Thank you so much for taking the times to educate me; I've happily 
adopted all your suggestions.

> When you use a path in a shell script, you should always quote the path. In 
> AppleScript you can conveniently do that by using "quoted form of thePath". 
> 
> Also, there can be multiple linked files. So getting "POSIX path of linked 
> files" gives you a *list* of strings. So you should get something like "POSIX 
> path of linked file 1". And you should first check whether there are linked 
> files, and also whether the Local-Url isn't overwritten or unnecessarily set.

Ah. At first I had tried the quoted form (inspired by the script of 
Andreas who is careful to do that), but it hadn't worked, and I believe 
your second remark above explains exactly why. With both changes in 
place it works fine.

I don't intend to ever have an entry with more than one linked file. 
btw. To me this contradicts the whole spirit of bibliography: if we have 
two different documents, then they should each have its own entry.

On the other hand, I *do* often have several entries linking to the same 
file, for practical reasons. (E.g., I've got volumes of collected works, 
or Encyklopädie, from archive.org. Each has its own entry under the 
editor's or collectee's name; but I often have entries for individual 
articles. These link again to the whole volume -- since it's there.) I 
mention this because it's easy to do by manually in the Local-Url field, 
whereas... how would I do it under the new filing system?

> Yes, this is the decomposed S + caron and the combined S-with-caron 
> respectively. What you get is fairly unpredictable, different methods will 
> give different results, all kinds of processes can change the composition. 
> That's the problem with Unicode. I wouldn't really know how to normalize 
> that. I don't know why it would bite you, most methods you would use should 
> not make a difference between different decompositions.

Well, the reason we want these paths in plain text is that they are also 
being used on a Linux server (which mirrors the .bib and Archive). I can 
see how this sort of issue could bite, because it's the whole reason why 
Unison (file synchronizer) was not usable between Mac and Linux systems 
as soon as non-ASCII file names were involved -- until they found a way 
to normalize.

Don't get me wrong: I'm not complaining -- only trying to explain my use 
case and perspective here. I'm very happy to work with (or around) what 
I'm getting for free, and your program manages perhaps the most crucial 
file in my work.

Thanks again,
Francois

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure 
contains a definitive record of customers, application performance, 
security threats, fraudulent activity, and more. Splunk takes this 
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
Bibdesk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bibdesk-users

Reply via email to