Searching String for Multiple Occurring

[bvr]

Basically I gather data from google trends rss all in one string. I tell it to gather everything from the "<ol" so it fetches the ordered list of the 10 trending keywords all on a different line. The problem is that is all one string. I want to be able to search the string for multiple occurring words or strings of words.

Of course there won't be multiple trends in the first hour, but I plan on gathering keywords every hour and comparing new keywords to old ones and finding the matches and separating them. It would be easy to "search" for keywords, the problem is I never know what the keyword is going to be, so I can't really specify what to search for. Could I search each line separate from the string, or make each line a sub string of a string, or use an array? Kind of confused on how to handle the data.

[JLogan3o13]

Hi, bvr. You're not the only one confused here Hard to give suggestions on how to perform your search when you state you don't know what you'll be searching for.

[bvr]

Local $Url = 'http://www.google.com/trends/hottrends/atom/hourly'
Local $Html = BinaryToString(InetRead($Url))


;Get the <ol> code from google
$Html = StringMid($Html,StringInStr($Html,'<ol'))


;Remove the javascripts
$Html = StringRegExpReplace($Html,'(?s)(?i)<script.+?script>','')


;Retrieve the visible text
Local $Text = StringRegExpReplace($Html,'(?s)<.+?>','')

;gets rid of the "]]>" thats left when we get visible text
$Text1 = StringReplace($Text, "]]>", " ")


;Display the visible text
FileWrite("keywords.txt",@CRLF&$Text1)

So after I pull the string of keywords, it saves is in a long string in a text file. I want to break that string up into 1 keyword per line, then be able to search through those keywords everytime I get the new keyword list, and place the multipules together to see which keywords are valuable. Then after I figure that out, I'm going to check competition for keywords and strings. But this is meant for SEO, I thought it would be fun to try after I found out how to pull data from the web.

Edited February 10, 2012 by bvr

[Blue_Drache]

Are the keywords delimited in any way? Like a comma ... or a space?

orisitallonebigfatlumpofrobotext?

Edited February 10, 2012 by Blue_Drache

[Spiff59]

StringSplit() gets you a nice array to play with. Am not sure I'm clear enough on your goal to suggest whether another array column would be useful for keeping counts of how many times the strings were encountered, or if you'd be wanting to merge newer scans into this array, etc.

#include <Array.au3>
Local $Url = 'http://www.google.com/trends/hottrends/atom/hourly'
Local $Html = BinaryToString(InetRead($Url))

;Get the <ol> code from google
$Html = StringMid($Html,StringInStr($Html,'<ol'))

;Remove the javascripts
$Html = StringRegExpReplace($Html,'(?s)(?i)<script.+?script>','')

;Retrieve the visible text
Local $Text = StringRegExpReplace($Html,'(?s)<.+?>','')
;gets rid of the "]]>" thats left when we get visible text
$Text1 = StringStripWS(StringReplace($Text, "]]>", ""), 3)
$aText = StringSplit($text1, @CRLF)

;Display the visible text
_ArrayDisplay($aText)
;FileWrite("keywords.txt",@CRLF&$Text1)

Edited February 10, 2012 by Spiff59

[bvr]

here is googles code: view-source:http://www.google.com/trends/hottrends/atom/hourly

The only way to tell the difference in keywords would be the classes. Some are medium, no change, low and so on. So maybe I could just grab the medium or low keywords and then search for occurring with new scans?