[Mifos-developer] Why are XML datasets compressed in GIT repo?
Artur Siekielski
There's a big disadvantage that datasets for acceptance tests in
acceptanceTests/src/test/resources/dataSets are ZIP files - GIT doesn't
track changes in binary files so you don't see diffs, don't have merging
functionality and all versions are stored inefficiently as separate
copies. Is there some reason for zipping these files, other that
reducing disk space? I think that currently all versions take more space
in Git database than unpacked XMLs (120MB).
Can we unpack these files? Or maybe we should create a separate
repository for test data to keep Mifos source code repo small?
Regards,
Artur
------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
http://p.sf.net/sfu/nokia-dev2dev
Udai Gupta
You might already know that Mifos has switched to GIT just few months
ago, before that Mifos code was in SVN.
SVN is worst at doing compression.
"The Mozilla project's CVS repository is about 3 GB; it's about 12 GB
in Subversion's fsfs format. In Git it's around 300 MB. "
https://git.wiki.kernel.org/index.php/GitSvnComparison
At that time datasets were making it hard to do merging, checkout,
branching etc. So we decided to compress dataset on our own which made
things little better.
Now, I don't remember any other reasons for zipping datasets. (let's
see if someone does)
Cheers,
Udai
Adam Feuer
> Now, I don't remember any other reasons for zipping datasets. (let's
> see if someone does)
Artur,
I think it was because svn was so slow. Now that we use git, we should
try going back to uncompressed.
Will you change the datasets to be uncompressed? This may require
changing some code, since I think the acceptance test fixture loads
the zip file directly... it may be able to use uncompressed data sets
too, I just don't remember.
-adam
--
Adam Feuer <adamf at pobox dot com>
Adam Monsen
Unpack 'em!