Re: texi2dvi: locale-dependent error in egrep [A-z]

[Martin von Gagern]

On 31.03.2010 09:47, Eli Zaretskii wrote:
> Does it work to use [a-Z] instead?

[a-Z] works in de_DE.utf8 but not in C.

> I think this has to do with a
> different collating order under UTF-8, which is aAbB...zZ.

aAbB...zZ would include a valid range a-Z. It seems to be more like
ab...zAB...Z, for reasons I cannot begin to fathom.

>> Of course I know that [A-z] includes the ASCII characters between Z
>> and a, namely  [\]^_`  which technically aren't allowed as DOS drive
>> letters

In locales other than "C", the new sequence [A-Za-z] might include even
stranger letters:

$ echo Ä | LC_ALL=de_DE.utf8 egrep '[A-Z]'
Ä

So if you want to exclude these, you'd have to go for LC_ALL=C. Dunno if
it's worth the trouble, though.

Greetings,
 Martin

signature.asc
Description: OpenPGP digital signature