Comments on: JGit-Matlab integration

By: Inter-Matlab data transfer with memcached | Undocumented Matlab

Inter-Matlab data transfer with memcached | Undocumented Matlab — Wed, 27 Aug 2014 12:27:07 +0000

[…] Once again I welcome guest blogger Mark Mikofski. Mark has written here last year about JGIT-Matlab integration and earlier this year on JSON-Matlab integration. Today, Mark shows how to use the well-known memcached library […]

By: JSON-Matlab integration | Undocumented Matlab

JSON-Matlab integration | Undocumented Matlab — Wed, 27 Aug 2014 07:19:42 +0000

[…] would like to once again welcome guest blogger Mark Mikofski. Mark has written here last year about JGIT-Matlab integration (p.s., in the recently-released R2014a, MathWorks added GIT support to Simulink projects, although […]

By: mark mikofski

mark mikofski — Thu, 10 Oct 2013 21:52:48 +0000

The latest update is version 0.6 (Falcon):

It allows (subset of) Git syntax to be used.

>> jgit add file1 file2

>> jgit commit -m 'enter msg in single quotes'

>> jgit pull

>> jgit push origin feature

Bugs fixes.
`javaclasspath.txt` is now in prefdir instead of userpath (thanks Volker).
Merge conflicts also output *.BASE, *.LOCAL, *.REMOTE and *.orig for use in merge tools such as meld and p4m.
Documentation is now at http://mikofski.github.io/JGit4MATLAB/.

Please submit bugs to https://github.com/mikofski/JGit4MATLAB.

Enjoy!
–Mark

By: mark mikofski

mark mikofski — Sat, 28 Sep 2013 09:34:23 +0000

Just wanted to post an update – I have made some major improvements to JGit4MATLAB. Now it can use SSH passphrases (or not), and can pull, push and fetch. I put utilities in to change or view global config and to save your passphrase. Still more to come! Also check out new website: http://mikofski.github.io/JGit4MATLAB/.

By: Mark Mikofski

Mark Mikofski — Thu, 11 Jul 2013 18:05:45 +0000

In reply to Malcolm Lidierth. @Malcolm, Thanks for your comment. In response to ... "...and I am guessing with JGit/JGit4MATLAB too (?)", yes, JGit/JGit4MATLAB works with with all Git repositories, both local and remote (i.e.: Github, Bitbucket), regardless of what Git client they were generated with or maintained by, and it inter-operates with the original (C/Perl/shell) Git and other Git-bindings as well (such as Python's Dulwich and Java's JGit). That's cool that NetBeans also has Git integration. I use both Eclipse and Spyder a lot, and they both have full SCM integration, e.g.: Git, SVN, Hg, etc. Maybe this will put some pressure on TMW to finally introduce SCM integration into their IDE, recognizing it's importance in best coding practices. In response to your comment about open-source collaboration and FEX, I made a similar comment on CSSM here. Frankly, IMHO Github's window's app is not the best Git client, I prefer to use TortoiseGit on windows, it has a very intuitive and rich feature set that will be very familiar to TortoiseSVN users. Another client I think is better than Github's is SourceTree by Atlassian, maintainers of BitBucket, another online Git/Hg repository host. BTW: I think the Waterloo graphics are awesome! Can't wait to try them out.

By: Malcolm Lidierth

Malcolm Lidierth — Thu, 11 Jul 2013 12:11:31 +0000

@Mark
Thanks for this article. I have been using Git for a while now to support the Waterloo code I blogged about over the past couple of weeks. The Java code is developed in NetBeans which has full Git support – it would be good to see that also in MATLAB.
Being far from expert in Git, I find the GitHub app GUIs useful (it is not specific to the GitHub site: I use SourceForge). It shows me the Git setup, supports SSH and it’s easy to exclude specific files/folders from the public repo. As all changes are made in the Git repo, they are honoured also in NetBeans/other IDEs/from the command line and I am guessing with JGit/JGit4MATLAB too (?).
MATLAB makes use of NetBeans RCP components such as the editor. It would be good to see the MATLAB file manager provide the sort of version control support that NetBeans provides. That could encourage fuller open-source collaboration amongst the MATLAB community too: a problem with the MATLAB FEX is that it does not presently support that.