refactor code to avoid flake8 warning

NatLibFi · osma · Dec 20, 2024 · Nov 12, 2024 · Nov 12, 2024 · Nov 12, 2024
commit 66f577d5983e9fe8c7043d410d9186b6e4a3afef
diff --git a/annif/analyzer/estnltk.py b/annif/analyzer/estnltk.py
@@ -2,24 +2,30 @@
 
 from __future__ import annotations
 
+import importlib
+
 from . import analyzer
 
 
 class EstNLTKAnalyzer(analyzer.Analyzer):
     name = "estnltk"
 
+    @staticmethod
+    def is_available() -> bool:
+        # return True iff EstNLTK is installed
+        return importlib.util.find_spec("estnltk") is not None
+
     def __init__(self, param: str, **kwargs) -> None:
         self.param = param
         super().__init__(**kwargs)

    def tokenize_words(self, text: str, filter: bool = True) -> list[str]:
        import estnltk
 
         txt = estnltk.Text(text.strip())
         txt.tag_layer()
-        lemmas = [
+        return [
             lemma
-            for lemma in [l[0] for l in txt.lemma]
+            for lemma in [lemmas[0] for lemmas in txt.lemma]
             if (not filter or self.is_valid_token(lemma))
         ]
-        return lemmas
diff --git a/tests/test_analyzer_estnltk.py b/tests/test_analyzer_estnltk.py
@@ -3,18 +3,22 @@
 import pytest
 
 import annif.analyzer
+import annif.analyzer.estnltk
 
-estnltk = pytest.importorskip("estnltk")
+pytestmark = pytest.mark.skipif(
+    not annif.analyzer.estnltk.EstNLTKAnalyzer.is_available(),
+    reason="EstNLTK is required",
+)
 
 
 def test_estnltk_tokenize_words():
    analyzer = annif.analyzer.get_analyzer("estnltk")
    words = analyzer.tokenize_words(
        """
        Aga kõik juhtus iseenesest. Ka köögis oli kõik endine.
        """
    )
    assert words == [
        "aga",
        "kõik",
        "juhtuma",
@@ -27,14 +31,14 @@


 def test_estnltk_tokenize_words_no_filter():
    analyzer = annif.analyzer.get_analyzer("estnltk")
    words = analyzer.tokenize_words(
        """
        Aga kõik juhtus iseenesest. Ka köögis oli kõik endine.
        """,
        filter=False,
    )
    assert words == [
        "aga",
        "kõik",
        "juhtuma",