On Sun, Aug 06, 2006, Dmitry Morozovsky wrote: > On Sun, 6 Aug 2006, Ralf S. Engelschall wrote: > > RSE> > DM> It seems at least on amd64 size_t (strlen() result) is not int. > RSE> > > RSE> > Or, maybe, the following would be less ugly: > > RSE> > tdone_str = ctime(&tdone); > RSE> > + tdone_str[(strlen(tdone_str) - 1)] = '\0'; > > Well, next thought: ctime(3) described as POSIX.1 function having fixed length > of 26 chars. Is it safe and standards-compliant to save strlen(3) call and just > use tdone_str[24] = '\0' ? This was my first thought also after reading the ctime(3) manual page. But after quickly checking the sources it didn't looked to me that the string is always really exactly 26 characters only. I think the strlen(3) call is acceptable as this way we are really doing what we want: we strip the trailing newline. If we hard-code a "26" here we would just obfuscate the source even more as the reader certainly doesn't understand at the first spot from where the magic "26" comes from. Ralf S. Engelschall rse_at_engelschall.com www.engelschall.comReceived on Sun Aug 06 2006 - 08:16:03 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:58 UTC