| Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
Handle both file sizes larger than SIZE_MAX and allocation
failures when reading the data of a tarmember.
If using the go() function with a callback, pass None as the data
parameter for files that are too large. For extractdata, raise
a MemoryError if the file too extract is too large.
Also check for an existing error first in extractdata, before
raising a new one.
Test the whole thing on 32-bit platforms with a sample deb that
contains a zeroed file that is 5GB large (compressed using xz
at level 1 to a few kb).
|
|
In a multi-line field, a comma might be followed by a newline instead
of a space.
Closes: #751770
|
|
This makes our support very flexible, and open for future additions
of new compression algorithms.
debfile_extract_tar() is almost identical to ExtractTarMember() of
apt's debDebFile.
|
|
Patch by Michael, tests derived from the ones provided by the
reporter.
I included the test debs without the source package, as for our
use case, all needed "source code" is included in the package
itself anyway.
Reported-By: Francois Gouget <fgouget@free.fr>
Closes: #717859
Closes: #717861
|
|
|
|
- use APT::Configuration::getCompressionTypes() instead of duplicating
the supported methods here
* tests/test_debfile.py:
- add test for raise on unknown data.tar.xxx
|
|
- add support for .xz archives
* tests/test_debfile.py:
- add test for xz compression
|
|
in the control file (LP: #624290) and add test
|
|
|
|
|
|
|
|
|
|
|
|
- fix crash in DscFile handling and add regression test
|
|
- fix crash in DscFile handling and add regression test
|
|
|
|
|