On 4/5/07, Darren Garvey <lists.drrngrvy@googlemail.com> wrote:
Hi Cory,
On 05/04/07, Cory Nelson <phrosty@gmail.com > wrote:
On 4/5/07, Darren Garvey <lists.drrngrvy@googlemail.com> wrote:
[cc'ed from boost-users list]
Hello all,
This may seem off-topic, but I assure you it isn't. ;)
This is an initial probe for ideas about an API for [any] upcoming CGI library submission. Of particular interest: *should GET/POST variables be parsed by default? (not _strictly_ API, but relevant)
Perhaps parsing them on first use might give a perf boost, but I don't think it would make much difference.
The other part to this question is with regards to the stdin stream (ie. POST variables). This can be huge and there may be no need to parse it at all. So, do you parse GET vars and not POST vars by default (possibly breaking a program if there is a change of a form from GET to POST); parse nothing (making code that bit uglier) or parse everything (potentially dangerous)? Try a more complex way?
I think GET should be parsed immediately, and POST should be read in by the user. Maybe have an adapter class that can decode POST data if the user wants it like that?
[noted, agreed, snipped]
The reason I'm asking such an open-ended question is that I'm assuming lots of people here will have experience using CGI libraries (possibly in other languages, possibly your own) and may have specific ideas about what should or shouldn't move from them into a Boost.CGI library, if one were accepted.
I think a CGI library should also include transparent FastCGI support.
I agree that shipping FastCGI support is a must - I'm not sure making it completely transparent is the best option - but I agree in principle. This discussion is for a separate thread though, perhaps (I've been thinking of this for the summer of code (or just the summer...), so input is more than welcome)?
Well, normal CGI is already phased out by scripting langs, I imagine FastCGI will be the primary use of this library. Innards can be kept to other threads but I think not taking it into consideration while discussing the basics of the library would be a mistake. I do think it could be mostly transparent: the user should be able to setup a request handler and either give it to a FastCGI listener or attach it to stdin/stdout without code changes.
Something I would also like to see is async support, with an API
similar to asio.
By 'async support' you mean for input/output?
Yes, like an asio socket class, with begin_read/begin_write etc. This is one thing I always see lacking in current web development kits: so many of them spend time accessing databases etc, things that block and should be done in an async manner, and I bet suffer performance because of it. It would be nice for a change to have an API that allows such scalability when it is wanted and ease of use when it doesn't matter. This is not so important for regular CGI but in FastCGI it could be a great boon.
Thanks for the comments, Darren _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
-- Cory Nelson http://www.int64.org