[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Issue 1870 in lilypond: replace DIGIT token class with UNSIGNED
From: |
lilypond |
Subject: |
Issue 1870 in lilypond: replace DIGIT token class with UNSIGNED |
Date: |
Mon, 05 Sep 2011 20:11:16 +0000 |
Status: Started
Owner: address@hidden
Labels: Type-Other Patch-needs_work Warning
New issue 1870 by address@hidden: replace DIGIT token class with UNSIGNED
http://code.google.com/p/lilypond/issues/detail?id=1870
replace DIGIT token class with UNSIGNED
Do _not_, I repeat, do _not_ merge. This patch passes regression
testing. It removes the token class DIGIT (single-digit unsigned
constants) and replaces it by UNSIGNED (previously multiple-digit
unsigned constants, now any unsigned constants). Every previous
occurence of UNSIGNED in the grammar has been _removed_: instead every
occurence of DIGIT has been _replaced_ with UNSIGNED.
The result is that single-digit unsigned constants are interpreted by
the grammar just like they were before. For example,
xxx=-3
creates a variable containing a _fingering_ event instead of a numeric
value of -3.
Multiple-digit unsigned numbers, however, will be interpreted as
strange as previously just single-digit constants were. So
xxx=-30
also creates a variable containing a fingering event rather than a
numeric constant.
I consider the resulting grammar bordering on insane but have been
unable to come up with a grammar without additional shift/reduce
conflicts that would have resolved this particular nuisance the other
way round.
So this is just food for thought.
http://codereview.appspot.com/4311041/
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Issue 1870 in lilypond: replace DIGIT token class with UNSIGNED,
lilypond <=