Re: svn commit: r109832 -

2004-12-05 Thread Paul Querna
Joe Orton wrote: On Sun, Dec 05, 2004 at 10:35:11AM -0700, Paul Querna wrote: Jeff Trawick wrote: On Sun, 05 Dec 2004 00:01:05 -0700, Paul Querna <[EMAIL PROTECTED]> 1) write_full() has the wrong semantics for apr_file_writev(); apr_file_writev() and apr_file_write() are not supposed to block Than

Re: svn commit: r109832 -

2004-12-05 Thread Joe Orton
On Sun, Dec 05, 2004 at 10:35:11AM -0700, Paul Querna wrote: > Jeff Trawick wrote: > >On Sun, 05 Dec 2004 00:01:05 -0700, Paul Querna <[EMAIL PROTECTED]> > >1) write_full() has the wrong semantics for apr_file_writev(); > >apr_file_writev() and apr_file_write() are not supposed to block > > Thank

Re: svn commit: r109843 - /apr/apr/trunk/CHANGES /apr/apr/trunk/file_io/unix/fullrw.c /apr/apr/trunk/file_io/unix/readwrite.c /apr/apr/trunk/include/apr_file_io.h /apr/apr/trunk/test/Makefile.in /apr/

2004-12-05 Thread Joe Orton
On Sun, Dec 05, 2004 at 02:17:31AM -, Paul Querna wrote: > --- apr/apr/trunk/file_io/unix/fullrw.c (original) > +++ apr/apr/trunk/file_io/unix/fullrw.c Sat Dec 4 18:17:30 2004 > @@ -60,3 +60,37 @@ > > return status; > } > + > +APR_DECLARE(apr_status_t) apr_file_writev_full(

Re: svn commit: r109832 -

2004-12-05 Thread Garrett Rooney
Paul Querna wrote: 2) fix the code formatting Can this please be more specific? I reviewed the style guide at http://httpd.apache.org/dev/styleguide.html and I believe all of the code I have modified here is within those guidelines. I believe what he's referring to is the lack of spaces between

Re: svn commit: r109832 -

2004-12-05 Thread Paul Querna
Jeff Trawick wrote: On Sun, 05 Dec 2004 00:01:05 -0700, Paul Querna <[EMAIL PROTECTED]> wrote: Jeff Trawick wrote: > we're hosed here if apr_file_write() can only write part of the data (tbytes is < vec[i].iov_len and rv is APR_SUCCESS); don't you also need to bail if tbytes < vec[i].iov_len? Comm

Newbie: use in already-threaded application?

2004-12-05 Thread Kevin P. Fleming
I'm looking into using apr for an existing application, so that I can use its hash tables, pools, rings, atomic operations, etc. The concern I have is that the application is already multi-threaded (using pthreads) and I'm not ready to re-engineer the entire codebase... I need to start small an

Re: svn commit: r109832 -

2004-12-05 Thread Jeff Trawick
On Sun, 05 Dec 2004 00:01:05 -0700, Paul Querna <[EMAIL PROTECTED]> wrote: > Jeff Trawick wrote: > > we're hosed here if apr_file_write() can only write part of the data > > (tbytes is < vec[i].iov_len and rv is APR_SUCCESS); don't you also > > need to bail if tbytes < vec[i].iov_len? > > Commit

Re: apr_sha1

2004-12-05 Thread Joe Orton
On Sun, Dec 05, 2004 at 12:53:58PM +, Thom May wrote: > Guys, > any reason why apr_sha1_* are declared as void, and apr_md*_* are > apr_status_t? Any objections (I have patch plus tests ready to go) for > fixing this on HEAD? You don't say which you propose to change, but either way it's an AP

apr_sha1

2004-12-05 Thread Thom May
Guys, any reason why apr_sha1_* are declared as void, and apr_md*_* are apr_status_t? Any objections (I have patch plus tests ready to go) for fixing this on HEAD? -T

Re: svn commit: r109832 -

2004-12-05 Thread Paul Querna
Jeff Trawick wrote: > we're hosed here if apr_file_write() can only write part of the data (tbytes is < vec[i].iov_len and rv is APR_SUCCESS); don't you also need to bail if tbytes < vec[i].iov_len? Committed r109865 which uses apr_file_write_full() to hopefully write the entire iovec. According

Re: svn commit: r109832 - /apr/apr/trunk/CHANGES /apr/apr/trunk/file_io/unix/readwrite.c /apr/apr/trunk/test/Makefile.in /apr/apr/trunk/test/data /apr/apr/trunk/test/testfile.c

2004-12-05 Thread Jeff Trawick
On 4 Dec 2004 23:40:37 -, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: > Author: pquerna > Date: Sat Dec 4 15:40:37 2004 > New Revision: 109832 > > URL: http://svn.apache.org/viewcvs?view=rev&rev=109832 > Log: > * test/testfile.c: Add a test for apr_file_writev(). > * file_io/unix/readwrite.c