[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: How to extract a Changed Files Report for a single SVN User?

From: Erik Huelsmann <ehuels_at_gmail.com>
Date: 2007-06-08 11:15:37 CEST

On 6/8/07, Luis Motta Campos <l.campos@blinck.com> wrote:
> On Jun 8, 2007, at 10:21 AM, Erik Huelsmann wrote:
> > Hi Luis,
> > Well, you can retrieve the changed files and the user who changed them
> > all at once by running the command:
> >
> > svn log -vq <url of the root of your repository>
> >
> > It will give you log formatted output, with a list of added / changed
> > paths, but won't include the log message. From the header line you can
> > retrieve the username.
> >
> > If you do want to see the log messages, run 'svn log -v <url>'
> > instead.
>
> Hi Erik
>
> Thanks for the quick answer. "svn log" was almost all I wanted. A
> quicck look at the manual gave me
>
> svn log -vqr COMMITTED </uri/to/repo>
>
> as a slightly better solution. The log messages aren't exactly
> machine-parseable.

Actually, the messages may not be machine parseable, but the log
output has been designed to be machine parseable. If you look at the
line below the separator line (the one with the =-signs), you see at
the end the number of lines. It's not the number of lines in the
change, but the number of lines in the log item to follow :-) This
means you know exactly how many lines to read until the next
separator.

Also, if you request the changed files to be listed, the order is
defined to be the changed paths first and than the log message. The
format of the changed paths is also fixed in order to be machine
parseable. (It's, if I'm not mistaken, the same format as used by svn
status.)

The user name of the one who committed the change can be read from the
line below the separator, as can the date of the commit.

> Anyway, with a bit of patience, I was able to locate the version
> where the changes begin.
>
> Does anyone knows about a way of having a report of changed files
> per revision, per user, one entry per line, so I can parse it easyly
> using perl? I've been reading about a perl interface to SVN, this is
> a good starting point for what I want?
>
> I wonder if someone else have problems like this, and if making a
> tool for this stuff could be of any interest to other people (so this
> could become my first module on Perl's CPAN repository).
>
> Thanks a lot again. :-)

There are tools to generate statistics from your svn repository, they
might include functionality which help analyse scenario. Have a look
at http://subversion.tigris.org/links.html .

bye,

Erik.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Fri Jun 8 11:16:27 2007

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.