neilrose

I am a long time Max user and have avoided regexp for sometime.

I am attempting to write a patch that does the following:

1. generates a random search term that is derived from inputted text

4. searches the downloaded HTML for the short text descriptors and forms them into a message/symbol

5. these descriptors will then be spoken (using say~)

It is at point 4 that I am heavily stumbling.

I know I need to use the jit.str.regexp object and the problem I am facing is that the downloaded results are so flexible that my extremely limited knowledge of using jit.str.regexp makes it impossible.

I have extensively read many (too many) threads on this forum about parsing HTML - particularly the web scraping for images. The difference with the scraping for images patch is that it is possible to look for a format, the web address, with what I require there is no formatting pre or suffixing the text to anchor the expression.

Any assistance would be very gratefully received.

I attach the bones of stage 4 of the patch.

Hi all,
I am a long time Max user and have avoided regexp for sometime.
I am attempting to write a patch that does the following:
1. generates a random search term that is derived from inputted text
2. uses ask.com to search the internet
3. downloads the results from ask.com
4. searches the downloaded HTML for the short text descriptors and forms them into a message/symbol
5. these descriptors will then be spoken (using say~)

It is at point 4 that I am heavily stumbling.
I know I need to use the jit.str.regexp object and the problem I am facing is that the downloaded results are so flexible that my extremely limited knowledge of using jit.str.regexp makes it impossible.

I have extensively read many (too many) threads on this forum about parsing HTML - particularly the web scraping for images. The difference with the scraping for images patch is that it is possible to look for a format, the web address, with what I require there is no formatting pre or suffixing the text to anchor the expression. 

I attach the bones of stage 4 of the patch.


using-jit-str-regexp-to-find-text-in-dl-web-search-results

Please use copy compressed ; i couldn't open the patcher on my computer. It is very probable that the forum text editor replaces some characters eve in the codebox.

Hey,
Please use copy compressed ; i couldn't open the patcher on my computer. It is very probable that the forum text editor replaces some characters eve in the codebox.


In my limited knowledge of expression usage I tried this - [jit.str.regexp @re /abstract(.*?)div/] and this [jit.str.regexp @re "abstract more-link [[^]]+]"]

as each of the search results - within the web page - are prefixed on a line above with "" - the r in id="r0_a" counts up sequentially based on the number of results.

Been trying too much to very little results.

Hi VICHUG
apologies - have repasted the code,
hopefully it works.
In my limited knowledge of expression usage I tried this - [jit.str.regexp @re /abstract(.*?)div/] and this [jit.str.regexp @re "abstract more-link [[^]]+]"]
as each of the search results - within the web page - are prefixed on a line above with "" - the r in id="r0_a" counts up sequentially based on the number of results.
Been trying too much to very little results.
Thanks for taking the time.


hah, alas you've repasted it not compressed, and not even in a codebox... If the codebox didn't work, this method will most definitely break everything. See here : 

https://cycling74.com/forums/how-to-post-patches-to-the-forum/

I'll (or someone else) probably be able to help once we can open the patch, but before that !..

hah, alas you've repasted it not compressed, and not even in a codebox... If the codebox didn't work, this method will most definitely break everything. See here : https://cycling74.com/forums/how-to-post-patches-to-the-forum/
I'll (or someone else) probably be able to help once we can open the patch, but before that !..


Cripes! One day I will get there... Really sorry...

https://cycling74.com/forums/regexp-line-feed/

Felt like a step forward but to no avail.

What I would really like is to be able to get the text from to .

Really I only want the text itself but am confident about filtering out extraneous code bits after sorting. If that makes sense.

Cripes! One day I will get there... Really sorry...
So I have been reading this: https://cycling74.com/forums/regexp-line-feed/
Felt like a step forward but to no avail.
What I would really like is to be able to get the text from  to .
Really I only want the text itself but am confident about filtering out extraneous code bits after sorting. If that makes sense.


Also - so you don't have to mess about, unnecessarily, with generating search terms I put a message box in with a typical search term so you can click on that to make it go.

Using jit.str.regexp to find text in dl web search results