Stas Bekman wrote:
Geoffrey Young wrote:
hi
I've implemented apr_table_compress, which was recently added to the APR table API.
the only thing I'm unsure about is what our position is on APR changes - can we rely on the autogeneration code for this (and perhaps other recent additions), breaking back compat, or do we need some kind of wrapper to protect users with older httpd/apr versions?
since we don't have a wrapper, this is not a problem for the code generation, as it'll be just ignored.
ignored how? 2.0.38 compiles cleanly but then throws a runtime error when I try and use the function (which I guess that is how it always works when you call an undefined C function)
If we do have a wrapper then we need #if/#else/#end logic, which is not the case with this function.
ok.
So the only thing you need is to fix the test to run the sub-tests only if apache version is the minimal one that has this feature.
yeah, that was my thought after I posted. I'll try and figure out in which revision it was added and make the appropriate change via skip().
The pod is broken a bit, instead of empty lines you have lines full of spaces, which will break the pod parser. Time to fix/replace your editor?
+Compress the data in C<$table>.
probably add a comment of what does it mean, while you are at it. I'm not quite sure what does it do from reading this.
ok.
+When C<$flags> == C<APR_OVERLAP_TABLES_SET>, each key will be +set to the last value seen for that key.
should this be:
APR::OVERLAP_TABLES_SET?
and below?
well, the APR_OVERLAP_* stuff was already there - it didn't make sense to me but I figured you had a reason for it :) at any rate, I'll fix the whole doc for that while I'm at it.
--Geoff
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
