Add Kurdish Kurmanji language

spacy/lang/kmr/__init__.py

@@ -0,0 +1,12 @@
+from ...language import BaseDefaults, Language
+from .stop_words import STOP_WORDS
+
+class KurmanjiDefaults(BaseDefaults):
+    stop_words = STOP_WORDS
+
+
+class Kurmanji(Language):
+    lang = "kmr"
+    Defaults = KurmanjiDefaults
+
+__all__ = ["Kurmanji"]

spacy/lang/kmr/examples.py

@@ -0,0 +1,17 @@
+"""
+Example sentences to test spaCy and its language models.
+
+>>> from spacy.lang.kmr.examples import sentences
+>>> docs = nlp.pipe(sentences)
+"""
+
+sentences = [
+    "Berê mirovan her tim li geşedana pêşerojê ye",  # People's gaze is always on the development of the future
+    "Kawa Nemir di 14 salan de Ulysses wergerand Kurmancî.",  # Kawa Nemir translated Ulysses into Kurmanji in 14 years.
+    "Mem Ararat hunermendekî Kurd yê bi nav û deng e.",  # Mem Ararat is a famous Kurdish artist
+    "Firat Cewerî 40 sal e pirtûkên Kurdî dinivîsîne.",  # Firat Ceweri has been writing Kurdish books for 40 years
+    "Rojnamegerê ciwan nûçeyeke balkêş li ser rewşa aborî nivîsand",  # The young journalist wrote an interesting news article about the economic situation
+    "Sektora çandiniyê beşeke giring a belavkirina gaza serayê li seranserê cîhanê pêk tîne",  # The agricultural sector constitutes an important part of greenhouse gas emissions worldwide
+    "Xwendekarên jêhatî di pêşbaziya matematîkê de serkeftî bûn",  # Talented students succeeded in the mathematics competition
+    "Ji ber ji tunebûnê bavê min xwişkeke min nedan xwendin ew ji min re bû derd û kulek.",  # Because of poverty, my father didn't send my sister to school, which became a pain and sorrow for me
+]

spacy/lang/kmr/stop_words.py

@@ -0,0 +1,44 @@
+STOP_WORDS = set(
+    """
+û
+li
+bi
+di
+da
+de
+ji
+ku
+ew
+ez
+tu
+em
+hûn
+ew
+ev
+min
+te
+wî
+wê
+me
+we
+wan
+vê
+vî
+va
+çi
+kî
+kê
+çawa
+çima
+kengî
+li ku
+çend
+çiqas
+her
+hin
+gelek
+hemû
+kes
+tişt
+""".split()
+)

spacy/tests/lang/test_initialize.py

@@ -10,7 +10,7 @@ LANGUAGES = ["af", "am", "ar", "az", "bg", "bn", "ca", "cs", "da", "de", "el",
              "hr", "hu", "hy", "id", "is", "it", "kn", "ky", "lb", "lt", "lv",
              "mk", "ml", "mr", "nb", "ne", "nl", "pl", "pt", "ro", "ru", "sa",
              "si", "sk", "sl", "sq", "sr", "sv", "ta", "te", "ti", "tl", "tn",
-             "tr", "tt", "uk", "ur", "xx", "yo"]
+             "tr", "tt", "uk", "ur", "xx", "yo", "kmr"]
 # fmt: on
 
 

spacy/tests/tokenizer/test_explain.py

@@ -57,6 +57,7 @@ LANGUAGES = [
     pytest.param("tr", marks=pytest.mark.slow()),
     pytest.param("tt", marks=pytest.mark.slow()),
     pytest.param("ur", marks=pytest.mark.slow()),
+    pytest.param("kmr", marks=pytest.mark.slow()),
 ]
 
 

website/meta/languages.json

@@ -480,6 +480,12 @@
             ],
             "example": "这是一个用于示例的句子。",
             "has_examples": true
+        },
+        {
+            "code": "kmr",
+            "name": "Kurdish Kurmanji",
+            "example": "Ev hevokek e",
+            "has_examples": true
         }
     ],
     "licenses": [