Closed
Bug 476075
Opened 17 years ago
Closed 17 years ago
Pushlog: including/excluding revision issues with a revision range
Categories
(Developer Services :: Mercurial: hg.mozilla.org, defect)
Tracking
(Not tracked)
RESOLVED
WORKSFORME
People
(Reporter: sgautherie, Unassigned)
Details
0) FromChange: including or excluding ?
|hg log| include the "from" revision.
pushlog excludes it.
Fwiw, bonsai includes it:
http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=MozillaTinderboxAll&branch=HEAD&branchtype=match&sortby=Date&hours=2&date=explicit&mindate=2009-01-29+07%3A33&maxdate=2009-01-29+13%3A55&cvsroot=%2Fcvsroot
I kind of prefer to include it, so I can check it.
But if you think it's better to exclude it, maybe I won't complain.
1) Example of too few results:
hg log -r f706be55329e:31d411bd26bc
{
changeset: 24003:f706be55329e
user: Neil Deakin <neil@mozilla.com>
date: Tue Jan 20 12:46:52 2009 -0500
summary: Bug 474336, updating max on progressmeter doesn't update display, r+sr=neil
changeset: 24004:31d411bd26bc
user: Neil Deakin <neil@mozilla.com>
date: Tue Jan 20 12:46:55 2009 -0500
summary: Bug 467360, support buttons with child panels, r+sr=neil
}
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=f706be55329e&tochange=31d411bd26bc
{
(Nothing)
}
31d411bd26bc should be listed.
2) Example of too many results:
hg log -r e416d62b69a5:f706be55329e
{
changeset: 24002:e416d62b69a5
user: Alexander Surkov <surkov.alexander@gmail.com>
date: Wed Jan 21 00:19:10 2009 +0800
summary: fix for bug 468034
changeset: 24003:f706be55329e
user: Neil Deakin <neil@mozilla.com>
date: Tue Jan 20 12:46:52 2009 -0500
summary: Bug 474336, updating max on progressmeter doesn't update display, r+sr=neil
}
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=e416d62b69a5&tochange=f706be55329e
{
neil@mozilla.com
Tue Jan 20 09:47:59 2009 -0800
31d411bd26bc Neil Deakin — Bug 467360, support buttons with child panels, r+sr=neil
f706be55329e Neil Deakin — Bug 474336, updating max on progressmeter doesn't update display, r+sr=neil
}
31d411bd26bc should not be listed.
Comment 1•17 years ago
|
||
"fromchange / tochange" queries explicitly exclude the "fromchange" parameter. The query is designed to be used to find regression ranges. Given a good build from an earlier changeset, and a bad build from a later changeset, the query is designed to answer the question "What new changes were introduced after the good build, up to and including the bad build?"
Status: NEW → RESOLVED
Closed: 17 years ago
Resolution: --- → WORKSFORME
Reporter | ||
Comment 2•17 years ago
|
||
I can understand that and it answers point 0.
Are you saying that points 1 and 2 work for you ??
Comment 3•17 years ago
|
||
The pushlog deals with pushes, not really individual changesets.
in 1) & 2), both were pushed at the same time:
http://hg.mozilla.org/mozilla-central/pushloghtml?changeset=31d411bd26bc
so we display them both. It doesn't seem relevant to ask questions about individual changesets within a push, as they both appeared in the repo at the same time, so there's no way a build could contain one but not the other.
Reporter | ||
Comment 4•17 years ago
|
||
(In reply to comment #3)
> The pushlog deals with pushes, not really individual changesets.
True, I see...
> in 1) & 2), both were pushed at the same time:
> http://hg.mozilla.org/mozilla-central/pushloghtml?changeset=31d411bd26bc
> so we display them both. It doesn't seem relevant to ask questions about
Then, point 2 would be as best as it can be.
Yet, could point 1 be "fixed / special case'd" to give the same output as point 2, instead of nothing.
> individual changesets within a push, as they both appeared in the repo at the
Yet, the parameters are '*change', not '*push'; (and I'm waiting for bug 452882).
I mean, I don't know how to do this search (to get an online result) otherwise:
http://hg.mozilla.org/mozilla-central/log doesn't seem to accept theses parameters.
> same time, so there's no way a build could contain one but not the other.
That's true for tinderboxes, but not for local builds when narrowing down a regression timeframe.
(And I'm mainly building SeaMonkey/1.9.2, which currently has no tinderboxes. :-<)
Assignee | ||
Updated•12 years ago
|
Product: mozilla.org → Release Engineering
Assignee | ||
Updated•11 years ago
|
Product: Release Engineering → Developer Services
You need to log in
before you can comment on or make changes to this bug.
Description
•