Tagging Inconsistencies 4.0.1 (2014-02-27)

[Bex]

I just thought it was to add them? What do I have to do to make it work?

[ZvezdanD]

Well, I didn't tried with VB scripts, but VB classic (<= 6.0) had serious problem with Unicode characters. VB source files are ANSI files, even if VB supports Unicode with its functions. Here is one excellent tutorial about this problem (it is very long, if you have a patience to read it): http://www.cyberactivex.com/UnicodeTutorialVb.htm

I suppose you should use ChrW function, instead of directly entered characters. For example: instead of Русский, you should write ChrW$(1056) & ChrW$(1091) & ChrW$(1089) & ChrW$(1089) & ChrW$(1082) & ChrW$(1080) & ChrW$(1081).

[Bex]

Yes that was indeed a long tutorial.
Could you (or someone else) try to add some Cyrillic letters to the options and see if it works?

[ZvezdanD]

You don't believe me, right? I told you, VBS files are ANSI files, each character is only one byte long. You could not enter Unicode characters in VBS file directly.

Now, I see you have one more problem. Where are you storing those options? You know, INI file is also ANSI, not Unicode. Please, read mentioned tutorial, there is an explanation about that. This is one more reason why I am thinking about storing Magic Nodes in next major upgrade into XML or similar format, not in MM INI file.

[NLS]

so this is getting more complex

UNICODE is the way to go whether we want it or not (for the whole project)

[ZvezdanD]

I just tried to enter Cyrillic characters directly into VBS file, saved it as Unicode text file and start MM - I got an error message about that file. Also happens if I save VBS file as UTF-8 file.

Now a good news - ChrW is supported by VB scripts. I tried it and it works (on the next address you could find a list of unsupported Visual Basic keywords inside of VB scripts, but I think it is not very accurate: http://www.htmlgoodies.com/beyond/asp/v ... hp/3458631).

Anyway, I didn't tried to follow a logic of your source code, but I think that you should not enter all needed Greek, Cyrillic, Latin 2, ... characters directly into VBS or INI or whatever file. If you want more elegant and complete solution, instead of that you should programatically test Unicode ranges. Please, take a look on 22. chapter of mentioned tutorial named "Unicode Blocks". There you could find all character ranges for different character groups. For example, Cyrillic is from U+0400 to U+04FF. So, you could put testing into a loop for UCase(ChrW(i)) = ChrW(i) if you want to know if some character from that range is upper case.

[Bex]

Thanks for all the help!
Yes, going with Unicode ranges seems to be the way. I'll have a look when I'm done with the update of my duplicate script.

[NLS]

thank you in advance man

[chrisjj]

Just tried this - very nice - thanks!

Suggestions for the UI labelling

1 Refer to tracks as tracks rather than sometimes songs
2 Refer to the Date property as Date rather than Year

Can you explain what properties is SingleArtist in "SingleArtist <> AlbumArtist"? Some derivative of track Artist?

Also what should one get "Same Disc# & Track#"? 'Cos I get e.g.

Disc# Track# Title Artist
1 1 Bajo el cono Alfredo De Angelis
1 2 Como se muere de amor (pal... Alfredo De Angelis
1 3 Corazon de oro [CJJ introless] Francisco Canaro
1 4 La zandunga Francisco Canaro

Thanks.

[NLS]

Bex don't forget me eh?

[Bex]

@ chrisjj

Thanks!
1. Yes, it makes little sense to have inconsistencies(!) in this script. So I'll correct it in the next release.
2. I actually retrieve the year from the date and use it in the node. But perhaps it would make sense not to do that. We'll see what happens.

"Single Artist <> AlbumArtist"
Finds all single Artist Albums where the Artist is not the same as the Album Artist.

"Same Disc# & Track#"
Should return dups on Disc# and Track#. But I'll have to check to code since you get strange result here.

@NLS
No, but I haven't had any time to work with this script lately. Perhaps in a week or two...

[chrisjj]

> 2. I actually retrieve the year from the date and use it in the node. But
> perhaps it would make sense not to do that.

I think it would be clearer and more useful to use the name and date of the standard property.

> "Single Artist <> AlbumArtist" Finds all single Artist Albums where the
> Artist is not the same as the Album Artist.

"single Artist Albums" == Albums all tracks of which have the same Artist value??

> "Same Disc# & Track#"
> Should return dups on Disc# and Track#.

Ah, I get it.

> But I'll have to check to code since you get strange result here.

Bex, now I try again, I get zero, so please treat that report as false. Sorry.

BTW, I use <album>\disc>\<track>.<ext> and it would be good to have a node to verify an album folder has uniform Album and Album Artist.

[NLS]

ok man

[Bex]

"single Artist Albums" == Albums all tracks of which have the same Artist value??

Yes.

BTW, I use <album>\disc>\<track>.<ext> and it would be good to have a node to verify an album folder has uniform Album and Album Artist.

I've been thinking of such node since day one but it wasn't really possible to do in MM2. In MM3 it would be quite easy so I'll put it on my todo-list.

[chrisjj]

>> "single Artist Albums" == Albums all tracks of which have the same
>> Artist value??

> Yes.

Thanks. Then I think "Albums With... single Artist Albums <> AlbumsArtist" is not clear, nor actually accurate (e.g. with Artist="Gilbert and Sullivan"), and better would be e.g. "Albums With... uniform Artist <> Album Artist"?

>> a node to verify an album folder has uniform Album and Album Artist.[
> In MM3 it would be quite easy so I'll put it on my todo-list.

Great! Thanks, Bex.