Hi.
I've just finished writing the dparser code.
It should be compatible with dtcl's, except:
<?= $variable ?> is the same as <? hputs $variable ?>
<?dtcl code ?> is the same as <? code ?>
The first thing is that it does NOT create Tcl code, it's only purpose is
to create a list of elements in the page.
Ie it does the following:
This is a <? code ?>, which might be <?=$useful ?> to <?dtcl dtcl::code ?>
It cuts it in a list of:
type prefix data suffix
str '' 'This is a ' ''
tcl '' ' code ' ''
str '' ', which might be ' ''
tcl 'hputs ' ' $useful ' ''
str '' ' to ' ''
tcl '' ' dtcl::code ' ''
eof '' '' ''
So basically this needs to be parsed to plain Tcl. That's what David wrote
in a perfect way - the code in test.c is a big mess and only exists for
testing purposes :).
I used such a description because I want to use the same code in
AOLserver's dtcl port.
If David finds it useful and fast enough (don't have a clue how fast/slow
this is - but I tried to make it quite as fast as possible :), I wish it
was included in the core. But after some small rewrites - but this one is
a '5 minute hack' I wrote this evening, I want to know if this might be
useful before finishing it up...
The URL is http://dtcl.zoro2.org/dparser-0.1.tar.gz
There's also a quick-hack testing code :>
The code was tested and should have no memory leaks and no bugs - after
all, this is just a small piece of code. I didn't have time to test binary
data, though. I didn't use any str*() function, but only mem*() ones, so
this code should work on binary files as well.
--
Wojtek Kocjan
[EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
