- Jul 23, 2013
-
-
Sébastien Luttringer authored
-
Sébastien Luttringer authored
-
- Jun 28, 2013
-
-
Aurélien Dunand authored
Signed-off-by: Sébastien Luttringer <sebastien.luttringer@smartjog.com>
-
Sébastien Luttringer authored
This avoid conflict with os.path.isfile. Also add must_exists parameter to check local file existance
-
- Jun 27, 2013
-
-
Aurélien Dunand authored
As added to documentation [1], delete essential items from sys.modules may cause Python to fail. Import machinery is complicated [2] and not well documented before Python 3.3 [3] and the import system rework. In our case, at least two issues was caused by clearing sys.modules. - The first one impact import mechanism in our libs. If a lib import a function (or other) with the 'from module import function' syntax, this function will be loaded, but the module will not be properly loaded, this will result to TypeError. For example if in our lib we use: >>> from re import search >>> search('(?<=-)\w+', 'spam-egg') This will raise the following error: File "/usr/lib/python2.7/re.py", line 142, in search return _compile(pattern, flags).search(string) TypeError: 'NoneType' object is not callable - The second one is raised by importing a gzipped tarball in a script and store them in 'namespace', namespace.tar for example. If we try to extract a file from namespace.tar in subsequent scripts, we've got this error: File "/usr/lib/python2.7/tarfile.py", line 2112, in extractfile tarinfo = self.getmember(member) File "/usr/lib/python2.7/tarfile.py", line 1794, in getmember tarinfo = self._getmember(name) File "/usr/lib/python2.7/tarfile.py", line 2357, in _getmember members = self.getmembers() File "/usr/lib/python2.7/tarfile.py", line 1805, in getmembers self._load() # all members, we first have to File "/usr/lib/python2.7/tarfile.py", line 2380, in _load tarinfo = self.next() File "/usr/lib/python2.7/tarfile.py", line 2319, in next tarinfo = self.tarinfo.fromtarfile(self) File "/usr/lib/python2.7/tarfile.py", line 1239, in fromtarfile buf = tarfile.fileobj.read(BLOCKSIZE) File "/usr/lib/python2.7/gzip.py", line 240, in read raise IOError(errno.EBADF, "read() on write-only GzipFile object") IOError: [Errno 9] read() on write-only GzipFile object [1] http://hg.python.org/cpython/rev/4f8160e45cb7 [2] http://python-notes.boredomandlaziness.org/en/latest/python_concepts/import_traps.html [3] http://docs.python.org/dev/reference/import.html
-
Aurélien Dunand authored
Signed-off-by: Sébastien Luttringer <sebastien.luttringer@smartjog.com>
-
Aurélien Dunand authored
Image and version are required, so an error is raised if missing. Description, author and is_min_version are optional, so update their default value. Signed-off-by: Sébastien Luttringer <sebastien.luttringer@smartjog.com>
-
Sébastien Luttringer authored
This let place clean when error occurs in building a payload when a previous successful build have been done.
-
Sébastien Luttringer authored
-
Sébastien Luttringer authored
- 2 new packages (image and repository). - Split big modules into little one under packages. - Replace full module import by needed objects - Avoid wildcard imports - Fix a lot of pylint error and warnings on modules
-