There's a bug in the 1.0.2 tokenizer.c that makes it run wild if it
receives a last line that doesn't end in \n. Here's the fix (again?):
===================================================================
RCS file: /hosts/voorn/ufs/guido/python-RCS/new/Parser/RCS/tokenizer.c,v
retrieving revision 2.14.2.3
diff -c -1 -r2.14.2.3 tokenizer.c
*** 2.14.2.3 1994/04/28 12:31:46
--- tokenizer.c 1994/05/23 12:40:51
***************
*** 269,272 ****
(int)(tok->end - tok->inp),
! tok->fp) == NULL)
! break;
tok->inp = strchr(tok->inp, '\0');
--- 269,275 ----
(int)(tok->end - tok->inp),
! tok->fp) == NULL) {
! /* Last line does not end in \n,
! fake one */
! strcpy(tok->inp, "\n");
! }
tok->inp = strchr(tok->inp, '\0');
--Guido van Rossum, CWI, Amsterdam <Guido.van.Rossum@cwi.nl>
URL: <http://www.cwi.nl/cwi/people/Guido.van.Rossum.html>