From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9564 invoked by alias); 17 Sep 2012 14:57:58 -0000 Mailing-List: contact zsh-users-help@zsh.org; run by ezmlm Precedence: bulk X-No-Archive: yes List-Id: Zsh Users List List-Post: List-Help: X-Seq: 17272 Received: (qmail 19476 invoked from network); 17 Sep 2012 14:57:56 -0000 X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on f.primenet.com.au X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,RCVD_IN_DNSWL_NONE, SPF_HELO_PASS autolearn=ham version=3.3.2 Received-SPF: none (ns1.primenet.com.au: domain at bewatermyfriend.org does not designate permitted sender hosts) From: Frank Terbeck To: Ioannis Koutras Cc: zsh-users@zsh.org, Seth House Subject: Re: vcs_info on a hg repository with a git subrepository; actionformats works ok, formats does not In-Reply-To: (Ioannis Koutras's message of "Mon, 17 Sep 2012 15:37:34 +0200") References: User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.2.50 (gnu/linux) Date: Mon, 17 Sep 2012 16:50:14 +0200 Message-ID: <87boh4bt5l.fsf@ft.bewatermyfriend.org> MIME-Version: 1.0 Content-Type: text/plain X-Df-Sender: [pbs]NDMwNDQ0 Ioannis Koutras wrote: [...] > Generally this setup works well, there is just one case where it does > not, a mercurial repository with a git sub-repository. More > specifically, when no change has been done I get a prompt like this: > >> (hg) [git --version ] > > On the other hand, prompt is ok if there is a change, e.g.: > >> (hg) [5! default ] > > I have found out that %i is replaced by "git" and %b by "--version" in > formats, but in actionformats they seem to be properly replaced. > > Apparently this is an issue of the formatting, but in previous > versions it was working. E.g. the same configuration works on another > PC with zsh v4.3.17, mercurial v2.0.2 and git v1.7.9.5. Any idea what > could be the problem? There were no changes in the mercurial backend between 4.3.17 and 5.0.0, so I guess mercurial changed its output. The call responsible for the information with your setup is probably: % hg --debug id -i -n -b -r. Try those in the working and the non-working setup, is there a difference? Regards, Frank (CC:ing Seth, who wrote most of the advanced hg-backend code)