Sort of a repost from last week, but it was probably just lost within the
torrent of goto posts...
I finally got a bit of time away from work and whatnot to finally fix the
oddities with the browscap/ini parser.
A tarball with the patch and the new files for HEAD is available at
http://bugs.tutorbuddy.com/download.php/browscap.patch.tar.gz
It's basically just a rip of some of the code from Zend/zend_ini_(parser
scanner).* with a few modifications to prevent operator handling and that
sort of thing. It still uses the Zend ini code for pretty everything except
the actual parser and scanner.
I know this probably won't get into the 5.0.x branch, but I'd like to get it
into 5.1.0 if possible. I've only had a chance to test it out on linux, so
I have no idea if it builds on Windows or anything else for that matter (my
Win2k dev box was recently reappropriated at work so I can't test it at the
moment) so if somebody could take a look at that I'd appreciate it. I have
no idea how to manipulate the newer Windows builds, and have no Win32
compiler at the moment anyways...
If everything checks out, I'd like to commit it to HEAD sometime in the near
future...
thx
J
While anything would be better then the current solution I wonder is perhaps
going the INI way is not the best idea due to the potential confusion. The
fellow who supplies the ini files with browser information also has a csv
file with the very same data that is much easier to parse.
Ilia
Sort of a repost from last week, but it was probably just lost within the
torrent of goto posts...I finally got a bit of time away from work and whatnot to finally fix the
oddities with the browscap/ini parser.A tarball with the patch and the new files for HEAD is available at
http://bugs.tutorbuddy.com/download.php/browscap.patch.tar.gz
It's basically just a rip of some of the code from Zend/zend_ini_(parser
scanner).* with a few modifications to prevent operator handling and that
sort of thing. It still uses the Zend ini code for pretty everything except
the actual parser and scanner.I know this probably won't get into the 5.0.x branch, but I'd like to get
it into 5.1.0 if possible. I've only had a chance to test it out on linux,
so I have no idea if it builds on Windows or anything else for that matter
(my Win2k dev box was recently reappropriated at work so I can't test it at
the moment) so if somebody could take a look at that I'd appreciate it. I
have no idea how to manipulate the newer Windows builds, and have no Win32
compiler at the moment anyways...If everything checks out, I'd like to commit it to HEAD sometime in the
near future...thx
J
Ilia Alshanetsky wrote:
While anything would be better then the current solution I wonder is
perhaps going the INI way is not the best idea due to the potential
confusion. The fellow who supplies the ini files with browser information
also has a csv file with the very same data that is much easier to parse.Ilia
I would think that sticking with the current ini files would probably lead
to less confusion. Just from the perspective that there may be people using
customized ini files or even just to avoid the initial "why is browscap
broken, my ini file is in the right place" bug reports and such. I wouldn't
expect any sort of register_globals-level "wtf", but it would probably be
enough to be annoying...
This modified parser isn't all that complicated anyways. The grammar and
lexer are both pretty small now, with the extraneous parsing normally done
on PHP's own ini file stripped out.
J
At 07:10 10/08/2004, Jay Smith wrote:
Ilia Alshanetsky wrote:
While anything would be better then the current solution I wonder is
perhaps going the INI way is not the best idea due to the potential
confusion. The fellow who supplies the ini files with browser information
also has a csv file with the very same data that is much easier to parse.Ilia
I would think that sticking with the current ini files would probably lead
to less confusion.
I second that...
Zeev