Stripping Characters from Strings

[4b0082 1]

I'm looking for a way to strip part of a string, but I'm not sure how to go about doing it. I want to remove everything before the 0, leaving me with only "0.93" as my result.

&t129boo;0.93

Is there a way to ignore part of a string?

[4b0082 1]

Nevermind, I think I found what I'm looking for using this! Should have just looked an extra couple of minutes before making a post, sorry.

#include <MsgBoxConstants.au3>

Local $sString = StringTrimRight("This is a sentence with whitespace.", 5) ; Remove the 5 rightmost characters from the string.
MsgBox($MB_SYSTEMMODAL, "", $sString)

[guinness 1,459]

Is there other scenarios where the character length might vary?

[4b0082 1]

Is there other scenarios where the character length might vary?

Not in this case, I'm pulling data from a JSON output and needed to delete a specific portion of one of the values.

Incidentally this might be related, so rather than making a new thread I'll post it here.

At the moment I'm looking at a problem like this this:

<div>
. . . a bunch of code . . .
<a href="http://somewebsite.com/123-some-random-numbers-and-text">Link Name</a>
. . . a bunch of code . . .
Item: 33 ;this is the line of code I was searching for in my string.
</div>

What I want to be able to do is from "33" read the code from right to left to find the closest instance of "<a," then copy the hyperlink into a variable. Is something like this possible?

[4b0082 1]

I just found out about StringReverse and it might help get me to my goal, it seems like the hyperlink is always a specific number of characters away, so I'm thinking I might be able to cut the characters left to the end of the hyperlink, then reverse my string and read it until the " appears, then re-reverse it and write it into a variable.

It seems like an ass-backwards way of handling the problem, but at least it's something.. Now, I just need to figure out how to get from theory to actually having the code work. 

[guinness 1,459]

I would use a regular expression, personally speaking.

[4b0082 1]

I would use a regular expression, personally speaking.

What do you mean? I'll be honest, I'm still very new to programming and am just taking shots in the dark to get my programs to function in some kind of expected way.

[guinness 1,459]

Ah, sorry. Perhaps grasp the basics and then move forward with complicated programming concepts.

Edited November 1, 2014 by guinness

[Melba23 2,733]

4b0082,

Regular Expressions are a very clever way of extracting small sections of data from larger sets when there is a suitable pattern to identify the required section. In this case you could do something like this:

#include <MsgBoxConstants.au3>

$sString = "<div>" & @CRLF & _
    '. . . a bunch of code . . .' & @CRLF & _
    '<a href="http://somewebsite.com/123-some-random-numbers-and-text">Link Name</a>' & @CRLF & _
    '. . . a bunch of code . . .' & @CRLF & _
    'Item: 33 ;this is the line of code I was searching for in my string.' & @CRLF & _
    '</div>'

$aExtract = StringRegExp($sString, '(?s)<div>.*href="(.*)".*Item:\s\d+.*<\/div>', 3)

MsgBox($MB_SYSTEMMODAL, "Extract", $aExtract[0])

which gets you the link.

The RegEx works like this:

(?s)          - Ignore line endings
<div>         - Look for '<div>'...
.*            - and then any number of characters until...
href="        - we find 'href="'.
(.*)          - Now capture the characters until...
"             - the next double-quote.
.*            - Then there will be other characters until...
Item:\s\d+    - we find 'Item: ' followed by one or more digits (you said this was important).
.*            - Then there will be other characters until...
<\/div>       - we find '</div>'

The "3" means put all matches into an array

As you can see RegExes are not for the faint-hearted - I rate them as one of the hardest things I have ever tried to learn and they still make my brain hurt. But they can be incredibly powerful and are well worth the effort to get at least a working knowledge of how they are constructed - the real gurus around here can do magic with them.

I always recommend this site as a good place to get started. But if you are a novice programmer I would suggest you concentrate on the basics for a while yet.

M23

[mikell 648]

Versatile way of Melba's

$var = "Item: 33"
$aExtract = StringRegExp($sString, '(?s)<div>.*href="(.*)".*\Q' & $var & '\E.*<\/div>', 3)

[4b0082 1]

<snip> 

I get the basic concept of how to use this function, but I'm not sure I'm going to get it to work in the way I need it to. The hyperlink I'm trying to extract is one of several dozen other links in an extracted InetRead.

Local $dData = InetRead("http://marketplace.com/") ; Read page source code
Local $sData = BinaryToString($dData) ; Convert source from binary to characters
Local $sScan = StringInStr($sData, "3.00") ; Find the number "3.00"

Basically I'm searching a dynamically changing marketplace for the price of $3.00, and once I find it I need to open the hyperlink relevant to the price. This could be sandwiched several items down the page, though. I'm not having any problems finding the price which currently works to trigger a message box letting me know something's been found at my designated threshold, but I want to be able to provide a direct link using the ShellExecute command.

The hyperlink does follow a specific pattern, but the problem is that the other 9 links on the page also follow that same pattern with the difference being the specific item name, price, and hyperlink, which is why I was looking to flip the source code so I could locate the nearest "<a".

I might not be understanding the functionality of StringRegExp though, so maybe someone can tell me if this would work:

StringRegExp ( "test", "pattern" [, flag = 0 [, offset = 1]] )

I assume offset is the line number in the case of my extract source code, so would it be possible to detect the line number using FileReadLine (or a similar command), then subtract the number of lines back to the nearest anchor tag and then use StringRegExp?

 

Ah, sorry. Perhaps grasp the basics and then move forward with complicated programming concepts.

 

I'm trying, but I also like to get my hands dirty with complicated concepts too because it helps me learn more quickly.

Edited November 1, 2014 by Melba23
Huge quotes removed

[Melba23 2,733]

4b0082,

When you reply, please use the "Reply to this topic" button at the top of the thread or the "Reply to this topic" editor at the bottom rather than the "Quote" button - we know what we wrote and it just pads the thread unnecessarily.

I suggest posting a representative data dump from that site with an explanation of which link from within it you require to extract and what makes it the target - we might then be able to work out a suitable pattern and get it working for you.

M23

[4b0082 1]

Right now I'm trying to avoid using the site url because I don't really want my script available to people who might have similar ideas for it, but if it comes down to being unable to find a solution, I'll give my code. Sorry, I know, frustrating. Anyway, I'm trying to play around with StringRegExp and I'm getting an error when I try to run the code..

==> Subscript used on non-accessible variable.:
MsgBox($MB_SYSTEMMODAL, "Extract", $aExtract[0])
MsgBox($MB_SYSTEMMODAL, "Extract", $aExtract^ ERROR
>Exit code: 1    Time: 4.675

#include <MsgBoxConstants.au3>
#RequireAdmin

$pCurrent = "4"
$dData = InetRead("http://marketplace.com")
$sData = BinaryToString($dData)
$aExtract = StringRegExp($sData, '(?s)<a>.*href="(.*)".*\Q&#36;' & $pCurrent & '..*<\/a>', 3)
MsgBox($MB_SYSTEMMODAL, "Extract", $aExtract[0])

Edited November 1, 2014 by 4b0082

[Melba23 2,733]

4b0082,

 

I'm trying to avoid using the site url

Then make a redacted copy of the data you get from the site and post that. All you need to do is make sure that the relevant parts (the links and the cost from what details you have already posted) are not too different from reality so that the pattern will still work. Otherwise it is going to be a bit difficult to help.

M23

[4b0082 1]

After tinkering with the code more, I still can't figure out why I'm getting an error..
This is working code from a test.au3 I've made:

#include <MsgBoxConstants.au3>
#include <StringConstants.au3>

$pCurrent = "4"
Local $aArray = StringRegExp('<a href="hello">a&#36;4.94</a>', '(?i)<a.*href="(.*?)".*>.*&#36;' & $pCurrent & '..*</a>', $STR_REGEXPARRAYFULLMATCH)
For $i = 0 To UBound($aArray) - 1
    MsgBox($MB_SYSTEMMODAL, "RegExp Test with Option 2 - " & $i, $aArray[$i])
 Next

So I can't figure out why replacing "'<a href="hello">a&#36;4.94</a>'" with a $variable is giving me an error.

I think it has something to do with InetRead being broken into different lines; I'm doing all the tests I can think of and nothing is yielding an output.

Does using BinaryToString that outputs a string with several lines affect my array in some way?
I'm going nuts trying to get this to run.

Edit: The more I mess with this the more I'm certain that it's because my string uses multiple lines. Can someone help me find a fix for this?

Edited November 1, 2014 by 4b0082

[mikell 648]

Regular expressions are tricky and need precision

If you don't post the concerned part of the source code as Melba asked you twice, it's nearly impossible to give a correct and appropriate answer

[4b0082 1]

Regular expressions are tricky and need precision

If you don't post the concerned part of the source code as Melba asked you twice, it's nearly impossible to give a correct and appropriate answer

I'm to the point where I feel like giving up, so here's my source. I did contact Melba after I was asked, but he was unable to help any further for personal reasons and I'm still hitting a lot of walls. I think I'd like to approach it from a different angle by collecting all of the prices and names at one time and then processing them, rather than processing them as they're being searched.

I'm still unable to get StringRegExp to work the way I'm expecting it to, though. I found a solution to converting the entire source code into one line (didn't seem to help), and I was able to start getting results by removing important chunks of my search parameters (but that's not going to cut it).

At this point I just need something that works so I can move on.

 

Edited November 1, 2014 by 4b0082

[Melba23 2,733]

4b0082,

As I mentioned when we exchanged PMs, I see no problem with what you are trying to do - the question is simply one of interaction with the content of a website and has nothing to do with gaming per se. So if anyone wants to help, please feel free to do so.

M23

[mikell 648]

Not sure about the links you want to get

Please try this

#include <Array.au3>

Local $data = BinaryToString(InetRead("http://steamcommunity.com/market/search?q=") )
$var = "0.03" 
$res = StringRegExp($data, '(?s)listing_row_link"\h*href="(.*?)".*?\Q' & $var & '\E\h*USD', 3)

_ArrayDisplay($res)

[4b0082 1]

For some reason I can't get the code you posted to display an array. However, after using a post you made in another thread and at the advice of Melba, and beating my head on my desk, I FINALLY got something to work! It almost makes those 6 hours of frustration worth it!

At this point, I'm wondering if there's a way I can break my array results into a multidimensional array (10 rows 3 columns) so the data is coupled with it's relevant information. It's probably not necessary to do what I'm trying to do, but I might try to export data to an Excel document down the road and it might be more useful to have it arranged in that way.

#include <Array.au3>

 $sData = BinaryToString(InetRead("http://steamcommunity.com/market/"))
 $aItems = StringRegExp($sData, '(?sU)row_link.*href="(.*)".*&#36;(\d+\.\d{2}).*item_name".*>(.*)<\/span>.*<\/a>', 3)

 _ArrayDisplay($aItems)

Calling multidimensional array results is as simple as $aItems[1][1] to output the information stored in that spot, correct?