额度不够，使用qianwen修改，未验证

data_analysis_agent.py

@@ -29,11 +29,30 @@ from prompts import data_analysis_system_prompt, final_report_system_prompt, dat
 
 # Regex patterns that indicate a data-context error (column/variable/DataFrame issues)
 DATA_CONTEXT_PATTERNS = [
+    # KeyError - missing key/column
     r"KeyError:\s*['\"](.+?)['\"]",
-    r"ValueError.*(?:column|col|field)",
+    # ValueError - value-related issues
-    r"NameError.*(?:df|data|frame)",
+    r"ValueError.*(?:column|col|field|shape|axis)",
-    r"(?:empty|no\s+data|0\s+rows)",
+    # NameError - undefined variables
+    r"NameError.*(?:df|data|frame|series)",
+    # Empty/missing data
+    r"(?:empty|no\s+data|0\s+rows|No\s+data)",
+    # IndexError - out of bounds
     r"IndexError.*(?:out of range|out of bounds)",
+    # AttributeError - missing attributes
+    r"AttributeError.*(?:DataFrame|Series|object)\s+has\s+no\s+attribute",
+    # Pandas-specific errors
+    r"pd\.errors\.(?:EmptyDataError|ParserError|MergeError)",
+    r"MergeError: No common columns",
+    # Type errors
+    r"TypeError.*(?:unsupported operand|expected string|cannot convert)",
+    # UnboundLocalError - undefined local variables
+    r"UnboundLocalError.*referenced before assignment",
+    # Syntax errors
+    r"SyntaxError: invalid syntax",
+    # Module/Import errors for data libraries
+    r"ModuleNotFoundError.*(?:pandas|numpy|matplotlib)",
+    r"ImportError.*(?:pandas|numpy|matplotlib)",
 ]
 
 

tests/test_unit.py

@@ -26,12 +26,32 @@ from config.app_config import AppConfig
 # Helpers
 # ---------------------------------------------------------------------------
 
+# Updated patterns matching data_analysis_agent.py
 DATA_CONTEXT_PATTERNS = [
+    # KeyError - missing key/column
     r"KeyError:\s*['\"](.+?)['\"]",
-    r"ValueError.*(?:column|col|field)",
+    # ValueError - value-related issues
-    r"NameError.*(?:df|data|frame)",
+    r"ValueError.*(?:column|col|field|shape|axis)",
-    r"(?:empty|no\s+data|0\s+rows)",
+    # NameError - undefined variables
+    r"NameError.*(?:df|data|frame|series)",
+    # Empty/missing data
+    r"(?:empty|no\s+data|0\s+rows|No\s+data)",
+    # IndexError - out of bounds
     r"IndexError.*(?:out of range|out of bounds)",
+    # AttributeError - missing attributes
+    r"AttributeError.*(?:DataFrame|Series|object)\s+has\s+no\s+attribute",
+    # Pandas-specific errors
+    r"pd\.errors\.(?:EmptyDataError|ParserError|MergeError)",
+    r"MergeError: No common columns",
+    # Type errors
+    r"TypeError.*(?:unsupported operand|expected string|cannot convert)",
+    # UnboundLocalError - undefined local variables
+    r"UnboundLocalError.*referenced before assignment",
+    # Syntax errors
+    r"SyntaxError: invalid syntax",
+    # Module/Import errors for data libraries
+    r"ModuleNotFoundError.*(?:pandas|numpy|matplotlib)",
+    r"ImportError.*(?:pandas|numpy|matplotlib)",
 ]
 
 
@@ -75,11 +95,11 @@ class TestErrorClassifier:
     def test_index_out_of_range(self):
         assert classify_error("IndexError: index 10 is out of range") == "data_context"
 
-    def test_syntax_error_is_other(self):
+    def test_syntax_error_is_data_context(self):
-        assert classify_error("SyntaxError: invalid syntax") == "other"
+        assert classify_error("SyntaxError: invalid syntax") == "data_context"
 
-    def test_type_error_is_other(self):
+    def test_type_error_is_data_context(self):
-        assert classify_error("TypeError: unsupported operand") == "other"
+        assert classify_error("TypeError: unsupported operand") == "data_context"
 
     def test_generic_text_is_other(self):
         assert classify_error("Something went wrong") == "other"
@@ -87,6 +107,55 @@ class TestErrorClassifier:
     def test_empty_string_is_other(self):
         assert classify_error("") == "other"
 
+    # ===========================================================================
+    # Additional tests for improved error classifier
+    # ===========================================================================
+
+    def test_attributeerror_dataframe(self):
+        assert classify_error("AttributeError: 'DataFrame' object has no attribute 'xxx'") == "data_context"
+
+    def test_attributeerror_series(self):
+        assert classify_error("AttributeError: 'Series' object has no attribute 'xxx'") == "data_context"
+
+    def test_pd_emptydataerror(self):
+        assert classify_error("pd.errors.EmptyDataError: No data") == "data_context"
+
+    def test_pd_parsererror(self):
+        assert classify_error("pd.errors.ParserError: Error tokenizing data") == "data_context"
+
+    def test_pd_mergeerror(self):
+        assert classify_error("MergeError: No common columns to merge") == "data_context"
+
+    def test_typeerror_unsupported_operand(self):
+        assert classify_error("TypeError: unsupported operand type(s) for +: 'int' and 'str'") == "data_context"
+
+    def test_typeerror_expected_string(self):
+        assert classify_error("TypeError: expected string or bytes-like object") == "data_context"
+
+    def test_unboundlocalerror(self):
+        assert classify_error("UnboundLocalError: local variable 'df' referenced before assignment") == "data_context"
+
+    def test_syntaxerror(self):
+        assert classify_error("SyntaxError: invalid syntax") == "data_context"
+
+    def test_modulenotfounderror(self):
+        assert classify_error("ModuleNotFoundError: No module named 'pandas'") == "data_context"
+
+    def test_importerror(self):
+        assert classify_error("ImportError: cannot import name 'xxx' from 'pandas'") == "data_context"
+
+    def test_valueerror_shape(self):
+        assert classify_error("ValueError: shape mismatch") == "data_context"
+
+    def test_valueerror_axis(self):
+        assert classify_error("ValueError: axis out of bounds") == "data_context"
+
+    def test_nameerror_series(self):
+        assert classify_error("NameError: name 'series' is not defined") == "data_context"
+
+    def test_no_data_message(self):
+        assert classify_error("No data available for analysis") == "data_context"
+
 
 # ===========================================================================
 # Task 12.1 continued: Unit tests for column extraction and lookup
@@ -189,8 +258,7 @@ class TestTemplateSystem:
         for name in TEMPLATE_REGISTRY:
             t = get_template(name)
             assert t.name  # has a display name
-            steps = t.build_steps()
+            assert len(t.steps) > 0  # template has steps
-            assert len(steps) > 0
 
     def test_get_invalid_template_raises(self):
         with pytest.raises(ValueError):