scan() -> replace_escape_sequences() ...

Convert a unit test to pytest. In the year 2022, unit tests like this
are deprecated - pytest wins.

Author: rocky

mathics_scanner/prescanner.py

@@ -46,7 +46,7 @@ def incomplete(self):
             raise IncompleteSyntaxError()
         self.code += line
 
-    def scan(self) -> str:
+    def replace_escape_sequences(self) -> str:
         """
         Replace escape sequences in self.code. The replacement string is returned.
         Note: self.code is not modified.

mathics_scanner/tokeniser.py

@@ -357,7 +357,7 @@ def __init__(self, feeder):
         self.pos = 0
         self.feeder = feeder
         self.prescanner = Prescanner(feeder)
-        self.code = self.prescanner.scan()
+        self.code = self.prescanner.replace_escape_sequences()
         self._change_mode("expr")
 
     def _change_mode(self, mode):
@@ -371,7 +371,7 @@ def _change_mode(self, mode):
     def incomplete(self):
         "Get more code from the prescanner and continue."
         self.prescanner.incomplete()
-        self.code += self.prescanner.scan()
+        self.code += self.prescanner.replace_escape_sequences()
 
     def sntx_message(self, pos=None):
         """Send a message to the feeder."""

test/test_prescanner.py

@@ -1,88 +1,105 @@
 # -*- coding: utf-8 -*-
-
-import unittest
+import pytest
 
 from mathics_scanner import IncompleteSyntaxError, ScanError
 from mathics_scanner.prescanner import Prescanner
 from mathics_scanner.feed import SingleLineFeeder
 
 
-class PrescannerTest(unittest.TestCase):
-    def prescan(self, code):
-        prescanner = Prescanner(SingleLineFeeder(code))
-        return prescanner.scan()
-
-    def invalid(self, code):
-        self.assertRaises(ScanError, self.prescan, code)
-
-    def incomplete(self, code):
-        self.assertRaises(IncompleteSyntaxError, self.prescan, code)
-
-    def equal(self, code, result):
-        assert self.prescan(code) == result
-
-    def equal_length(self, code, length):
-        assert len(self.prescan(code)) == length
-
-    def test_named_characters(self):
-        self.equal(r"\[Theta]", "\u03B8")
-        self.equal(r"\[CapitalPi]", "\u03A0")
-        self.equal(r"\[Fake]", r"\[Fake]")
-        self.equal("z \\[Conjugate]", "z \uF3C8")
-        self.equal("z \\[Integral]", "z \u222b")
-        self.equal("z \\\\[Integral]", "z \\\\[Integral]")
-        self.equal("z \\\\\\[Integral]", "z \\\\\u222b")
-        self.equal("abc\\\\", "abc\\\\")
-
-    def test_lengths(self):
-        self.equal_length(r'"\[Integral]"', 3)
-        # Prescanner keep both slashes and quotes.
-        # The tokenizer brings \\ into \ if it appears
-        # inside a string.
-        self.equal_length(r'"\\[Integral]"', 14)
-
-    def test_oct(self):
-        self.equal(r"\051", ")")
-
-    def test_hex_dot(self):
-        self.equal(r"\.30", "0")
-
-    def test_hex_colon(self):
-        self.equal(r"\:0030", "0")
-        self.equal(r"\:03B8", "\u03B8")
-        self.equal(r"\:03b8", "\u03B8")
-
-    def test_hex_vbar(self):
-        self.equal(r"\|01D451", "\U0001D451")
-
-    def test_incomplete(self):
-        self.incomplete(r"\[")
-        self.incomplete(r"\[Theta")
-
-    def test_invalid_oct(self):
-        self.invalid(r"\093")
-        self.invalid(r"\01")
-
-    def test_invalid_colon(self):
-        self.invalid(r"\:")
-        self.invalid(r"\:A")
-        self.invalid(r"\:01")
-        self.invalid(r"\:A1")
-        self.invalid(r"\:ak")
-        self.invalid(r"\:A10")
-        self.invalid(r"\:a1g")
-        self.invalid(r"\:A1g9")
-        self.invalid(r"\:01-2")
-
-    def test_invalid_dot(self):
-        self.invalid(r"\.")
-        self.invalid(r"\.0")
-
-    def test_combined(self):
-        self.equal(r"\:03B8\[Theta]\.30\052", "\u03B8\u03B80*")
-
-    def test_nested(self):
-        self.equal(r"\[Thet\141]", r"\[Thet\141]")
-
-    def test_trailing_backslash(self):
-        self.incomplete("x \\")
+def replace_escape_sequences(code):
+    prescanner = Prescanner(SingleLineFeeder(code))
+    return prescanner.replace_escape_sequences()
+
+
+def invalid(code):
+    with pytest.raises(ScanError):
+        replace_escape_sequences(code)
+
+
+def incomplete(code):
+    with pytest.raises(IncompleteSyntaxError):
+        replace_escape_sequences(code)
+
+
+def equal(code, result):
+    assert replace_escape_sequences(code) == result
+
+
+def equal_length(code, length):
+    assert len(replace_escape_sequences(code)) == length
+
+
+def test_named_characters():
+    equal(r"\[Theta]", "\u03B8")
+    equal(r"\[CapitalPi]", "\u03A0")
+    equal(r"\[Fake]", r"\[Fake]")
+    equal("z \\[Conjugate]", "z \uF3C8")
+    equal("z \\[Integral]", "z \u222b")
+    equal("z \\\\[Integral]", "z \\\\[Integral]")
+    equal("z \\\\\\[Integral]", "z \\\\\u222b")
+    equal("abc\\\\", "abc\\\\")
+
+
+def test_text_lengths():
+    equal_length(r'"\[Integral]"', 3)
+    # Prescanner keep both slashes and quotes.
+    # The tokenizer brings \\ into \ if it appears
+    # inside a string.
+    equal_length(r'"\\[Integral]"', 14)
+
+
+def test_oct():
+    equal(r"\051", ")")
+
+
+def test_hex_dot():
+    equal(r"\.30", "0")
+
+
+def test_hex_colon():
+    equal(r"\:0030", "0")
+    equal(r"\:03B8", "\u03B8")
+    equal(r"\:03b8", "\u03B8")
+
+
+def test_hex_vbar():
+    equal(r"\|01D451", "\U0001D451")
+
+
+def test_incomplete():
+    incomplete(r"\[")
+    incomplete(r"\[Theta")
+
+
+def test_invalid_oct():
+    invalid(r"\093")
+    invalid(r"\01")
+
+
+def test_invalid_colon():
+    invalid(r"\:")
+    invalid(r"\:A")
+    invalid(r"\:01")
+    invalid(r"\:A1")
+    invalid(r"\:ak")
+    invalid(r"\:A10")
+    invalid(r"\:a1g")
+    invalid(r"\:A1g9")
+    invalid(r"\:01-2")
+
+
+def test_invalid_dot():
+    invalid(r"\.")
+    invalid(r"\.0")
+
+
+def test_combined():
+    equal(r"\:03B8\[Theta]\.30\052", "\u03B8\u03B80*")
+
+
+def test_nested():
+    equal(r"\[Thet\141]", r"\[Thet\141]")
+
+
+def test_trailing_backslash():
+    incomplete("x \\")