1- *** See the ChangeLog file for bugs fixed since 1.2 BETA 2 ***
2-
3- Sorry, this list does not claim completeness. If I fixed a bug
4- immediately upon receiving the first complaint I usually did not
5- nother to make an entry in this file, unless it was a serious bug
6- (core dump or infinite loop).
1+ THIS LIST DOES NOT CLAIM COMPLETENESS.
72
83==> Status indicators: (-) not fixed; (*) fixed; (?) not sure.
94
105======================================================================
116
12- (-) "f()=0" generates syntax error msg without line number
13-
14- ======================================================================
15-
16- Known BUGS in 1.2 BETA 1, fixed in 1.2 BETA 2
17- ---------------------------------------------
18-
19- (*) list slice assignment contains reantrancy bug
20-
21- (*) range() and xrange() break on Alpha's
22-
23- (*) interactive use of vars() gives barrage of output due to recursive
24- inclusion of '_'
25-
26- (*) configure.in contains bogus name to check for inet library
27-
28- Known BUGS in 1.1.1 and 1.2 BETA 3
29- ----------------------------------
30-
31- (-) still a memory leak in threads when thread.exit_thread() is used
7+ Problems that are difficult to solve
8+ ------------------------------------
329
33- Problems in 1.1.1 that are difficult to solve
34- ---------------------------------------------
10+ (-) "f()=0" generates syntax error msg without line number
3511
3612(-) tkinter seems to leave an exception around sometime which breaks
3713unmarshalling code objects [hard to reproduce, have added a trap to
@@ -49,9 +25,6 @@ called [problem: what if other threads are still active?]
4925Known portability problems
5026--------------------------
5127
52- (-) tkinter doesn't seem to see any declaration of malloc on sunos
53- 4.1.3?
54-
5528(-) arraymodule doesn't compile under Ultrix (FPROTO macro)
5629
5730(-) makesetup assumes CCC is the C++ compiler -- not portable
@@ -67,6 +40,44 @@ LIBS=/usr/lib/libdld.sl) [hard to test without a HP machine handy]
6740BUGS present in 1.1.1 and fixed in 1.2
6841--------------------------------------
6942
43+ (*) extraneous fclose() in run_script() in pythonrun.c for .pyc file
44+
45+ (*) __str__ is called if it exists (and then fails) when applying
46+ str() to a class
47+
48+ (*) mem leaks in inittime() in timemodule.c
49+
50+ (*) mem leak in optimize() in compile.c
51+
52+ (*) mem leak in func_dealloc() in funcobject.c
53+
54+ (*) missing DECREF for result of run_string in exec_statement() in
55+ ceval.c
56+
57+ (*) missing INCREF in RAISE_EXCEPTION case after gettupleitem() in
58+ ceval.c
59+
60+ (*) posix.utime gives problems on problems on platforms where struct
61+ utime members are bitfields
62+
63+ (*) leak in regex module.c:reg_dealloc() -- should free compiled pattern
64+
65+ (*) many uses of macros from <ctype.h> fail with signed characters
66+
67+ (*) compilation on NeXT requires manual editing of the Makefile
68+
69+ (*) tkinter should cast malloc() result
70+
71+ (*) marshal.c (w_object()) triggers GCC bug on DEC Alpha
72+
73+ (*) int/long size bug in range() and xrange() on DEC Alpha
74+
75+ (*) memory leaks in dbm and gdbm modules
76+
77+ (*) refcnt bug in select.select([f], [f], [f])
78+
79+ (*) Should fflush(stdout) before printing traceback to stderr
80+
7081(*) Linux uses GNU getopt by default which is broken
7182
7283(*) make sharedinstall references to machdep directory but doesn't
@@ -85,8 +96,8 @@ cmd line forgets a newline
8596(*) C-level coerce() doesn't call __coerce__ when it should (and
8697similar for __cmp__)
8798
88- (*) struct module aligns doubles wrongly when compiled with -DDEBUG on
89- sparc
99+ (*) struct module assigns unaligned doubles when compiled with -DDEBUG
100+ on sparc
90101
91102(*) memory leak (namebuf) in initmodule2
92103
@@ -234,14 +245,15 @@ Bugs found in 1.0.2 and not yet fixed
234245-------------------------------------
235246
236247(?) compiler warnings about argument type of uname() on ULTRIX
237- machines (don't know what to do about it)
248+ machines (don't know what to do about it) [could be fixed by fix for
249+ bitfields in struct uname]
238250
239251(?) syntax error for interactive input prints garbage instead of last
240252source line on some systems (e.g. AIX) (impossible to test/reproduce)
241- !!! I think I've found this one -- a missing INCREF in print_error!
253+ [ I think I've found this one -- a missing INCREF in print_error]
242254
243255(?) (maybe) a bad .pyc file (with old magic number) causes the .py
244- file to be ignored
256+ file to be ignored [should be fixed by rewrite of import.c]
245257
246258(?) Sunos4.0.2 / 386 configure bugs:
247259 - timelocal instead of mktime
@@ -314,7 +326,7 @@ Bugs found in 1.0.1 and fixed in 1.0.2
314326BUGS found in 1.0.0 and not yet fixed
315327-------------------------------------
316328
317- (? ) On NeXT, need to define _POSIX_SOURCE.
329+ (* ) On NeXT, need to define _POSIX_SOURCE.
318330
319331(?) there appears to be something wrong with gcc and -ldl on some
320332SunOS 4.1.3 systems
0 commit comments