On Mon, Jan 29, 2007 at 02:31:13PM -0500, Bill Sommerfeld wrote: > On Mon, 2007-01-29 at 11:22 -0800, Stephen Lau wrote: > > > 1) There will be no restriction on number of changesets fixed per > > changegroup. You can address as many changesets in a putback as you like. > > > > 2) While there will be no restriction on number of bugs fixed per > > changeset, it would behoove you (and whoever looks at the SCM history > > after you) if bugs were separated out as much as possible. i.e.: if you > > are fixing bugs which are completely independent and easily-separable, > > please do so in separate changesets. If fixes for bugs are intertwined, > > (i.e.: you are putting back a project which, in turn, fixes 9 other bugs > > or something) then they may be combined into one changeset. > > Will the tools permit manipulating a workspace such that specific deltas > can be moved between changesets?
Can you clarify what you mean by "delta" in a mercurial world? But I think the answer to your question is "no". > If not, I see myself defaulting to "N bugs, one workspace, one changeset, > one changegroup" since I often find that dependencies between different > fixes change over time as the fixes under development evolve. That's fine. If you're sufficiently anal, mq can help you keep individual bugfixes separated, but I don't think it's worth enshrining in policy or any of the tools -- it's complicated, requires a rigorous mental model of what's going on, and doesn't work much of the time, anyway. Danek _______________________________________________ tools-discuss mailing list [email protected]
