File: KNOWN_BUGS

package info (click to toggle)
oo2c 1.5.5-3.1
  • links: PTS
  • area: main
  • in suites: woody
  • size: 6,268 kB
  • ctags: 692
  • sloc: ansic: 4,220; sh: 507; makefile: 349; perl: 57; lisp: 20
file content (81 lines) | stat: -rw-r--r-- 2,310 bytes parent folder | download | duplicates (5)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
This file lists all known problems and bugs that should be fixed in
oo2c.  It doesn't address any non-oo2c related installation problems.
For every item a short description, a workaround, and the email
address of the responsible developer is listed.

Any help to solve these problems is appreciated.

-- Michael van Acken <acken@informatik.uni-kl.de>


------------------------------------------------------------------------


Date: Jan 13 1998
Problem: oo2c 1.3.1+ floating point troubles on SunOS 4(?) 
Reported by: grieblm@trt.allied.com

Description:
During installation the compiler aborts with an error message "Number
out of range" in LowReal.Mod.  This is most likely caused by an error
in SunOS's libc functions for conversion from string to real values.
There were no further reports for other systems.  oo2c is known to
compile on a Solaris 2.6 machine.

Workaround:
Take the oo2c distrib and replace all RealConv files (.Mod, .c, .d,
.h) with their counterparts from 1.3.0 before running "make".

Responsible: grieblm@trt.allied.com

--------

Date: since ancient times
Problem: Error reporting of math libraries does not match the manual

Description:
The current implementation of the math modules does not report errors
as described in the OOC Reference Manual.  It uses a proprietary
mechanism to signal errors, instead of raising the required
exceptions.

Workaround:
None.

Responsible: grieblm@trt.allied.com

--------

Date: since ancient times(?)
Problem: On an Alpha machine, the C compiler reports fp underflows

Description:
On machines that do not implement IEEE gradual underflow, there are
still two warnings about numerical underflows in the math modules.
Nevertheless the math functions seem to work as advertised.

Workaround:
None.

Responsible: grieblm@trt.allied.com

--------

Date: Jul 9 1999
Problem: Modules `Rider', `LongRider', and `UnicodeRider' do not exist

Description:
The OOC Reference Manual describes the modules `Rider', `LongRider',
and `UnicodeRider', but they are not part of the compiler
distribution.  The reason for this is, that the module implementations
do not exist yet.

Workaround:
Write them.

Responsible: acken@informatik.uni-kl.de
(Well, technically I am responsible, but I am waiting for someone else
to put these modules together.)

--------