Mango, DSource, and CGIHTML

[Morgan McDermott]

pragma wrote:
> Morgan McDermott wrote:
>> I've been experimenting around with D for a couple of days now, and 
>> I'm really starting to like it. And, although the available code-base 
>> is not as plentiful as for other languages (C), porting code to D from 
>> C is feasible.
>>  Right now I'm trying to get some basic features for a web-application 
>> designed in D. From the little I've read about the Mango project, it 
>> appears that people have already put a lot of time into generating 
>> useful code for D-web-applications, but sadly DSource.org seems to be 
>> down.
>>   I was unable to locate any mirrors, so I looked about for other 
>> options and found the CGIHTML project for C. At this point I'm not 
>> sure whether I should port CGIHTML to D, or find a way to get mango.. 
>> Any suggestions?
>>
>> Links:
>>     http://www.dsource.org/
>>     http://www.eekim.com/software/cgihtml/
> 
> I see that another D'er in this group hooked you up with Mango.  So I'll 
> offer my $0.02 instead. :)
> 
> FWIW, I've done some research on the same front (DSP and DDL).  Another 
> fellow was working on a FastCGI implementation, and I vaguely recall 
> someone else asking about composing an Apache module, here on the DNG. 
> There may be other undisclosed solutions out there... who knows.
> 
> You're dead on about the D codebase though.  What is clear to me is that 
> the critical missing piece are some serious man-hours spent testing the 
> various D libraries out there for their usefulness in making web 
> applications.  This goes well beyond Phobos and Mango and touches on 
> things like smart string handling, i18n, XML, XSLT, JSON, WDDX, SOAP and 
> so forth.  It's like this massive bowl of alphabet soup that's just 
> waiting for someone to dive in.
> 
> So regardless of how you, or any of us proceed, having a solution now is 
> key so that the community can take a crack at getting a D web toolkit 
> put together.  CGIHTML seems as good a solution as any to get us there, 
> so I say it's quite feasible.  Go for it.

  Hmm.. The CGIHTML port would be a quick-and-easy solution for me right 
now, so I may go in that direction, but I've been considering writing an 
Apache module for a while. I have the one book I can find on the subject 
  (O'Reilly's Writing Apache Modules with Perl and C), and I might turn 
to that solution simply because I was planning to use Apache for this 
project anyway, and Apache has neat features to hook into [ storing 
state information in web-server memory, etc].