[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [Gcl-devel] Re: Time/work remaining?
From: |
Mike Thomas |
Subject: |
RE: [Gcl-devel] Re: Time/work remaining? |
Date: |
Thu, 18 Mar 2004 17:18:17 +1000 |
Hi Camm.
| > 2. I discovered this afternoon that si::info doesn't wait for
| a response
| > from the keyboard anymore when more than one option exists. It seems
| > related to the problem I reported with Maxima doc the other day, but I
| > haven't chased it up yet. Does this also happen with Linux?
| >
|
| I think this is windows specific. Below is an example on Linux. Also
| works after (si::readline-off). Is this related to the removal of
| that extra CR? Is this change working otherwise? One would think
| something is still providing an extra CR somewhere. Can you pinpoint?
| How many windows specific i/o code snippets do we have?
I'm pretty bogged down with other stuff at the moment but what I have found
out is that the problem is with (read-line). It doesn't seem to be related
to recent changes as a version of GCL from January 5 exhibits the same
problem.
I'm having trouble debugging as even if I don't remove read.c in the build
process the debugger uses read.d and gets lost due to the nasty macros eg:
Breakpoint 1, Lread_line () at read.d:2112
2112 if (strm == Cnil)
(gdb) p strm
No symbol "strm" in current context.
(gdb) n
2119 @(return eof_value)
(gdb) bt
#0 Lread_line () at read.d:2119
#1 0x0041f487 in eval (form=0x650a80) at eval.c:1090
gdb) n
120 else
gdb) n
122 }
gdb) n
123 i = 0;
gdb) n
124 for (;;) {
gdb) n
126 if (char_code(c) == '\n') {
gdb) p c
o symbol "c" in current context.
gdb) n
127 c = Cnil;
gdb) p c
o symbol "c" in current context.
gdb)
Sorry to slow you down on the release but I'm overwhelmed with work at the
moment.
Mike Thomas.