summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorGuido van Rossum <guido@python.org>1994-08-12 13:18:41 (GMT)
committerGuido van Rossum <guido@python.org>1994-08-12 13:18:41 (GMT)
commitfda5fb21583934d4a82f61e58a6706eeda0aca87 (patch)
tree5dd766298e75c3c805f92a2b6928cf62e15ede26
parenta0e9a77183c14683d2da2f6ce601db0c12952d33 (diff)
downloadcpython-fda5fb21583934d4a82f61e58a6706eeda0aca87.zip
cpython-fda5fb21583934d4a82f61e58a6706eeda0aca87.tar.gz
cpython-fda5fb21583934d4a82f61e58a6706eeda0aca87.tar.bz2
The usual
-rw-r--r--BUGS31
-rw-r--r--ChangeLog72
-rw-r--r--TODO39
3 files changed, 128 insertions, 14 deletions
diff --git a/BUGS b/BUGS
index 9a232af..fe9231a 100644
--- a/BUGS
+++ b/BUGS
@@ -10,6 +10,31 @@ nother to make an ehtry in this file, unless it was a serious bug
BUGS found in 1.0.3
-------------------
+(-) unwanted entries in stack trace if err_clear() clears an error
+that also set a stack trace
+
+(-) i, x[i] = a, b assigns b to x[a] rather than to x[i] as expected
+(if we don't fix this, it should be documented with a warning!)
+
+(-) various memory leaks (see purify report from anthony.baxter@aaii.oz.au)
+
+(*) etags no longer supports -t flag
+
+(-) compile.c:com_argdefs() references unalloc'ed memory for def
+f(a=1,): ...
+
+(-) Syntax errors are reported in a silly way if multi-line tokens are
+involved.
+
+(-) SyntaxError exception for compile('...') are reported wrongly
+(lineno is always zero and offset is offset into the whole string).
+
+(-) If you have a python binary in your path like
+/ufs/guido/bin/sgi/python then the default prefix option computed by
+the configure script is bogus!
+
+(-) Make rule for making lib*.a should remove the lib*.a file first.
+
(*) vars() error message is wrong (copied from dir() obviously).
(*) socket.gethostname() is undocumented.
@@ -19,9 +44,9 @@ BUGS found in 1.0.3
(-) urllib caching is wrong (should use date from Expires header)
(*) On a related matter: regexpr.c still has two malloc()s the results
- of which are not tested for being NULL (lines 1253 and 1530). There
- are also some in rgbimagemodule.c. Am I overlooking something or is
- this a crasher?
+of which are not tested for being NULL (lines 1253 and 1530). There
+are also some in rgbimagemodule.c. Am I overlooking something or is
+this a crasher?
(*) strop.rindex('abc', '') returns 0 instead of 3
diff --git a/ChangeLog b/ChangeLog
index cc19e53..bbd266a 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,71 @@
+Fri Aug 12 15:00:20 1994 Guido van Rossum (guido@voorn.cwi.nl)
+
+ * Doc/libfuncs.tex: don't use $math$ in description of pow(x,y,z);
+ describe tuple()
+
+ * Doc/libposixfile.tex: use tableiii instead of tableii, so
+ partparse will work again (I know, chicken!)
+
+ * Doc/libthread.tex: Added get_ident(); updated text on module
+ availability
+
+ * Doc/myformat.perl: Added sub do_cmd_Cpp
+
+ * Python/compile.c (com_argdefs, com_arglist): avoid referencing
+ CHILD(n,i) for i >= NCH(n)
+
+ * Python/bltinmodule.c: added tuple() builtin
+
+Thu Aug 11 16:41:14 1994 Guido van Rossum (guido@voorn.cwi.nl)
+
+ * Objects/classobject.c, Include/classobject.h: added __getattr__
+ and __setattr__ support to override getattr(x, name) and
+ setattr(x, name, value) for class instances. This uses a special
+ hack whereby the class is supposed to be static: the __getattr__
+ and __setattr__ methods are looked up only once and saved in the
+ instance structure for speed
+
+Wed Aug 10 13:42:29 1994 Guido van Rossum (guido@voorn.cwi.nl)
+
+ * configure.in, Makefile.in: remove the AC_PREFIX() call -- it's
+ more trouble than it's worth at CWI and most other people seem to
+ install Python in the default (/usr/local) anway. Changed comment
+ describing --prefix in Makefile.in
+
+ * Lib/urllib.py (ftpcache): remove debug print statement
+
+Tue Aug 9 14:32:45 1994 Guido van Rossum (guido@voorn.cwi.nl)
+
+ * Objects/{int,long,float}object.c, Include/object.h,
+ Python/bltinmodule.c: mods by Andrew Kuchling to implement
+ pow(x,y,z) == pow(x,y)%z, but without incurring overflow
+
+ * Python/import.c: if initializing a module did not enter the
+ module into sys.modules, it may have raised an exception -- don't
+ override this exception.
+
+ * Include/rename1.h: added PyMethodDef and PyObject
+
+Mon Aug 8 09:51:08 1994 Guido van Rossum (guido@voorn.cwi.nl)
+
+ * Doc/{Makefile,*.tex}: Changes by Andrew Kuchling:
+ * Added dependencies to the Makefile
+ * Added \optional{} commands all over the place (possibly not
+ finished yet)
+ * Defined \optional and \Cpp in myformat.sty
+ * Changed all occurrences of C++ to \Cpp{}, for nicely
+ formatting the + signs.
+ * Fixed typos, sentence structure in lots of places
+ * Added documentation for the rotor module
+ * Added a commented-out paragraph to libcrypto.tex that can be
+ added once the Python Cryptography Kit is released.
+ * Altered the table in libposixfile.tex a bit.
+ * Documented socket.gethostname()
+ * Started completely rewriting the Extending manual.
+
+ * Modules/{Setup.in, gdbmmodule.c}, Doc/{lib,libgdbm}.tex: added
+ Anthony Baxter's gdbm module (derived from Jack's dbm module)
+
Fri Aug 5 11:43:16 1994 Guido van Rossum (guido@voorn.cwi.nl)
* BUGS: new file (merger of unofficial BUGS1.0.x files)
@@ -23,7 +91,9 @@ Mon Aug 1 01:28:29 1994 Guido van Rossum (guido@voorn.cwi.nl)
* Makefile.in (TAGS): Call etags w/o -t option
- * Lib/rfc822.py: fix two bugs
+ * Lib/rfc822.py: fix two bugs: error in readheaders interpreting
+ regex.match() result, and wrong logic in getfirstmatchingheader()
+ when the same header occurs twice consecutively
* Lib/test/test_types.py (6.4.1): test for particular bug in
integer multiply
diff --git a/TODO b/TODO
index ab0ea26..78308a8 100644
--- a/TODO
+++ b/TODO
@@ -1,3 +1,22 @@
+(-) redo __xxx__ operators for class instances; add __getattr__ and
+__setattr__.
+
+(-) many things that take strings should also take arrays of chars
+
+(-) fix pow(x, y, z) for integers
+
+(*) add tuple(seq) to turn any sequence into a tuple
+
+(-) add list of existing extensions to FAQ
+
+(-) update "recent additions" chapter in tutorial
+
+(-) rewrite "output formatting" chapter in tutorial
+
+(-) add a make rule to build .so files (unfortunately need to figure
+ out what ld option is needed) (more appropriate for makesetup
+ script?)
+
(-) write a script and add a "Make" rule (perhaps) that changes
#!/usr/local/bin/python to something else in all scripts around.
@@ -11,19 +30,19 @@
----------------------------------------------------------------------
(from BUGS1.0.1)
----------------------------------------------------------------------
-document addpack, urllib, ...
+(-) document addpack, urllib, ...
(*) import.c from JaapV
-document os.exec*
+(-) document os.exec*
-name sunaudiodevmodule.c is too long
+(*) name sunaudiodevmodule.c is too long
(*) play with / include SUIT interface
-make regsub.[g]sub() optionally case insensitive
+(-) make regsub.[g]sub() optionally case insensitive
-handle printing of errors in lines containing multiline strings
+(-) handle printing of errors in lines containing multiline strings
======================================================================
@@ -31,21 +50,21 @@ handle printing of errors in lines containing multiline strings
(*) errors in __repr__() are handled wrong when called from format
-- long(0x80000000) has wrong value!
+(*) long(0x80000000) has wrong value!
-- hex(0x80000000) shouldn't have sign (?)
+(-) hex(0x80000000) shouldn't have sign (?)
(*) need way to set buffering at file open time
(*) need way to force stdout unbuffered
-- document new modules (builtin as well as Lib)
+(-) document new modules (builtin as well as Lib)
-- restart CVS tree
+(*) restart CVS tree
(?) build shared libs for SunOS 4.1.3
-- Mac X... code resources as shared libs?
+(-) Mac X... code resources as shared libs?
(*) X patches from Adrian Phillips