forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ross Gardler <rgard...@apache.org>
Subject Re: svn commit: r367799 - /forrest/branches/dispatcher/
Date Wed, 11 Jan 2006 10:51:21 GMT
Thorsten Scherler wrote:
> El mié, 11-01-2006 a las 08:36 +0000, Ross Gardler escribió:
> 
>>Thorsten Scherler wrote:
>>

...

>>>I thought I should use a branch for it. That is what you suggested the
>>>other time (or better said I understood). 
>>>
>>>Is there a problem with the chosen method?
>>
>>I did suggest a *branch*, but this is not a branch.
>>
>>I foresee a few problems:
>>
>>1) Non standard use of SVN therefore difficult to trace what is actually 
>>done (especially historically)
>>
> 
> 
> Hmm, all is in the svn history (see commit mails) but yeah in different
> branches.

Yes, but it would be easy to make "small" and "insignificant" change in 
trunk rather than copy over the relevant file and edit it. Then history 
is lost. Furthermore, keeping your "branch" in synch will be difficult.

My point is that being in the "branches" part of svn people will assume 
they can check it out and work with it, or look at the history and see 
everything together. That is what a branch usually is.

>>2) "ln -s" does not work on windows platforms, therefore a number of 
>>devs are prevented from participating (even as testers)
> 
> 
> Yeah, that is indeed a problem. There is cygwin but we cannot force
> people to use it or linux. Point taken.

For future reference, even when using Cygwin "ln -s" does not work for 
Java on Windows. There is no equivalent of a symbolic link in windows. 
Cygwin simulates them (using shortcuts), but Java does not.

>>How much of a problem these will become I am not sure, but at the very 
>>least (2) is working against the community.

One more problem, it will be too difficult to create a test in our zone 
for this "branch" since the scrpts are set up to work with svn in the 
standard way, not to create symbolic links between different parts of 
the tree.

> yeah agree, so what do you suggest? Like said I do not see the point in
> a 100% copy if I change <5%.

That is not how branching works. When you create a branch it only 
creates copies of the parts that are changed. Therefore, doing an "svn 
switch" is really quick and easy.

Since you have found your requred changes will affect trunk, I'd suggest 
using a proper branch.

Ross


Mime
View raw message