another #! !# problem

(begin (display 1) #! (display 2) !# ) gets: ERROR: In procedure scm_lreadr: ERROR: tmp34.scm:6:2: unexpected ) ABORT: (read-error) whereas (begin #! (display 2) !# (display 1) ) is ok. This is the case in both the current CVS guile and version 1.6.4.

Documentation error: srfi-13, string-tokenize

The description of string-tokenize in 1.6.4 (doc/ref/srfi-modules.texi) does not seem to be quite accurate. The part under the Scheme Procedure heading looks like it was fixed in head in cvs, but the synopsis just under the miscellaneous heading still implies that the token_char argument can be

Re: another #! !# problem

Bill Schottstaedt [EMAIL PROTECTED] writes: (begin (display 1) #! (display 2) !# ) gets: ERROR: In procedure scm_lreadr: ERROR: tmp34.scm:6:2: unexpected ) Looks like #! !# doesn't work as the last thing in a list. Maybe that comment should be detected and discarded in

Re: Documentation error: srfi-13, string-tokenize

Mike Small [EMAIL PROTECTED] writes: The description of string-tokenize in 1.6.4 (doc/ref/srfi-modules.texi) does not seem to be quite accurate. The part under the Scheme Procedure heading looks like it was fixed in head in cvs, Thanks, I brought that across.

Re: [patch] 1.6.4: aclocal 1.8 compatibility

Kevin Ryde [EMAIL PROTECTED] writes: (Not sure what automake version will actually be used for the next release though.) NB: unless there's some compelling reason not to, I generally use whatever's in current Debian unstable. -- Rob Browning rlb @defaultvalue.org and @debian.org; previously

Re: gethost error display

Kevin Ryde [EMAIL PROTECTED] writes: so that we pass the bad hostname as one of the exception extra data items rather than making it the first format string item. Or alternately, perhaps just quietly drop it. Other system errors don't include the offending filename or whatever in the throw.