directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Zheng, Kai" <kai.zh...@intel.com>
Subject RE: Git workflow
Date Thu, 19 Nov 2015 14:20:58 GMT
Hi Steve,

Thanks for raising this!

>> would it be easier for you if I just attached patches to the issues in Jira (until
it's deemed worth letting me commit directly)?  If you're manually performing the step in
the middle, I think Git is going to hurt us more than it helps us.

Yes patches will be more convenient for us, either committer or contributor. If you can generate
patch and upload to JIRA, then Jiajia and other committers can help look at it and commit,
as most of us have done in the past year. I thought this process will go on for some while
until we have made the GH for JIRA integration happen at the time this project can receive
many PR requests. Before that happen, JIRA based work flow would still be the best way for
us to collaborate. 

Regards,
Kai

-----Original Message-----
From: Steve Moyer [mailto:smoyer@psu.edu] 
Sent: Thursday, November 19, 2015 10:02 PM
To: Apache Directory Developers List <dev@directory.apache.org>
Subject: Git workflow

Kai and Jia,

As much as I like GitHub and pull requests, I'm being slowed down by what I'm guessing is
a workflow issue.  Here's the path the code is currently taking:

+------------------------+
|  Apache Kerby GitHub   |
+------------------------+
            |
            | Pull
            V
+------------------------+
|   Local Workstation    |
+------------------------+
            |
            | Push
            V
+------------------------+
| PennState Kerby GitHub |
+------------------------+
            |
            | Magic
            V
+------------------------+
|   Apache Kerby Git     |
+------------------------+
            |
            | Mirror
            V
+------------------------+
|  Apache Kerby GitHub   |
+------------------------+

The problem with this workflow is that almost every pull we make is causing merge conflicts.
 I suspect (from how the code I've sent is merged into the baseline) that you're effectively
applying a patch using the contents of the pull request during the step I've labelled "Magic".
 This seems like it's harder than it should be for you - and it's certainly means we're constantly
"fudging" our code back to the baseline.

So the big question - would it be easier for you if I just attached patches to the issues
in Jira (until it's deemed worth letting me commit directly)?  If you're manually performing
the step in the middle, I think Git is going to hurt us more than it helps us.

Thanks,

Steve

--

“The mark of the immature man is that he wants to die nobly for a cause, while the mark
of the mature man is that he wants to live humbly for one.” - Wilhelm Stekel
Mime
View raw message