On Mon, Mar 15, 2004 at 11:58:00PM +0000, Julian Foad wrote:
>
> What do you mean by "svn grep -r X:Y" ? Perhaps something like:
>
> for R = X to Y:
> svn cat -r R [file] | grep [args]
>
> or
>
> svn cat -r X [file] | grep [args]
> for R = X+1 to Y:
> svn diff -r R-1:R [file] | grep [args]
I meant the latter, with some way of indicating which revision
the match occurred in. Something like:
svn cat -r X [file] | grep [args]
for R = X+1 to Y:
svn diff -r R-1:R [file] | grep [args] | sed -e "s/^/$R: /"
Like blame, but more along the lines of "Where and when did this regex
occur?". I plan on diving back into the blame code in order to get it
to pull deltas from the repository, rather than whole fulltexts; there
may be room for a little refactoring so that the same logic can drive
a pattern-match-accumulator as well as a blame-accumulator. Actually,
those are orthogonal decisions.
>
> It might be worth creating an "enhancement" or "wish" issue to track this,
> so that if the idea comes up again the discussion about it doesn't have to
> re-start from nothing.
>
Let's see how far this thread goes, then we can create an issue that
points to the thread?
--ben
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Mar 16 04:14:36 2004