wireproto: properly report server Abort during 'getbundle'
Previously Abort raised during 'getbundle' call poorly reported (HTTP-500 for
http, some scary messages for ssh). Abort error have been properly reported for
"push" for a long time, there is not reason to be different for 'getbundle'. We
properly catch such error and report them back the best way available. For
bundle, we issue a valid bundle2 reply (as expected by the client) with an
'error:abort' part. With bundle1 we do as best as we can depending of http or
ssh.
#require symlink
$ origdir=`pwd`
$ hg init repo
$ cd repo
$ ln -s nothing dangling
avoid tar warnings about old timestamp
$ hg ci -d '2000-01-01 00:00:00 +0000' -qAm 'add symlink'
$ hg archive -t files ../archive
$ hg archive -t tar -p tar ../archive.tar
$ hg archive -t zip -p zip ../archive.zip
files
$ cd "$origdir"
$ cd archive
$ readlink.py dangling
dangling -> nothing
tar
$ cd "$origdir"
$ tar xf archive.tar
$ cd tar
$ readlink.py dangling
dangling -> nothing
#if unziplinks
zip
$ cd "$origdir"
$ unzip archive.zip > /dev/null 2>&1
$ cd zip
$ readlink.py dangling
dangling -> nothing
#endif
$ cd ..