help="enable debug output")
+    args = parser.parse_args()
+
+    if args.debug:
+        logger.setLevel(logging.DEBUG)
+
+    pdf_data = PdfReader(args.input_filename)
+
+    objects_names = resolve_objects_names(pdf_data, args.objects_ids)
+    pdf_data = strip_objects(pdf_data, objects_names)
+
+    PdfWriter().write(args.output_filename, pdf_data)
+
+    return 0
+
+
+if __name__ == '__main__':
+    sys.exit(main())