[Reproducible-commits] [debbindiff] 02/02: Add support for PDF files

[Jérémy Bobbio]

This is an automated email from the git hooks/post-receive script.

lunar pushed a commit to branch master
in repository debbindiff.

commit b3fbe5658789e01fa6ee034c65900c90b3e65356
Author: Jérémy Bobbio <lunar at debian.org>
Date:   Thu Jan 8 12:00:03 2015 +0100

    Add support for PDF files
---
 README                             | 36 ++++++++++++------------
 debbindiff/comparators/__init__.py |  2 ++
 debbindiff/comparators/pdf.py      | 56 ++++++++++++++++++++++++++++++++++++++
 debian/control                     |  2 ++
 4 files changed, 79 insertions(+), 17 deletions(-)

diff --git a/README b/README
index 1b4b115..ff6bd98 100644
--- a/README
+++ b/README
@@ -26,23 +26,25 @@ External dependencies
 The various comparators rely on these external commands being available
 in the path:
 
-| command  | Debian package     |
-+----------|--------------------|
-| ar       | binutils-multiarch |
-| bzip2    | bzip2              |
-| file     | file               |
-| ghc      | ghc                |
-| gpg      | gnupg              |
-| gzip     | gzip               |
-| msgunfmt | gettext            |
-| objdump  | binutils-multiarch |
-| readelf  | binutils-multiarch |
-| showttf  | fontforge-extras   |
-| sng      | sng                |
-| vim      | vim                |
-| xxd      | vim-common         |
-| xz       | xz-utils           |
-| zipinfo  | unzip              |
+| command   | Debian package     |
++-----------|--------------------|
+| ar        | binutils-multiarch |
+| bzip2     | bzip2              |
+| file      | file               |
+| ghc       | ghc                |
+| gpg       | gnupg              |
+| gzip      | gzip               |
+| msgunfmt  | gettext            |
+| objdump   | binutils-multiarch |
+| pdftk     | pdftk              |
+| pdftotext | poppler-utils      |
+| readelf   | binutils-multiarch |
+| showttf   | fontforge-extras   |
+| sng       | sng                |
+| vim       | vim                |
+| xxd       | vim-common         |
+| xz        | xz-utils           |
+| zipinfo   | unzip              |
 
 Authors
 -------
diff --git a/debbindiff/comparators/__init__.py b/debbindiff/comparators/__init__.py
index 4997a7d..bf22bbb 100644
--- a/debbindiff/comparators/__init__.py
+++ b/debbindiff/comparators/__init__.py
@@ -32,6 +32,7 @@ from debbindiff.comparators.fonts import compare_ttf_files
 from debbindiff.comparators.gettext import compare_mo_files
 from debbindiff.comparators.gzip import compare_gzip_files
 from debbindiff.comparators.haskell import compare_hi_files
+from debbindiff.comparators.pdf import compare_pdf_files
 from debbindiff.comparators.png import compare_png_files
 from debbindiff.comparators.text import compare_text_files
 from debbindiff.comparators.tar import compare_tar_files
@@ -78,6 +79,7 @@ COMPARATORS = [
      compare_elf_files),
     (r'^application/(x-font-ttf|vnd.ms-opentype)(;|$)', r'\.(ttf|otf)$', compare_ttf_files),
     (r'^image/png(;|$)', r'\.png$', compare_png_files),
+    (r'^application/pdf(;|$)', r'\.pdf$', compare_pdf_files),
     (None, r'\.a$', compare_static_lib_files),
     ]
 
diff --git a/debbindiff/comparators/pdf.py b/debbindiff/comparators/pdf.py
new file mode 100644
index 0000000..38a1d4c
--- /dev/null
+++ b/debbindiff/comparators/pdf.py
@@ -0,0 +1,56 @@
+# -*- coding: utf-8 -*-
+#
+# debbindiff: highlight differences between two builds of Debian packages
+#
+# Copyright © 2015 Jérémy Bobbio <lunar at debian.org>
+#
+# debdindiff 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.
+#
+# debbindiff 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 debbindiff.  If not, see <http://www.gnu.org/licenses/>.
+
+import subprocess
+from debbindiff.comparators.utils import binary_fallback
+from debbindiff.difference import Difference, get_source
+
+
+def uncompress(path):
+    output = subprocess.check_output(
+        ['pdftk', path, 'output', '-', 'uncompress'],
+        shell=False, close_fds=True)
+    return output.decode('latin-1').encode('ascii', 'backslashreplace')
+
+
+def pdftotext(path):
+    return subprocess.check_output(
+        ['pdftotext', path, '-'],
+        shell=False, close_fds=True)
+
+
+ at binary_fallback
+def compare_pdf_files(path1, path2, source=None):
+    differences = []
+    src = get_source(path1, path2) or 'FILE'
+    text1 = pdftotext(path1)
+    text2 = pdftotext(path2)
+    if text1 != text2:
+        differences.append(
+            Difference(text1.splitlines(1), text2.splitlines(1),
+                       text1, text2,
+                       source="pdftotext %s" % src))
+    uncompressed1 = uncompress(path1)
+    uncompressed2 = uncompress(path2)
+    if uncompressed1 != uncompressed2:
+        differences.append(
+            Difference(uncompressed1.splitlines(1), uncompressed2.splitlines(1),
+                       path1, path2,
+                       source="pdftk %s output - uncompress" % src))
+    return differences
diff --git a/debian/control b/debian/control
index deb0074..4eb0979 100644
--- a/debian/control
+++ b/debian/control
@@ -24,6 +24,8 @@ Depends: binutils-multiarch,
          gettext,
          ghc,
          gnupg,
+         pdftk,
+         poppler-utils,
          sng,
          unzip,
          vim,

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/reproducible/debbindiff.git