What I was doing : I was comparing 2 rather large files with many
differences. Most of the differing lines were identical EXCEPT for
some upper/lower case differences. And I don't care about them,
because for all intents and purposes, Tomato and tomato and TOMATO
are all the same to me.
So I checked the "Ignore case changes" option. I would have expected
this to hide all the changes that were only different in their case,
allowing me to easily move around to the important changes.
Instead, Merge changed the highlighting so that only the WORDS that
had the case differences were highlighted, rather than the whole line,
and in a less noticeable color. But Next and Previous difference
still stopped on these lines. This does not help me get to the
relevant changes so much. There might as well not be an "Ignore
case" option if this is all that it does.
Please don't take the next section in the wrong way... I include it
for two reasons... 1) the possibility that a language gap is
contributing to the issue; and 2) I read a post from June 2006 by
Brad Stiles where he brought up the same problem. His report seems
to have been dismissed without any real consideration, as if his
expectation of the feature was wrong...
I looked up the word "ignore" in three different online dictionaries.
I think these represent the universally accepted meaning of the word:
- to refrain from noticing or recognizing
- To refuse to pay attention to; disregard
- to take no notice of; to pay no attention to
By these definitions, TortoiseMerge is doing no such thing. It is, in
fact, RECOGNIZING the differences (evidenced by highlighting them),
and is _paying_attention_to_them_ by stopping at each and every of
them, which in turn forces me to pay attention to them. A better
title for the feature might be "Make Case Changes a Little Less
Obvious" or "Attenuate the Look of Case Changes".
OK, legalism and humor aside... My contention is that by any
meaningful interpretation of the term "Ignore Case Changes" (whether
technical, traditional, or intuitive), the behavior of "ignoring case
changes" should be to act as if those changes were never there. The
diff program should highlight and navigate as if both files had been
run through a ' tolower ' filter before comparing.
The feature as it exists is pretty much useless. At least I can't
think of a good use for it, which is why I assume it's a bug.
Final thought... The TortoiseSVN help file wording more closely
matches my expectations than it matches the actual behavior:
"... hides changes which are due solely to case changes within
the text. ..."
I hope I've argued my case strongly enough, but not too strongly ;)
To unsubscribe, e-mail: firstname.lastname@example.org
For additional commands, e-mail: email@example.com
Received on Thu Feb 15 02:06:13 2007