NZB Creation Bug?

[haileris]

I'm using 6.80RC7. I've bulk extracted a bunch of downloads using the 'Create NZB' option to load onto another newsbin instance - I called it notcomstar.nzb. When I import it onto the other computer all is well but every file created is called notcomstar.fileextension!

[Quade]

If the files you're downloading are obscured meaning "ABFKJKJSKS444....mkv" then Newsbin will rename them using the NZB Filename. Many post these days contain MKV's with just numbers and letters for the MKV name.

It sounds to me like you're downloading these obscured files.

[haileris]

But the filename is correct when I download them 'normally' as part of a NZB watch folder configuration In some circumstances the file name is obscured but the folder name is 'correct'. So the create NZB option should have the option to look at the download descriptions to use as the file subject name as opposed to the bulk name? Either that or export them individually.

[Quade]

The issue is that when you download them in the watch folder it's probably one NZB per set and the NZB Filename is probably related to the set you're downloading. When you're manually generating NZB's with a generic name, the NZB Filename matches that generic name.

The subject is never used to rename these files. Just the NZB Filename.

There is no "probe inside and use that filename" when you unrar. It either uses the filename of the actual files inside the rar or it uses the NZB Filename. You could try looking inside one of these RAR with WinRAR to see what the filename of the files inside the RAR is.

[haileris]

As I said, either the folder name or the file name will be correct if I download the file normally. In a lot of cases the file name is a garbage name, and therefore I assume the filename parameter I used to name the folder is taken from the NZB name. However, this hasn't happened in this bulk export/import scenario so either I've been very unlucky with the 180 odd files I've downloaded or there is behaviour difference.

Irrespective of this, what you are telling me is that I bulk create a NZB from a list of download files in a single file name I'm in danger of destroying the file naming opportunity. My take on the bulk create NZB feature is that in today's "obscured" world the function is useless. Who wants to have 180 odd media files all with the same name?

So my question still stands. Can for example you create multiple NZB files with the filename created from the download name? Or do I have to individually create a NZB and call it an understandable name manually...

[Quade]

Let me guess what you're trying to do because, you're not really telling me with enough detail to determine what you're trying to do.

1 -You're generating NZB's from the search tab which are automatically de-obscured by the search engine.

2 - You're feeding multiple sets into a single NZB on a generically named file?

3 - You then feed this NZB somehow into Newsbin?

4 - You're expecting the output files to be renamed based on the de-obscured subjects from the search engine?

I can tell you that that won't work. If you want the MKV files to be renamed, you should generate 1 NZB per set with the de-obscured part of the filename in the name of the NZB.

There is no mechanism to take the de-obscured name of the file from the subject and rename the output MKV from it. On the other hand, there is a mechanism to run a scripts post-unrar which could be used to do this rename.

Taking the de-obscured part of the subject and using it to rename obscured MKV's isn't a bad idea. I could probably look into it but there's no way to do it directly right now.

[haileris]

My first post told you what I was doing (but OK perhaps not clear enough):

"I'm using 6.80RC7. I've bulk extracted a bunch of downloads using the 'Create NZB' option to load onto another newsbin instance "

1. So no not search tab entries, downloads that already exist in the original newsbin download list and already have a correct name in the "download descriptions" field.
2. If I leave the downloads on the original newsbin instance then (typically) either the folder name or the file name will contain that "download descriptions" field contents
3. If I create a bulk NZB and use the standard newsbin NZB import facilties on my second machine, the filenames are normalised to a single name

So the point I'm trying (and I guess failing) to make is that newsbin has a valid file name to use as a data source at the point in time the export is generated. It assume it loses this data source / link during the createNZB file option.

I'm expecting a way (perhaps in the future) to use the data that exists in the first copy of newsbin to inform the export NZB process. Or allow the mass export with an option to individually export the NZB as a filename. Otherwise the mass export is pretty much useless now.

Does this make sense now?

[Quade]

So on the original PC, where are these files coming from? Headers or NZB? The download descriptions don't always match the final filenames.

[haileris]

NZB imports. I don't mind if the final filenames don't match the download descriptions, although that is obviously the preferred option. But it would be helpful if some element of the final path contained the 'download description' contents (d:\decode\$(filename)\filename or d:\decode\$(nzbfile)\filename for example), such that I have a known point to rename the actual filename.

[Quade]

NZB imports.

Right, which goes back to what I originally said.

When you download on the first system, the NZB Filenames are being used to rename the obscured files. When you use these NZB's to generate a NEW NZB with a generic name, you strip this filename information out. It's not included in the new NZB. Neither is the download or unrar path settings.

Why aren't you just using the source NZB's on the second machine instead of loading the NZB's on the first machine then generating a new NZB? Your current mechanism, using a single generic NZB to transport the information from multiple NZB's, is stripping the filename information out before you feed this NZB to the second machine. Basically it's not going to work for you if you want obscured MKV filenames to be renamed.

You could use "$(SUBJECT)" or "$(RAWSUBJECT)" in the unrar path on the second machine if all you care about is a pointer to the real filename,

You're better off copying the source NZB's to the second machine but how you choose to download is your business. I can only offer advice.

[haileris]

OK so I've written a post-processing script against SABNZB to do what I wanted to do... but FYI...

I know the export is stripping out required information; that much I've learnt from you.

I have asked twice below, is it not possible to code the bulk extract NZB option to write a series of individual files rather than one big one? Then you could call each individual NZB by the name in the download files column which would then give the NZB a proper name which could then be reused? You've not once replied to that suggestion - would that work or not?

Why am I not just using the source NZB's = because the second newsbin is a backup, I only want to try that if the first one fails to download. I have multiple servers, multiple news server accounts and multiple ISPs. I'd expected it to be easier to export the NZB failures from the download screen in bulk rather than to individually export them them or find each individual failed one.

[Quade]

have asked twice below, is it not possible to code the bulk extract NZB option to write a series of individual files rather than one big one? Then you could call each individual NZB by the name in the download files column which would then give the NZB a proper name which could then be reused? You've not once replied to that suggestion - would that work or not?

Anything is possible. It's just software. It's not going to happen any time soon though. 6.80 is just released.

One thought would be instead trying to make NZB's do something they're not particularly good at, you could copy the Downloads.db3 from Newsbin #1 to Newsbin #2. That would copy the entire download, failed and Wish list over. The only downside is that the pathing between the two machines would have to be the same. It's not that hard in Windows to have the same drive letter on different machines. For example my work drive is Z not matter what machine I'm on. You could make the download drive the same letter on both instances. You make make your download drive Y: on all machines.