[9fans] no error checking in strtol and strtoul

[9fans] no error checking in strtol and strtoul

[Saroj Mahapatra]

Hi,

Was the decision to leave out error checking (like ANSI C strtol and
strtoul do) deliberate? The first type of error (wrong syntax) is easy
to check (if ret value == 0 and rest == beginning of string), but not
handling overflow situation is unsatisfactory. I'd like to hear plan 9
implementators' comment on this.

Thanks,
Saroj Mahapatra

Re: [9fans] no error checking in strtol and strtoul

[rob pike, esq.]

[-- Attachment #1: Type: text/plain, Size: 269 bytes --]

Please be more specific.  I just read the spec and, within the
constraints of a different model (no errno), the Plan 9 routines
seem to handle overflow fine.  If you're asking why it doesn't
set errstr, well, it could and that's probably just an oversight.

-rob

[-- Attachment #2: Type: message/rfc822, Size: 2108 bytes --]

From: Saroj Mahapatra <saroj@optonline.net>
To: 9fans@cse.psu.edu
Subject: [9fans] no error checking in strtol and strtoul
Date: Mon, 24 Feb 2003 10:00:21 GMT
Message-ID: <6d3220b4.0302221730.6677da7@posting.google.com>

Hi,

Was the decision to leave out error checking (like ANSI C strtol and
strtoul do) deliberate? The first type of error (wrong syntax) is easy
to check (if ret value == 0 and rest == beginning of string), but not
handling overflow situation is unsatisfactory. I'd like to hear plan 9
implementators' comment on this.

Thanks,
Saroj Mahapatra