Clarification: is the hgmerge program always called?

Vadim Gelfer vadim.gelfer at gmail.com
Tue May 16 18:19:49 CDT 2006


On 5/16/06, Matt Mackall <mpm at selenic.com> wrote:

> Interesting. This is about what I was thinking would happen, but it's
> definitely nice to see it quantified.

yes. i have not seen numbers to compare 3-way merge with history
sensitive merge before.

> Is your repo a public project?

no. sorry.

> If not, can we run this against one? Eg, the Linux kernel history?

yes. runs fine there. trouble with using linux kernel history is most
merges are clean because that is how people work.

> It'd also be good to characterize/classify what those improved cases
> look like.

am running now. conflict rate is low, like i expect. first 5000 revs
have only 5 conflicts. just 1 conflict is different in histmerge and
3-way. i have attached files with conflicts so you can see example.
file with ".hist." in name is histmerge output. file with ".local." in
name is 3-way merge output.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: arch_ia64_kernel_fsys-2436.hist.S
Type: application/octet-stream
Size: 27618 bytes
Desc: not available
Url : http://www.selenic.com/pipermail/mercurial/attachments/20060516/aec7a3de/arch_ia64_kernel_fsys-2436.hist-0001.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: arch_ia64_kernel_fsys-2436.local.S
Type: application/octet-stream
Size: 27605 bytes
Desc: not available
Url : http://www.selenic.com/pipermail/mercurial/attachments/20060516/aec7a3de/arch_ia64_kernel_fsys-2436.local-0001.obj


More information about the Mercurial mailing list