Hi Ben!
On So, 10 Jun 2012, Ben Fritz wrote:
> On Sunday, June 10, 2012 7:52:37 AM UTC-5, Christian Brabandt wrote:
> > How did you generate the aligned.patch file? No matter what I do, for me
> > the plugin always generates:
> >
> > 2,7c2,4
> > < a
> > < b
> > < cb
> > < db
> > < eb
> > < f
> > ---
> > > cc
> > > dd
> > > ee
> >
> > But perhaps, I'm using your plugin wrongly, e.g. how many alignemnt
> > points do I have to set and where? But even after I manually copied the
> > aligned.patch file over the resulting file, Vim still gets the alignment
> > wrong.
> >
>
> Oh, sorry.
>
> enter the text shown in the screenshots, place the cursor on line 4 of
> file_a.txt ("cb"), press <Leader>mda to "mark diff alignment". Go to
> line 2 of file_b.txt and again type <Leader>mda> to indicate that
> those two points correspond to each other. Then run :diffupdate.
I did that. I still get the above patch. Ah, I found the problem. You
are parsing the output of sign place, and on my localized version it
prints:
:sign place
--- Zeichen ---
Zeichen für ../diff1.txt:
Zeile=2 id=1339437602 Name=manDiffAlign01
Zeichen für ../diff2.txt:
Zeile=4 id=1339437599 Name=manDiffAlign01
So it is probably better to look for the first equal sign.
After changing my Vim to an english localization I could reproduce your
issue. I think the problem is, that Vim thinks, the 2 diff regions are
overlapping. This looks like an "off-by 1" error for me, and indeed this
patch fixes it for me:
diff --git a/src/diff.c b/src/diff.c
--- a/src/diff.c
+++ b/src/diff.c
@@ -1318,8 +1318,8 @@
}
if (dp != NULL
- && lnum_orig <= dp->df_lnum[idx_orig] + dp->df_count[idx_orig]
- && lnum_orig + count_orig >= dp->df_lnum[idx_orig])
+ && lnum_orig < dp->df_lnum[idx_orig] + dp->df_count[idx_orig]
+ && lnum_orig + count_orig > dp->df_lnum[idx_orig])
{
/* New block overlaps with existing block(s).
* First find last block that overlaps. */
> > BTW: attached is a patch, that prevents the use of using diffexpr
> > for checking if diff really works (using the "line1" vs. "line2"
> > check):
> >
>
> I don't think that's a very good idea, what if somebody uses diffexpr
> to specify the full path to diff? Or to use a separate program
> entirely, and they don't even have diff installed on their system?
> Then the test will always fail, right?
Good point, indeed.
> But it should be documented by diffexpr how the test actually works,
> it took me quite by surprise.
Yes, this looks unsurprisingly when first using 'diffexpr'.
regards,
Christian
--
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php
