Signed-off-by: Mattia Rizzolo <mattia@debian.org>

debian: reinstate oggvideotools and procyon-decompiler test dependencies, now that are not buggy anymore

Signed-off-by: Mattia Rizzolo <mattia@debian.org>

Signed-off-by: Mattia Rizzolo <mattia@debian.org>

tests: with DIFFOSCOPE_TESTS_FAIL_ON_MISSING_TOOLS=1, actually fail only tests that are missing the required tools

Signed-off-by: Mattia Rizzolo <mattia@debian.org>

tests: allow specifying which tools are known missing using DIFFOSCOPE_TESTS_MISSING_TOOLS, to override a DIFFOSCOPE_TESTS_FAIL_ON_MISSING_TOOLS

Signed-off-by: Mattia Rizzolo <mattia@debian.org>

Signed-off-by: Mattia Rizzolo <mattia@debian.org>

Signed-off-by: Mattia Rizzolo <mattia@debian.org>

Fix execution of symbolic links that point to the "bin/diffoscope" entry point by fully resolving the location as part of dynamically calculating Python's include path.

Gbp-Dch: ignore

Add support for GnuPG "keybox" files. Thanks to Daniel Kahn Gillmor for the suggestion. (Closes: #871244, reproducible-builds/diffoscope#23)

Vibhu Agrawal authored Apr 27, 2019 and Chris Lamb committed Apr 28, 2019

We catches the ENOSPC OSError at the top level and exit diffoscope without
printing the complete traceback. Other errors are raised and handled normally.

Signed-off-by: Chris Lamb <lamby@debian.org>

Treat missing tools as failures if the DIFFOSCOPE_TESTS_FAIL_ON_MISSING_TOOLS environment variable is exported. (Closes: #905885, reproducible-builds/diffoscope#35)

Copying the comment from
reproducible-builds/diffoscope@659d1917:

    @Vibhu-guest also I'm reverting this one, mostly because the
    authoring information is screwed (root <root@localhost.localdomain> -
    does this mean you are doing your work as root??), and should have
    probably be gone through an MR instead of being pushed to master.

    I'm also at odds about the changes themselves.  You are catching
    that OSError in those two fairly specific positions, but it feels
    rather arbitrary.  Surely ENOSPC could happen anywhere?

    Finally, I think the user at the end should at least getting
    something that tells them what the problem is.  I think this way the
    program terminates without printing anything?

This reverts commit 659d1917.

This commit isn't what the message says, and it breaks `black`

This reverts commit 465176e0.

Added a try except block to handle OSError 28 and make the diffoscope
fail more gracefully.

Require that "-" is explicitly specified to read a single diff from standard input to avoid non-intuitive behaviour when diffoscope is called without any arguments. (Closes: reproducible-builds/diffoscope#54)

This reverts commit afca0488.

See reproducible-builds/diffoscope@afca0488 (comment 71853)

Gbp-Dch: ignore

diffoscope/comparators/zip.py: added another except block to handle
OSError.
Fixes https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=874582

ENOSPC errors.
diffoscope/comparators/zip.py: added an except condition to track ENOSPC
errors.

Both the above changes were made to ensure diffoscope fails more
gracefully in case ENOSPC is hit. These changes were tested with a
small tmpfs. The files used for testing were initrd.img and
initrd.img.old for first case and two .aia files for second case.

Fixes https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=874582

Make --use-dbgsym a ternary operator to make it easier to totally disable. (re. reproducible-builds/diffoscope#2)

Always warn if tlsh module is not available (not just if a specific fuzziness threshold is specified) to match the epilog of the --help output. This prevents missing support for file rename detection. (Closes: reproducible-builds/diffoscope#29, #888237)

Milena Boselli Rosa authored Mar 19, 2019 and Chris Lamb committed Mar 20, 2019

Signed-off-by: Chris Lamb <lamby@debian.org>

Milena Boselli Rosa authored Mar 19, 2019 and Chris Lamb committed Mar 20, 2019

Prevent empty values for the "name" attribute name on HTML anchor tags, and add an "id" to its parent "div" container.

Signed-off-by: Chris Lamb <lamby@debian.org>

Milena Boselli Rosa authored Mar 19, 2019 and Chris Lamb committed Mar 20, 2019

Signed-off-by: Chris Lamb <lamby@debian.org>

Milena Boselli Rosa authored Mar 13, 2019 and Chris Lamb committed Mar 20, 2019

Fix a "table column x established by element 'col' has no cells beginning in it" HTML validation error.

The first row after the header in a table can't use the "colspan" attribute.

Signed-off-by: Chris Lamb <lamby@debian.org>

Revert "Correction to HTML markup warnings/errors: fix empty value for attributte name on 'a' tags and add an id to the nearest div containers."

This reverts commit a6fd6b33; not sure how this
ended up on master (?) given it's on an MR..?

Gbp-Dch: ignore

Correction to HTML markup warnings/errors: fix empty value for attributte name on 'a' tags and add an id to the nearest div containers.


(cherry picked from commit c7c487cc7ac5bef7a6781853d4b714f1752e936c)

Provide explicit help when the libarchive system package is missing or "incomplete". (Closes: reproducible-builds/diffoscope#50)

Explicitly mention when the guestfs module is missing at runtime and we are falling back to a binary diff. (Closes: reproducible-builds/diffoscope#45)

Improve the displayed comment when falling back to a binary diff to include the file type. (Closes: #49)