Newsbin 6.53 refuses to download fill posts

[Shine]

I'm trying to download fills for a previously broken post, but Newsbin simply skips downloading them.

Debug log:

Code: Select all

[12:35:50]  DEBUG HFC: File seems to have already been downloaded - hand it to autopar : [post title]
[12:35:50]  DEBUG AutoPARPlugin: Have a listing for all files but, not enough blocks to repair [file name]
[12:35:50]  DEBUG PostAssembly - Unable to get PAR Status - Getting largest PAR block: [file name]

I'm using "Download - bypass filters". Duplicate detection is off.
AutoPAR is off. Why does NB try to use it anyway?!
All PARs are downloaded and available.

So I'm stuck now. All parts are available, but Newsbin simply skips downloading. WTF?!
How can I download these files? Why does Newsbin overrule my decision to (re-)download these files?!

[Quade]

Newsbin won't re-download a file that's already sitting in download folder. You need to move the existing broken file out of the way.

[Shine]

That's not good. Especially since it appears to look only at the file name. I should be able to download files with the same file name to the same folder, like it was possible before, ie. when Newsbin appended "-(000x)" to the file name. Just think about generic file names like "info.txt". In fact, Newsbin should never ever overrule my decision to download anything at all (especially if I use the "bypass filter" option). Such wannabe-intelligent behavior is more harmful than useful, as can be clearly seen in the above example.

[Quade]

It looks at the filename and the first 600K or so of the file.

If the file was actually different, it would rename and download

[Shine]

OK, then that's the reason why some few files actually DID re-download.

Two questions:
- why does it still skip the file if I turn the duplicate detector off?
- why does it skip the file if I queue it using the "bypass filters" option? It was possible in earlier versions to re-download a file this way even if the duplicate detector was on.

It should always be possible to force re-downloading a file. There are multiple reasons why this can be necessary, for example fill posts or reposts. Also sometimes, downloading two different incomplete posts of one file results in enough blocks for repair, while each one on its own doesn't.

[Quade]

The answer is the same no matter how many questions you ask. If you're trying to replace files you already downloaded, move the damaged files out of the download folder OR you can download them to a different folder. That's how it works.

It doesn't have to be how it works in the future but, that's how it works now.

The code that's checking for the existence of the file isn't the duplicate detector, which only applies to the first 20-30K of the file. The duplicate detector works whether the file is in the download folder or not.