[issue742] have fetch merge with current branch instead of failing

Charles Samuels mercurial-bugs at selenic.com
Tue Sep 18 15:31:16 CDT 2007

New submission from Charles Samuels <charles at vmware.com>:

If you have an alternate head, you're probably going to still want to continue 
development on your main head which means that you're going to still want to do 
pushes and pulls on the main branch.

I would like it if `hg fetch` were to automatically merge on the currently 
selected head. If there's an orphaned head somewhere, it shouldn't change my 
entire "modus operandi" - especially considering that it may be a while before 
I can merge that other head, or I may never be able to merge that head.

Here's a demo:

$ hg fetch
abort: multiple heads in this repository (use "hg heads" and "hg merge" to 
# Please just fetch and merge into what I'm currently working on (the working 

$ hg pull
(run 'hg heads' to see heads, 'hg merge' to merge)

$ hg merge
abort: repo has 3 heads - please merge with an explicit rev
# oh come on, why can't you just merge what I downloaded with what I am 
currently "updated" to - it's obvious I'm going to be merging those two

This also has a nice advantage of how I can't accidentally forget to do 
the "pull" and as a result, end up merging the branch I didn't want to merge 
with my local version (because hg merge assumes that if you have two heads, you 
should merge them)

messages: 4009
nosy: charles
priority: wish
status: unread
title: have fetch merge with current branch instead of failing

Mercurial issue tracker <mercurial-bugs at selenic.com>

More information about the Mercurial-devel mailing list