Merge pull request #44 from unlv-evol/dev

[FEAT]: Add boolean and delimiters regex for major programming languages

CONTRIBUTING.md

@@ -61,5 +61,5 @@ When you're finished with the changes, create a pull request, also known as a PR
 
 ### Your PR is merged!
 
-Congratulations :tada::tada: <br />
+Congratulations :tada: :tada: <br />
 The PyReprism team thanks you :sparkles:.

README.md

@@ -9,7 +9,7 @@
 [![codecov](https://codecov.io/gh/unlv-evol/PyReprism/graph/badge.svg?token=J2JV31837H)](https://codecov.io/gh/unlv-evol/PyReprism)
 ![PyPI - Version](https://img.shields.io/pypi/v/pyreprism)
 ![GitHub last commit (branch)](https://img.shields.io/github/last-commit/unlv-evol/pyreprism/main)
-
+[![SWH](https://archive.softwareheritage.org/badge/origin/https://pypi.org/project/PyReprism//)](https://archive.softwareheritage.org/browse/origin/?origin_url=https://pypi.org/project/PyReprism/)
 
 # PyReprism
 

dev-requirements.txt

@@ -0,0 +1,9 @@
+pytest
+flake8
+pytest-cov
+build
+twine
+hatchling
+hatch-vcs
+sphinx
+sphinx_rtd_theme

docs/Makefile

@@ -1,4 +1,4 @@
- Minimal makefile for Sphinx documentation
+# Minimal makefile for Sphinx documentation
 #
 
 # You can set these variables from the command line.

docs/conf.py

@@ -10,9 +10,9 @@
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 #
-# import os
-# import sys
-# sys.path.insert(0, os.path.abspath('.'))
+import os
+import sys
+sys.path.insert(0, os.path.abspath('../src'))
 
 
 # -- Project information -----------------------------------------------------
@@ -60,4 +60,4 @@
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ["_static"]
+html_static_path = ["_static"]

docs/normalizer.rst

@@ -2,4 +2,18 @@
 
 ============
 Normalizer
-============
+============
+
+.. automodule:: PyReprism.utils.normalizer
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+.. automodule:: PyReprism.utils.extension
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__

docs/reference.rst

@@ -5,7 +5,106 @@ Reference
 ============
 
 Version
---------
+---------
 .. py:data:: src.PyReprism.__version__
 
-   Current PyReprism version.
+   Current PyReprism version.
+
+Python
+-------
+.. automodule:: PyReprism.languages.python
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+Java
+-------
+.. automodule:: PyReprism.languages.java
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+C++
+-------
+.. automodule:: PyReprism.languages.cpp
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+C
+-------
+.. automodule:: PyReprism.languages.c
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+Scala
+-------
+.. automodule:: PyReprism.languages.scala
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+Dart
+-------
+.. automodule:: PyReprism.languages.dart
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+Rust
+-------
+.. automodule:: PyReprism.languages.rust
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+Kotlin
+-------
+.. automodule:: PyReprism.languages.kotlin
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+Swift
+-------
+.. automodule:: PyReprism.languages.swift
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+JavaScript
+----------
+.. automodule:: PyReprism.languages.javascript
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__
+
+JSX
+-------
+.. automodule:: PyReprism.languages.jsx
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :special-members:
+   :exclude-members: __dict__, __weakref__, __eq__, __str__, __hash__

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "PyReprism"
-version = "0.0.3"
+version = "0.0.4"
 authors = [
   { name= "UNLV EVOL LAB", email="ogenrwot@unlv.nevada.edu" },
 ]

requirements.txt

@@ -1,7 +1,3 @@
 pytest
 flake8
-pytest-cov
-build
-twine
-hatchling
-hatch-vcs
+pytest-cov

src/PyReprism/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.0.4"

src/PyReprism/languages/c.py

@@ -8,36 +8,106 @@ def __init__():
 
     @staticmethod
     def file_extension() -> str:
+        """
+        Return the file extension used for C files.
+
+        :return: The file extension for C files.
+        :rtype: str
+        """
         return extension.c
 
     @staticmethod
     def keywords() -> list:
+        """
+        Return a list of C keywords and built-in functions.
+
+        :return: A list of C keywords and built-in function names.
+        :rtype: list
+        """
         keyword = '_Alignas|_Alignof|_Atomic|_Bool|_Complex|_Generic|_Imaginary|_Noreturn|_Static_assert|_Thread_local|asm|typeof|inline|auto|break|case|char|const|continue|default|do|double|else|enum|extern|float|for|goto|if|int|long|register|return|short|signed|sizeof|static|struct|switch|typedef|union|unsigned|void|volatile|while|define|defined|elif|else|endif|error|ifdef|ifndef|if|import|include|line|pragma|undef|using|__FILE__|__LINE__|__DATE__|__TIME__|__TIMESTAMP__|__func__|EOF|NULL|SEEK_CUR|SEEK_END|SEEK_SET|stdin|stdout|stderr'.split('|')
         return keyword
 
     @staticmethod
-    def comment_regex():
+    def comment_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify different types of comments and non-comment code in C source files.
+
+        :return: A compiled regex pattern with named groups to match single-line comments, multiline comments, and non-comment code elements.
+        :rtype: re.Pattern
+        """
         pattern = re.compile(r'(?P<comment>//.*?$|/\*[^*]*\*+(?:[^/*][^*]*\*+)*?/)|(?P<noncomment>[^/]+)', re.DOTALL | re.MULTILINE)
         return pattern
 
     @staticmethod
-    def number_regex():
+    def number_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify numeric literals in C code.
+
+        :return: A compiled regex pattern to match C numeric literals, including integers, floats, and complex numbers.
+        :rtype: re.Pattern
+        """
         pattern = re.compile(r'(?:\b0x[\da-f]+|(?:\b\d+\.?\d*|\B\.\d+)(?:e[+-]?\d+)?)[ful]*')
         return pattern
 
     @staticmethod
-    def operator_regex():
+    def operator_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify C operators.
+
+        :return: A compiled regex pattern to match various C operators and logical keywords.
+        :rtype: re.Pattern
+        """
         pattern = re.compile(r'-[>-]?|\+\+?|!=?|<<?=?|>>?=?|==?|&&?|\|\|?|[~^%?*\/]')
         return pattern
 
     @staticmethod
-    def keywords_regex():
+    def keywords_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify C keywords.
+
+        :return: A compiled regex pattern to match C keywords.
+        :rtype: re.Pattern
+        """
         return re.compile(r'\b(' + '|'.join(C.keywords()) + r')\b')
 
+    @staticmethod
+    def boolean_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify C boolean literals.
+
+        :return: A compiled regex pattern to match C boolean literals.
+        :rtype: re.Pattern
+        """
+        return re.compile(r'\b(?:true|false)\b')
+
+    @staticmethod
+    def delimiters_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify C and C++ delimiters.
+
+        :return: A compiled regex pattern to match C and C++ delimiters.
+        :rtype: re.Pattern
+        """
+        return re.compile(r'[()\[\]{}.,:;@<>*&]')
+
     @staticmethod
     def remove_comments(source_code: str) -> str:
+        """
+        Remove comments from the provided C source code string.
+
+        :param str source_code: The C source code from which to remove comments.
+        :return: The source code with all comments removed.
+        :rtype: str
+        """
         return C.comment_regex().sub(lambda match: match.group('noncomment') if match.group('noncomment') else '', source_code).strip()
 
     @staticmethod
-    def remove_keywords(source: str):
+    def remove_keywords(source: str) -> str:
+        """
+        Remove all C keywords from the provided source code string.
+
+        :param str source: The source code string from which to remove C keywords.
+        :return: The source code string with all C keywords removed.
+        :rtype: str
+        """
         return re.sub(re.compile(C.keywords_regex()), '', source)