Graham Dumpleton
graham.dumpleton at gmail.com
Tue Aug 19 07:43:39 EDT 2008
Changes tested and works as expected as far as allowing chunking to be disabled for HTTP/1.1 when no content length supplied for response. Fixes in 687024 of trunk: https://svn.apache.org/repos/asf/quetzalcoatl/mod_python/trunk Have fun. Graham 2008/8/19 Graham Dumpleton <graham.dumpleton at gmail.com>: > Haha, maybe I should scroll down. :-) > > You do setup the setattr after all. > > Graham > > 2008/8/19 Graham Dumpleton <graham.dumpleton at gmail.com>: >> 2008/8/19 Samuele Kaplun <Samuele.Kaplun at cern.ch>: >>> Dear Grahm, >>> >>> Il Tuesday 12 August 2008 12:17:37 Graham Dumpleton ha scritto: >>>> >>>> You would need the following to be implemented: >>>> >>>> https://issues.apache.org/jira/browse/MODPYTHON-238 >>>> >>>> If it is your own application on your own server, you could perhaps >>>> modify mod_python source code to address the issue. >>>> >>>> Graham >>> >>> pardon my ignorance, but do you mean that if I patch mod_python in order to >>> have a setter for req.chunked and req.connection.keepalive, I could be able >>> to disable chunked encoding even if content length is not specified, >> >> That is what my analysis in JIRA issues suggests. >> >>> thus being able to manually build multirange responses? >> >> Don't know about the multirange response side of things. :-) >> >>> I tried to patch mod_python, as you suggested, creating the patches I'm >>> attaching. For the time being I'm still encountering some errors, that I >>> still need to investigate (they are probably related to my implementation of >>> multirange). >>> >>> If you think my patches are enough for enabling writing support for chunked >>> and keeplive (thus fulfilling MODPYTHON-238 issue), I'd be glad if you would >>> like to accept them for integration into mod_python. >> >> Wouldn't you also need to change: >> >> {"chunked", (getter)getreq_recmbr, NULL, "Sending chunked >> transfer-coding", "chunked"}, >> >> So that setter, rather than being NULL is: >> >> (setter)setreq_recmbr >> >> The connection object hasn't previously had writable members so not >> setup same way. Thus not sure how to enable member as writable without >> trying it. >> >> Not on machine where can play with source code at moment, can only >> view it via web from repository. >> >> Do you changes actually allow you to set values, then read back to >> confirm changed? >> >> Graham >> >
|