fix: Fix error caused by running all tests at once

src/textdescriptives/components/__init__.py

@@ -4,3 +4,4 @@
 from .pos_proportions import POSProportions  # noqa: F401
 from .quality import Quality  # noqa: F401
 from .readability import Readability  # noqa: F401
+from .information_theory import InformationTheory  # noqa: F401

src/textdescriptives/extractors.py

@@ -72,11 +72,13 @@ def extract_dict(
         extracted_metrics["text"] = docs.text
     for component in metrics:
         if component == "quality":
-            extracted_metrics.update(__get_quality(docs))
+            metric = __get_quality(docs)
         elif component == "descriptive_stats":
-            extracted_metrics.update(__get_descriptive_stats_dict(docs))
+            metric = __get_descriptive_stats_dict(docs)
         else:
-            extracted_metrics.update(getattr(docs._, component))
+            metric = getattr(docs._, component)
+        if metric:
+            extracted_metrics.update(metric)
 
     return [extracted_metrics]
 

src/textdescriptives/load_components.py

@@ -28,29 +28,7 @@ def __call__(self, doc: Doc):
         return doc
 
 
-@Language.factory(
-    "textdescriptives/all",
-    assigns=[
-        "doc._.readability",
-        "doc._.dependency_distance",
-        "doc._.token_length",
-        "doc._.sentence_length",
-        "doc._.n_sentences",
-        "doc._.n_tokens",
-        "doc._.descriptive_stats",
-        "doc._.pos_proportions",
-        "doc._.quality",
-        "doc._.syllables",
-        "doc._.counts",
-        "doc._.coherence",
-        "doc._.first_order_coherence_values",
-        "doc._.second_order_coherence_values",
-        "doc._.passed_quality_check",
-        "doc._._n_sentences",
-        "doc._._n_tokens",
-        "doc._._n_syllables",
-    ],  # intentionally not assigning span attributes
-)
+@Language.factory("textdescriptives/all")
 def create_textdescriptives_component(nlp: Language, name: str):
     components = [
         k

src/textdescriptives/utils.py

@@ -8,6 +8,21 @@
 from wasabi import msg
 
 
+def get_valid_metrics() -> set:
+    """Get valid metrics for extractor.
+
+    Returns:
+        set: Set of valid metrics
+    """
+
+    # extract textdescriptive components from the list of spacy Language factory
+    return {
+        k.split("/")[1]
+        for k in Language.factories.keys()
+        if k.startswith("textdescriptives")
+    }
+
+
 def get_doc_assigns(metric: str) -> List[str]:
     """Get doc extension attributes for a given metric.
 
@@ -16,6 +31,15 @@ def get_doc_assigns(metric: str) -> List[str]:
     """
     # extract the assign names from the factory meta (this assumes that the doc._.
     # only includes elements which are also extracted as a part of the dataframe.
+    if metric == "all":
+        return [
+            col[6:]
+            for component in get_valid_metrics()
+            for col in Language.get_factory_meta(
+                f"textdescriptives/{component}",
+            ).assigns
+            if col.startswith("doc._.")
+        ]
     return [
         col[6:]
         for col in Language.get_factory_meta(f"textdescriptives/{metric}").assigns
@@ -49,21 +73,6 @@ def get_token_assigns(metric: str) -> List[str]:
     ]
 
 
-def get_valid_metrics() -> set:
-    """Get valid metrics for extractor.
-
-    Returns:
-        set: Set of valid metrics
-    """
-
-    # extract textdescriptive components from the list of spacy Language factory
-    return {
-        k.split("/")[1]
-        for k in Language.factories.keys()
-        if k.startswith("textdescriptives")
-    }
-
-
 def load_sms_data():
     """Load the sms dataset
     https://archive.ics.uci.edu/ml/datasets/SMS+Spam+Collection."""

tests/test_information.py

@@ -1,12 +1,14 @@
 import pytest
 import spacy
+
 import textdescriptives as td
 from textdescriptives.components.information_theory import (
     entropy_getter,
     per_word_perplexity_getter,
     perplexity_getter,
     set_lexeme_prob_table,
 )
+from textdescriptives.utils import _remove_textdescriptives_extensions
 
 
 @pytest.fixture
@@ -31,6 +33,7 @@ def test_unigram_information_metrics(nlp):  # noqa F811
 
 
 def test_extract_df(nlp):  # noqa F811
+    _remove_textdescriptives_extensions()
     nlp.add_pipe("textdescriptives/information_theory")
     doc = nlp("This is a very likely sentence.")