19 replies to this topic

[thatguyandrew1992]

Hey guys,
I wanted to work on a new project just for fun, so I thought a program that downloaded imgur galleries would be cool. I am working on it, but the program will run through duplicates after 51 images usually.

I've looked at all i can and can't find the problem. My guess is that it's the server resetting me. I wanted a challenge so i tried to get past it. I tried creating a new url connection each time, I also created a seperate class that did the downloading and creating a new object of that class everytime but it didn't work.

Thoughts?

Source
http://shayconcepts....der/imgurDL.zip

[Aethec]

You are parsing the HTML page. Don't do that. Use imgur's API instead - from the documentation, you can just append ".xml" to any gallery page and it returns the gallery as an XML file.

In general, don't parse HTML pages unless there is no other way. It's not easy and it can change any day, making your program useless.

[+Asik]

Aethec, on 04 October 2012 - 20:30, said:

In general, don't parse HTML pages unless there is no other way.

Especially not with regex.

[thatguyandrew1992]

Aethec, on 04 October 2012 - 20:30, said:

You are parsing the HTML page. Don't do that. Use imgur's API instead - from the documentation, you can just append ".xml" to any gallery page and it returns the gallery as an XML file.

In general, don't parse HTML pages unless there is no other way. It's not easy and it can change any day, making your program useless.

Thanks! I'll take a look.

EDIT: it seems it only returns 56 images though

Dr_Asik, on 05 October 2012 - 01:51, said:

Especially not with regex.

I learned about regex and it has worked really well for me! I'll take a look at your link and find a better way to parse HTML. Thanks!

[+Asik]

thatguyandrew1992, on 05 October 2012 - 04:14, said:

I learned about regex and it has worked really well for me! I'll take a look at your link and find a better way to parse HTML. Thanks!

If you really need to parse HTML, you should use a proper HTML library; for Java there is HtmlUnit and Jsoup at least. Better would be to not parse any HTML and use imgur's API as Aethec suggested.

[thatguyandrew1992]

Dr_Asik, on 05 October 2012 - 06:05, said:

If you really need to parse HTML, you should use a proper HTML library; for Java there is HtmlUnit and Jsoup at least. Better would be to not parse any HTML and use imgur's API as Aethec suggested.

But when using the API, I append the .xml but it doesn't show ALL the images in the gallery

[Aethec]

To get other pages, append /page/XXX to an existing gallery URL. XXX is the page number. Don't forget the .xml, too.

[thatguyandrew1992]

Aethec, on 05 October 2012 - 14:05, said:

To get other pages, append /page/XXX to an existing gallery URL. XXX is the page number. Don't forget the .xml, too.

Thank you so much!! How did you find this? The API doc?

[ZakO]

thatguyandrew1992, on 05 October 2012 - 19:59, said:

Thank you so much!! How did you find this? The API doc?

http://api.imgur.com/gallery

[thatguyandrew1992]

ZakO, on 05 October 2012 - 20:02, said:

http://api.imgur.com/gallery

Thanks!

[thatguyandrew1992]

So Im using an XML parser and the /page/#.xml and it's working perfectly! Superb help! Thanks guys!

Now if a server disconnects me, how can I reestablish a connection, as if I had restarted my program.

[thatguyandrew1992]

Sometimes I get the java.net.ConnectException: Connection time out...

[ArialBlue]

try {
// le magic
} catch (Exception e)
{
// re-establish connection
}

[+Asik]

ArialBlue, on 09 October 2012 - 02:00, said:

try {
// le magic
} catch (Exception e)
{
// re-establish connection
}

catch (ConnectException e) ... always catch the specific exception you want, and never the base class of all exceptions.

[thatguyandrew1992]

If I catch that ConnectException, how do I reetasblish the connection?