Rietveld Code Review Tool
Help | Bug tracker | Discussion group | Source code | Sign in
(7830)

Issue 577710043: Issue 5864: improve Rational infinity initialization (Closed)

Can't Edit
Can't Publish+Mail
Start Review
Created:
4 years, 1 month ago by Dan Eble
Modified:
4 years ago
Reviewers:
lemzwerg, hanwenn
CC:
lilypond-devel_gnu.org
Visibility:
Public.

Description

https://sourceforge.net/p/testlilyissues/issues/5864/ 1: take advantage of C++11 in real.hh (this is a tangent) 2: handle init by (nonzero, 0) as +/- infinity 3: handle init by double +/- infinity This work grew out of concern that not supporting these cases was the source of a problem in some code I was experimenting with. I no longer believe that, but I thought it would be beneficial to submit anyway. I also spent some time trying to make 0/0 and NaN work in a way that seemed consistent to me, but I gave up on that when I started getting regression test differences.

Patch Set 1 : take advantage of Cplusplus11 in real.hh #

Patch Set 2 : handle init by (nonzero, 0) as +/- infinity #

Patch Set 3 : handle init by double +/- infinity #

Unified diffs Side-by-side diffs Delta from patch set Stats (+37 lines, -35 lines) Patch
M flower/include/real.hh View 1 chunk +12 lines, -14 lines 0 comments Download
M flower/rational.cc View 1 2 3 chunks +9 lines, -3 lines 0 comments Download
D flower/real.cc View 1 chunk +0 lines, -10 lines 0 comments Download
M flower/test-rational.cc View 1 2 3 chunks +16 lines, -8 lines 0 comments Download

Messages

Total messages: 5
Dan Eble
handle init by (nonzero, 0) as +/- infinity
4 years, 1 month ago (2020-03-26 02:20:33 UTC) #1
Dan Eble
handle init by double +/- infinity
4 years, 1 month ago (2020-03-26 02:21:05 UTC) #2
lemzwerg
LGTM
4 years, 1 month ago (2020-03-26 05:58:58 UTC) #3
hanwenn
LGTM
4 years ago (2020-03-31 20:24:10 UTC) #4
hanwenn
4 years ago (2020-03-31 20:24:10 UTC) #5
LGTM
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld f62528b