Re: [frogs] PATCH for getting output-suffix to work (Trackers 714 and 404) |
[ Thread Index |
Date Index
| More lilynet.net/frogs Archives
]
- To: Ian Hulin <ian@xxxxxxxxxxxx>, Reinhold Kainhofer <reinhold@xxxxxxxxxxxxx>
- Subject: Re: [frogs] PATCH for getting output-suffix to work (Trackers 714 and 404)
- From: Carl Sorensen <c_sorensen@xxxxxxx>
- Date: Thu, 13 Aug 2009 10:02:46 -0600
- Accept-language: en-US
- Acceptlanguage: en-US
- Cc: "frogs@xxxxxxxxxxx" <frogs@xxxxxxxxxxx>
- Thread-index: AcocC2iniT0RqgnuRQe8yqtC/4aYeQAJBbCT
- Thread-topic: [frogs] PATCH for getting output-suffix to work (Trackers 714 and 404)
Ian,
This patch can't be applied because it has lots of changes that are outside
of your scope.
It appears that you made your changes on an old version of master, then
created your patch against the current version.
Can you do the following:
If are working on the branch master in your local repository, do
git pull -r
This will update master, but still keep all of your local changes
separate and in the most recent commit.
Then make a new copy of the patch.
Also, please give one more try on the regexp. The hyphen in your [ ]
expression should have a \ before it so it will stand for a hyphen, instead
of a range marker.
I'd recommend that you use [^\w\-] as I think it's the simplest
non-alphanumeric and non-hyphen regex.
Thanks,
Carl
On 8/13/09 5:44 AM, "Ian Hulin" <ian@xxxxxxxxxxxx> wrote:
> Reinhold and Carl,
> I've implemented Carl's comments and also investigated Reinhold's feedback
> here. We can used Western European accented characters: acutes, graves,
> umlauts/diaereses on all vowels, scharfes s, Ash, Edh and thorn c cedilla and
> n tilde, but can't use the Extended Latin stuff like c hracek or A macron.
>
> I've spent too long trying to get regexp to work here and I can't find a
> satisfactory fix (sigh). I've marked the code with a FIXME comment block.
>
> I seem to have picked up some unrelated line changes from git again. The
> difference lines refer to split-at-predicate which I haven't touched. My
> patch should only have changes to print-book-with and the new get-outfile-name
> function. Carl, could yuo clean the patch up please? I don't know what to do
> here.
>
> I've also attached the test file I was using to demonstrate the eight-bit code
> problem.
>
> Cheers,
>
> Ian
>
> Reinhold Kainhofer wrote:
>>
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Am Mittwoch, 12. August 2009 14:32:37 schrieb Ian Hulin:
>>
>>
>>>
>>> Reinhold Kainhofer wrote:
>>>
>>>
>>>>
>>>>>
>>>>> + (if (string? output-suffix)
>>>>> + (set! result (format "~a-~a" base (string-regexp-substitute
>>>>> + "[^a-zA-Z0-9-]" "_"
>>>>> output-suffix))))
>>>>>
>>>>>
>>>>
>>>> This rules out any accented or other foreigh letters! E.g. all chinese
>>>> suffixes will simply turn into ______. Also, German instrument names like
>>>> "Fl���te" or "H���rner" would become "Fl_te" and "H_rner" in the suffix.
>>>>
>>>> I supose it would be better to use
>>>> "[^[:alnum:]-+]"
>>>> as the regexp to allow all alpha-numeric letters.
>>>>
>>>>
>>>
>>> Reinhold -
>>> There's a restriction with the gs back-end. Marek asked abourt this and
>>> Han-Wen confirmed that's why we're restricted to ujsing the ASCII set
>>> for filenames.
>>>
>>>
>>
>>
>> Hmm, that's strange, since using a file name like "������������������.ly"
>> works just
>> fine. What causes problems is multi-byte unicode characters (i.e. anything
>> outside the ANSI charset), which allows umlaute and other accented vowels....
>>
>> Cheers,
>> Reinhold
>>
>>
>> - --
>> - ------------------------------------------------------------------
>> Reinhold Kainhofer, reinhold@xxxxxxxxxxxxx, http://reinhold.kainhofer.com/
>> * Financial & Actuarial Math., Vienna Univ. of Technology, Austria
>> * http://www.fam.tuwien.ac.at/, DVR: 0005886
>> * LilyPond, Music typesetting, http://www.lilypond.org
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1.4.9 (GNU/Linux)
>>
>> iD8DBQFKgs/mTqjEwhXvPN0RAly0AJ9AonsFrzYTcuyqnTlLlyeRTKHrqgCgoEZn
>> OlBMs15FKdSspzMh+iaCwGY=
>> =oRBG
>> -----END PGP SIGNATURE-----
>>
>> ______________________________________________
>> This email has been scanned by Netintelligence
>> http://www.netintelligence.com/email
>>
>>
>>
>