[webkit-dev] I *HATE* CHANGELOGS!!!

Mark Rowe mrowe at apple.com
Wed Aug 26 17:38:02 PDT 2009

On 2009-08-26, at 17:29, Brady Eidson wrote:

> On Aug 26, 2009, at 5:26 PM, Mark Rowe wrote:
>> On 2009-08-26, at 17:19, Brady Eidson wrote:
>>> On Aug 26, 2009, at 4:20 PM, Mark Rowe wrote:
>>>> On 2009-08-26, at 15:43, Maciej Stachowiak wrote:
>>>>>> On top of that sometimes the ChangeLog merges cleanly but puts  
>>>>>> your entry underneath others, and then you have to open the  
>>>>>> file and move your entry back to the top.  Sometimes I have not  
>>>>>> noticed this and then I land with someone else's commit message.
>>>>>> I just don't get why people are willing to put up with this.   
>>>>>> It's really driving me crazy.
>>>>> One possibility is to have a commit hook that builds the  
>>>>> ChangeLog entry and includes it in the commit atomically - that  
>>>>> way there are no races.
>>>> Hooks in Subversion cannot mutate the transaction that is taking  
>>>> place so I don't think this is an option.
>>> Is there any sort of guarantee that a post commit hook for  
>>> revision N will complete before the post commit hook for revision  
>>> N + 1 executes?
>>> Couldn't that accomplish this?
>> Generating two SVN revisions for each change would be awful.
> And - I honestly don't know the answer to this - there's actually  
> *NO* way to alter a previous commit in SVN, even when you have full  
> access to the server?

It may be doable through some amount of hackery, but it wouldn't be a  
good idea.  For example, let's assume that I made a commit that the  
server recorded as r35000. If r35000 was somehow rewritten to include  
modifications to ChangeLogs then suddenly the view of r35000 in my  
working copy differs from the view of anyone else looking at the  
repository.  For this to be a workaround solution Subversion would  
need the ability to allow a pre-commit hook to modify the transaction  
that is currently underway, and to be able to push any modifications  
made back to the committer.

> (I ask this with the risk of you bringing up GIT, and me smacking  
> you upside the head for the non-helpful suggestion in getting SVN to  
> work better for us)  ;)

There's no need to be such a git about it ;-)

I have no interest in pushing Git on other people working on WebKit.   
It works well for my uses, but I can easily see why the vastly  
different model, complexity of interface, etc. make some people balk  
at the idea of using it.

- Mark

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3166 bytes
Desc: not available
URL: <http://lists.webkit.org/pipermail/webkit-dev/attachments/20090826/115e3ada/attachment-0001.bin>

More information about the webkit-dev mailing list