Add HUC Braille support

[Andre9642]

HUC Braille is another representation for undefined chars designed by @DrSooom. See the official documentation for more info.

Using this HUC-Braille converter, I'd like to implement this alternative in BrailleExtender.

In this scope, we need review the option — Show undefined characters as... —. I suggest:

• A new label: Representation of undefined characters
• A list box with the following choices:
  • Use braille table behavior (default)
  • Dots 1-8 (⣿)
  • Dots 1-6 (⠿)
  • Empty cell (⠀)
  • Question mark (depending output table)
  • Other dot patterns (to be filled in an edit box)
  • Hexadecimal, Liblouis style
  • Hexadecimal, HUC8
  • Hexadecimal, HUC6

Also I'd like to try to implement the possibility to — Use HUC8/HUC6 for braille input —. Maybe a special temporary mode for that (?).

Limitation: due to the end of Python 2 in January and other technical issues with Python 2 and Unicode, This feature will require Python 3.

Any though @DrSooom, @zstanecic?

[zstanecic]

Hi Andre! I discourage this feature, because it will break existing character definitions. It will be god, if a new opcode from Daniel will be merged to Ll. See the discussion from 640, if I recall it correctly from my heart… From: André-Abush Clause <notifications@github.com> Sent: Wednesday, November 27, 2019 12:24 AM To: Andre9642/BrailleExtender <BrailleExtender@noreply.github.com> Cc: zstanecic <zvonimirek222@yandex.com>; Mention <mention@noreply.github.com> Subject: [Andre9642/BrailleExtender] Add HUC Braille support (#36) HUC Braille is another representation for undefined chars designed by @DrSooom <https://github.com/DrSooom> . See the official documentation <https://danielmayr.at/huc/en.html> for more info. Using this HUC-Braille converter <https://github.com/Andre9642/HUC-braille-converter> , I'd like to implement this alternative in BrailleExtender. In this scope, we need review the option — Show undefined characters as... —. I suggest: * A new label: Representation of undefined characters * A list box with the following choices: * Use braille table behavior (default) * Dots 1-8 (⣿) * Dots 1-6 (⠿) * Empty cell (⠀) * Question mark (depending output table) * Other dot patterns (to be filled in an edit box) * Hexadecimal, Liblouis style * Hexadecimal, HUC8 * Hexadecimal, HUC6 Also I'd like to try to implement the possibility to — Use HUC8/HUC6 for braille input —. Maybe a special temporary mode for that (?). Limitation: due the end of Python 2 in January and other technical issues with Python 2 and Unicode, This feature will require Python 3. Any though @DrSooom <https://github.com/DrSooom> , @zstanecic <https://github.com/zstanecic> ? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#36?email_source=notifications&email_token=ACVCDE6GUG45HO45ECPR2ETQVWVYNA5CNFSM4JR7DYH2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4H4IJMNQ> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACVCDE2BL62F4FD5NCBRNGLQVWVYNANCNFSM4JR7DYHQ> . <https://github.com/notifications/beacon/ACVCDE6EDJPRBXWDE5HYDQ3QVWVYNA5CNFSM4JR7DYH2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4H4IJMNQ.gif>

[DrSooom]

@zstanecic: The implementation of the HUC Braille Tables into the BrailleExtender will and MUST be a different one than the way I describe here (= just including the complete tbl/tbi files). Due to @Andre9642's script it's (theoretically) already possible to limited the HUC8 and HUC6 definitions only to a limited code point range. Furthermore it could be also possible that @Andre9642 is able to merge existing braille tables with parts of the HUC Braille Tables by using a script. So the definitions in the original braille tables always win, as this process will no longer produce multiple definitions – which is currently the case with my HUC Braille Tables if you are completely including them into other braille tables.

See also: liblouis/liblouis#868 and liblouis/liblouis#869

[Andre9642]

@zstanecic In addition to Daniel's answer, I think I'll do as follows on the first attempt:

First, add an 'undefined' rule to the current table. E.g.:
undefined 78-12345678-78

Then, for each string to translate (i.e. each string sent to liblouis):

1. Retrieve the liblouis translation.
2. If the braille string contains at least one "⣀⣿⣀" occurrence, replace all these occurrences by their corresponding HUC representation (using the original string and — HUC-braille converter —).
3. Show the string on the braille display.

[DrSooom]

For historical reasons see also: Wayback Machine link for the English documentation and nvaccess/nvda#8702

[Andre9642]

@DrSooom Thank you for the clarification.
I also planned to implement the possibility to:

• Limit the HUC8 and HUC6 definitions only to a limited code point ranges.
• Customize prefix braille characters.