python-negar 1.2.15__tar.gz → 1.3__tar.gz

{python_negar-1.2.15 → python_negar-1.3}/Changelog.txt

@@ -1,3 +1,6 @@
+1.3 - 2025-01-01
+-- Enhanced the handling of immutable words to achieve greater accuracy in results.
+
 1.2.15 - 2024-12-31
 -- Fixed issue #24 and added several new immutable words.
 

{python_negar-1.2.15 → python_negar-1.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: python-negar
-Version: 1.2.15
+Version: 1.3
 Summary: Negar is a spell corrector and Persian text editor
 Home-page: http://github.com/shahinism/python-negar
 Author: Shahin Azad
@@ -13,7 +13,7 @@ Classifier: Environment :: Console
 Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
 Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3
-Requires-Python: >=3.6
+Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 License-File: AUTHORS

{python_negar-1.2.15 → python_negar-1.3}/negar/constants.py

@@ -1,7 +1,7 @@
 import platform
 from pathlib import Path
 
-__version__ = "1.2.15"
+__version__ = "1.3"
 
 DATAFILE = Path(__file__).parent.absolute() / "data/immutable.words"
 APPDATA = "AppData/Roaming/" if platform.system() == "Windows" else "."

{python_negar-1.2.15 → python_negar-1.3}/negar/data/immutable.words

@@ -92,6 +92,7 @@
 الشان
 الکتر
 الکترومتر
+الکترومتری
 الکتری
 الماس‌نشان
 المان
@@ -331,6 +332,7 @@
 بیایی
 بیایید
 بیاییم
+بیاییم
 بیبسکو
 بیبل
 بی‌بها
@@ -1169,6 +1171,7 @@
 کودکستان
 کوهستان
 کیلومتر
+کیلومتری
 کیها
 گالوانومتر
 گانگستر
@@ -1200,6 +1203,8 @@
 لیموعمان
 ماتریس
 متان
+متر
+متری
 متشابها
 متکلمان
 متها

{python_negar-1.2.15 → python_negar-1.3}/negar/virastar.py

@@ -57,7 +57,7 @@ class PersianEditor:
         self.cleanup()
 
     def cleanup(self):
-        self._handle_urls(State.save)
+        self.__handle_urls__(State.save)
         # fix punctuation spaces at first
         # : ; , ! ? and their Persian counterparts should have one space after and no space before
         # An exception is triple dots, which should be handled first
@@ -77,10 +77,12 @@ class PersianEditor:
         self.char_validator()       if self._fix_misc_non_persian_chars else None
         self.fix_arabic_numbers()   if self._fix_arabic_numbers else None
         self.fix_english_numbers()  if self._fix_english_numbers else None
+        self.__handle_immutable_words__(State.save)
         self.fix_prefix_spacing()   if self._fix_prefix_spacing else None
         self.fix_prefix_separate()  if self._fix_prefix_separate else None
         self.fix_suffix_spacing()   if self._fix_suffix_spacing else None
         self.fix_suffix_separate()  if self._fix_suffix_separate else None
+        self.__handle_immutable_words__(State.restore)
         self.fix_spacing_for_braces_and_quotes() \
             if self._fix_spacing_for_braces_and_quotes else None
         self.cleanup_redundant_zwnj()
@@ -92,7 +94,7 @@ class PersianEditor:
         )
         if self._trim_leading_trailing_whitespaces:
             self.text = "\n".join([line.strip() for line in self.text.split("\n")])
-        self._handle_urls(State.restore)
+        self.__handle_urls__(State.restore)
         return self.text
 
     def __repr__(self):
@@ -100,7 +102,7 @@ class PersianEditor:
 
     __str__ = __repr__
 
-    def _handle_urls(self, state):
+    def __handle_urls__(self, state):
         """Remove URLs and putting them back at the end of the process."""
         if state == State.save:
             self.urls = list(set(re.findall(URLREGX, self.text, re.M | re.I | re.X)))
@@ -111,6 +113,19 @@ class PersianEditor:
             for i, url in enumerate(self.urls):
                 self.text = re.sub(f"__URL__#{i}__", url, self.text)
 
+    def __handle_immutable_words__(self, state):
+        """Remove URLs and putting them back at the end of the process."""
+        if state == State.save:
+            self.immutable_words = {}
+            words = self.text.split()
+            for i, word in enumerate(words):
+                if (word:=word.strip()) in ImmutableWords.get():
+                    self.immutable_words[i] = word
+                    self.text = regex.sub(rf"\b{word}\b", rf"__IMM__#{i}__", self.text)
+        if state == State.restore:
+            for imm_key, imm_word in self.immutable_words.items():
+                self.text = re.sub(f"__IMM__#{imm_key}__", imm_word, self.text)
+
     def fix_dashes(self):
         """Replace double and triple dashes with `ndash` and `mdash`, respectively."""
         self.text = re.sub(r"-{3}", r"—", self.text)
@@ -193,24 +208,8 @@ class PersianEditor:
         """,
             re.VERBOSE,
         )  #  \b for words like سهمیه
-
-        # Replace backslashes with a temporary placeholder
-        text = self.text.replace("\\", "PN__BACKSLASH__PN")
-        wlist = text.split()
-        for word in wlist:
-            if word.strip() in ImmutableWords.get():
-                continue
-            regx_iter = regx.finditer(word)
-            for p in regx_iter:
-                # Checks that the prefix (mi* nemi* bi*) is part a a word or not, like میلاد.
-                if p.group() not in ImmutableWords.get():
-                    text = re.sub(
-                        re.escape(p.group()),
-                        p.group(1) + "\u200c" + p.group(2),
-                        text,
-                    )
-        # Restore the original backslashes
-        self.text = text.replace("PN__BACKSLASH__PN", "\\")
+        # Checks that the prefix (mi*, nemi*, and bi*) is part a a word or not, like میلاد.
+        self.__immutability_check__(regx)
 
     def fix_suffix_spacing(self):
         """Put ZWNJ between a word and its suffix (*ha[ye] *tar[in])."""
@@ -258,21 +257,22 @@ class PersianEditor:
             )\b""",
             re.VERBOSE,
         )
+        # Checks that the suffix (tar*, haye*) is part of a word or not, like بهتر.
+        self.__immutability_check__(regx)
+
+    def __immutability_check__(self, regx: re.Pattern):
         # Replace backslashes with a temporary placeholder
         text = self.text.replace("\\", "PN__BACKSLASH__PN")
         wlist = text.split()
         for word in wlist:
-            if word.strip() in ImmutableWords.get():
-                continue
             regx_iter = regx.finditer(word)
             for p in regx_iter:
-                # Checks that the suffix (tar* haye*) is part of a word or not, like بهتر.
-                if p.group() not in ImmutableWords.get():
-                    text = re.sub(
-                        re.escape(p.group()),
-                        p.group(1) + "\u200c" + p.group(2),
-                        text,
-                    )
+                # if p.group() not in ImmutableWords.get():
+                text = re.sub(
+                    rf"\b{re.escape(p.group())}\b",
+                    p.group(1) + "\u200c" + p.group(2),
+                    text,
+                )
         # Restore the original backslashes
         self.text = text.replace("PN__BACKSLASH__PN", "\\")
 

{python_negar-1.2.15 → python_negar-1.3}/python_negar.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: python-negar
-Version: 1.2.15
+Version: 1.3
 Summary: Negar is a spell corrector and Persian text editor
 Home-page: http://github.com/shahinism/python-negar
 Author: Shahin Azad
@@ -13,7 +13,7 @@ Classifier: Environment :: Console
 Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
 Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3
-Requires-Python: >=3.6
+Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 License-File: AUTHORS

{python_negar-1.2.15 → python_negar-1.3}/setup.py

@@ -33,7 +33,7 @@ setup(
     },
     long_description_content_type="text/markdown",
     long_description=open("README.md", encoding="utf8").read(),
-    python_requires=">=3.6",
+    python_requires=">=3.8",
     classifiers=[
         "Environment :: Console",
         "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",