[TxMt] Re: Irrational folding behavior when syncing a file between two TextMates via .Mac

Jacob Rus jacobolus at gmail.com
Sat Jan 20 01:55:32 UTC 2007


Hi,

John Bergmayer <bergmayer at ...> writes:
> I keep a bunch of markdown-formatted text files on my iDisk.  I use
> folding extensively in them, like this:
> 
> 	Some words{
> 
> 	*  Text that is folded.
> 
> 	}
> 
> To the best of my knowledge, curly braces are the only
> way to fold in Markdown.

Yeah, that's because folding patterns aren't currently sophisticated enough to
allow folding for markdown lists, sections (between headings), quotations, etc.
 Hopefully this will be different in TextMate 2.0.  We'll have to see how Allan
decides to make folding work.  If it's based on scope, I hope we'll be able to
fold all of the above.

If you want to have some other folding mechanism, you could make the folding
pattern more complicated, but it would require mirrored beginning and ending
markers of some sort.

> Sometimes I've been opening a file and seeing that the folding goes
> from the top brace to some random point *inside* the folded text,
> instead of to the closing brace or anything that could possibly be
> construed as a folding marker.
> 
> I'm pretty sure that this is because when I create the file with one
> TextMate on my iMac, edit it while away on my Macbook, and then open
> up the changed file again on my iMac later (all the syncing and such
> has happened in the meantime), somehow the metadata  hasn't come along
> with it.  So when I open the file, I have the data on one but the
> saved folding info for the old version, still.  So the top brace folds
> to the line number of where the closing brace *used to be*.

Yeah, that sounds about right.  Somehow the extended attributes are getting
mixed up somewhere in your process.

> network iDIsk.  Either a ._something file is being skipped, or
> something is causing extended attributes to be lost.

Exactly right.

> Is anyone else familiar with this kind of folding weirdness?  Can any
> one think of a solution besides turning extended attributes off?  (For

No, it's a problem I've never encountered.  And no, I can't think of other
solutions, but that one should cause the foldings to just be ditched altogether,
so I'm not sure it's a great solution.

> example, if it is indeed that extended attributes are being lost, is
> there a way to just force TextMate to always use ._something files?
> Presumably, this would sync fine.)

Hmm, I don't think so.  You can make it not ever use `._*` files, but I don't
think you can force it to always use them.

-Jacob

PS: if there are no responses to this thread that adequately answer your
questions, please ask again when Allan is back in 3 weeks, or file a ticket at
macromates.com/ticket




More information about the textmate mailing list