Unofficial Arabic pipeline for the spaCy framework

About

Basic information on this release can be found in the README of the package https://github.com/gtoffoli/spacy-cameltokenizer, which constitutes a prerequisite, together with the CAMeL Tools library by CAMeL-Lab (https://github.com/CAMeL-Lab/camel_tools).

Further information on the the problems encountered and on the motivations of some choices can be found in the discussion https://github.com/explosion/spaCy/discussions/7146

Installation

I assume that you work in a Python "virtual environment" (venv), where possibly you already installed spaCy. You also need a local git directory to clone 2 packages from GitHub:

git clone https://github.com/gtoffoli/spacy-cameltokenizer.git
git clone https://github.com/gtoffoli/spacy-ar_core_news_md.git

In the site-packages directory of your venv, create 2 symbolic links:

• cameltokenizer, linking to the cameltokenizer sub-directory of the local spacy-cameltokenizer repository;
• ar_core_news_md, linking to the ar_core_news_md sub-directory of the local spacy-ar_core_news_md repository.

In the site-packages directory, create also the sub-directory ar_core_news_md-1.1.0.dist-info; in said sub-directory, copy the METADATA file from the top-level folder of the spacy-ar_core_news_md repository.

Finally, install spaCy (if needed) and the CAMeL Tools library:

pip install spacy
pip install camel-tools

spaCy customization

Replace 2 modules in the spacy/lang/ar subdirectory of the spaCy directory in site-packages, taking the new ones from the spacy_lang_ar_custom sub-directory of the local spacy-ar_core_news_md repository:

• __init__.py
• punctuation.py

Pipeline initialization

In a settings module of your applications (in my case it is the settings.py of a Django app), put the following code:

	import spacy
	from cameltokenizer import tokenizer

	ar = spacy.load('ar_core_news_md')
	cameltokenizer = tokenizer.CamelTokenizer(ar.vocab)

	@Language.component("cameltokenizer")
	def tokenizer_extra_step(doc):
		return cameltokenizer(doc)

	ar.add_pipe("cameltokenizer", name="cameltokenizer", first=True)