Re: [gentoo-portage-dev] [PATCH] eapply: Output verbosely only if patch fails to apply with -F0

[Michał Górny]

On Wed, 2019-11-27 at 11:13 -0800, Zac Medico wrote:
> On 11/27/19 10:54 AM, Michał Górny wrote:
> > 12d0c48ad disabled silent output for eapply, in order to obtain fuzz
> > factors in build logs.  However, this also causes eapply to report all
> > patched files which can make logs unreadable when there are no fuzz
> > factors to be reported.  Instead, use verbose output only when applying
> > the patch with -F0 fails.
> >
> > To achieve that, attempt to apply each patch with -F0 --dry-run first.
> > If this succeeds, just silently apply the patch for real.  If it
> > doesn't, output an explicit eqawarn that the patch does not apply
> > cleanly and retry with the default fuzz factor and verbose output.
> > Non-silenced output applies both to successful application with fuzz
> > and to failure.
> >
> > Signed-off-by: Michał Górny <mgorny@g.o>
> > ---
> >  bin/phase-helpers.sh | 16 ++++++++++++++--
> >  1 file changed, 14 insertions(+), 2 deletions(-)
> >
> > diff --git a/bin/phase-helpers.sh b/bin/phase-helpers.sh
> > index 60f8d3243..6f9b52678 100644
> > --- a/bin/phase-helpers.sh
> > +++ b/bin/phase-helpers.sh
> > @@ -995,8 +995,20 @@ if ___eapi_has_eapply; then
> >  			# -f to avoid interactivity
> >  			# -g0 to guarantee no VCS interaction
> >  			# --no-backup-if-mismatch not to pollute the sources
> > -			${patch_cmd} -p1 -f -g0 --no-backup-if-mismatch \
> > -				"${patch_options[@]}" < "${f}"
> > +			local all_opts=(
> > +				-p1 -f -g0 --no-backup-if-mismatch
> > +				"${patch_options[@]}"
> > +			)
> > +			# try applying with -F0 first, output a verbose warning
> > +			# if fuzz factor is necessary
> > +			if ${patch_cmd} "${all_opts[@]}" --dry-run -s -F0 \
> > +					< "${f}" &>/dev/null; then
> > +				all_opts+=( -s -F0 )
> > +			else
> > +				eqawarn "    patch failed to apply without a fuzz factor, please rebase"
>
> The earlier ebegin call only outputs the basename, so there could be
> some confusion about the origin of the patch. This function is also
> called by eapply_user, it's possible that people could get confused and
> report bugs about fuzz in their local patches. Other than that it looks
> good.
>

Ok, I will add a full path.

-- 
Best regards,
Michał Górny