I agree with Michael. Some of these warnings can be stopped with trivially small code modifications, and they do try to prevent human error. Consider this line (from my 3.5.9 amalgamation):
X1 = 365.25*(Y+4716); //X1 is an int Which produces the following warning in VC++2008: warning C4244: '=' : conversion from 'double' to 'int', possible loss of data This is useful. It's the compiler telling you "Watch out, a truncation is taking place here. Was that your original intention?" It can easily be suppressed by changing the line into: X1 = (int) (365.25*(Y+4716)); //Yes, truncation is taking place, that's how I want it. So why not fix this? I can understand (and even recommend ;) using _CRT_SECURE_NO_WARNINGS on VC++, but fixing these small ones won't hurt anyone, and makes the code slightly easier to read for a new comer. - Sherief > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:sqlite-users- > [EMAIL PROTECTED] On Behalf Of Michael Ruck > Sent: Tuesday, October 07, 2008 11:33 AM > To: 'General Discussion of SQLite Database' > Subject: Re: [sqlite] Compile SQLite3 for MS Windows Driver Kit user- > modeapplication > > Dan, > > While I do understand the position chosen by SQLite, these warnings do > try > to prevent human error. In case of Visual C++, Microsoft has added lots > of > tests in order to make porting code from 32-bits to 64-bits easier. And > those > problems Mark wrote about were certainly reported as warnings since > 2002 - > maybe even earlier. > > The point I'm trying to make is that generally turning off warnings is > a bad > idea. If you dislike a warning for "stylistic purposes", that's fine - > then > only turn of those warnings with appropriate pragmas. > > Generally hiding warnings is a mistake IMHO. > > Mike > > > -----Ursprüngliche Nachricht----- > > Von: [EMAIL PROTECTED] > > [mailto:[EMAIL PROTECTED] Im Auftrag von Dan > > Gesendet: Dienstag, 7. Oktober 2008 16:30 > > An: General Discussion of SQLite Database > > Betreff: Re: [sqlite] Compile SQLite3 for MS Windows Driver > > Kit user-modeapplication > > > > > The flood of warnings is a pain. SQLite dev claims they are all > > > spurious, but with so many I wouldn't venture to guess how they can > > > tell. > > > > http://www.sqlite.org/faq.html#q17 > > > > > > _______________________________________________ > > sqlite-users mailing list > > sqlite-users@sqlite.org > > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > > > > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
