temporal-normalization-spacy 2.1.0__tar.gz → 2.2.1__tar.gz

{temporal_normalization_spacy-2.1.0/temporal_normalization_spacy.egg-info → temporal_normalization_spacy-2.2.1}/PKG-INFO

@@ -1,29 +1,35 @@
 Metadata-Version: 2.4
 Name: temporal_normalization_spacy
-Version: 2.1.0
-Summary: A spaCy plugin for identifying and parsing historical data in Romanian texts
+Version: 2.2.1
+Summary: A spaCy plugin for temporal normalization and extraction of historical dates in Romanian narrative texts.
 Home-page: https://github.com/iliedorobat/timespan-normalization-spacy
 Author: Ilie Cristian Dorobat
+Keywords: spacy,nlp,temporal normalization,timex,historical dates,romanian
 Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
-Requires-Python: >=3.6
+Requires-Python: >=3.9,<3.13
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: spacy>=3.0
-Requires-Dist: py4j
-Requires-Dist: langdetect
+Requires-Dist: spacy<4.0.0,>=3.8.7
+Requires-Dist: py4j>=0.10.9.9
+Requires-Dist: langdetect>=1.0.9
 Dynamic: author
 Dynamic: classifier
 Dynamic: description
 Dynamic: description-content-type
 Dynamic: home-page
+Dynamic: keywords
 Dynamic: license-file
 Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# Temporal Expressions Normalization for spaCy (TeNs)
+# Temporal Expression Normalization for spaCy (TeNs)
 
 <b>Temporal Expressions Normalization spaCy (TeNs)</b> is a powerful pipeline component for spaCy that seamlessly
 identifies and parses date entities in text. It leverages the <b>[Temporal Expressions Normalization Framework](
@@ -75,8 +81,8 @@ quality and reliability of date-related information extracted from text.
 To integrate TeNs into spaCy pipelines you need the following:
 
 ### Prerequisites
-- Python 3.x
-- JRE 11+
+- Python 3.9-3.12
+- JRE 11-17
 - spaCy 3.x
 - py4j 0.10.9.9
 - langdetect 1.0.9
@@ -101,7 +107,6 @@ import subprocess
 import spacy
 
 from temporal_normalization.commons.print_utils import console
-from temporal_normalization.index import create_normalized_component, TemporalNormalization  # noqa: F401
 
 LANG = "ro"
 MODEL = "ro_core_news_sm"
@@ -121,7 +126,7 @@ try:
 except OSError:
     console.warning(f'Started downloading {MODEL}...')
     # Download the Romanian model if it wasn't already downloaded
-    subprocess.run(["python", "-m", "spacy", "download", MODEL])
+    subprocess.run(["python3", "-m", "spacy", "download", MODEL])
     # Load the spaCy model
     nlp = spacy.load(MODEL)
 

{temporal_normalization_spacy-2.1.0 → temporal_normalization_spacy-2.2.1}/README.md

@@ -1,4 +1,4 @@
-# Temporal Expressions Normalization for spaCy (TeNs)
+# Temporal Expression Normalization for spaCy (TeNs)
 
 <b>Temporal Expressions Normalization spaCy (TeNs)</b> is a powerful pipeline component for spaCy that seamlessly
 identifies and parses date entities in text. It leverages the <b>[Temporal Expressions Normalization Framework](
@@ -50,8 +50,8 @@ quality and reliability of date-related information extracted from text.
 To integrate TeNs into spaCy pipelines you need the following:
 
 ### Prerequisites
-- Python 3.x
-- JRE 11+
+- Python 3.9-3.12
+- JRE 11-17
 - spaCy 3.x
 - py4j 0.10.9.9
 - langdetect 1.0.9
@@ -76,7 +76,6 @@ import subprocess
 import spacy
 
 from temporal_normalization.commons.print_utils import console
-from temporal_normalization.index import create_normalized_component, TemporalNormalization  # noqa: F401
 
 LANG = "ro"
 MODEL = "ro_core_news_sm"
@@ -96,7 +95,7 @@ try:
 except OSError:
     console.warning(f'Started downloading {MODEL}...')
     # Download the Romanian model if it wasn't already downloaded
-    subprocess.run(["python", "-m", "spacy", "download", MODEL])
+    subprocess.run(["python3", "-m", "spacy", "download", MODEL])
     # Load the spaCy model
     nlp = spacy.load(MODEL)
 

temporal_normalization_spacy-2.2.1/setup.py

@@ -0,0 +1,38 @@
+from setuptools import setup, find_packages
+
+setup(
+    name="temporal_normalization_spacy",
+    version="2.2.1",
+    author="Ilie Cristian Dorobat",
+    description="A spaCy plugin for temporal normalization and extraction of "
+    "historical dates in Romanian narrative texts.",
+    keywords="spacy, nlp, temporal normalization, timex, historical dates, romanian",
+    long_description=open("README.md").read(),
+    long_description_content_type="text/markdown",
+    url="https://github.com/iliedorobat/timespan-normalization-spacy",
+    packages=find_packages(),
+    include_package_data=True,
+    entry_points={
+        "spacy_factories": [
+            "temporal_normalization=temporal_normalization_spacy.factory:create_component",
+        ],
+    },
+    package_data={
+        "temporal_normalization.libs": ["temporal-normalization-2.1.0.jar"],
+    },
+    install_requires=[
+        "spacy>=3.8.7,<4.0.0",
+        "py4j>=0.10.9.9",
+        "langdetect>=1.0.9"
+    ],
+    classifiers=[
+        "Programming Language :: Python :: 3",
+        "Programming Language :: Python :: 3.9",
+        "Programming Language :: Python :: 3.10",
+        "Programming Language :: Python :: 3.11",
+        "Programming Language :: Python :: 3.12",
+        "License :: OSI Approved :: MIT License",
+        "Operating System :: OS Independent",
+    ],
+    python_requires=">=3.9,<3.13",
+)

{temporal_normalization_spacy-2.1.0 → temporal_normalization_spacy-2.2.1}/temporal_normalization/commons/print_utils.py

@@ -47,10 +47,13 @@ class console:
 
     @staticmethod
     def lang_warning(query: str, target_lang: str):
-        if detect(query) != target_lang:
-            console.warning(
-                f'Detected language: "{detect(query)}" but required: "{target_lang}"'
-            )
+        try:
+            if detect(query) != target_lang:
+                console.warning(
+                    f'Detected language: "{detect(query)}" but required: "{target_lang}"'
+                )
+        except Exception as e:
+            console.warning(f'⚠️ Language could not be detected: {e}')
 
     @staticmethod
     def tokens_table(document):

{temporal_normalization_spacy-2.1.0 → temporal_normalization_spacy-2.2.1}/temporal_normalization/index.py

@@ -1,8 +1,11 @@
+import gc
 import re
 import subprocess
+import time
 from pathlib import Path
 
 from py4j.java_gateway import JavaGateway
+from py4j.protocol import Py4JNetworkError
 from spacy import Language
 from spacy.tokens import Doc, Span
 from spacy.tokens._retokenize import Retokenizer
@@ -48,6 +51,7 @@ class TemporalNormalization:
 
         Span.set_extension(TemporalNormalization.__FIELD, default=None, force=True)
         self.nlp = nlp
+        self.count = 0
 
         root_path = str(Path(__file__).resolve().parent.parent)
         java_process, gateway = start_conn(root_path)
@@ -68,15 +72,41 @@ class TemporalNormalization:
             Doc: The modified Doc object with temporal expressions processed.
         """
 
-        expressions: list[TemporalExpression] = extract_temporal_expressions(
-            self.gateway, doc.text
-        )
-        str_matches: list[str] = _prepare_str_patterns(expressions)
-        _retokenize(doc, str_matches, expressions)
+        self.count += 1
+
+        if self.count % 1000 == 0:
+            self.count = 1
+
+            gc.collect()
+            time.sleep(0.01)
+
+        try:
+            expressions: list[TemporalExpression] = extract_temporal_expressions(
+                self.gateway, doc.text
+            )
+            str_matches: list[str] = _prepare_str_patterns(expressions)
+            _retokenize(doc, str_matches, expressions)
+        except Py4JNetworkError as e:
+            print(f"⚠️ Py4J network error: {e}")
+        except Exception as e:
+            print(f"⚠️ Unexpected error during extract_temporal_expressions: {e}")
 
         return doc
 
     def __del__(self):
+        """
+        Clean up resources when the TemporalNormalization component is destroyed.
+
+        This method is automatically called by Python's garbage collector when the
+        `TemporalNormalization` instance is about to be deleted. It ensures that the
+        external Java process and the Py4J gateway connection used for temporal
+        expression extraction are properly closed.
+
+        By explicitly terminating the Java subprocess and shutting down the gateway,
+        the method prevents resource leaks such as orphaned Java processes or open
+        network sockets that might otherwise persist after the Python process ends.
+        """
+
         close_conn(self.java_process, self.gateway)
 
 
@@ -118,7 +148,6 @@ def _retokenize(
                                                 each with time series metadata.
     """
 
-    # TODO: WIP
     regex_matches: list[str] = [rf"{re.escape(item)}" for item in str_matches]
     pattern = f"({'|'.join(regex_matches)})"
     matches = (

{temporal_normalization_spacy-2.1.0 → temporal_normalization_spacy-2.2.1}/temporal_normalization/process/java_process.py

@@ -1,13 +1,17 @@
+import io
 import re
 import shutil
 import subprocess
+import threading
 
-from py4j.java_gateway import JavaGateway
+from py4j.java_gateway import JavaGateway, GatewayParameters, CallbackServerParameters
 from py4j.protocol import Py4JNetworkError
 
 from temporal_normalization.commons.print_utils import console
 
 
+gateway_started = threading.Event()
+
 def start_conn(root_path: str) -> tuple[subprocess.Popen, JavaGateway]:
     """
     Starts the Java temporal normalization process and establishes a Py4J gateway connection.
@@ -33,6 +37,11 @@ def start_conn(root_path: str) -> tuple[subprocess.Popen, JavaGateway]:
         f"{root_path}/temporal_normalization/libs/temporal-normalization-2.1.0.jar"
     )
 
+    def stdout_callback(line: str):
+        if "Gateway Server Started" in line:
+            gateway_started.set()
+        print(line.strip())
+
     java_process = subprocess.Popen(
         ["java", "-jar", jar_path, "--python"],
         stdout=subprocess.PIPE,
@@ -40,12 +49,18 @@ def start_conn(root_path: str) -> tuple[subprocess.Popen, JavaGateway]:
         text=True,
     )
 
-    for line in java_process.stdout:
-        if "Gateway Server Started." in line:
-            print(line.strip())
-            break
+    threading.Thread(target=drain_stream, args=(java_process.stdout, stdout_callback), daemon=True).start()
+    threading.Thread(target=drain_stream, args=(java_process.stderr,), daemon=True).start()
+
+    if not gateway_started.wait(timeout=10.0):
+        java_process.terminate()
+        raise RuntimeError("Java Gateway did not start within 10 seconds")
+
+    gateway = JavaGateway(
+        gateway_parameters=GatewayParameters(auto_convert=True, read_timeout=None),
+        callback_server_parameters=None,
+    )
 
-    gateway = JavaGateway()
     print("Python connection established.")
 
     return java_process, gateway
@@ -73,13 +88,45 @@ def close_conn(java_process: subprocess.Popen, gateway: JavaGateway) -> None:
     try:
         # Proper way to shut down Py4J
         gateway.shutdown()
-        print("Python connection closed.")
+        print("✅ Python connection closed.")
     except Py4JNetworkError:
-        print("Java process already shut down.")
+        print("⚠️ Java process already shut down.")
+    except Exception as e:
+        print(f"⚠️ Error shutting down gateway: {e}")
 
     # Terminate Java process
     java_process.terminate()
-    print("Java server is shutting down...")
+    java_process.wait()
+    print("✅ Java process terminated.")
+
+
+def drain_stream(stream: io.TextIOBase, callback=None) -> None:
+    """
+    Consumes the output from a given stream until a specific marker is found,
+    then closes the stream.
+
+    This function is typically used to monitor the stdout or stderr of a subprocess
+    (e.g., a Java process started from Python) and detect when a certain event occurs,
+    such as the initialization of a gateway server. Once the marker line is encountered,
+    the function prints it (or logs it) and terminates the stream reading.
+
+    Args:
+        stream (io.TextIOBase): A text-based stream object to read from, usually
+                                subprocess.stdout or subprocess.stderr.
+        callback (callable, optional): A function that takes a single string argument (line).
+                                       It will be called for every line read from the stream.
+                                       Useful for detecting specific markers, logging, or
+                                       triggering events when certain output appears.
+
+    Raises:
+        AttributeError: If the provided `stream` does not have `readline` or `close` methods.
+    """
+    for line in iter(stream.readline, ""):
+        line = line.strip()
+        if callback:
+            callback(line)
+
+    stream.close()
 
 
 def check_java_version() -> None:

{temporal_normalization_spacy-2.1.0 → temporal_normalization_spacy-2.2.1/temporal_normalization_spacy.egg-info}/PKG-INFO

@@ -1,29 +1,35 @@
 Metadata-Version: 2.4
 Name: temporal_normalization_spacy
-Version: 2.1.0
-Summary: A spaCy plugin for identifying and parsing historical data in Romanian texts
+Version: 2.2.1
+Summary: A spaCy plugin for temporal normalization and extraction of historical dates in Romanian narrative texts.
 Home-page: https://github.com/iliedorobat/timespan-normalization-spacy
 Author: Ilie Cristian Dorobat
+Keywords: spacy,nlp,temporal normalization,timex,historical dates,romanian
 Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
-Requires-Python: >=3.6
+Requires-Python: >=3.9,<3.13
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: spacy>=3.0
-Requires-Dist: py4j
-Requires-Dist: langdetect
+Requires-Dist: spacy<4.0.0,>=3.8.7
+Requires-Dist: py4j>=0.10.9.9
+Requires-Dist: langdetect>=1.0.9
 Dynamic: author
 Dynamic: classifier
 Dynamic: description
 Dynamic: description-content-type
 Dynamic: home-page
+Dynamic: keywords
 Dynamic: license-file
 Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# Temporal Expressions Normalization for spaCy (TeNs)
+# Temporal Expression Normalization for spaCy (TeNs)
 
 <b>Temporal Expressions Normalization spaCy (TeNs)</b> is a powerful pipeline component for spaCy that seamlessly
 identifies and parses date entities in text. It leverages the <b>[Temporal Expressions Normalization Framework](
@@ -75,8 +81,8 @@ quality and reliability of date-related information extracted from text.
 To integrate TeNs into spaCy pipelines you need the following:
 
 ### Prerequisites
-- Python 3.x
-- JRE 11+
+- Python 3.9-3.12
+- JRE 11-17
 - spaCy 3.x
 - py4j 0.10.9.9
 - langdetect 1.0.9
@@ -101,7 +107,6 @@ import subprocess
 import spacy
 
 from temporal_normalization.commons.print_utils import console
-from temporal_normalization.index import create_normalized_component, TemporalNormalization  # noqa: F401
 
 LANG = "ro"
 MODEL = "ro_core_news_sm"
@@ -121,7 +126,7 @@ try:
 except OSError:
     console.warning(f'Started downloading {MODEL}...')
     # Download the Romanian model if it wasn't already downloaded
-    subprocess.run(["python", "-m", "spacy", "download", MODEL])
+    subprocess.run(["python3", "-m", "spacy", "download", MODEL])
     # Load the spaCy model
     nlp = spacy.load(MODEL)
 

{temporal_normalization_spacy-2.1.0 → temporal_normalization_spacy-2.2.1}/temporal_normalization_spacy.egg-info/SOURCES.txt

@@ -14,5 +14,6 @@ temporal_normalization/process/java_process.py
 temporal_normalization_spacy.egg-info/PKG-INFO
 temporal_normalization_spacy.egg-info/SOURCES.txt
 temporal_normalization_spacy.egg-info/dependency_links.txt
+temporal_normalization_spacy.egg-info/entry_points.txt
 temporal_normalization_spacy.egg-info/requires.txt
 temporal_normalization_spacy.egg-info/top_level.txt

temporal_normalization_spacy-2.2.1/temporal_normalization_spacy.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[spacy_factories]
+temporal_normalization = temporal_normalization_spacy.factory:create_component

temporal_normalization_spacy-2.2.1/temporal_normalization_spacy.egg-info/requires.txt

@@ -0,0 +1,3 @@
+spacy<4.0.0,>=3.8.7
+py4j>=0.10.9.9
+langdetect>=1.0.9

temporal_normalization_spacy-2.1.0/setup.py

@@ -1,24 +0,0 @@
-from setuptools import setup, find_packages
-
-setup(
-    name="temporal_normalization_spacy",
-    version="2.1.0",
-    author="Ilie Cristian Dorobat",
-    description="A spaCy plugin for identifying and parsing historical data "
-    "in Romanian texts",
-    long_description=open("README.md").read(),
-    long_description_content_type="text/markdown",
-    url="https://github.com/iliedorobat/timespan-normalization-spacy",
-    packages=find_packages(),
-    include_package_data=True,
-    package_data={
-        "temporal_normalization.libs": ["temporal-normalization-2.1.0.jar"],
-    },
-    install_requires=["spacy>=3.0", "py4j", "langdetect"],
-    classifiers=[
-        "Programming Language :: Python :: 3",
-        "License :: OSI Approved :: MIT License",
-        "Operating System :: OS Independent",
-    ],
-    python_requires=">=3.6",
-)

temporal_normalization_spacy-2.1.0/temporal_normalization_spacy.egg-info/requires.txt

@@ -1,3 +0,0 @@
-spacy>=3.0
-py4j
-langdetect