On Apr 22, 2016, at 11:34 PM, Ron W <ronw.m...@gmail.com> wrote:

> My team's wrapper for Fossil does the following (from memory, so might not be 
> exactly right):
> 
> On "branch new":
>    fossil commit --allow-empty --branch $bname -m "$comment (Issue [$issue])"
>    fossil tag add --propagate issue $bname $issue 
> 
> On "commit":
>    $issue=`fossil tag ls $revision|perl -n -e 'if /issue=(\w+)/ { print $1; }`
>    fossil commit -m "$comment (Issue [$issue])" $args


I have been perusing all the docs all evening, I can’t actually see any way to 
tell fossil which branch to commit to, so there must be some implicit behavior 
here that I am overlooking.  I can see that “checkout” is the command that 
checks out a workspace to a given VERSION, which exists on either the trunk or 
some branch.  

When you do the 

    commit —branch

does that change the workspace implicitly to be checked out to that branch 
instead of the trunk?  Does the workspace then become tied implicitly to the 
branch unless specifically checked out back to the trunk or another branch?

and thus a subsequent commit would be to the end of that branch rather then the 
trunk?  Am I right to conclude that at this point if the dev issues a checkout 
command again against a VERSION on the trunk, then subsequent commits would go 
to the end of the trunk rather then the branch?

Do I have that about right?

if so, not a big deal then, just have to use wrapper scripts to make sure to 
follow your work flow, always create a new branch on the first commit for any 
ticket.  I will probably use a cookie or something to make sure while working 
on a ticket, all commits go to that branch, and tag the commit with the 
[nnnnnnnnnn] number as well.

 If they need to work on more than one ticket concurrently, then no big deal, 
create another branch and switch to that one for a while.

if they need to do something directly to the trunk, then checkout trunk and 
just do it without code review, which is sometimes the case for some sys admin 
stuff.

I’m starting to get my head around fossil I think..












_______________________________________________
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Reply via email to