Hey,

2011/7/8 Matthew Sherborne <[email protected]>:
> I don't think it'd work directly as there are so many different ways to use
> and configure Wt.

The idea is of course compelling, but one of the challenges is that
many people use Wt in entirely different ways. There is also the risk
of packaging boost with Wt but then getting into trouble when users
also use boost but a different version.

> What about adding them on as plugins ? But then, what would be the
> difference between using your system's graphics magick and your Wt graphicks
> magic plugin ? just that the Wt one would be tested against a certain
> version of Wt ? Or would it provide some kind of GM-WT binding features ..
> like  WImageWithGMOptions .. like a WImage, but you can 'resize and rotate'
> and all that stuff to generate new WImage Resource ?
> That might be a nice feature.

I agree there is indeed no value to bind a third party into Wt api
just for the sake of it. Only when the integration is beneficial
(feature wise, non-trivial integration, improved portability), could
such plugins make sense.

> Personally I'm using:
>  * yahoo grids for layout (CSS library)

I haven't actual experience with any of the CSS grid layout systems --
anything you can comment on how it is working out?

>  * mongo-db
>  * podofo for pdf parsing (not generation, actually parsing incoming ones)
>  * imagemagick (for image resizing)
>  * a bunch of boost libraries.
> Probably for me mongo-db is the one I'm binding closest to Wt, basically
> making my own gui-db layer for a lot of things. I can't use the DBO stuff
> because mongo has no SQL and is not Relational.

Actually ... I already have a local git branch "dbo-mongo". The idea
is to abstract away the Sql in Dbo so that it could also be used for a
document store (or other types of database).

MongoDB's features map quite well on Dbo API, but of course the query
language (in session.query() and session.find() will be different),
and some things do not apply to MongoDB, and probably some things from
MongoDB would not apply to Sql.

The mapping I had in mind would be to define a "root" class (or
perhaps more than one?) which corresponds to documents, and in which
related objects are stored (in JSON syntax).

I have not worked on it for a while though, and alot of work is needed
! But the result would be pretty awesome as an experiment (and as a
way to learn about MongoDb).

>> But what do you think about integrating into Wt all of these libraries
>> (and others in the future for other functions (like 7zip)) to be called
>> directly by Wt classes?!
>>
>> eg: #include <Wt/...> // WXml, WEmail, WDownload, WGraphic, WZip, etc...
>>
>> Thus will be much simpler to increase site functionality (and fast
>> development), without searching and installing several different types
>> of libraries (and there are many for each function).

On the other hand, well documented "How-to use XYZ library with Wt"
information with code examples might perhaps be just as valuable ?

Regards,
koen

------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security 
threats, fraudulent activity, and more. Splunk takes this data and makes 
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2d-c2
_______________________________________________
witty-interest mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/witty-interest

Reply via email to