| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
This should help to clean up the config files considerably.
Change-Id: I8c994f4b1654bd5ec648715b3ccdbdd4e02a6fca
Reviewed-by: Cristian Maureira-Fredes <cristian.maureira-fredes@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
|
| |
The tool opened the file for each branch-specific key and each key. Store
the values in a hash instead.
Change-Id: I6719e490a4ad78da70e9db0ce8016b0db2ea3ed5
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add option -i which performs a quick test on Linux or tries to
locate ibjom on Windows. Move the code checking for make
into a subroutine so that it can be called after the options
parsing.
Add elapsed time output.
Change-Id: Ie7b965b53d5e06521fb3eedee463c7178a96267c
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
initially, git-gpush was just a glorified git-push, and hence it started
out accepting the same command line. however, one of the main
motivations for the script's existence was _avoiding_ the need to
specify things we already know, and therefore the regular push syntax
actually isn't very useful here in the common case.
to that effect, we deprecated the bare remote syntax over two years ago,
in favor of the (very rarely used) --remote option. this allows us now
to interpret a bare word as something more useful: the source of the
push, avoiding the need to always use a colon.
following the same logic, we now introduce the --branch option for
specifying the target branch, expecting that it will not be used much.
these two together allow us to deprecate the old way of specifying both
source and target.
Change-Id: I0964252d911b6e9fe32458cf309c6b5ee4f646ae
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
| |
this code was in for two years now; it clearly outlived its usefulness.
Change-Id: I19310f0afc852c24544ff5f33180c1a9f4d76a9c
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
|
|
| |
in particular, don't do it for "WIP", as most people expect these
changes to be visible. so now only obvious "garbage subjects" are
marked.
Change-Id: I1704e56208a87f42941332773fce18b7c79abbbc
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
|
| |
this is cleaner (syntax-specific data has a shorter life-time) and
allows for adding more git-receive-pack arguments at a later point.
Change-Id: I1287b7c7a9b4b4bf932bf38eb6d1b7c690b94034
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
|
| |
this was once used to pass -n and -v to git, but we do that explicitly
now. passing on unknown options just serves to obfuscate usage errors.
Change-Id: I6d971e7c5ba8d3be7093f4171b4a7a03cf63ec90
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
| |
Change-Id: Icb5c60a9fb68a17b023f66a12f8da1d2fe27fc5b
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
|
| |
this sounds less clumsy. the terminology is used interchangeably by the
git documentation, and there is the @{upstream} ref-spec.
Change-Id: Id5f5e03b4d5011c38317ed9847aec0d92173d7a3
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
| |
This is faster and also works in the case of gerrit hickups.
Change-Id: Id990cc16731f4ef6543a1a55c416ebb84bbe0e5b
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
| |
Previously, the environment variable QMAKESPEC
was checked, which should no longer be set.
Change-Id: I1f9925fb2c468472dd1789289ada0727f1d3c441
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
Only shows author date instead of commit date for changes. And
better to make sure rerere.enabled is disabled when analyzing
the conflicts.
Change-Id: I4268b55efb39cc94572117e46686cd5eaa6233c6
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
| |
Create a WIP change when conflicts arise.
Change-Id: Ifd9f4c78df3491642241754b256ac623399158cf
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
| |
Previously, the tool would not create the tracking branch of qt5.git.
Change-Id: I13b29852d71e3f56a0a3735c0141bb1366ebf1ec
Reviewed-by: Jan Arve Sæther <jan-arve.saether@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Iaa8a4ff67e532b65d27c239f5623a00e4151a7cb
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
Stray semicolon at end of line, testing len rather than the thing
itself. Use elif rather than if nested within else; append to
existing lists rather than creating new ones by doing arithmetic.
Prefer .startswith() over comparing [0]; it can take a tuple of
candidates.
Change-Id: I8272e21cb237fe376d6ec097c6ae322adae00c65
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
| |
Swap the two branches of an if/else clause so that we can test a
positive condition that's simpler and easier to read than its negation.
Change-Id: I72aa0b6ba7eabfee4d9cb62cc9ff3daa4acafb76
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
| |
This makes it easier to document that you need one of them !
That wasn't obvious before ...
Change-Id: Id35ed832721781d26d02fd642162f9ec78821f8c
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
| |
It just duplicates the argparse.Namespace class that's used for the
default return from parse_args() anyway.
Change-Id: Id306acd5e0367105c9fdabd3806567158b06852f
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
| |
Change-Id: Ib237cd0dea97d6a750e490bf257de9f056533299
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The call to get_submodules() produced output distracting to those
needing help; move it later and simply set args.modules if unset,
instead of supplying the list as default to the parser.
Moved banner line after arg-parse, too, as it duplicates the help
message's description. Added an epilog to the help message saying how
to invoke this script.
Change-Id: Ib970f7a9ffa235c9523256d4ccfee448afd37ab2
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@qt.io>
|
|
|
|
|
|
|
|
|
| |
Without supermodule integration, the changes are potentially harmful
for dependent modules.
Change-Id: Ie092275822fe3bf0e060c31bcfb7d520ac568b2a
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
|
|
|
|
|
|
|
| |
This makes the growing skip-list manageable again.
Change-Id: If418ed7d22411fda50a9e2ef8446954434524a2e
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
Update old header.LGPL3 to header.LGPL
Change-Id: Ic9b7de3ff8ad0e6f1c8477ced3d07f7e90f3dae9
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I2717a81cd4a1f3d2ca54e642cf65c334392bf72f
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I4fda2bd1e7908af453b7c4e1b30b22093ff7d08a
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
Repetitive white space in configure argument lists will then no longer
cause errors.
Change-Id: I747c410106dddc9d4161464e06332f9c24649620
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
| |
Having to specify the modules in 'configureArguments' causes
overlong configuration files lines as Qt grows (-skip qtquick1 ...)
Introduce as eparate key 'skip' for this, in which only the module
names are listed (qtquick1 ...).
Change-Id: I00f72630e1fc69086c76d67985fea1601b9b3f3a
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com>
|
|
|
|
|
|
|
|
|
| |
Add newly introduced directory 'usr/bin' to the search path
and always perform search in order to account for all installation
options (bin/cmd).
Change-Id: I23cd835d3d574aacd9c641ddf7f7ec5aae927983
Reviewed-by: Jan Arve Sæther <jan-arve.saether@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I6689c204ae4bf44984aa520f456c2f24ece8e8f0
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
|
|
|
|
|
|
|
| |
saves subsequent git-config invocations, which are slow.
Change-Id: I18d4ab4c383d612c55e45cd26218a3293ad65261
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
use open3() to be windows-friendly and fully flexible (a bit too much for
the moment - there are several unused functions).
always die on unexpected errors and subprocess crashes.
this also disables io warnings, as they are redundant with the explicit
error handling (and ugly at that).
Change-Id: I894336ffee8a23b1cd0db57c7c1d2a712b211f5c
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
| |
the purpose is allowing easy site-wide configuration.
i'm not renaming the file, as some people have symlinked both the binary
and the file, and the change would (more or less silently) break their
setups.
Change-Id: I4b40042131bdea73e9fcefb29814113321a1c2ca
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
it is very unlikely that anyone uses custom symbolic refs, in particular
as a source of a push.
Change-Id: I674506f9be0ab99bafcc6134b3852de3b7f74445
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
|
|
|
|
|
|
|
|
|
| |
Remove broken option to rebuild configure.exe and unused
$optModuleBranchArgument.
Change-Id: I7a66e5ac72acde3a3f9f8d55f367ae90417d3b1e
Reviewed-by: Tobias Hunger <tobias.hunger@theqtcompany.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I14cffe56652ada9d1b0c38a22868ee702a3b9d07
Reviewed-by: Tobias Hunger <tobias.hunger@theqtcompany.com>
|
|
|
|
|
|
|
| |
Use new option syntax and exclude qtwebengine by default.
Change-Id: I3a3f082df5137c8fe787f44039a87f9c3fcc1ce7
Reviewed-by: Tobias Hunger <tobias.hunger@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
Replace hash %preferredBranches by a function, making it possible
to return values depending on the Qt branch.
Change-Id: I2271cb945dabeb544290f5152ab8ad6c628d8c33
Reviewed-by: Jan Arve Sæther <jan-arve.saether@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
they are fixed terminology, and especially in the case of 'Changes' it's
sometimes confusing without some marker.
Change-Id: I75ce8d65940ea2f373cb86488fc4ac2aac2f9fde
Reviewed-by: Sergio Ahumada <sahumada@texla.cl>
|
|
|
|
|
|
|
|
|
|
| |
Qt copyrights are now in The Qt Company, so we could update the source
code headers accordingly. In the same go we should also fix the links to
point to qt.io.
Change-Id: If3c8d221bcf8e5010048ed5f1c50cd35c04e7199
Reviewed-by: Jani Heikkinen <jani.heikkinen@theqtcompany.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I4614ad0e1e2d5d0743a4ff5ebe24eea406f92f9b
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
|
|
|
|
|
|
|
| |
the nick resolves properly to the new account.
Change-Id: I797dde054efd8ded9b1a249c90ae4386c206740d
Reviewed-by: Sergio Ahumada <sahumada@texla.cl>
|
|
|
|
|
|
|
| |
Option f is ambiguous (force, from)
Change-Id: Ie4d02f22afc143e8f7488ac4f45ed7742a480301
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Currently some of the scripts uses env and others don't. This patch
aims to unify these so all scripts behave the same and follow git
rules.
From git's mailing list, by default, e.g. /usr/bin/perl should be used.
That path can be customized when installing git.
Change-Id: If2cb48e6bd3d630e86a0a275077609a8e90fc6c1
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
overriding the remote should be an extraordinarily uncommon action with
gpush, so there is no need to make it so prominent in the UI.
in the next step (in a few months), we can re-use the bare syntax to
allow specifying the source commit(s) without appending a colon to them.
Change-Id: Iaa814a6be8dc7ce82a1530d0c4eb577a6885079c
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
| |
later on, we will do additional things after pushing, so we need to be
properly aware of the operation mode.
Change-Id: Ie7ab6c8ce4e1623d2d0abffa4c0ffb47fe3fc463
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
| |
that's where it really belongs. it's not useful for the automatically
determined fallback anyway.
Change-Id: I870efbfb07ff7921a42d8be0215fe92cd90a703f
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
| |
Change-Id: Ia6c920ea31befbaf6d3c6a90d635db2004fce6f0
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
| |
these structures are empty (not uninitialized) by default.
Change-Id: Iec8ab8bfc6f08f2e0fa3712c559cf69cdf02a93e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|