Push PerCharacterEscaper (#27 closes #21)

Author: nedtwigg

python/selfie-lib/selfie_lib/PerCharacterEscaper.py

@@ -0,0 +1,95 @@
+from typing import List
+
+
+class PerCharacterEscaper:
+    def __init__(
+        self,
+        escape_code_point: int,
+        escaped_code_points: List[int],
+        escaped_by_code_points: List[int],
+    ):
+        self.__escape_code_point = escape_code_point
+        self.__escaped_code_points = escaped_code_points
+        self.__escaped_by_code_points = escaped_by_code_points
+
+    def __first_offset_needing_escape(self, input_string: str) -> int:
+        length = len(input_string)
+        for offset in range(length):
+            codepoint = ord(input_string[offset])
+            if (
+                codepoint == self.__escape_code_point
+                or codepoint in self.__escaped_code_points
+            ):
+                return offset
+        return -1
+
+    def escape(self, input_string: str) -> str:
+        no_escapes = self.__first_offset_needing_escape(input_string)
+        if no_escapes == -1:
+            return input_string
+        else:
+            result = []
+            result.append(input_string[:no_escapes])
+            for char in input_string[no_escapes:]:
+                codepoint = ord(char)
+                if codepoint in self.__escaped_code_points:
+                    idx = self.__escaped_code_points.index(codepoint)
+                    result.append(chr(self.__escape_code_point))
+                    result.append(chr(self.__escaped_by_code_points[idx]))
+                else:
+                    result.append(char)
+            return "".join(result)
+
+    def unescape(self, input_string: str) -> str:
+        if input_string.endswith(
+            chr(self.__escape_code_point)
+        ) and not input_string.endswith(chr(self.__escape_code_point) * 2):
+            raise ValueError(
+                "Escape character '{}' can't be the last character in a string.".format(
+                    chr(self.__escape_code_point)
+                )
+            )
+
+        no_escapes = self.__first_offset_needing_escape(input_string)
+        if no_escapes == -1:
+            return input_string
+        else:
+            result = [input_string[:no_escapes]]
+            skip_next = False
+            for i in range(no_escapes, len(input_string)):
+                if skip_next:
+                    skip_next = False
+                    continue
+                codepoint = ord(input_string[i])
+                if codepoint == self.__escape_code_point and (i + 1) < len(
+                    input_string
+                ):
+                    next_codepoint = ord(input_string[i + 1])
+                    if next_codepoint in self.__escaped_by_code_points:
+                        idx = self.__escaped_by_code_points.index(next_codepoint)
+                        result.append(chr(self.__escaped_code_points[idx]))
+                        skip_next = True
+                    else:
+                        result.append(input_string[i + 1])
+                        skip_next = True
+                else:
+                    result.append(chr(codepoint))
+            return "".join(result)
+
+    @classmethod
+    def self_escape(cls, escape_policy):
+        code_points = [ord(c) for c in escape_policy]
+        escape_code_point = code_points[0]
+        return cls(escape_code_point, code_points, code_points)
+
+    @classmethod
+    def specified_escape(cls, escape_policy):
+        code_points = [ord(c) for c in escape_policy]
+        if len(code_points) % 2 != 0:
+            raise ValueError(
+                "Escape policy string must have an even number of characters."
+            )
+        escape_code_point = code_points[0]
+        escaped_code_points = code_points[0::2]
+        escaped_by_code_points = code_points[1::2]
+        return cls(escape_code_point, escaped_code_points, escaped_by_code_points)

python/selfie-lib/selfie_lib/__init__.py

@@ -1,2 +1,3 @@
 from .LineReader import LineReader as LineReader
 from .Slice import Slice as Slice
+from .PerCharacterEscaper import PerCharacterEscaper as PerCharacterEscaper

python/selfie-lib/tests/PerCharacterEscaper_test.py

@@ -0,0 +1,88 @@
+import pytest
+
+from selfie_lib import PerCharacterEscaper
+
+
+class TestPerCharacterEscaper:
+    def test_performance_optimization_self(self):
+        escaper = PerCharacterEscaper.self_escape("`123")
+        abc = "abc"
+        # Correct use of 'is' for checking object identity.
+        assert (
+            escaper.escape(abc) is abc
+        ), "Escape should return the original object when no change is made"
+        assert (
+            escaper.unescape(abc) is abc
+        ), "Unescape should return the original object when no change is made"
+
+        # Use '==' for checking value equality.
+        assert (
+            escaper.escape("1") == "`1"
+        ), "Escaping '1' should prepend the escape character"
+        assert (
+            escaper.escape("`") == "``"
+        ), "Escaping the escape character should duplicate it"
+        assert (
+            escaper.escape("abc123`def") == "abc`1`2`3``def"
+        ), "Escaping 'abc123`def' did not produce the expected result"
+
+        assert escaper.unescape("`1") == "1", "Unescaping '`1' should produce '1'"
+        assert escaper.unescape("``") == "`", "Unescaping '``' should produce '`'"
+        assert (
+            escaper.unescape("abc`1`2`3``def") == "abc123`def"
+        ), "Unescaping 'abc`1`2`3``def' did not produce the expected result"
+
+    def test_performance_optimization_specific(self):
+        escaper = PerCharacterEscaper.specified_escape("`a1b2c3d")
+        abc = "abc"
+        # Correct use of 'is' for object identity.
+        assert (
+            escaper.escape(abc) is abc
+        ), "Escape should return the original object when no change is made"
+        assert (
+            escaper.unescape(abc) is abc
+        ), "Unescape should return the original object when no change is made"
+
+        # Use '==' for value equality.
+        assert escaper.escape("1") == "`b", "Escaping '1' should produce '`b'"
+        assert escaper.escape("`") == "`a", "Escaping '`' should produce '`a'"
+        assert (
+            escaper.escape("abc123`def") == "abc`b`c`d`adef"
+        ), "Escaping 'abc123`def' did not produce the expected result"
+
+        assert escaper.unescape("`b") == "1", "Unescaping '`b' should produce '1'"
+        assert escaper.unescape("`a") == "`", "Unescaping '`a' should produce '`'"
+        assert (
+            escaper.unescape("abc`1`2`3``def") == "abc123`def"
+        ), "Unescaping 'abc`1`2`3``def' did not produce the expected result"
+
+    def test_corner_cases_self(self):
+        escaper = PerCharacterEscaper.self_escape("`123")
+        with pytest.raises(ValueError) as excinfo:
+            escaper.unescape("`")
+        assert (
+            str(excinfo.value)
+            == "Escape character '`' can't be the last character in a string."
+        ), "Unescaping a string ending with a single escape character should raise ValueError"
+        assert escaper.unescape("`a") == "a", "Unescaping '`a' should produce 'a'"
+
+    def test_corner_cases_specific(self):
+        escaper = PerCharacterEscaper.specified_escape("`a1b2c3d")
+        with pytest.raises(ValueError) as excinfo:
+            escaper.unescape("`")
+        assert (
+            str(excinfo.value)
+            == "Escape character '`' can't be the last character in a string."
+        ), "Unescaping a string ending with a single escape character should raise ValueError"
+        assert escaper.unescape("`e") == "e", "Unescaping '`e' should produce 'e'"
+
+    def test_roundtrip(self):
+        escaper = PerCharacterEscaper.self_escape("`<>")
+
+        def roundtrip(str):
+            assert (
+                escaper.unescape(escaper.escape(str)) == str
+            ), f"Roundtrip of '{str}' did not return the original string"
+
+        roundtrip("")
+        roundtrip("<local>~`/")