On 08/01/2023 18:49, Holger Wansing wrote:
Roland Clobus <rclobus@rclobus.nl> wrote (Sat, 7 Jan 2023 11:31:29 +0100):I agree. But the work for the translators can be reduced by automatically parsing the work they have already done (i.e. the translations). That would leave only some characters that have not been used in any translated text, but might be present i.e. on the keyboard. @Translators: It that something you might want to use?For a translator, it's just a matter of 1 minute, to write down the alphabet for his language, I guess. But anyway, we can still use your script, if we cannot get such info from translator (if there is no active translator, for example).Attached: 1) chars_v2.py: The updated script 2) sample.summary: Its output to the console 3) ka.utf: Automatically generated Georgian list of use charactersThe ka.utf you attached is conspicuously (nearly) identical to what I grabbed from Wikipedia, so: good work!!!
Note the 'nearly'... :-)In the translations the following code points are used, which are not in your list yet:
10f2 (Georgian Letter Hie), used in 'retriever/media/loadnow' and 'partman-auto-raid/notenoughparts'
201c (Left Double Quotation Mark) 16x used 201d (Right Double Quotation Mark) 38x used 201e (Double Low-9 Quotation Mark) 56x used21b5 (Downwards Arrow With Corner Leftwards), used in 'partman-md/deleteverify' and 'mdcfg/deleteverify'
Note that 10f1 (Georgian Letter He) and 10f3-10fe are not used in any translated text.
The last four code points will be included by other languages, so it's probably not important to have them missing in ka.utf.
With kind regards, Roland ClobusPS: The strings with the missing letters were found be enabling lines 36 and 37 of the script.
Attachment:
OpenPGP_signature
Description: OpenPGP digital signature