Open salewski opened 3 years ago
Poor man's workaround for the time being (see also issue #32 for a different workaround): When you hit problematic repos, re-run ads-github-fetch-all-upstreams
plugging-in the name of the most recent problematic repo encountered into the sed
expresion:
$ ads-github-fetch-all-upstreams -v $(find . -mindepth 1 -maxdepth 1 -type d | sed -ne 's,^[.]/,,p' | sort | sed -e '0,/^PROBLEMATIC_REPO_NAME$/d')
That has the effect of skipping over all of the repos through the one named. You will need to do the above once for every problematic repo you encounter, so with more than a few it will get tedious fast.
This issue pertains to
ads-github-fetch-all-upstreams
, as shipped withads-github-tools-0.3.3
(and all previous versions).This issue is similar in spirit to issue #32, but approaches the problem with a less sophisticated (but perhaps easier to use) mechanism.
When you have a significant number (or pehaps just an unknown number) of problematic repos (e.g., forked repos where the original upstream repo is no longer accessible on GitHub), you sometimes want to tell the
ads-github-fetch-all-upstreams
to "just deal with the repos that can be dealt with; complain about problematic repos (onstderr
), but otherwise keep going".We should add a command line option to
ads-github-fetch-all-upstreams
to allow the the user to request the above behavior. For consistence with theads-github-merge-all-upstreams
program, we should name the option-k
(--keep-going
).