Reports language coverage given a set of unicode values
Smörgåsbord tests sets of unicode characters to determine what language support is covered.
This library deveoped as a tool for type designers only as an approximate guide for planning purposes. Regional variation in letterforms is left as a exercise for the designer, and detection isn’t a planned feature.
The Smorgasbord class inherits from UnicodeSet and supports the same features.
Supports Python 2.6 – 3.x
from smorgasbord import Smorgasbord >>> bord = Smorgasbord([97, "b", "c", u"ü", u"\u0660"]) Smorgasbord([u"a", u"c", u"b", u"\xfc", u"\u0660"]) # Reports are accessed though the "reports" dict using the language code >>> en = bord.reports["en"] # Basic information about the report's language is accessible >>> en.language.code "en" >>> en.language.name "English" >>> en.language.characters FrozenUnicodeSet([u"a", u"b", u"c", ...]) # Amount of coverage is availbe as float and string representations >>> en.coverage 0.057 >>> en.coverage.percentage u"5.7%" # Sets of glyphs can be accessed >>> en.covered FrozenUnicodeSet([u"a", u"b", u"c"]) >>> en.uncovered FrozenUnicodeSet([u"d", u"e", u"f", ...]) # Reports can also return a boolean value for completeness: >>> en.complete False >>> en.incomplete True
Each language is defined in a text file. Characters are separated by spaces, and lines starting with a # are ignored as comments. Line breaks can and should be used to wrap lines to 80 characters maximum.
For example, the en.txt definition for English:
# Language: English a b c d e f g h i j k l m n o p q r s t u v w x y z A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
The first line is a special comment that will be parsed as the language name. Other special comments may be added in the future, but for now only Language is supported.
Custom Language Files
Custom langauges files can be set at runtime by supplying a path to a folder.
Language files are searched for in each succesive folder, using the first matching file. Language files included with this package can be overriden by prepending a local folder; an en.txt language file in the local folder will be used in preference to the internal en.txt.
Alternatively, the language_paths array can be replaced to only use locally available language files:
Smorgasbord.language_paths = ["/my/path/to/language/files/dir"]
- Fork this repository
- Add your language file to the languages directory or modify an existing file
- Add you language to the list of supported languages in this README
- Commit the change with a brief description (e.g. created en.txt, added ñ to es.txt)
- Create a pull request
This is a quick list of features that will need to be added in the near future (and will probably comprise a 1.0 release).
- Lazily evaluate reports. Currently the library loads all language files when a Smorgasbord is initialized, which will get slow, fast. This should happen at the latest possible moment.
- Unicode ranges in language files. Adding support for unicode ranges will probably be necesary for languages with large character sets.
- Better language support. Probably starting with languages using the latin alphabet, and continue in subjective order of easiest to define and number of speakers.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, Size & Hash SHA256 Hash Help||File Type||Python Version||Upload Date|
(10.6 kB) Copy SHA256 Hash SHA256
|Egg||2.7||Sep 18, 2014|
(5.7 kB) Copy SHA256 Hash SHA256
|Source||None||Sep 18, 2014|