Improve heuristic to warn about passing lexer/formatter class (#2123)

Don't rely on the error message since 'missing 1 required positional
argument' can give false positives.  Instead, use issubclass().

2 files changed, 28 insertions, 7 deletions

diff --git a/pygments/__init__.py b/pygments/__init__.py
index 52ff035d..1c107f33 100644
--- a/pygments/__init__.py
+++ b/pygments/__init__.py
@@ -39,9 +39,9 @@ def lex(code, lexer):
     try:
         return lexer.get_tokens(code)
     except TypeError as err:
-        if (isinstance(err.args[0], str) and
-            ('unbound method get_tokens' in err.args[0] or
-             'missing 1 required positional argument' in err.args[0])):
+        # Heuristic to catch a common mistake.
+        from pygments.lexer import RegexLexer
+        if isinstance(lexer, type) and issubclass(lexer, RegexLexer):
             raise TypeError('lex() argument must be a lexer instance, '
                             'not a class')
         raise
@@ -63,9 +63,9 @@ def format(tokens, formatter, outfile=None):  # pylint: disable=redefined-builti
         else:
             formatter.format(tokens, outfile)
     except TypeError as err:
-        if (isinstance(err.args[0], str) and
-            ('unbound method format' in err.args[0] or
-             'missing 1 required positional argument' in err.args[0])):
+        # Heuristic to catch a common mistake.
+        from pygments.formatter import Formatter
+        if isinstance(formatter, type) and issubclass(formatter, Formatter):
             raise TypeError('format() argument must be a formatter instance, '
                             'not a class')
         raise
@@ -80,4 +80,3 @@ def highlight(code, lexer, formatter, outfile=None):
     it is returned as a string.
     """
     return format(lex(code, lexer), formatter, outfile)
-
diff --git a/tests/test_basic_api.py b/tests/test_basic_api.py
index 1767572f..62ea4895 100644
--- a/tests/test_basic_api.py
+++ b/tests/test_basic_api.py
@@ -15,6 +15,7 @@ import pytest
 from pygments import lexers, formatters, lex, format
 from pygments.token import _TokenType, Text
 from pygments.lexer import RegexLexer
+from pygments.formatter import Formatter
 from pygments.formatters.img import FontNotFound
 from pygments.util import ClassNotFound
 
@@ -250,6 +251,27 @@ def test_bare_class_handler():
     else:
         assert False, 'nothing raised'
 
+    # These cases should not trigger this heuristic.
+    class BuggyLexer(RegexLexer):
+        def get_tokens(self, text, extra_argument):
+            pass
+        tokens = {'root': []}
+    try:
+        list(lex('dummy', BuggyLexer()))
+    except TypeError as e:
+        assert 'lex() argument must be a lexer instance' not in str(e)
+    else:
+        assert False, 'no error raised by buggy lexer?'
+
+    class BuggyFormatter(Formatter):
+        def format(self, tokensource, outfile, extra_argument):
+            pass
+    try:
+        format([], BuggyFormatter())
+    except TypeError as e:
+        assert 'format() argument must be a formatter instance' not in str(e)
+    else:
+        assert False, 'no error raised by buggy formatter?'
 
 class TestFilters: