Just a curiousity: what do you mean that PDL is not 64bit safe? I have been using it on a 64bit computer for years, is there something I should be worried about?
Joel On Thu, Nov 10, 2011 at 8:55 AM, Chris Marshall <[email protected]>wrote: > Some tips that I have found useful for git development: > > - Always start a development session with git pull > > - Keep fixes small and coherent (e.g., fix one problem > at a time), this makes each commit message good > documentation for the development process and it allows > backtracking for bugs to be isolated to a single commit > which makes it easier to fix things with git > > - Always test your fixes with a complete PDL build and > tests (including the configure unless you know the change > is not affected/does not affect the configure process) before > any commit or patch submission. That keeps the pdl git > basically working which makes it easier to develop for all. > > - I also do a final git pull just before commit so that any > merges that need to take place can happen before the > push or patch. > > As far as the printf format specs fixes, the %p sounds > good to me. One thing to look out for is format specs > that might be affected by 64bit issues. PDL is not > yet 64bit safe but that is planned. Thanks for helping > with the cleanup. > > --Chris > > On Thu, Nov 10, 2011 at 7:21 AM, David Mertens <[email protected]> > wrote: > > Hi Tim - > > There are lots of options depending on your familiarity with git. If you > > were a registered PDL developer, you would make your edits, commit, and > > push. If you were very familiar with git, you would make your edits, > commit, > > and tell us where we could pull your changes (i.e. you'd make a pull > > request). However, as you are not familiar with git, here is what I > suggest > > you do: > > > > Make your edits, save the files. Do *not* commit via git. > > git diff > fix-pointers.patch > > Attach the patch to an email. One of the porters will apply it in your > name. > > > > Preferably these sorts of communications would be sent to > > [email protected], but that's not a strong preference. We'll > take > > patches however we can get them. :-) > > David > > P.S. Aren't you one of the Chicago Perlmongers? > > On Thu, Nov 10, 2011 at 2:04 AM, Tim Haines <[email protected] > > > > wrote: > >> > >> Greetings, all. > >> > >> This bug report is some general cleanup of several printf format flags. > >> Currently, pointer values are being printed out like this > >> > >> int *i; > >> printf("0x%x",i); > >> > >> which generates a warning since %x wants an unsigned int instead of a > >> pointer. I propose changing this to > >> > >> int *i; > >> printf("%p",(void*)i); > >> > >> As best I can tell, the %p flag is ANSI compliant. Does anyone have any > >> objection to using it? > >> > >> Since this bug fix affects several files and I don't have any experience > >> using Git, is there a tutorial somewhere that outlines your preferred > method > >> of doing a check-in of multiple files and their associated comments? I > used > >> `git clone git://pdl.git.sourceforge.net/gitroot/pdl/pdl my_dir` to > fetch > >> the latest repository. > >> > >> Thanks. > >> > >> - Tim > >> > >> _______________________________________________ > >> Perldl mailing list > >> [email protected] > >> http://mailman.jach.hawaii.edu/mailman/listinfo/perldl > >> > > > > _______________________________________________ > > Perldl mailing list > > [email protected] > > http://mailman.jach.hawaii.edu/mailman/listinfo/perldl > > > > > > _______________________________________________ > Perldl mailing list > [email protected] > http://mailman.jach.hawaii.edu/mailman/listinfo/perldl >
_______________________________________________ Perldl mailing list [email protected] http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
