Some thoughts on duck typing

[Alejandro Tejada]

Hi David,
 

Yes - well to be accurate revamping it. I've been storing scripts and staks
in CVS / SVN and now GIT for the last 10 years. Originally I used a full XML
based file structure, but did not need it, so for years I have been storing
only the full stacks and the scripts (so I can diff the scripts).


Great! :-D
If I remember well, Andre is using Dropbox features for
a simplified Version control of his stacks:
https://www.dropbox.com/help/11

Certainly, compiling stacks from text sources would make
a lot easier to colaborate and test libraries, projects
and widgets.

When i had show this platform to developers in Delphi, C++
or .NET, some of the inevitable questions are:
"Could you work in your code without first
building an interface?"
"How could work a developer team's (at the same time)
in the same stack and scripts?"

If we could compile or rebuild stacks from text sources,
all these questions could have a simple and straight answer...


David Bovill-5 wrote:
> 
> With recent changes to engine so that you can set id properties of
> objects full text based stack representations are possible. As there are
> now
> a number of fairly large open source projects within the community with
> compatible licenses (revignitor / MetaCard / glxApp framework and GLX
> script
> editor), and I have a need to store the full behavior based widget
> components in a robust way - I'm putting in a bunch of work to release
> this
> for the conference.
> 

Excellent news! :-D


David Bovill-5 wrote:
> 
> We are starting to cover some of these topics on www.livecode.tv, and also
> in ChatRev, and there is an open invitation for anyone interested in
> developing community resources to take part, show work, and contribute
> code.
> I'm putting the finishing touches to the file format now. It is not XML
> based, it is array based with a default text and folder based export
> format.
> There is an optional XML version of the format for projects that would
> want
> an XML based version. The format is optimised for speed and for utility
> when
> working with revision control systems.
> 

This would be a landmarking moment in this platform.
A complete game changer.

Hopefully I would find enough time to learn
the framework and participate.


David Bovill-5 wrote:
> 
> Alejandro, I want to show and share the code for interfacing with
> WikiPedia some time. How did you get along with your offline WikiPedia
> project - maybe we could work together on a presentation - even if it is
> work-in-progress?
> 

The offline Wikipedia project is stored away (with some other ideas
and projects) in some backup DVD.

I would like to work on this Wikipedia Offline reader
as part of a team, but I do not have the resources to
create the required infrastructure for this project.

Hopefully, someone with enough resources and time
would open this project and then I could participate.

These were the functional, but not optimized, scripts
for this Wikipedia Offline Reader project,

1) The Gzip Compressed XML Database.
2) Scripts to create multiple Indexes
from this database
3) Scripts to search the Indexes
4) Scripts to convert and display the XML article
inside a text field
5) Scripts to convert the XML article to HTML
and display it inside a browser window
(using the DLL within Livecode)
6) Scripts to enable clicking a link inside the article
and open the linked article.
7) Scripts to keep an history of previous visited articles.
(and store them for faster access)

Probably there was more, but these are the
features that I remember well.

Alejandro

-- 
View this message in context: http://runtime-revolution.278305.n4.nabble.com/Some-thoughts-on-duck-typing-tp3213779p3216692.html
Sent from the Revolution - User mailing list archive at Nabble.com.