PyRuSH 1.0.11__tar.gz → 1.0.12.dev1__tar.gz

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyRuSH
-Version: 1.0.11
+Version: 1.0.12.dev1
 Summary: PyRuSH is the python implementation of RuSH (Rule-based sentence Segmenter using Hashing), which is originally developed using Java. RuSH is an efficient, reliable, and easy adaptable rule-based sentence segmentation solution. It is specifically designed to handle the telegraphic written text in clinical note. It leverages a nested hash table to execute simultaneous rule processing, which reduces the impact of the rule-base growth on execution time and eliminates the effect of rule order on accuracy.
 Home-page: https://github.com/jianlins/PyRuSH
 Author: Jianlin
@@ -77,3 +77,16 @@ Start from version 1.0.3, PyRuSH adds Spacy compatible Sentencizer component: Py
 A Colab Notebook Demo
 ---------------------------
 Feel free to try this runnable `Colab notebook Demo <https://colab.research.google.com/drive/1gX9MzZTQiPw8G3x_vUwZbiSXGtbI0uIX?usp=sharing>`_
+
+Revision History
+----------------
+
+**1.0.11 (2025-09-02)**
+
+- Improved sentence splitting logic: Sentences are now split at the last token before exceeding the max length, ensuring no chunk exceeds the specified limit.
+- Edge case handling: Trailing whitespaces (caused by spacy sentence labeling mechanism) can be optionally split into a separate sentence (merge_gaps=False) to avoid necessarily long sentences.
+
+**1.0.9 (2024-10-27)**
+
+- Initial release with spaCy 3.x compatibility and core RuSH logic.
+- Added Spacy-compatible PyRuSHSentencizer component.

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/PyRuSH/__init__.py

@@ -30,7 +30,7 @@
 
 from .PyRuSHSentencizer import PyRuSHSentencizer
 from .RuSH import RuSH, BEGIN, END
-__version__ = '1.0.11'
+__version__ = '1.0.12dev1'
 
 
 

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/PyRuSH.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyRuSH
-Version: 1.0.11
+Version: 1.0.12.dev1
 Summary: PyRuSH is the python implementation of RuSH (Rule-based sentence Segmenter using Hashing), which is originally developed using Java. RuSH is an efficient, reliable, and easy adaptable rule-based sentence segmentation solution. It is specifically designed to handle the telegraphic written text in clinical note. It leverages a nested hash table to execute simultaneous rule processing, which reduces the impact of the rule-base growth on execution time and eliminates the effect of rule order on accuracy.
 Home-page: https://github.com/jianlins/PyRuSH
 Author: Jianlin
@@ -77,3 +77,16 @@ Start from version 1.0.3, PyRuSH adds Spacy compatible Sentencizer component: Py
 A Colab Notebook Demo
 ---------------------------
 Feel free to try this runnable `Colab notebook Demo <https://colab.research.google.com/drive/1gX9MzZTQiPw8G3x_vUwZbiSXGtbI0uIX?usp=sharing>`_
+
+Revision History
+----------------
+
+**1.0.11 (2025-09-02)**
+
+- Improved sentence splitting logic: Sentences are now split at the last token before exceeding the max length, ensuring no chunk exceeds the specified limit.
+- Edge case handling: Trailing whitespaces (caused by spacy sentence labeling mechanism) can be optionally split into a separate sentence (merge_gaps=False) to avoid necessarily long sentences.
+
+**1.0.9 (2024-10-27)**
+
+- Initial release with spaCy 3.x compatibility and core RuSH logic.
+- Added Spacy-compatible PyRuSHSentencizer component.

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/PyRuSH.egg-info/SOURCES.txt

@@ -21,6 +21,7 @@ PyRuSH/../conf/rush_rules.tsv
 conf/rush_rules.tsv
 tests/test_PyRuSHSentencizer_param.py
 tests/test_PyRushSentencizer.py
+tests/test_PyRushSentencizer2.py
 tests/test_Rush.py
 tests/test_Rush_w_Logger.py
 tests/test_cpredict_split_gaps.py

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/README.rst

@@ -52,3 +52,16 @@ Start from version 1.0.3, PyRuSH adds Spacy compatible Sentencizer component: Py
 A Colab Notebook Demo
 ---------------------------
 Feel free to try this runnable `Colab notebook Demo <https://colab.research.google.com/drive/1gX9MzZTQiPw8G3x_vUwZbiSXGtbI0uIX?usp=sharing>`_
+
+Revision History
+----------------
+
+**1.0.11 (2025-09-02)**
+
+- Improved sentence splitting logic: Sentences are now split at the last token before exceeding the max length, ensuring no chunk exceeds the specified limit.
+- Edge case handling: Trailing whitespaces (caused by spacy sentence labeling mechanism) can be optionally split into a separate sentence (merge_gaps=False) to avoid necessarily long sentences.
+
+**1.0.9 (2024-10-27)**
+
+- Initial release with spaCy 3.x compatibility and core RuSH logic.
+- Added Spacy-compatible PyRuSHSentencizer component.

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/conf/rush_rules.tsv

@@ -20,7 +20,7 @@
 
 #stbegin is the marker for sentence begin, the span of sentence will start at the begin of the captured group
 #stbegin has two scores 0, 1: 0 for true sentence begin clues, 1 for false sentence begin clues which will overwrite 0-scored rules when they are overlapping.
-#stend is the marker for sentence end, the span of sentence will end at the end of the captured group
+#stend is the marker for sentence begin, the span of sentence will end at the end of the captured group
 #stend also has two scores 2, 3: 2 for true sentence end clues, 3 for false sentence end clues which will overwrite 2-scored rules when they are overlapping
 
 # \b the begin of an input
@@ -47,12 +47,6 @@
 \b\s+(\C	0	stbegin
 \b\s+(\d	0	stbegin
 \c.\s+(\C)	0	stbegin
-Dr.\s+(\C)	1	stbegin
-Mr.\s+(\C)	1	stbegin
-Ms.\s+(\C)	1	stbegin
-Miss.\s+(\C)	1	stbegin
-Mrs.\s+(\C)	1	stbegin
-dr.\s+(\C)	1	stbegin
  mL.\s+(\C)	0	stbegin
 *)	1	stbegin
 \c\c.\s+(\C)	0	stbegin
@@ -245,7 +239,7 @@ dr.\s+(\C)	1	stbegin
 \n(? \C	0	stbegin
 \n(? \c	0	stbegin
 \n(. \C	0	stbegin
-\n(\+ \C	0	stbegin
+\n(+ \C	0	stbegin
 \n(/ \C	0	stbegin
 \n+\d\d-\d\d\s+(\C	0	stbegin
 \n+\d+-\d\d-\d\d\s+(\C	0	stbegin

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/setup.cfg

@@ -1,7 +1,6 @@
 [metadata]
 readme = README.md
 license = MIT
-license_files = LICENSE
 
 [bdist_wheel]
 python_tag = py3

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/tests/test_PyRuSHSentencizer_param.py

@@ -6,7 +6,7 @@ from PyRuSH.PyRuSHSentencizer import PyRuSHSentencizer
 text_short = "Sentence one. Sentence two!"
 text_long = "This is a very long sentence that should be split at whitespace before the max length is reached. " * 5
 text_whitespace = "First sentence.    Second sentence after spaces.\nThird sentence after newline."
-rule_path = os.path.join(os.path.dirname(os.path.dirname(__file__)), "conf", "rush_rules.tsv")
+rule_path = os.path.join(os.path.dirname(__file__), "rush_rules.tsv")
 
 def make_nlp(merge_gaps, max_sentence_length):
     nlp = English()

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/tests/test_PyRushSentencizer.py

@@ -10,7 +10,7 @@ from spacy.lang.en import English
 class TestRuSH(unittest.TestCase):
 
     def setUp(self):
-        pwd = os.path.dirname(os.path.abspath(__file__))
+        self.pwd = os.path.dirname(os.path.abspath(__file__))
 
     def test_doc(self):
         nlp = English()
@@ -51,14 +51,14 @@ I will see her in a month to six weeks.  She is to follow up with Dr. X before t
  End Ezoic - MTSam Sample Bottom Matched Content - native_bottom
 '''
         nlp = English()
-        nlp.add_pipe("medspacy_pyrush")
+        nlp.add_pipe("medspacy_pyrush", config={"rules_path": os.path.join(self.pwd, 'rush_rules.tsv')})
         doc = nlp(input_str)
         sents = [s for s in doc.sents]
         for sent in sents:
             print('>' + str(sent) + '<\n\n')
 
         # New expected count includes whitespace-only sentences
-        assert (len(sents) == 53)
+        assert (len(sents) == 51)
         # For content checks, filter out whitespace-only sentences
         content_sents = [s for s in sents if s.text.strip()]
         assert (content_sents[0].text == 'Ms. ABCD is a 69-year-old lady, who was admitted to the hospital with chest pain and respiratory insufficiency.')
@@ -74,7 +74,7 @@ I will see her in a month to six weeks.  She is to follow up with Dr. X before t
         from loguru import logger
         logger.add(sys.stdout, level="DEBUG")
         nlp = English()
-        nlp.add_pipe("medspacy_pyrush")
+        nlp.add_pipe("medspacy_pyrush", config={"rules_path": os.path.join(self.pwd, 'rush_rules.tsv')})
         doc = nlp(input_str)
         sents = [s for s in doc.sents]
         for sent in sents:
@@ -116,4 +116,5 @@ I will see her in a month to six weeks.  She is to follow up with Dr. X before t
 
         # SpaCy has no control of sentence end. Thus, it ends up with sloppy ends.
         assert (sents[1].text == 'Ms. ABCD is a 69-year-old lady, who was admitted to the hospital with'
-                                 ' chest pain and respiratory insufficiency.')
+                                 ' chest pain and respiratory insufficiency.')
+                

pyrush-1.0.12.dev1/tests/test_PyRushSentencizer2.py

@@ -0,0 +1,45 @@
+import unittest
+import os
+import sys
+sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
+
+from PyRuSH import PyRuSHSentencizer
+from spacy.lang.en import English
+
+
+class TestRuSH(unittest.TestCase):
+
+    def setUp(self):
+        self.pwd = os.path.dirname(os.path.abspath(__file__))
+
+    # def test_doc(self):
+    #     nlp = English()
+    #     nlp.add_pipe("medspacy_pyrush")
+    #     doc = nlp("This is a sentence. This is another sentence.")
+    #     print('\n'.join([str(s) for s in doc.sents]))
+    #     print('\nTotal sentences: {}'.format(len([s for s in doc.sents])))
+    #     print('\ndoc is an instance of {}'.format(type(doc)))
+
+    def test_doc4(self):
+        input_str='''Ms. [**Known patient lastname 2004**] was admitted on [**2573-5-30**]. Ultrasound
+at the time of admission demonstrated pancreatic duct dilitation and
+edematous gallbladder. She was admitted to the ICU.
+Discharge Medications:
+1. Miconazole Nitrate 2 % Powder Sig: One (1) Appl Topical  BID
+(2 times a day) as needed.
+2. Heparin Sodium (Porcine) 5,000 unit/mL Solution Sig: One (1)
+Injection TID (3 times a day).
+3. Acetaminophen 160 mg/5 mL Elixir Sig: One (1)  PO Q4-6H
+(every 4 to 6 hours) as needed.'''
+        nlp = English()
+        nlp.add_pipe("medspacy_pyrush", config={"rules_path": os.path.join(self.pwd, 'rush_rules.tsv')})
+        nlp.initialize()
+        doc = nlp(input_str)
+        sents = [s for s in doc.sents]
+        for sent in sents:
+            print('>' + str(sent) + '<\n\n')
+        assert(sents[-1].text=='''Sig: One (1)  PO Q4-6H
+(every 4 to 6 hours) as needed.''')
+
+if __name__ == '__main__':
+    unittest.main()

{pyrush-1.0.11 → pyrush-1.0.12.dev1}/tests/test_Rush.py

@@ -95,7 +95,7 @@ class TestRuSH(unittest.TestCase):
         sentences = rush.segToSentenceSpans(input_str)
         self.printDetails(sentences, input_str)
 
-    def test_doc2(self):
+    def test8(self):
         input_str = '''  
 9.  Advair b.i.d.
 10.  Xopenex q.i.d. and p.r.n.
@@ -109,12 +109,44 @@ I will see her in a month to six weeks.  She is to follow up with Dr. X before t
         sent = sentences[1]
         assert (input_str[sent.begin:sent.end] == '10.  Xopenex q.i.d. and p.r.n.')
 
-
-    def test_doc11(self):
+    def test9(self):
         input_str='  This is a sentence. This is another sentence.'
-        sentences=self.rush.segToSentenceSpans(input_str)
-        for sent in sentences:
-            print('>' + input_str[sent.begin:sent.end] + '<\n')
+        self.rush = RuSH(str(os.path.join(self.pwd, 'rush_rules.tsv')), min_sent_chars=2, enable_logger=True)
+        sentences = self.rush.segToSentenceSpans(input_str)
+        self.printDetails(sentences, input_str)
+
+    def test10(self):
+        input_str='''Ms. [**Known patient lastname 2004**] was admitted on [**2573-5-30**]. Ultrasound
+at the time of admission demonstrated pancreatic duct dilitation and
+edematous gallbladder. She was admitted to the ICU.
+Discharge Medications:
+1. Miconazole Nitrate 2 % Powder Sig: One (1) Appl Topical  BID
+(2 times a day) as needed.
+2. Heparin Sodium (Porcine) 5,000 unit/mL Solution Sig: One (1)
+Injection TID (3 times a day).
+3. Acetaminophen 160 mg/5 mL Elixir Sig: One (1)  PO Q4-6H
+(every 4 to 6 hours) as needed.'''
+        self.rush = RuSH(str(os.path.join(self.pwd, 'rush_rules.tsv')), min_sent_chars=2, enable_logger=True)
+        sentences = self.rush.segToSentenceSpans(input_str)
+        self.printDetails(sentences, input_str)
+        assert (sentences[0].begin == 0 and sentences[0].end == 173)
+        assert (sentences[1].begin == 174 and sentences[1].end == 202)
+        assert (sentences[2].begin == 203 and sentences[2].end == 225)
+        assert (sentences[3].begin == 226 and sentences[3].end == 258)
+        assert (sentences[4].begin == 259 and sentences[4].end == 316)
+        assert (sentences[5].begin == 317 and sentences[5].end == 367)
+        assert (sentences[6].begin == 368 and sentences[6].end == 411)
+        assert (sentences[7].begin == 412 and sentences[7].end == 447)
+        assert (sentences[8].begin == 448 and sentences[8].end == 502)
+
+    def test11(self):
+        input_str = '''Patient doesn't have heart disease or high blood pressure, but their dad did have
+diabetes. Pt is a 63M w/ h/o metastatic carcinoid tumor, HTN and hyperlipidemia.'''
+        self.rush = RuSH(str(os.path.join(self.pwd, 'rush_rules.tsv')), min_sent_chars=2, enable_logger=True)
+        sentences = self.rush.segToSentenceSpans(input_str)
+        self.printDetails(sentences, input_str)
+        assert (sentences[0].begin == 0 and sentences[0].end == 91)
+        assert (sentences[1].begin == 92 and sentences[1].end == 162)
 
 if __name__ == '__main__':
     unittest.main()