Hrm, I'm getting the same error message you reported here: http://labs.divx.com/WebPlayerCodeGenerator. There may be a problem with the file you've posted.

Let's try a smaller video from DIVX.com and see if that works. Try uploading the video in the ZIP file and see if that works any different.

here is this test2.divx - http://libar-libar.com/libar/vb/show...=2405#post2405


works ... i hope ... it is few sec but work ...


my test.divx file is from here http://download.divx.com/divxlabs/Ap...20-12Mbps.divx
and i downloaded and just change name to test.divx and upload to my root!


reg

macc

edit: now is link for downloaded divx ok

Yup, that's working properly. Large DIVX files, at least in my experience, have a tendency to get corrupted easily. Try downloading the file again and see if you get the same error, because the modification appears to be working properly.

i tested this link http://download.divx.com/divxlabs/Ap...20-12Mbps.divx on codegenerator and works fine (mean divx file )


i will try download again and upload again for test ...

thanks and regards

macc

can you add supernovatube.com into the list ?

Hey there! (waves "hi" from Ireland)

I've just installed AME2.0 on my vB 3.7.4. Works just fine without the additional definitions.

However, just downloaded the master XML definition list, and on attempting to import it using the AME cp, I'm getting:

"XML Error: no error at line 0"

Any thoughts as to what this might mean?

Thanks! -- D.

I'll add it to the list of sites to include, provided there are no problems getting it to work.


I've not seen that before, but it could be the download got corrupted and you got an incomplete file. Try downloading it again.

DJ, may be i'm stupid, but what the hell does AME with ampersands in regular expressions? I need to define "&q=" pаttern in thе middle of URL. Somerthing like "(.+)(&q=)(.*)". But it trigger only when i set "(.+)(q=)(.*)". Tryed (.*)(&q=)(.*) with the same result.

What's wrong? Pattern is 100% correct, and test samples pass ok in preg debugger. I've rolled back to 2.0.2 but situation is the same. Any ideas?

Link example:
http://www.google.ru/search?hl=ru&newwindow=1&q=%D0%BF%D0%BE%D0%B8%D1%8 1%D0%BA+%D0%B2+%D0%B3%D1%83%D0 %B3%D0%BB%D0%B5&btnG=%D0%9F%D0%BE%D0%B8%D1%81%D0%B A&lr=&aq=f&oq=

http://www\.google\.[\w]{2,4}/search\?(.+)(q=[\w\d%+]+)(.*) - works, but catches wrong parameter (&aq=f)

http://www\.google\.[\w]{2,4}/search\?(.+)(&q=[\w\d%+]+)(.*) - doesn't works (brokes parceing and even shows plaint text instead of link).

Is this the link? I notice there's a space in the one you posted.

http://www.google.ru/search?hl=ru&newwindow=1&q=%D0%BF%D0%BE%D0%B8%D1%8 1%D0%BA+%D0%B2+%D0%B3%D1%83%D0%B3%D0%BB%D0%B5&btnG =%D0%9F%D0%BE%D0%B8%D1%81%D0%BA&lr=&aq=f&oq=

For ampersands, I always have to use the ampersand command, but then make it optional: &[amp;]* I never get the links to parse unless I change all my ampersands to this.

Also, I avoid using brackets if I can. [\w]+ will match any of Google's extensions without going past the backslash /search?. Brackets have a tendency to cause a parsing error, which is what the problem with IGN definition was.

Try this instead. I presume your trying to get three groupings?

DJ, thanks a lot.

Your example was not exact, but very usefull. I've waisted 8 hours without it for experiments :)

&[apms;]*(q=... can skip spoilers between "&" and "q" but this one seems to works as required:

(?:&|&)(q=...

Do I understand right, that preg must match BOTH link text and address in "URL"? Is it the reason, that pattern should allow both single "&" and encoded "&" ?

I've attached "letmegooglethatforyou" macros. Enjoy! type any URL with google search results and see what happens :) . Expression can be a bit improved, will do it later.

In nearest days will see, as promissed, what can i fo with google maps.