| Message ID | a8bc0958-6c46-4622-96ef-b1e55c15f30f@cherry.de (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | [b4] prep-pre-flight-checks doesn't seem to work | expand |
On Fri, Jan 17, 2025 at 12:35:16PM +0100, Quentin Schulz wrote: > Hi all, > > I believe prep-pre-flight-checks doesn't work. > > I've added > > [b4] > prep-pre-flight-checks = disable-all > > in my local git repo .b4-config, but b4 send still complains. Right, we only pay attention to a small subset of configuration options coming from .b4-config and prep-pre-flight-checks isn't one of them, because it's not something that should be configured project-wide. What problems are you seeing that makes you want to disable pre-flight checks for everyone? -K
Hi Konstantin, On 1/17/25 4:20 PM, Konstantin Ryabitsev wrote: > On Fri, Jan 17, 2025 at 12:35:16PM +0100, Quentin Schulz wrote: >> Hi all, >> >> I believe prep-pre-flight-checks doesn't work. >> >> I've added >> >> [b4] >> prep-pre-flight-checks = disable-all >> >> in my local git repo .b4-config, but b4 send still complains. > > Right, we only pay attention to a small subset of configuration options > coming from .b4-config and prep-pre-flight-checks isn't one of them, because > it's not something that should be configured project-wide. > > What problems are you seeing that makes you want to disable pre-flight checks > for everyone? > We do not have checks nor auto-to-cc commands, so there's no point in telling the user to run b4 prep --auto-to-cc and b4 prep --check before sending the patches as they'll do nothing. Cheers, Quentin
On Fri, Jan 17, 2025 at 04:49:36PM +0100, Quentin Schulz wrote: > > What problems are you seeing that makes you want to disable pre-flight checks > > for everyone? > > > > We do not have checks nor auto-to-cc commands, so there's no point in > telling the user to run b4 prep --auto-to-cc and b4 prep --check before > sending the patches as they'll do nothing. Okay, I'm more inclined to treat this as a bug, because we shouldn't be invoking obviously irrelevant pre-flight checks. I don't advise turning it off entirely, because some pre-flight checks (like the dependency checker) are still valid in your case. I'll work to better recognize when certain pre-flight checks are irrelevant and disable them automatically. bugspray tag me -K
Hello: This conversation is now tracked by Kernel.org Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=219705 There is no need to do anything else, just keep talking.
Hi Konstantin, On 1/17/25 5:17 PM, Konstantin Ryabitsev wrote: > On Fri, Jan 17, 2025 at 04:49:36PM +0100, Quentin Schulz wrote: >>> What problems are you seeing that makes you want to disable pre-flight checks >>> for everyone? >>> >> >> We do not have checks nor auto-to-cc commands, so there's no point in >> telling the user to run b4 prep --auto-to-cc and b4 prep --check before >> sending the patches as they'll do nothing. > > Okay, I'm more inclined to treat this as a bug, because we shouldn't be > invoking obviously irrelevant pre-flight checks. I don't advise turning it off > entirely, because some pre-flight checks (like the dependency checker) are > still valid in your case. > Indeed, I actually only disable needs-checking or needs-auto-to-cc wherever appropriate. It's just that disable-all was shorter to write and check if things were happening the way I expected them to :) > I'll work to better recognize when certain pre-flight checks are irrelevant > and disable them automatically. > Thanks! Cheers, Quentin
diff --git a/src/b4/__init__.py b/src/b4/__init__.py index 814413e..b514556 100644 --- a/src/b4/__init__.py +++ b/src/b4/__init__.py @@ -2883,7 +2883,7 @@ def _setup_main_config(cmdargs: Optional[argparse.Namespace] = None) -> None: # some options can be provided via the toplevel .b4-config file, # so load them up and use as defaults topdir = git_get_toplevel() - wtglobs = ['prep-*-check-cmd', 'send-*', '*mask', '*template*', 'trailer*', 'pw-*'] + wtglobs = ['prep-*-check-cmd', 'send-*', '*mask', '*template*', 'trailer*', 'pw-*', 'prep-pre-flight-checks']
Hi all, I believe prep-pre-flight-checks doesn't work. I've added [b4] prep-pre-flight-checks = disable-all in my local git repo .b4-config, but b4 send still complains. Adding --debug to b4 shows the following (stripped down to what I believe is important): Loading worktree configs from /home/qschulz/work/upstream/yocto/yocto-docs/.b4-config Running git --no-pager config --file /home/qschulz/work/upstream/yocto/yocto-docs/.b4-config -z --get-regexp b4\..* wtcfg={'send-series-cc': 'yocto-docs@lists.yoctoproject.org', 'prep-pre-flight-checks': 'disable-all'} wtcfg: send-series-cc=yocto-docs@lists.yoctoproject.org Now if I dump cfg_checks from https://git.kernel.org/pub/scm/utils/b4/b4.git/tree/src/b4/ez.py?id=33ea697cc3d0bc77d2f3fb58176116608a0f1332#n1923 I only get "enable-all" the default value if the prep-pre-flight-checks key is missing from the config dict. """ multivals = ['keyringsrc', 'am-perpatch-check-cmd', 'prep-perpatch-check-cmd'] if topdir: wtcfg = os.path.join(topdir, '.b4-config') """ patching b4 with the above makes the property persist. No clue if that is the proper way to fix it nor what makes a setting worthy of being in that list. Have I missed/misconfigured something somehow and this should work? If it helps in anyway, running Fedora 41, Python 3.13.1. b4 tested from master (33ea697cc3d0 ("b4: stop generating full object IDs for non-binary diffs")) with b4.sh and also from v0.14.2 from pypi. Let me know if there's anything I can do. Cheers, Quentin