summaryrefslogtreecommitdiff
path: root/TODO
diff options
context:
space:
mode:
authorMario Lang <mlang@delysid.org>2008-09-08 23:41:22 +0200
committerMario Lang <mlang@delysid.org>2008-09-08 23:41:22 +0200
commit57cfaebd90f0eecc5bc28ad9d46892e5f5d086af (patch)
treeb3166cc3c2d52caa8b25d241986ebaa9e16e5747 /TODO
parent10482f2760dcda6df099fb03f610bf4e6c31dc65 (diff)
Added a TODO item about chess-scid-get-result failing on some machines.
Diffstat (limited to 'TODO')
-rw-r--r--TODO11
1 files changed, 11 insertions, 0 deletions
diff --git a/TODO b/TODO
index 5b3db0d..d68aefb 100644
--- a/TODO
+++ b/TODO
@@ -6,6 +6,17 @@ EMACS-CHESS -*- mode: org; fill-column: 78 -*-
#+TAGS: FEATURE(f) DOCS(d) BUILD(b)
#+CATEGORY: Chess
+* TODO scid: chess-scid-get-result fails on some machines (timing?)
+ It happens on a fast 8-CPU machine of mine that chess-scid-get-result returns
+ the empty string. The reason for this is that sometimes, the prompt
+ is not output immediately after a result, so upon next call to
+ chess-scid-get-result the first output we see is the prompt ("% ") from
+ the previous command which due to the current algo results in
+ chess-scid-get-result returning "". It would need to continue reading if it
+ sees a prompt as first output. Waiting until we see a prompt to make sure we
+ have all the output results in *very slow* behaviour, maybe due to
+ line-buffering (the prompt has no final \n, obviously), so we probably shoud
+ skip over prompt and continue reading.
* TODO [#B] display: Sometimes final event of game doesn't change modeline
- Note taken on [2008-09-03 Wed 04:18] \\
I just had it happen again tonight, when I resigned with C-c C-r, and it