How can I rebase without destroying changeset metadata?

dukeofgaming dukeofgaming at gmail.com
Mon Jan 14 11:21:45 UTC 2013


Hi,

I did not mean disrespect. I figured I would get high-lever and low-level
answers depending on the lists, I'm not shouting for attention.

So rebase does recreate the changesets with the metadata?

Thanks,

David



On Mon, Jan 14, 2013 at 4:47 AM, Mads Kiilerich <mads at kiilerich.com> wrote:

> On 01/14/2013 11:00 AM, dukeofgaming wrote:
>
>> From:     dukeofgaming <dukeofgaming at gmail.com>
>> To:     mercurial at selenic.com <mercurial at selenic.com>,
>> mercurial-devel at selenic.com Devel <mercurial-devel at selenic.com>
>>
>
> Please don't cross-post. That will give noise and a fragmented discussion
> where somebody else than you will waste their time. That is not fair.
>
>
>  Hi,
>>
>> I am importing some Perforce stuff through Perfarce. Everything well
>> until I met with the delightful end of line problem since Perforce stores
>> everything as LF.
>>
>> So I solved that with the EOL extension. Now I wanted to move the
>> changeset where I version the .hgeol file to the very beginning of the
>> imported repo, no problem:
>>
>> $ hg update -C null
>> $ hg add .hgeol
>> $ hg commit -m "Adding .hgeol file"
>> $ hg id -n
>> 123
>> $ hg rebase --source 0 --dest 123
>>
>> So that leaves my newest commit as the root commit, effectively rewriting
>> history to include that .hgeol file with all changes... problem is now,
>> that I cannot pull from Perforce anymore, why?: there used to be a
>> "Perforce" field in my imported changesets and is now gone.
>>
>> Now, I get why the changesets must be rewriten: to recalculate the hash
>> (surely among other things). But what I don't get is why does the Perfarce
>> metadata die with rebase?
>>
>
> Probably because it references changesets using the hashes and not the
> metadata. I guess it maintains a mapping between Perforce revisions and
> Mercurial changesets - you will have to update that.
>
> Perfarce has a small user base and the hit rate on the Mercurial lists
> might be low. I suggest you contact the Perfarce project or its author and
> ask how you can help getting the issue resolved.
>
> /Mads
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial/attachments/20130114/48788bf2/attachment-0002.html>


More information about the Mercurial mailing list