My StringRegExp doesn't get it

[leuce]

G'day everyone

Please can someone just tell me what's wrong with my StringRegExp query here?

I'm querying an file downloaded from the Wikipedia. I want the script to find this:

nl.wikipedia.org/wiki/Lounge">Nederlands

and add this to the array:

Lounge

Here's my non-working code:

$result = StringRegExp($wikipage2, '(?:nl\.wikipedia\.org/wiki/)(.+)(?:">Nederlands)', 1)

Please tell me what wrong with it?

Thanks

Samuel

[Xenobiologist]

Hi,

try this pattern (?<=nl\.wikipedia\.org/wiki/).*(?=">Nederlands)

So long,

Mega

[leuce]

Try this pattern (?<=nl\.wikipedia\.org/wiki/).*(?=">Nederlands)

Thanks. I really don't understand it, though. The AutoIt beta help file does not mention "?<=" or "?=" at all. Where did you get those, and what do they mean?

[Xenobiologist]

Thanks. I really don't understand it, though. The AutoIt beta help file does not mention "?<=" or "?=" at all. Where did you get those, and what do they mean?

HI,

look at wiki or any perl explanation.

It is very simple.

(?<=nl\.wikipedia\.org/wiki/).*(?=">Nederlands)

This means:

(<=...) = Look that this is ahead of the search pattern

.* = the search pattern in this case anything

(?=...) = Look that this comes after the search pattern.

Keywords are : lookahead and lookbehind. (these are postive) they are also possible to create negative.

So long,

Mega

[leuce]

(?<=nl\.wikipedia\.org/wiki/).*(?=">Nederlands)

This means:

(<=...) = Look that this is ahead of the search pattern

.* = the search pattern in this case anything

(?=...) = Look that this comes after the search pattern.

Thanks. Well, AutoIt doesn't seem to recognise it. But I finally figured out the correct regex:

'(nl\.wikipedia\.org/wiki/)(.*)(">Nederlands)'

Don't ask me why, but it works. According to the AutoIt help file I have to specifically exclude groups by using (?: ... ), and in the example above I did not exclude any groups, and yet AutoIt only adds the middle group to the array. Weird, but as long as it works:

$result = StringRegExp('<li class="interwiki-nl"><a href="http://nl.wikipedia.org/wiki/Wetenschap">Nederlands</a></li>', '(nl\.wikipedia\.org/wiki/)(.+)(">Nederlands)', 1)
MsgBox (0, "", $result[1], 10)

The above MsgBox yields "Wetenschap", which was what I was hoping for.

Thanks for your help.

Samuel

[leuce]

Now for my next problem (or rather, the original one):

Here is the script which should work, but gives me an array error:

$pageurl = FileReadLine("URLs.txt", $i)
$wikipage = InetGet($pageurl, "foo.txt")
$wikipage2 = FileOpen ("foo.txt", 0)
$wikipage3 = FileRead ("foo.txt", 0)

Sleep ("2000")

$result = StringRegExp($wikipage3, '(nl\.wikipedia\.org/wiki/)(.*)(">Nederlands)', 1)
MsgBox (0, "", $result[1], 10)

In the above example, the file URLs.txt contains a URL of a page that definitely matches the regexp. I've checked it, and it does match it. So the problem seems to be that I can't get the script to recognise $wikipage3. You'll notice that I had added $wikipage2 and $wikipage3 later, because I initially thought I could pipe the InetGet contents directly into the StringRegExp... but that doesn't seem to be the case.

How can I tell AutoIt to do the StringRegExp on the page that was downloaded using InetGet?

Thanks

Samuel

[Xenobiologist]

Hi,

;Extract link

Global $str = '<li class="interwiki-nl"><a href="http://nl.wikipedia.org/wiki/Wetenschap">Nederlands</a></li>'

$link = StringRegExp($str, '(?<=nl\.wikipedia\.org/wiki/).*(?=">Nederlands)', 1)
MsgBox(64, 'Link', $link[0])

$result = StringRegExp($str, '(nl\.wikipedia\.org/wiki/)(.+)(">Nederlands)', 1)
MsgBox (64, "Link", $result[1], 10)

So long,

Mega

[leuce]

Global $str = '<li class="interwiki-nl"><a href="http://nl.wikipedia.org/wiki/Wetenschap">Nederlands</a></li>'

$link = StringRegExp($str, '(?<=nl\.wikipedia\.org/wiki/).*(?=">Nederlands)', 1)
MsgBox(64, 'Link', $link[0])oÝ÷ Ûú®¢×±çz£­*.®· {héÞyÛ-¢ë)äÓbr¬æjYr­ë®÷~íéî·«¡ø­zk"Ø^jºÚÉ©ÝÂ+a)Þ"wvÚ.±ébMp'!槲Ø^~*ì¶+^)íæ«­¬£*.z0¥êáj0«mç(®·¶Ì§µ¬p¢é]mëhvaÆ®¶­sdvÆö&Âb33c·7G"ÒfÆT÷VâgV÷C¶föòçGBgV÷C²Â

instead? My main hiccup now is to get AutoIt to do the regex search on the entire opened file.

I guess I could to FileReadLine one line at a time and do the regex search on that, but it will slow down the operation of the script tremendously.

Thanks again

Samuel

[Xenobiologist]

Hi,

soemthing like this:

$link = StringRegExp(FileRead(FileOpen('c:\test.txt', 0)), '(?<=nl\.wikipedia\.org/wiki/).*(?=">Nederlands)', 1)
If IsArray($link) Then MsgBox(64, "link", $link[0])

So long,

Mega