Smalltalk › Pharo › Pharo Smalltalk Users

Showcase of initial support for word clouds and data scrapping (and the usual newbie questions)

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

3 messages Options

Offray

Showcase of initial support for word clouds and data scrapping (and the usual newbie questions)

Hi all,

On [1] you can see a small video showing the support for word clouds and
data scrapping from Twitter for some public profiles (including mine).
As you can see the visualization is embedded on grafoscopio
trees/documents that provide context for the scripts and combine them
with other documentation. You can create your own visualization by
running the code at [2] and changing the profile name.

[1] https://archive.org/details/gfcp-wordclouds-initial.mp4
[2] http://ws.stfx.eu/DVG4C8VFON1L

So here are my usual questions:

a) The code that puts an image in a name cloud accepts only png images
and when the avatar has jpeg images I got an error. There is any way to
make the RTNameCloud to work with jpg or to convert it to png. I was
looking at Sven's documentation[3], but I can't make it work in my
context (jpeg is taken from the web and converted to png without using
the file system).

[3] https://medium.com/concerning-pharo/elegant-pharo-code-bb590f0856d0

b) There is any way to go beyond 19 tweets. In recent conversations with
Paul[4] he taught me how to extract tweets properly from a page using
ZnClient, but it gets the last 19. There is any way to make ZnClient to
go beyond that initial request and get, lets say 100 or more tweets?

[4]
http://forum.world.st/Data-scrapping-in-pharo-Extracting-tweets-contents-td4817746.html

Thanks as always for keeping Pharo/Moose and their communities awesome!

Cheers,

Offray

Nicolai Hess

Re: Showcase of initial support for word clouds and data scrapping (and the usual newbie questions)

2015-04-07 2:12 GMT+02:00 Offray Vladimir Luna Cárdenas <[hidden email]>:

Hi all,

On [1] you can see a small video showing the support for word clouds and data scrapping from Twitter for some public profiles (including mine). As you can see the visualization is embedded on grafoscopio trees/documents that provide context for the scripts and combine them with other documentation. You can create your own visualization by running the code at [2] and changing the profile name.

[1] https://archive.org/details/gfcp-wordclouds-initial.mp4
[2] http://ws.stfx.eu/DVG4C8VFON1L

So here are my usual questions:

a) The code that puts an image in a name cloud accepts only png images and when the avatar has jpeg images I got an error. There is any way to make the RTNameCloud to work with jpg or to convert it to png. I was looking at Sven's documentation[3], but I can't make it work in my context (jpeg is taken from the web and converted to png without using the file system).

Why do you need a png ? The R TNameCloud does not use a image on its own, its your code that adds a RTBitmap, and as far as I can see, the RTBitmap

only works with "Form".

[3] https://medium.com/concerning-pharo/elegant-pharo-code-bb590f0856d0

b) There is any way to go beyond 19 tweets. In recent conversations with Paul[4] he taught me how to extract tweets properly from a page using ZnClient, but it gets the last 19. There is any way to make ZnClient to go beyond that initial request and get, lets say 100 or more tweets?

[4] http://forum.world.st/Data-scrapping-in-pharo-Extracting-tweets-contents-td4817746.html

Thanks as always for keeping Pharo/Moose and their communities awesome!

Cheers,

Offray

Offray

Re: Showcase of initial support for word clouds and data scrapping (and the usual newbie questions)

Thanks Nicolai, you were right. The problem was not on the image format,
but on the parsing of the date. It was just that my non working test
were located at places where, coincidentally, the images were on jpg
format and I was misreading the signals :-). Now both, date
scrapping/formating and images in jpeg and png, are working properly.
Last version of Dataviz[1] reflect that changes.

[1] http://smalltalkhub.com/#!/~Offray/Dataviz

Cheers,

Offray

El 07/04/15 a las 03:18, Nicolai Hess escribió:

>
> 2015-04-07 2:12 GMT+02:00 Offray Vladimir Luna Cárdenas <[hidden email]
> <mailto:[hidden email]>>:
>
> Hi all,
>
> On [1] you can see a small video showing the support for word clouds and
> data scrapping from Twitter for some public profiles (including mine). As
> you can see the visualization is embedded on grafoscopio trees/documents
> that provide context for the scripts and combine them with other
> documentation. You can create your own visualization by running the code at
> [2] and changing the profile name.
>
> [1] https://archive.org/details/__gfcp-wordclouds-initial.mp4
> <https://archive.org/details/gfcp-wordclouds-initial.mp4>
> [2] http://ws.stfx.eu/DVG4C8VFON1L
>
> So here are my usual questions:
>
> a) The code that puts an image in a name cloud accepts only png images and
> when the avatar has jpeg images I got an error. There is any way to make the
> RTNameCloud to work with jpg or to convert it to png. I was looking at
> Sven's documentation[3], but I can't make it work in my context (jpeg is
> taken from the web and converted to png without using the file system).
>
>
> Why do you need a png ? The R TNameCloud does not use a image on its own, its
> your code that adds a RTBitmap, and as far as I can see, the RTBitmap
> only works with "Form".
>
>
> [3] https://medium.com/concerning-__pharo/elegant-pharo-code-__bb590f0856d0
> <https://medium.com/concerning-pharo/elegant-pharo-code-bb590f0856d0>
>
> b) There is any way to go beyond 19 tweets. In recent conversations with
> Paul[4] he taught me how to extract tweets properly from a page using
> ZnClient, but it gets the last 19. There is any way to make ZnClient to go
> beyond that initial request and get, lets say 100 or more tweets?
>
> [4]
> http://forum.world.st/Data-__scrapping-in-pharo-Extracting-__tweets-contents-td4817746.html
> <http://forum.world.st/Data-scrapping-in-pharo-Extracting-tweets-contents-td4817746.html>
>
> Thanks as always for keeping Pharo/Moose and their communities awesome!
>
> Cheers,
>
> Offray
>
>