feat: comments, boolean and delimiters regex

Author: danielogen

src/PyReprism/languages/kotlin.py

@@ -16,26 +16,58 @@ def keywords() -> list:
         return keyword
 
     @staticmethod
-    def comment_regex():
+    def comment_regex() -> 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:
         pattern = re.compile(r'\b(?:0[bx][\da-fA-F]+|\d+(?:\.\d+)?(?:e[+-]?\d+)?[fFL]?)\b')
         return pattern
 
     @staticmethod
-    def operator_regex():
+    def operator_regex() -> re.Pattern:
         pattern = re.compile(r'+[+=]?|-[-=>]?|==?=?|!(?:!|==?)?|[\/*%<>]=?|[?:]:?|\.\.|&&|\|\||\b(?:and|inv|or|shl|shr|ushr|xor)\b')
         return pattern
 
     @staticmethod
-    def keywords_regex():
+    def keywords_regex() -> re.Pattern:
         return re.compile(r'\b(' + '|'.join(Kotlin.keywords()) + r')\b')
 
+    @staticmethod
+    def delimiters_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify Kotlin language delimiters.
+
+        This function generates a regular expression that matches Kotlin language delimiters, which include parentheses `()`, brackets `[]`, braces `{}`, commas `,`, colons `:`, periods `.`, semicolons `;`, angle brackets `<`, `>`, and the question mark `?`.
+
+        :return: A compiled regex pattern to match Kotlin delimiters.
+        :rtype: re.Pattern
+        """
+        return re.compile(r'[()\[\]{}.,:;<>?]')
+
+    @staticmethod
+    def boolean_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify Kotlin boolean literals.
+
+        This function generates a regular expression that matches the Kotlin boolean literals `true`, `false`, and the special constant `null`.
+
+        :return: A compiled regex pattern to match Kotlin boolean literals and `null`.
+        :rtype: re.Pattern
+        """
+        return re.compile(r'\b(?:true|false|null)\b')
+
     @staticmethod
     def remove_comments(source_code: str, isList: bool = False) -> str:
+        """
+        Remove comments from the provided Kotlin source code string.
+
+        :param str source_code: The Kotlin source code from which to remove comments.
+        :return: The source code with all comments removed.
+        :rtype: str
+        """
+        return Kotlin.comment_regex().sub(lambda match: match.group('noncomment') if match.group('noncomment') else '', source_code).strip()
         result = []
         for match in Kotlin.comment_regex().finditer(source_code):
             if match.group('noncomment'):
@@ -45,5 +77,5 @@ def remove_comments(source_code: str, isList: bool = False) -> str:
         return ''.join(result)
 
     @staticmethod
-    def remove_keywords(source: str):
+    def remove_keywords(source: str) -> str:
         return re.sub(re.compile(Kotlin.keywords_regex()), '', source)

src/PyReprism/languages/rust.py

@@ -2,48 +2,80 @@
 from PyReprism.utils import extension
 
 
-class Dart:
+class Rust:
     def __init__():
         pass
 
     @staticmethod
     def file_extension() -> str:
-        return extension.dart
+        return extension.rust
 
     @staticmethod
     def keywords() -> list:
         keyword = 'abstract|alignof|as|be|box|break|const|continue|crate|do|else|enum|extern|false|final|fn|for|if|impl|in|let|loop|match|mod|move|mut|offsetof|once|override|priv|pub|pure|ref|return|sizeof|static|self|struct|super|true|trait|type|typeof|unsafe|unsized|use|virtual|where|while|yield'.split('|')
         return keyword
 
     @staticmethod
-    def comment_regex():
+    def comment_regex() -> re.Pattern:
         pattern = re.compile(r'(?P<comment>//.*?$|///.*?$|/\*[\s\S]*?\*/|/\*.*?$|^.*?\*/)|(?P<noncomment>\'(\\.|[^\\\'])*\'|"(\\.|[^\\"])*"|.[^/\'"]*)', re.DOTALL | re.MULTILINE)
         return pattern
 
     @staticmethod
-    def number_regex():
+    def number_regex() -> re.Pattern:
         pattern = re.compile(r'\b(?:0x[\dA-Fa-f](?:_?[\dA-Fa-f])*|0o[0-7](?:_?[0-7])*|0b[01](?:_?[01])*|(\d(?:_?\d)*)?\.?\d(?:_?\d)*(?:[Ee][+-]?\d+)?)(?:_?(?:[iu](?:8|16|32|64)?|f32|f64))?\b')
         return pattern
 
     @staticmethod
-    def operator_regex():
+    def operator_regex() -> re.Pattern:
         pattern = re.compile(r'[-+*\/%!^]=?|=[=>]?|@|&[&=]?|\|[|=]?|<<?=?|>>?=?')
         return pattern
 
     @staticmethod
-    def keywords_regex():
-        return re.compile(r'\b(' + '|'.join(Dart.keywords()) + r')\b')
+    def keywords_regex() -> re.Pattern:
+        return re.compile(r'\b(' + '|'.join(Rust.keywords()) + r')\b')
+
+    @staticmethod
+    def boolean_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify Rust boolean literals.
+
+        This function generates a regular expression that matches the Rust boolean literals `true`, `false`, and the special constant `None`.
+
+        :return: A compiled regex pattern to match Rust boolean literals and `None`.
+        :rtype: re.Pattern
+        """
+        return re.compile(r'\b(?:true|false|None)\b')
+
+    @staticmethod
+    def rust_delimiters_regex() -> re.Pattern:
+        """
+        Compile and return a regular expression pattern to identify Rust language delimiters.
+
+        This function generates a regular expression that matches Rust language delimiters, which include parentheses `()`, brackets `[]`, braces `{}`, commas `,`, colons `:`, periods `.`, semicolons `;`, angle brackets `<`, `>`, and the question mark `?`.
+
+        :return: A compiled regex pattern to match Rust delimiters.
+        :rtype: re.Pattern
+        """
+        return re.compile(r'[()\[\]{}.,:;<>?]')
 
     @staticmethod
     def remove_comments(source_code: str, isList: bool = False) -> str:
+        """
+        Remove comments from the provided Rust source code string.
+
+        :param str source_code: The Rust source code from which to remove comments.
+        :return: The source code with all comments removed.
+        :rtype: str
+        """
+        return Rust.comment_regex().sub(lambda match: match.group('noncomment') if match.group('noncomment') else '', source_code).strip()
         result = []
-        for match in Dart.comment_regex().finditer(source_code):
+        for match in Rust.comment_regex().finditer(source_code):
             if match.group('noncomment'):
                 result.append(match.group('noncomment'))
         if isList:
             return result
         return ''.join(result)
 
     @staticmethod
-    def remove_keywords(source: str):
-        return re.sub(re.compile(Dart.keywords_regex()), '', source)
+    def remove_keywords(source: str) -> str:
+        return re.sub(re.compile(Rust.keywords_regex()), '', source)