Removing bad, overdump etc from dats

[ChilliBean]

Hi everyone, how do I remove certain set of files from the dats using clrmames set information tool?
I know you can put in b;o in brackets etc but it seems to either deselt everything or does nothing.

love the TOSEC.

[PandMonium]

Hi,

i don't know what clrmame set information tool is since i don't use the tool for a long time now, i have to look at it one of these days.
Excluding flags may be a bit hard using it. We want to provide a way to create filtered dats based on the complete ones, which dats are you trying to use and excluding which flags?

Bye

[Symmo]

hi Chilli bean

I looked into the set information thing had me interested what u meant.
Anyways that is a uneeded mask ,from what i take is when the scanner removes things that dont belong in the folder
for eg say u had .jpg's in there and u dont want them removed it will mask (hide) them from the uneeded scanner.
*[b*]* would be the right mask? seeing there is like b1 and b2 delete as well.. its not a rebuilder exclude mask. Just look in the clrmame docs in the folder for more info.

If u dont want bad roms or anything else

1.from a console in windows del *[b*]* (test first)
2.use this http://symmo.net/tosec/ListTool.zip listtool its old but very handy only supports old version dats but tosec ones are right
I did a test on tosec sega megadrive games and loaded the dat in clicked filter and used *[b*]* and appears to work.
But click the down arrow and the end of the input line its got a lot codes from roms already done :-) just find the one that works for u.

Heres a link to wildcards http://www.linfo.org/wildcard.html (gonna have a refresh course as well lol)

Well hope it helps u i tried :-)

[PandMonium]

Hi Symmo

Good post full of information, i guess it will serve for most of cases but i have something to point since i can see it catching some false positives.
First of all i'm not sure if the * can represent no chars so [b*] can catch [ b] cases too.
Second it will catch any more info fields that match that pattern, cases like Title (19xx)(Publisher)[boot] for example.
Third and the last i can remember for now, there are flags that are even harder to catch (and i'm only talking about dump info flags). For example the translation and trained flags are very similar and a "*[t*]*" to catch trained sets only will also catch translated ones as well as any more info field starting with [t*].

Don't know if it handles regular expressions but if so you have better (and more complex ) ways to deal with it and i can help with it too.
I've a parser to our naming convention but unfortunately don't have much time currently to create some tool to filter dats right now, maybe in a couple of weeks or months

Anyway good work and keep posting your new findings, they can help others too.

[Symmo]

Hi Mate

Didn't realize there was boot etc.. then [ b] then *[b1]* or *[b?]* <-? etc... would just be done in multiple goes.
And expression are more complex would love to learn how to use them better for they can do fullon matches.

bye

[PandMonium]

edited your post just to change

Code:

[b] to [ b]

since it means bold in bbcode

Yes you could always split it in multiple cases [b1] is never used but splitting it could work for the [ b] one, the other part is harder since it could always be a more info label.
If you could use regular expressions (i never tried so i don't know what is supported there), you could create more complex patterns and catch them better, like any parser would do. For example "\[b[0-9]* [A-z0-9]\]" would catch it better (i know it will catch [b0] and only accepts chars and numbers as bad dump descriptor but this is just an untested example that can be wrong )