Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 634576 - sys-apps/portage: emerge could fail faster when given arguments for packages that don't exist
Summary: sys-apps/portage: emerge could fail faster when given arguments for packages ...
Status: RESOLVED OBSOLETE
Alias: None
Product: Portage Development
Classification: Unclassified
Component: Core - Interface (emerge) (show other bugs)
Hardware: All Linux
: Normal normal (vote)
Assignee: Portage team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 155723 835380
  Show dependency tree
 
Reported: 2017-10-17 20:40 UTC by Zac Medico
Modified: 2023-05-20 10:08 UTC (History)
5 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Zac Medico gentoo-dev 2017-10-17 20:40:24 UTC
Currently, emerge can get very far into the dependency calculation before realizing that it has been given one or more arguments for packages that don't exist. It would be better to fail faster.
Comment 1 Zac Medico gentoo-dev 2017-10-17 20:52:51 UTC
We want to detect the packages that don't exist in the depgraph._select_files method, around here:

> 	if "/" in x.split(":")[0]:
> 		args.append(AtomArg(arg=x, atom=Atom(x, allow_repo=True),
> 			root_config=root_config))
> 		continue
> 	expanded_atoms = self._dep_expand(root_config, x)
Comment 2 Berin Aniesh 2023-05-20 09:54:46 UTC
This bug seems to have been fixed at some point in 2021. Sam James and myself tried to track down the commit, but the results are inconsistent because of different EAPI and python versions. But rest assured, this bug has been fixed. Can be tested with the command `emerge www-client/chromium "<cython-3" libreoffice dev-lang/ghc dev-haskell/doctest dev-ruby/actionpack firefox tensorflow idonotexist`. Here, portage finds "idontexist" is not a valid package before dependency calculation of any package is done.
Comment 3 Sam James archtester Gentoo Infrastructure gentoo-dev Security 2023-05-20 10:08:22 UTC
I can't for the life of me see where this got fixed but I can't seem to reproduce this right now.

I feel a bit annoyed that I can't spot which commit did it, but I also absolutely can't hit it anymore.

I wonder if it's just because of all the caching we've added?