Re: Re: Web development

[Sergey Dolin]

> > > So, maybe someone on this list knows if there is a way to store
> > > continuations?
> > You can (best way) use XML. O! It's realy the best way
> Surely there's nothing to be gained by converting to XML instead of
> S-expressions?
I don't know what S-expressions are. Are they  have DTD?
 XML - is "standard" aproch. Now i make all "external" output in XML, and
 parse it for "visualisation". So, while i'm choose ncurses now, i will
 not have much troubles with representing same data through web. 

> >  But my problems are others:
> >  i.Guile
> >  1)I want to return to each http-query  continuation the same user
> >  has been working before.
> Why is this a problem?
I dont know how to store "pointer to current continuatin"... i know it
is a shame for "real schemer", but i am as i am. 

> >  2)User can "start transaction" during wich hi is modifing some
> >  "transactional objects". In the moment he (or system) may decide to
> >  "rollback" all chandes. But only for "transactional" objects no more.
> Now I can see why this could be tricky, but surely using a private database
> connection from a pool for that continutation would solve it? 
The problem is to rollback "cached objects".
 I've collected some data from database in goops objects. Afterwards
 I have modified it and rollback. It is too expecive getting objects
 from database everytime isn't it? So we are having two copies of
 objects -- in guile and database and they must by synchronisied.

> But you say:
> >    I've solved second task by inventing my goops-db system which
> >    implement (define-transact-class) macro and metaclass which created
> >    classes with "transactional" slots. But where (and how) should old
> >    values of slots be saved? Especially arrays, lists and so on...
> although I probably don't understand this yet.  ;-)
I stucked with complex structures, tried to invent container class...
Remembed words of russian avia-ingeneer Tupolev: "Ugly plane will not
fly", counted time and money and came to conclusion i could not to
continue in this way. So i do what i can -- old good client-server
impoved with telnet and cacheFS having guile-web as my hobby only.
(now supressed by guile db2 interface)

> > I think so... I've looked for for OpenSource multithreaded http server,
> > knowing about russian encodings, having FastCGI and working on AIX.
> > Failed.
> Why is multithreading important?
To share data between http-queries. Even if you can to share your data
beetwen processes via mmap, you can't do that with status of database
connections: is transacton active for ex.

>Can you comment on the quality of serveez
> or the slib http functions? 
No. I've looked it quickly and decided it is that sort of tasks which
can be better solved by own libraries of eachother: minium functionality
but maximum details.
 You need set of html-generating functions? Make it youself and you will
 never find youself in troubles of spending time in hacking another library.

 If you need a high-level schemish tool to make out entire web site --
 look for HB, as it was advised in the list. It is a great tool.

 I was tring to invent a big application server and this is another task i
 have no enough brains for.

> MJR
And, sorry, but i have to give up with this thread -- I lack 
good thoughts and have almost no free time :) -- but will read your
answer eventualy.

-- 
Sergey
Mutt/1.2.5i*FVWM2*[FreeBSD 4.3-RELEASE]*(486DX2-50/20/730 Digial HiNote cs450)