book file name MUST BE EXACT?

Post any problems / bugs / issues that are Mylar-related in here.
Post Reply
Posts: 14
Joined: Tue May 06, 2014 6:56 pm

book file name MUST BE EXACT?

Post by ben »

maybe I've just had a bad batch of books lately but I feel like Mylar has been SUPER particular about the naming convention regarding whether it's the correct book or not.

Just recently, I downloaded a collection of GROO comics and the artists name,Sergio Aragonés, is in the title.
Mylar would not process the books until I replaced the english E with the E-acute.
I appreciate proper naming but I want the software to do the work, Not me.

Publishing years have also been an issue lately. the year in the file name may be 1 or 2 years off from what is listed in the mylar list but it won't process until it is corrected.

Again, I understand the need to match the year if books have had multiple series over the years with the same name but maybe the search could have a Plus/Minus settings ei. 1988 +/- 2 [1986-1990]

Is there a 'soft' search switch or the option to have mylar keep books in a queue for review and approval?

Still LOVE mylar!!
THank you!
User avatar
Site Admin
Posts: 2587
Joined: Sat Apr 20, 2013 3:43 pm

Re: book file name MUST BE EXACT?

Post by evilhero »

Nothing has changed in regards to the way Mylar parses - if anything it's been more accommodating over the course of the last few months while we're trying to deal with various naming conventions that have appeared.

Under normal circumstances, Mylar will be able to parse the non-alphanumeric character as long as the CV title matches it should match up. Obviously there are some discrepancies as with anything, but from some brief testing of other titles with accents in various languages they seem to be picked up just fine. If you have really decent mismatching between what CV has and what your filenames are - you can always use the Alternate Search Name field and add in the name that's in the filenames and Mylar would then be able to pick it up.

Like in this particular case you could have added Sergio Aragonés Groo (or whatever it is called) in the Alternate Search Name field, and then Sergio Arargones Groo (without the accent) in the Alternate File-Naming a post-process run (or manual metatag, etc) it would rename each file as per the Alternate File-Naming which might not even differentiate much from the original naming convention.

If you have some examples of the filenames that you were trying to get Mylar to recognise that might help as far as me trying to understand what the parser is doing exactly. Debug logs would be even better ;)

There is an option to Fuzzy the year within the series detail page for a particular series - this will match it to roughly a year of the published date for a particular issue. Going any more outside of that is asking for some trouble, albeit it might be something to look into but I'm really hesitant to just open up the fuzzy aspect in that regards due to matching issues that would arise (and I've seen them first-hand just when it's fuzzied the year by just one).

I'd start to question the source of the files you've received if the issue dates are wrong, there might be other aspects that are wrong as well. If you have issues that don't have the right issue year in the filename, it sounds more like a quality control problem on that side as far as the naming conventions from where you got the files from. Are these just files that you got from somewhere that weren't named appropriately - or tagged even ? Are you sure (I know right?), that the series you're trying to match to is the exact series that the files pertain to, and not some random / unknown ones ?

We're obviously open to suggestions about what to improve upon - so if there's something that you're looking for, and obviously we can do it without causing more problems, we'd probably go about doing it as long as the time permits. Hop on github and log an issue there but label it as an enhancement request and see what happens.

(I'm assuming in all of this you're using Mylar3, and not the python2 version of Mylar since that's not really supported any longer).
Post Reply