test: clarify some output in `test-transaction-rollback-on-sigpipe`

We don't want a dirty transaction to remain, but need to check the transaction
was aborted. Otherwise it is easy to have no abandoned transaction if the
transaction succeed.

Differential Revision: https://phab.mercurial-scm.org/D10945

tests/test-transaction-rollback-on-sigpipe.t

@@ -1,61 +1,65
 #require bash
 Test that, when an hg push is interrupted and the remote side receives SIGPIPE,
 the remote hg is able to successfully roll back the transaction.
 
   $ hg init -q remote
   $ hg clone -e "\"$PYTHON\" \"$TESTDIR/dummyssh\"" -q ssh://user@dummy/`pwd`/remote local
 
   $ pidfile=`pwd`/pidfile
   $ >$pidfile
 
   $ script() {
   >     cat >"$1"
   >     chmod +x "$1"
   > }
 
 On the remote end, run hg, piping stdout and stderr through processes that we
 know the PIDs of. We will later kill these to simulate an ssh client
 disconnecting.
 
   $ killable_pipe=`pwd`/killable_pipe.sh
   $ script $killable_pipe <<EOF
   > #!/usr/bin/env bash
   > echo \$\$ >> $pidfile
   > exec cat
   > EOF
 
   $ remotecmd=`pwd`/remotecmd.sh
   $ script $remotecmd <<EOF
   > #!/usr/bin/env bash
   > hg "\$@" 1> >($killable_pipe) 2> >($killable_pipe >&2)
   > EOF
 
 In the pretxnchangegroup hook, kill the PIDs recorded above to simulate ssh
 disconnecting. Then exit nonzero, to force a transaction rollback.
 
   $ hook_script=`pwd`/pretxnchangegroup.sh
   $ script $hook_script <<EOF
   > #!/usr/bin/env bash
   > for pid in \$(cat $pidfile) ; do
   >   kill \$pid
   >   while kill -0 \$pid 2>/dev/null ; do
   >     sleep 0.1
   >   done
   > done
   > exit 1
   > EOF
 
   $ cat >remote/.hg/hgrc <<EOF
   > [hooks]
   > pretxnchangegroup.break-things=$hook_script
   > EOF
 
+  $ hg --cwd ./remote tip -T '{node|short}\n'
+  000000000000
   $ cd local
   $ echo foo > foo ; hg commit -qAm "commit"
   $ hg push -q -e "\"$PYTHON\" \"$TESTDIR/dummyssh\"" --remotecmd $remotecmd 2>&1 | grep -v $killable_pipe
   abort: stream ended unexpectedly (got 0 bytes, expected 4)
 
 The remote should be left in a good state
+  $ hg --cwd ../remote tip -T '{node|short}\n'
+  000000000000
   $ hg --cwd ../remote recover
   no interrupted transaction available
   [1]