Rici Lake wrote:
> I was taking a look at the implementation of the renamed (but still
> misleading) AP_MODE_EATCRLF, in an attempt to figure out what a
> conforming input filter ought to do if it's handed that mode.

I think that really depends on your filter.  I have two strategies
there (setting aside simply ignoring the mode, which is wrong but
works if you know where the filter's being used).

* If the filter is happy to work with lines, get data from the next
  filter using the same mode and process it that way.
* Otherwise, assume the filter is unwanted, and remove itself from
  the filter chain.

> Some comments:

Agreed, with the proviso that it doesn't really complicate things that
much, as a filter can uninsert itself.

Maybe we should have a declarative version of this, in the manner of
mod_filter's protocol handling for output filters.  So that any input
filter passes an OR of input modes it works in to util_filter, and
never gets inserted when called in a mode it doesn't want to support?

-- 
Nick Kew

Reply via email to