Re: Inconsistent behaviour of BufferLines

[Piotr Piastucki <leech miranda gmail com>]

Hi,

The attached patch seems to fix the issues I spotted in BufferLines 
behaviour so far. I tested it a bit and I did not notice any side 
effects, however, more testing will be needed. And of course I agree it 
is not a good idea to introduce such a change just before the 1.6 
release. It may come in handy once the new cycle is started though.

Regards,
Piotr

>From 2ed9057bd04fbeadcd62b60aa6c95c808986e60d Mon Sep 17 00:00:00 2001
From: Piotr Piastucki <leech miranda gmail com>
Date: Sat, 7 Apr 2012 15:55:53 +0200
Subject: [PATCH 3/3] Fix inconsistent behaviour of BufferLines

---
 meld/filediff.py |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/meld/filediff.py b/meld/filediff.py
index b3bedbf..2c1f405 100644
--- a/meld/filediff.py
+++ b/meld/filediff.py
@@ -110,7 +110,7 @@ class BufferLines(object):
         # newline. As splitlines() discards an empty line at the end, we need
         # to artificially add a line if the requested slice is past the end of
         # the buffer, and the last line in the slice ended in a newline.
-        if hi >= self.buf.get_line_count() and \
+        if hi >= self.buf.get_line_count() and lo < self.buf.get_line_count() and \
            (len(lines) == 0 or len(lines[-1]) != len(ends[-1])):
             lines.append(u"")
             ends.append(u"")
@@ -135,7 +135,7 @@ class BufferLines(object):
         return lines
 
     def __getitem__(self, i):
-        if i > len(self):
+        if i >= len(self):
             raise IndexError
         line_start = get_iter_at_line_or_eof(self.buf, i)
         line_end = line_start.copy()
-- 
1.7.5.4