Mailman 3 April 2003 - Python-Dev

backporting string changes to 2.2.3
by Neal Norwitz April 11, 2003

April 11, 2003

Just in case anybody missed it the first several times, there were several inconsistencies in the string methods/functions. This checkin should make everything consistent for 2.3. I'm planning to backport these string changes to 2.2.3. The reason is that methods on string objects already have the changes, only doc is being updated. The string module has the change for strip, but not lstrip/rstrip, and UserString doesn't have any. Modified Files: Doc/lib/libstring.tex: 1.49 Lib/UserString.py: 1.17 Lib/string.py: 1.68 Lib/test/string_tests.py: 1.31 Objects/stringobject.c: 2.208 Objects/unicodeobject.c: 2.187 Log Message: Attempt to make all the various string *strip methods the same. * Doc - add doc for when functions were added * UserString * string object methods * string module functions 'chars' is used for the last parameter everywhere. These changes will be backported, since part of the changes have already been made, but they were inconsistent. Neal

3 3

tp_clear return value
by Jeremy Hylton April 11, 2003

April 11, 2003

Why does tp_clear have a return value? All the code I've seen returns 0, but the only place that clear is called doesn't inspect its return value. Jeremy

9 25

tzset
by Tim Peters April 11, 2003

April 11, 2003

We seem to have added tzset() gimmicks to CVS Python. test_time now fails on Windows, simply because time.tzset raises AttributeError there. Now Windows does support tzset(), but not TZ values of the form test_time.test_tzset() is testing, like environ['TZ'] = 'US/Eastern' and environ['TZ'] = 'Australia/Melbourne' The rub here is that I haven't found *any* tzset man pages on the web that claim TZ accepts such values (other than to silently ignore them because they're not in a recognized format). The POSIX defn is typical: http://www.opengroup.org/onlinepubs/007904975/basedefs/xbd_chap08.html and search down for TZ. There's no way to read that as supporting the values we're testing. Anyone have a clue? not-all-pits-should-be-dived-into-ly y'rs - tim

5 17

Re: More socket questions
by Marangozov, Vladimir (Vladimir) April 11, 2003

April 11, 2003

Hi, inet_aton() is a pretty simple parser of an IP address string, but it is not available on all setups. Libraries relying on it usually provide a local version. So do the same. Search the Web for "inet_aton.c" and you'll hit a standard implementation, with all the niceties about the base encoding of each part of the IP address which follows the C convention: 0x - hex, 0 - octal, other - decimal. And thus, BTW, "ping 192.30.20.10" is not the same as "ping 192.030.020.010". So take that code, stuff it in my_inet_aton() and case closed. You could use my_inet_aton() before calling gethostbyname('name') to see whether 'name' is an IP address and return immediately, but as I said, decent resolvers should do that for you. After all, their job is to give you an IP address in return. If you feed an IP address as an input, you should get it as a reply. Not all resolvers are decent, though. On top of that, some have bugs :-). I can't answer the question about netdb's status quo. Cheers, Vladimir

1 0

More socket questions
by Mihai Ibanescu April 10, 2003

April 10, 2003

Hello, Since somebody mention inet_addr, here's something else that I can attempt to fix if we agree on it. In python 2.2.2: socket.inet_aton("255.255.255.255") Traceback (most recent call last): File "<stdin>", line 1, in ? socket.error: illegal IP address string passed to inet_aton Implementation: static PyObject* PySocket_inet_aton(PyObject *self, PyObject *args) { #ifndef INADDR_NONE #define INADDR_NONE (-1) #endif /* Have to use inet_addr() instead */ char *ip_addr; unsigned long packed_addr; if (!PyArg_ParseTuple(args, "s:inet_aton", &ip_addr)) { return NULL; } #ifdef USE_GUSI1 packed_addr = inet_addr(ip_addr).s_addr; #else packed_addr = inet_addr(ip_addr); #endif if (packed_addr == INADDR_NONE) { /* invalid address */ PyErr_SetString(PySocket_Error, "illegal IP address string passed to inet_aton"); Reason for this behaviour can be found in the man page for inet_addr: The inet_addr() function converts the Internet host address cp from numbers-and-dots notation into binary data in network byte order. If the input is invalid, INADDR_NONE (usually -1) is returned. This is an obsolete interface to inet_aton, described immediately above; it is obsolete because -1 is a valid address (255.255.255.255), and inet_aton provides a cleaner way to indicate error return. I propose that we use inet_aton to implement PySocket_inet_aton (am I clever or what). The part that I don't know, how portable is this function? Does it exist on Mac and Windows? Thanks, Misa

4 3

Why is spawn*p* not available on Windows?
by Shane Holloway (IEEE) April 10, 2003

April 10, 2003

Ok, so here's my story. I got curious as to why the various spawn*p* were not available on Windows. The conclusion I came to is that only "spawnv" and "spawnve" are exported by posixmodule.c, and os.py creates the other variants in terms of these functions. However, the "spawnvp" and "spawnvpe" python implementations are dependant upon the availablity of "fork". So, after all that, I looked in standard library header file for process.h and found function prototypes for the various _spawn functions. Would it make sense to add support for "spawnvp" and "spawnvpe" to posixmodule.c? Should it be guarded by the existing HAVE_SPAWNV, new HAVE_SPAWNVP, or by MS_WINDOWS definitions? Or, has someone already tried this with lessons learned? Thanks, -Shane

2 1

Re: _socket efficiency ideas
by Marangozov, Vladimir (Vladimir) April 10, 2003

April 10, 2003

Hi, About the DNS discussion, I'll chime in with some info. (I don't know what Python does about this and have no time to figure it out). The format of an Internet (IPv4) address is: a.b.c.d - with all parts treated as 8 bits a.b.c - with 'c' treated as 16 bits a.b - with 'b' treated as 24 bits a - with 'a' treated as 32 bits You can try this out with ping 127.1; ping 127, etc. Any decent DNS resolver first tries to figure out whether the requested name string is an IP address. If it is, it doesn't send a query and returns immediately the numeric value of the string representation of the IP address. How a DNS resolver detects whether it should launch a query for the name 'name' varies from resolver to resolver, but basically, it does the following: 1. check for local resolution of 'name' (ex. if 'name' == 'localhost', return 127.0.0.1) 2. if inet_aton('name') succeeds, 'name' is an IP address and return the result from inet_aton. 3. If caching is enabled, check the cache for 'name' If 1, 2 and 3 don't hold, send a DNS query. Caching is a separate/complementary issue and I agree that it should be left to the underlying resolver. Cheers, Vladimir

2 1

Unicode
by Huey Jiang April 10, 2003

April 10, 2003

Hi There, I wonder how can I get python to support Chinese language? I noticed python has Unicode feature in version 2.2.2, but as I tried: >>> str = " a_char_in_chinese_lan" I encountered UnicodeError. How can I make this to work? Thanks! Huey __________________________________________________ Do you Yahoo!? Yahoo! Tax Center - File online, calculators, forms, and more http://tax.yahoo.com

7 7

Minor issue with PyErr_NormalizeException
by Eric Huss April 10, 2003

April 10, 2003

We had a bug in one of our extension modules that caused a core dump in PyErr_NormalizeException(). At the very top of the function (line 133) it checks for a NULL type. I think it should have a "return" here so that the code does not continue and thus dump core on line 153 when it calls PyClass_Check(type). This should also make the comment not lie about dumping core. ;) Just thought I'd pass it on.. -Eric

4 3

Re: [Python-Dev] OT: Signal/noise ratio
by Zooko April 10, 2003

April 10, 2003

Aahz wrote: > > I am starting to > think that a better idea would be to make python-dev a closed list (only > subscribers may post), and have the subscription process include a > challenge/response with a code word embedded in the list rules. This is how we run p2p-hackers [1] with Mailman and it works quite well to quell off-topic posts without, as far as I can tell, deterring any valuable posts. Regards, Zooko http://zooko.com/ ^-- under re-construction: some new stuff, some broken links [1] http://zgp.org/mailman/listinfo/p2p-hackers

2 1