UILAAT: Unicode Interlingual Aesthetic Appropriation Toolkit

A Mounaiban mini-project

What's This?

UILAAT is a Unicode text processing library for working with decorative Unicode text, artistic communication devices produced by appropriation of graphemes between languages and superfluous use of non-word symbols and combining characters.

UILAAT is a part of yet another chapter in the history of an Internet tradition that resulted from the encounter between an ambitious goal, to proliferate telegraphic typography support for every known (and unknown) written language and dingbat, and the human fascination with the foreign and exotic.

UILAAT is free software, licensed to you under the Terms of the GNU General Public License, Version 3 or later. Please view the LICENSE file for full terms and conditions.

Unicode is a registered trademark of Unicode, Inc.

Usage

Also: 𝔀𝓱𝓪𝓽 𝓬𝓪𝓷 𝔀𝓮 𝓭𝓸 𝔀𝓲𝓽𝓱 𝓽𝓱𝓲𝓼？

The software can be broken down into three main parts, and its usage into five main steps.

Parts

• Databases (DBs)
• Repositories
• Text Processors (TPs)

Steps

1. Prepare the Repositories
2. Prepare the Text Processors, and attach the Repositories to the TPs
3. Load the translations from the DBs in to the TPs
4. Prepare operations in the TPs
5. Apply the operations to input strings to make fancy text!

Demo and Quick Start

There is a demo module demos/demo.py which contains a ready-to-play Text Processor. To run the demo, run the following command from the repo root (same directory as the main module uilaat.py):

python -im demos.demo

You should end up in the Python interactive prompt. There should be a TP called demo.

A database and repository has been prepared for you. The demonstration TP has been linked to a built-in database trans via a repository named jr.

Loading Translations and Preparing Operations

In order to start making fancy text, the translations have to be loaded and queued into an operation list.

Bring up a list of available translations with the list_trans() method:

demo.list_trans()

The list will contain fully-qualified names which include the name of the database of the translation.

Load the translations like this:

demo.add_trans_dict('trans:wip-squares', n=0)
demo.add_trans_dict('trans:wip-squares', n=2) # alternate translation
demo.add_trans_dict('trans:ascii-aesthetic')
demo.add_trans_dict('ascii-tiny-capitals')    # non-FQ name

Non-fully qualified names may be safely used when using only one repository, or if there aren't multiple databases from different repos sharing the same name. The n argument selects alternate mappings where available.

PROTIP: translations that lack alternate mappings will produce the same output regardless of n index.

Confirm that the translations have been added like this:

>>> for d in demo.trans_dicts.keys():
...     print(d)
...
trans:wip-squares.0
trans:wip-squares.2
trans:ascii-aesthetic.0
trans:ascii-tiny-capitals.0

Note the a dot-and-number suffixes; these correspond to the alternate translation index specified by the n argument when the translation was added to the TP.

Define the default translation operation by queuing one or more dictionaries like this:

demo.add_trans_ops('trans:wip-squares.0')
demo.add_trans_ops(3)

Translation names must match those in demo.trans_dicts.keys(). Numerical indices are also accepted. The 3 is a reference to trans:ascii-tiny-capitals.0.

Verify the operation by checking the contents of trans_ops_list:

>>> demo.trans_ops_list
['trans:ascii-tiny-capitals.0, 'trans:wip-squares.0']

Finally, use translate() to generate some text:

>>> print(demo.translate("kitsune express"))
ᴋ⃞  ɪ⃞  ᴛ⃞  s⃞  u⃞  ɴ⃞  ᴇ⃞  　ᴇ⃞  x⃞  ᴘ⃞  ʀ⃞  ᴇ⃞  s⃞  s⃞
# print() formats the string to show wide spaces and squares

How's that for a start? 🦊

Removing Operations

Remove a translation from the default operation with pop_trans_ops():

>>> demo.pop_trans_ops('trans:wip-squares.0')
'trans:wip-squares.0'
>>> demo.translate("kitsune express")
'ᴋɪᴛsuɴᴇ ᴇxᴘʀᴇss'

Overriding Operations with the order Argument

The order argument references the dictionaries in trans_dicts directly, overriding the default translation:

>>> print(demo.translate("KITSUNE EXPRESS", order=[2,1]))
Ｋ⃞Ｉ⃞Ｔ⃞Ｓ⃞Ｕ⃞Ｎ⃞Ｅ⃞　Ｅ⃞Ｘ⃞Ｐ⃞Ｒ⃞Ｅ⃞Ｓ⃞Ｓ⃞
# remember the contents of demo.trans_dicts.keys()

In this example, trans:ascii-aesthetic.0 was applied to the input text, followed by trans:wip-squares.2.

Wrapping Up and Starting All Over

To start over, simply clear the translations from the TP with:

>>> demo.clear_trans()

PROTIP: To exit the Python shell, press CTRL-D on an empty command line.

Application Support and Previewing Your Text

Web Sites and Applications

Most web sites or apps provide their own fonts, so levels of support will vary.

Page inspectors featured in most desktop web browsers may be used as a previewing tool, by making transient edits to content currently displayed in the browser:

Chrome

1. Press CTRL + Shift + C/Cmd + Opt + C to bring up the Inspect option. The DevTools pane will also appear.
2. Select some text of the same type as what you are going to post or insert (such as a comment or caption) with the element picker. Further navigation in Elements in the DevTools may be needed to reach the text.
3. Once you find the text, simply double-click on it in Elements to edit it, then copy your own text over to preview it.
4. Close DevTools by clicking on the X on the far upper right hand side of the pane.

Firefox

1. Press CTRL/Cmd + Shift + C to bring up the Element Picker. The Toolbox pane will also appear.
2. Select some text of the same type as what you are going to post or insert (such as a comment or caption) with the element picker. Further navigation in the inspector may be needed to reach the text.
3. Once you find the text, simply double-click on it in the inspector to edit it, then copy your own text over.
4. Close the Toolbox by clicking on the X on the far upper right hand side of the pane, or by pressing CTRL + Shift + I/Cmd + Opt + I

On devices with limited processing speed, inspectors may be rather unresponsive on long and complex pages. Try testing on a shorter page where possible.

Some apps or sites may restrict the use of fancy text.

For other browsers, consult the relevant documentation for instructions on how to do this, where available.

Rationale

Also: w⃞  ʜ⃞  ʏ⃞  　ᴍ⃞  ᴀ⃞  ᴋ⃞  ᴇ⃞  　ᴀ⃞  ɴ⃞  o⃞  ᴛ⃞  ʜ⃞  ᴇ⃞  ʀ⃞  　ꜰ⃞  ᴀ⃞  ɴ⃞  c⃞  ʏ⃞  　ᴛ⃞  ᴇ⃞  x⃞  ᴛ⃞  　ʟ⃞  ɪ⃞  ʙ⃞  ʀ⃞  ᴀ⃞  ʀ⃞  ʏ⃞  ?⃞

This project was inspired by contemporaries such as Lunicode.js and LingoJam, yet attempts a different take on the art and science of text mangling:

1. A stronger emphasis is placed on the linguistic aspects of fancy text. This project attempts to curate the relationships between grapheme substitutes, and possibly spark public interest and appreciation of language studies.
2. The software in this project is intended for on-device use; this is not a web API or any other kind of internet RPC software service.
   • Any hacker is still welcome to redistribute this software as a service (and get rich doing so), as the software is free and open-source under GPLv3 T&Cs, but such use is beyond the scope of this project.
3. Amusement value is not the highest priority, but language nerds can still get plenty of amusement out of the project regardless.

TODO

We've got 🅱🅸🅶 🅰🅼🅱🅸🆃🅸🅾🅽🆂, but here are the ones that matter most now:

• More Preset Databases: more types of fancy text, and also cleanup translations to convert fancy text back to clear text (really helpful for users of screen readers and text archivists)
• Database Creation Tools: creating databases is a laborious endeavour but helper tools such as translation managers and visualisers could make things easier.
• User Apps: desktop and mobile apps for GTK, maybe Android and iOS, that generate text automatically on-device at the touch of a button. Ideas include:
  • clipboard monitors that automatically mangles copied text
  • input methods to generate fancy text as they are typed
• Multi-Code Point Targets: a low-cost method of supporting multi-code point targets in translations would be really nice, as these are currently only possible with computationally-expensive regular expressions.

Want More?

Additional information that don't quite fit here may be found in the Wiki or the docs directory of the source tree.