[mwl]

Currently, the format for these package names is "4chan.org - [board name] - [date]". The date isn't particularly useful beyond ensuring JD puts each thread in a separate folder. To me, it would seem more useful to put the thread number in there. On the slower boards, this would mean you could re-add a thread at a later date and only get the newer images, rather than re-downloading them all. Benefits are a saving in bandwidth and a saving in time. Where any threads in the queue are likely to expire, the time saved may mean the difference between getting all the images and the thread and images 404ing before you get to them.

I couldn't find an option to change this in the options, so I took to the source. It looks like it comes from this bit of code in Brds4Chnrg.java:

                String domain = "4chan.org";
                String cat = br.getRegex("<div class=\"boardTitle\">/.{1,3}/ - (
.*?)</div>").getMatch(0);
                if (cat == null) cat = br.getRegex("<title>/b/ - (.*?)</title>")
.getMatch(0);
                if (cat != null) {
                    cat = cat.replace("&amp;", "&");
                } else {
                    cat = "Unknown Cat";
                }
                String date = new Date().toString();
                fp.setName(domain + " - " + cat + " - " + date);

The thread number for the image boards is after /res/ in the URL, and appears in the HTML like this:

<div class="thread" id="t12345678">

It also appears in the post form at the top, though this isn't shown if the thread is locked, so I wouldn't rely on it being there.

I'm happy to whip up a patch, though the prospect of making the change locally and getting it wiped out at each update is worrying. I suspect it's as simple as changing the last two lines above to:

                String threadno = br.getRegex("<div class=\"thread\" id=\"t([0-9]+)\">").getMatch(0);
                fp.setName(domain + " - " + cat + " - " + threadno);

What do you guys think?

[raztoki]

you will need a if and else statement with threadno isn't available otherwise presents a npe

[mwl]

It occurs to me that checking the URL may be faster than checking the entire HTML. This seems safe to me, since we can't get here unless the URL that matches the plugin conditions, the page doesn't 404, and the result contains at least one content image.

				// String suffix = br.getRegex("<div class=\"thread\" id=\"t([0-9]+)\">").getMatch(0);
				String suffix = new Regex(parameter, "/res/([0-9]+)").getMatch(0);
				if (suffix == null) {
					// Fall back to date if we can't resolve
					suffix = new Date().toString();
				}
				fp.setName(domain + " - " + cat + " - " + suffix);

For obvious reasons, it's difficult to put together a viable test case, so you'd have to actually visit the place to get URLs to test. It works for me for valid cases, though I've not really stress-tested any possible invalid ones, since I can't really see anything which could pass the checks to get to that block but not match the that URL regex.

[mwl]

After a bit of wrestling with the formatter, here's the change in patch form. Let me know if you want to do something with it - replacing the plug-in after JD over-writes it every time is getting tedious. As far as the IP goes, this fix is in the public domain and I assert no rights over it.

Index: src/jd/plugins/decrypter/Brds4Chnrg.java
===================================================================
--- src/jd/plugins/decrypter/Brds4Chnrg.java	(revision 20212)
+++ src/jd/plugins/decrypter/Brds4Chnrg.java	(working copy)
@@ -71,8 +71,13 @@
                 } else {
                     cat = "Unknown Cat";
                 }
-                String date = new Date().toString();
-                fp.setName(domain + " - " + cat + " - " + date);
+                // extract thread number from URL
+                String suffix = new Regex(parameter, "/res/([0-9]+)").getMatch(0);
+                if (suffix == null) {
+                    // Fall back to date if we can't resolve
+                    suffix = new Date().toString();
+                }
+                fp.setName(domain + " - " + cat + " - " + suffix);
                 for (String image : images) {
                     if (image.startsWith("/") && !image.startsWith("h")) image = image.replace("//", prot + "://");
                     DownloadLink dl = createDownloadlink(image);

[mwl]

No more word on this yet?

[pspzockerscene]

No...

GreeZ psp

[mwl]

That's a little unhelpful, especially given I've effectively done all the work for you (and even provided a patch ready to apply against the trunk). Is there anything else I can do to help you get this in?

[pspzockerscene]

Please just post the complete code and I'll apply it...

GreeZ psp

[mwl]

Originally Posted by pspzockerscene

Please just post the complete code and I'll apply it...

There's a svn diff attached to post #4. Team > Apply patch in Eclipse and you're done.

[pspzockerscene]

Please just give me the complete code as text, thx.

GreeZ psp

[mwl]

It's attached to post #4 ...

[tony2long]

Please provide some example links so we can make sure it works before commit.

[mwl]

These should be around for a few days at least, and shouldn't contain anything too objectionable (no guarantees though).
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
**External links are only visible to Support Staff****External links are only visible to Support Staff**
Shouldn't be any difference between SSL and non-SSL.

[tony2long]

It's done.