httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tony Finch <...@dotat.at>
Subject Re: more in the 1.3.X byteranges saga
Date Tue, 16 Jan 2001 00:33:44 GMT
Bill Stoddard <bill@wstoddard.com> wrote:
>> Tony Finch <dot@dotat.at> wrote:
>> >
>> >Reviews and testing, please, so we can get the +1s :-)
>>
>> Pretty-please?
>
>Is it significant that r->range gets nuked by parse_byterange?  Perhaps parse
>on a copy of r->range?

Check the old code. Handlers call ap_each_byterange to get the details
of each range that must be returned to the client; ap_each_byterange
calls internal_byterange with r_range pointing to r->range;
internal_byterange calls ap_getword to get each comma-separated range;
ap_getword modifies r->range.

Thanks for bringing this up; I had treated r->range as being owned by
the byteranges code so I could do what I liked with it, but it turns
out that although the new parsing code works in a completely different
way r->range should AFAICT end up pointing to the same stuff.

The API of the new code should be exactly the same as the old code.
The support for returning 416 errors has been added in a backwardly-
compatible way -- wrowe came up with that cleverness.

Tony.
-- 
f.a.n.finch    fanf@covalent.net    dot@dotat.at
"Plan 9 deals with the resurrection of the dead."

Mime
View raw message