[PATCH 3 of 3] cvsps: reproduce crash on bad branch name in CVS source (issue1615)

[Patrick Mézard]

Greg Ward a écrit :
> # HG changeset patch
> # User Greg Ward <greg-hg at gerg.ca>
> # Date 1240161140 14400
> # Node ID 1085b62d0fdfe10f2c725114ee1716e6280bc130
> # Parent  341830bc285a92aa3affb33d229872b2bbb8c7db
> cvsps: reproduce crash on bad branch name in CVS source (issue1615).

FYI, I will merge it with the one fixing the issue.

> diff --git a/tests/test-convert-cvs-detectmerge b/tests/test-convert-cvs-detectmerge
> --- a/tests/test-convert-cvs-detectmerge
> +++ b/tests/test-convert-cvs-detectmerge
> @@ -31,6 +31,7 @@
>  
>  echo "[extensions]" >> $HGRCPATH
>  echo "convert = " >> $HGRCPATH
> +echo "graphlog = " >> $HGRCPATH
>  echo "[convert]" >> $HGRCPATH
>  echo "cvsps=builtin" >> $HGRCPATH
>  echo "cvsps.mergefrom=\[MERGE from (\S+)\]" >> $HGRCPATH
> @@ -90,19 +91,26 @@
>  cvsci -m"add other text [MERGE from HEAD]" file1
>  
>  set -e
> +echo "% fix bug on v1_1, merge to trunk with error"
> +cvscall -Q update -rv1_1
> +echo "merge forward" >> file1
> +cvscall -Q tag unmerged
> +cvsci -m"fix file1"
> +cvscall -Q update -A
> +filterpath cvscall -Q update -junmerged -jv1_1
> +# note the typo in the commit log message
> +cvsci -m"fix file1 [MERGE from v1-1]"
> +cvs -Q tag -d unmerged
> +
> +set -e
>  echo % convert to hg
>  cd ..
> +rm -rf ~/.hg.cvsps

Is the rm -rf necessary here? Perhaps if you call run-test.py with a specific TMPDIR?

In this case, we could add an option to cvs convert to disable caching instead.