informatica-python 1.9.3__tar.gz → 1.9.4__tar.gz

{informatica_python-1.9.3 → informatica_python-1.9.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: informatica-python
-Version: 1.9.3
+Version: 1.9.4
 Summary: Convert Informatica PowerCenter workflow XML to Python/PySpark code
 Author: Nick
 License: MIT
@@ -430,7 +430,7 @@ The generated `helper_functions.py` provides a complete runtime library:
 - **Generated code formatting**: Consistent `# ---` section headers for Source Qualifiers, Transforms, and Target Writes; metadata comments (database type, field lists); column mapping and write operation comments; clean blank line handling
 - **Source/target detection**: Case-insensitive instance type matching
 - **Session→mapping inference**: Longest-suffix-match strategy for ambiguous mapping names
-- **646 tests** across unit, integration, expression, and formatting test suites
+- **663 tests** across unit, integration, expression, and formatting test suites
 
 ### v1.9.2 (Phase 8)
 - Mapping output files now use real mapping names (e.g., `mapping_m_customer_load.py`) instead of generic numeric indices (`mapping_1.py`)
@@ -495,7 +495,7 @@ The generated `helper_functions.py` provides a complete runtime library:
 cd informatica_python
 pip install -e ".[dev]"
 
-# Run tests (646 tests)
+# Run tests (663 tests)
 pytest tests/ -v
 ```
 

{informatica_python-1.9.3 → informatica_python-1.9.4}/README.md

@@ -403,7 +403,7 @@ The generated `helper_functions.py` provides a complete runtime library:
 - **Generated code formatting**: Consistent `# ---` section headers for Source Qualifiers, Transforms, and Target Writes; metadata comments (database type, field lists); column mapping and write operation comments; clean blank line handling
 - **Source/target detection**: Case-insensitive instance type matching
 - **Session→mapping inference**: Longest-suffix-match strategy for ambiguous mapping names
-- **646 tests** across unit, integration, expression, and formatting test suites
+- **663 tests** across unit, integration, expression, and formatting test suites
 
 ### v1.9.2 (Phase 8)
 - Mapping output files now use real mapping names (e.g., `mapping_m_customer_load.py`) instead of generic numeric indices (`mapping_1.py`)
@@ -468,7 +468,7 @@ The generated `helper_functions.py` provides a complete runtime library:
 cd informatica_python
 pip install -e ".[dev]"
 
-# Run tests (646 tests)
+# Run tests (663 tests)
 pytest tests/ -v
 ```
 

{informatica_python-1.9.3 → informatica_python-1.9.4}/informatica_python/__init__.py

@@ -7,7 +7,7 @@ Licensed under the MIT License.
 
 from informatica_python.converter import InformaticaConverter
 
-__version__ = "1.9.3"
+__version__ = "1.9.4"
 __author__ = "Nick"
 __license__ = "MIT"
 __all__ = ["InformaticaConverter"]

{informatica_python-1.9.3 → informatica_python-1.9.4}/informatica_python/generators/mapping_gen.py

@@ -757,7 +757,7 @@ def _generate_transformation(lines, tx, connector_graph, source_dfs, transform_m
     elif tx_type in ("joiner",):
         _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_dfs, connector_graph, data_lib)
     elif tx_type in ("lookup procedure", "lookup"):
-        _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs, data_lib)
+        _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs, connector_graph, data_lib)
     elif tx_type == "router":
         _gen_router_transform(lines, tx, tx_safe, input_df, source_dfs)
     elif tx_type in ("union",):
@@ -982,7 +982,7 @@ def _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_df
     source_dfs[tx.name] = f"df_{tx_safe}"
 
 
-def _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs, data_lib="pandas"):
+def _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs, connector_graph=None, data_lib="pandas"):
     lookup_table = ""
     lookup_sql = ""
     lookup_condition = ""
@@ -1012,6 +1012,11 @@ def _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs, data_lib="pa
 
     all_output_fields = return_fields + lookup_output_fields
 
+    port_to_col = {}
+    if connector_graph and tx.name in connector_graph.get("to", {}):
+        for conn in connector_graph["to"][tx.name]:
+            port_to_col[conn.to_field.lower()] = conn.from_field
+
     lines.append(f"    # Lookup: {lookup_table or tx.name}")
     if lookup_sql:
         _emit_sql_with_params(lines, f"lkp_sql_{tx_safe}", lookup_sql)
@@ -1020,10 +1025,13 @@ def _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs, data_lib="pa
         lines.append(f"    df_lkp_{tx_safe} = read_from_db(config, 'SELECT * FROM {lookup_table}', 'default')")
     else:
         empty_expr = lib_empty_df(data_lib)
-        lines.append(f"    df_lkp_{tx_safe} = {empty_expr}")
+        lines.append(f"    df_lkp_{tx_safe} = {empty_expr}  # WARNING: no lookup table/SQL override found")
 
     input_keys, lookup_keys = parse_lookup_condition(lookup_condition)
 
+    if input_keys and port_to_col:
+        input_keys = [port_to_col.get(k.lower(), k) for k in input_keys]
+
     if input_keys and lookup_keys:
         lines.append(f"    # Lookup condition: {lookup_condition}")
 
@@ -1078,12 +1086,23 @@ def _gen_router_transform(lines, tx, tx_safe, input_df, source_dfs):
         if "Group Filter Condition" in attr.name:
             group_conditions[attr.name] = attr.value
 
+    remaining_mask_parts = []
     if group_conditions:
         for i, (gname, cond) in enumerate(group_conditions.items()):
-            expr_py = convert_expression(cond) if cond else "True"
-            lines.append(f"    df_{tx_safe}_group{i} = {input_df}[{expr_py}].copy()  # {gname}")
+            if cond and cond.strip():
+                expr_py = convert_filter_vectorized(cond, input_df)
+            else:
+                expr_py = f"pd.Series(True, index={input_df}.index)"
+            mask_var = f"_router_mask_{tx_safe}_{i}"
+            lines.append(f"    {mask_var} = {expr_py}  # {gname}")
+            lines.append(f"    df_{tx_safe}_group{i} = {input_df}[{mask_var}].copy()")
             source_dfs[f"{tx.name}_group{i}"] = f"df_{tx_safe}_group{i}"
-    lines.append(f"    df_{tx_safe} = {input_df}.copy()  # Default group")
+            remaining_mask_parts.append(f"~{mask_var}")
+    if remaining_mask_parts:
+        lines.append(f"    _router_default_mask = {' & '.join(remaining_mask_parts)}")
+        lines.append(f"    df_{tx_safe} = {input_df}[_router_default_mask].copy()  # Default group")
+    else:
+        lines.append(f"    df_{tx_safe} = {input_df}.copy()  # Default group")
     source_dfs[tx.name] = f"df_{tx_safe}"
 
 

{informatica_python-1.9.3 → informatica_python-1.9.4}/informatica_python/utils/expression_converter.py

@@ -248,6 +248,7 @@ def _convert_infa_date_format(fmt_str):
     fmt = fmt.replace("Mon", "%b").replace("MON", "%b")
     fmt = fmt.replace("HH24", "%H").replace("HH12", "%I").replace("HH", "%H")
     fmt = fmt.replace("MI", "%M").replace("SS", "%S")
+    fmt = fmt.replace("US", "%f").replace("NS", "%f").replace("MS", "%f")
     return fmt
 
 
@@ -548,7 +549,7 @@ def _vec_recursive(expr, df_var):
                         'RTRIM': f'.str.rstrip("{char_arg}")',
                         'TRIM': f'.str.strip("{char_arg}")',
                     }
-                return f'{inner_val}{method_map[func_name.upper()]}'
+                return f'{inner_val}.astype(str){method_map[func_name.upper()]}'
 
     upper_result = _find_func_call(cleaned, 'UPPER')
     if upper_result and upper_result[0] == 0 and upper_result[1] == len(cleaned):
@@ -584,7 +585,7 @@ def _vec_recursive(expr, df_var):
         if len(args) >= 2:
             field_val = _vec_recursive(args[0], df_var)
             try:
-                start = int(args[1].strip()) - 1
+                start = max(int(args[1].strip()) - 1, 0)
             except ValueError:
                 start_val = _vec_recursive(args[1], df_var)
                 if len(args) >= 3:
@@ -722,7 +723,11 @@ def _vec_recursive(expr, df_var):
                 field_val = _vec_recursive(args[0], df_var)
                 pattern_val = args[1].strip().strip("'\"")
                 if func_name == 'REG_EXTRACT':
-                    return f'{field_val}.str.extract(r"({pattern_val})", expand=False)'
+                    if re.search(r'(?<!\\)\((?!\?)', pattern_val):
+                        extract_pat = pattern_val
+                    else:
+                        extract_pat = f'({pattern_val})'
+                    return f'{field_val}.str.extract(r"{extract_pat}", expand=False)'
                 elif func_name == 'REG_REPLACE':
                     replace_val = args[2].strip().strip("'\"") if len(args) >= 3 else ''
                     return f'{field_val}.str.replace(r"{pattern_val}", "{replace_val}", regex=True)'
@@ -894,7 +899,8 @@ def _vec_recursive(expr, df_var):
         'True', 'False', 'None', 'and', 'or', 'not', 'np', 'pd', 'get_variable',
         'str', 'int', 'float', 'bool', 'len', 'abs', 'round',
         'fillna', 'astype', 'isna', 'notna', 'where', 'errors', 'coerce',
-        'lookup_func',
+        'lookup_func', 'expand', 'extract', 'regex', 'contains', 'replace',
+        'upper', 'lower', 'strip', 'lstrip', 'rstrip', 'dt', 'copy',
     }
     converted = _substitute_fields(converted, df_var, skip_words)
 
@@ -904,6 +910,8 @@ def _vec_recursive(expr, df_var):
     converted = re.sub(r'<>', '!=', converted)
     converted = re.sub(r'(?<![<>!=])=(?!=)', '==', converted)
     converted = re.sub(r'\berrors\s*==\s*(["\'])', r'errors=\1', converted)
+    converted = re.sub(r'\bexpand\s*==\s*', 'expand=', converted)
+    converted = re.sub(r'\bregex\s*==\s*', 'regex=', converted)
 
     converted = re.sub(r'\s+', ' ', converted).strip()
 
@@ -1044,8 +1052,14 @@ def _vectorize_simple(part, df_var):
         'True', 'False', 'None', 'and', 'or', 'not', 'np', 'pd',
         'str', 'int', 'float', 'isna', 'notna', 'fillna',
         'get_variable', 'lookup_func', 'isin', 'eq',
+        'expand', 'extract', 'astype', 'errors', 'coerce', 'regex',
+        'contains', 'replace', 'upper', 'lower', 'strip', 'lstrip', 'rstrip',
+        'dt', 'len', 'copy', 'abs', 'round', 'where', 'bool',
     }
     c = _substitute_fields(c, df_var, skip_words)
+    c = re.sub(r'\bexpand\s*==\s*', 'expand=', c)
+    c = re.sub(r'\berrors\s*==\s*', 'errors=', c)
+    c = re.sub(r'\bregex\s*==\s*', 'regex=', c)
 
     return c
 

{informatica_python-1.9.3 → informatica_python-1.9.4}/informatica_python.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: informatica-python
-Version: 1.9.3
+Version: 1.9.4
 Summary: Convert Informatica PowerCenter workflow XML to Python/PySpark code
 Author: Nick
 License: MIT
@@ -430,7 +430,7 @@ The generated `helper_functions.py` provides a complete runtime library:
 - **Generated code formatting**: Consistent `# ---` section headers for Source Qualifiers, Transforms, and Target Writes; metadata comments (database type, field lists); column mapping and write operation comments; clean blank line handling
 - **Source/target detection**: Case-insensitive instance type matching
 - **Session→mapping inference**: Longest-suffix-match strategy for ambiguous mapping names
-- **646 tests** across unit, integration, expression, and formatting test suites
+- **663 tests** across unit, integration, expression, and formatting test suites
 
 ### v1.9.2 (Phase 8)
 - Mapping output files now use real mapping names (e.g., `mapping_m_customer_load.py`) instead of generic numeric indices (`mapping_1.py`)
@@ -495,7 +495,7 @@ The generated `helper_functions.py` provides a complete runtime library:
 cd informatica_python
 pip install -e ".[dev]"
 
-# Run tests (646 tests)
+# Run tests (663 tests)
 pytest tests/ -v
 ```
 

{informatica_python-1.9.3 → informatica_python-1.9.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "informatica-python"
-version = "1.9.3"
+version = "1.9.4"
 description = "Convert Informatica PowerCenter workflow XML to Python/PySpark code"
 readme = "README.md"
 license = {text = "MIT"}

{informatica_python-1.9.3 → informatica_python-1.9.4}/tests/test_integration.py

@@ -2246,3 +2246,242 @@ class TestJoinerFieldRemapping(unittest.TestCase):
             if "left_on" in line and "right_on" in line:
                 assert "Table_Name" in line, \
                     "Merge should use source column name Table_Name"
+
+
+class TestRegExtractConversion(unittest.TestCase):
+    """Tests for REG_EXTRACT capture group and expand parameter handling."""
+
+    def test_no_double_capture_group(self):
+        r = convert_expression_vectorized(r"REG_EXTRACT(col,'(\s+)')", "df")
+        assert r.count("(") - r.count("str.extract") <= 2
+        assert '((\\s+))' not in r
+
+    def test_adds_capture_group_when_missing(self):
+        r = convert_expression_vectorized(r"REG_EXTRACT(col,'\\d+')", "df")
+        assert 'expand=False' in r
+        assert '.str.extract' in r
+
+    def test_expand_is_boolean_not_series(self):
+        r = convert_expression_vectorized(r"REG_EXTRACT(col,'(\s+)')", "df")
+        assert 'expand=False' in r
+        assert 'expand==False' not in r
+        assert 'df["expand"]' not in r
+
+    def test_isnull_reg_extract_nested(self):
+        r = convert_expression_vectorized(
+            "IIF(ISNULL(REG_EXTRACT(PART_BIRTH_DTE,'(\\s+)')),PART_BIRTH_DTE,NULL)", "df_exp"
+        )
+        assert "np.where" in r
+        assert ".isna()" in r
+        assert "expand=False" in r
+        assert 'expand==False' not in r
+        assert 'df_exp["expand"]' not in r
+
+
+class TestDatetimeFormatMask(unittest.TestCase):
+    """Tests for datetime format mask conversion (US/microseconds)."""
+
+    def test_us_to_percent_f(self):
+        from informatica_python.utils.expression_converter import _convert_infa_date_format
+        fmt = _convert_infa_date_format("YYYY-MM-DD HH24.MI.SS.US")
+        assert "%f" in fmt
+        assert "US" not in fmt
+
+    def test_full_format_mask(self):
+        from informatica_python.utils.expression_converter import _convert_infa_date_format
+        fmt = _convert_infa_date_format("YYYY-MM-DD HH24:MI:SS")
+        assert fmt == "%Y-%m-%d %H:%M:%S"
+
+    def test_to_date_with_us_format(self):
+        r = convert_expression_vectorized(
+            "TO_DATE(x, 'YYYY-MM-DD HH24.MI.SS.US')", "df"
+        )
+        assert "%f" in r
+        assert "US" not in r
+
+
+class TestSubstrZeroIndex(unittest.TestCase):
+    """Tests for SUBSTR with 0-based start position."""
+
+    def test_substr_start_0(self):
+        r = convert_expression_vectorized("SUBSTR(x, 0, 11)", "df")
+        assert "str[0:" in r
+        assert "str[-1:" not in r
+
+    def test_substr_start_1(self):
+        r = convert_expression_vectorized("SUBSTR(x, 1, 5)", "df")
+        assert "str[0:" in r
+
+    def test_substr_start_5(self):
+        r = convert_expression_vectorized("SUBSTR(x, 5, 3)", "df")
+        assert "str[4:7]" in r
+
+
+class TestStringOpSafety(unittest.TestCase):
+    """Tests for string operations adding .astype(str) for safety."""
+
+    def test_ltrim_has_astype_str(self):
+        r = convert_expression_vectorized("LTRIM(name)", "df")
+        assert ".astype(str)" in r
+        assert ".str.lstrip()" in r
+
+    def test_rtrim_has_astype_str(self):
+        r = convert_expression_vectorized("RTRIM(name)", "df")
+        assert ".astype(str)" in r
+        assert ".str.rstrip()" in r
+
+    def test_trim_has_astype_str(self):
+        r = convert_expression_vectorized("TRIM(name)", "df")
+        assert ".astype(str)" in r
+        assert ".str.strip()" in r
+
+    def test_ltrim_with_char(self):
+        r = convert_expression_vectorized("LTRIM(name, '0')", "df")
+        assert ".astype(str)" in r
+        assert '.str.lstrip("0")' in r
+
+
+class TestRouterVectorized(unittest.TestCase):
+    """Tests for Router transformation generating vectorized conditions."""
+
+    ROUTER_XML = '''<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE POWERMART SYSTEM "powrmart.dtd">
+<POWERMART CREATION_DATE="01/01/2025" REPOSITORY_VERSION="1">
+<REPOSITORY NAME="repo" VERSION="1" CODEPAGE="UTF-8" DATABASETYPE="Oracle">
+<FOLDER NAME="TEST" OWNER="admin">
+  <SOURCE NAME="SRC" DATABASETYPE="Flat File" DBDNAME="SRC">
+    <FLATFILE DELIMITEDBY="COMMA" HEADERROWPRESENT="YES" PADBYTES="NO" ROWDELIMITER="\\n"/>
+    <SOURCEFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" NULLABLE="NOTNULL" KEYTYPE="PRIMARY KEY" FIELDNUMBER="1"/>
+    <SOURCEFIELD NAME="STATUS" DATATYPE="string" PRECISION="20" SCALE="0" NULLABLE="NULL" KEYTYPE="NOT A KEY" FIELDNUMBER="2"/>
+  </SOURCE>
+  <TARGET NAME="TGT" DATABASETYPE="Flat File">
+    <TARGETFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" NULLABLE="NULL" KEYTYPE="NOT A KEY" FIELDNUMBER="1"/>
+  </TARGET>
+  <MAPPING NAME="m_router_test" ISVALID="YES">
+    <TRANSFORMATION NAME="SQ_SRC" TYPE="Source Qualifier" REUSABLE="NO">
+      <TRANSFORMFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" PORTTYPE="OUTPUT"/>
+      <TRANSFORMFIELD NAME="STATUS" DATATYPE="string" PRECISION="20" SCALE="0" PORTTYPE="OUTPUT"/>
+    </TRANSFORMATION>
+    <TRANSFORMATION NAME="RTR_STATUS" TYPE="Router" REUSABLE="NO">
+      <TRANSFORMFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" PORTTYPE="INPUT/OUTPUT"/>
+      <TRANSFORMFIELD NAME="STATUS" DATATYPE="string" PRECISION="20" SCALE="0" PORTTYPE="INPUT/OUTPUT"/>
+      <TABLEATTRIBUTE NAME="Group Filter Condition_ACTIVE" VALUE="STATUS = 'ACTIVE'"/>
+      <TABLEATTRIBUTE NAME="Group Filter Condition_INACTIVE" VALUE="STATUS = 'INACTIVE'"/>
+    </TRANSFORMATION>
+    <INSTANCE NAME="SRC" TYPE="Source Definition" TRANSFORMATION_NAME="SRC"/>
+    <INSTANCE NAME="SQ_SRC" TYPE="Source Qualifier" TRANSFORMATION_NAME="SQ_SRC"/>
+    <INSTANCE NAME="RTR_STATUS" TYPE="Router" TRANSFORMATION_NAME="RTR_STATUS"/>
+    <INSTANCE NAME="TGT" TYPE="Target Definition" TRANSFORMATION_NAME="TGT"/>
+    <CONNECTOR FROMINSTANCE="SRC" FROMFIELD="ID" TOINSTANCE="SQ_SRC" TOFIELD="ID"/>
+    <CONNECTOR FROMINSTANCE="SRC" FROMFIELD="STATUS" TOINSTANCE="SQ_SRC" TOFIELD="STATUS"/>
+    <CONNECTOR FROMINSTANCE="SQ_SRC" FROMFIELD="ID" TOINSTANCE="RTR_STATUS" TOFIELD="ID"/>
+    <CONNECTOR FROMINSTANCE="SQ_SRC" FROMFIELD="STATUS" TOINSTANCE="RTR_STATUS" TOFIELD="STATUS"/>
+    <CONNECTOR FROMINSTANCE="RTR_STATUS" FROMFIELD="ID" TOINSTANCE="TGT" TOFIELD="ID"/>
+  </MAPPING>
+  <CONFIG NAME="default_session_config"/>
+  <WORKFLOW NAME="wf_router_test" ISVALID="YES">
+    <TASK NAME="Start" REUSABLE="NO" TYPE="Start"/>
+    <SESSION NAME="s_m_router_test" ISVALID="YES" REUSABLE="NO" MAPPINGNAME="m_router_test">
+      <CONFIGREFERENCE REFOBJECTNAME="default_session_config" TYPE="Session config"/>
+    </SESSION>
+    <TASKINSTANCE NAME="Start" TASKNAME="Start" TASKTYPE="Start"/>
+    <TASKINSTANCE NAME="s_m_router_test" TASKNAME="s_m_router_test" TASKTYPE="Session"/>
+    <WORKFLOWLINK FROMTASK="Start" TOTASK="s_m_router_test"/>
+  </WORKFLOW>
+</FOLDER>
+</REPOSITORY>
+</POWERMART>'''
+
+    def test_router_generates_group_filters(self):
+        converter = InformaticaConverter()
+        tmpdir = tempfile.mkdtemp()
+        try:
+            converter.convert_string(self.ROUTER_XML, output_dir=tmpdir)
+            for fn in os.listdir(tmpdir):
+                if fn.startswith("mapping_") and fn.endswith(".py"):
+                    with open(os.path.join(tmpdir, fn)) as f:
+                        code = f.read()
+                    assert "_router_mask_" in code or "group0" in code, \
+                        "Router should generate group filter masks"
+                    assert "Default group" in code
+                    break
+        finally:
+            shutil.rmtree(tmpdir)
+
+    def test_router_default_excludes_matched_rows(self):
+        converter = InformaticaConverter()
+        tmpdir = tempfile.mkdtemp()
+        try:
+            converter.convert_string(self.ROUTER_XML, output_dir=tmpdir)
+            for fn in os.listdir(tmpdir):
+                if fn.startswith("mapping_") and fn.endswith(".py"):
+                    with open(os.path.join(tmpdir, fn)) as f:
+                        code = f.read()
+                    assert "_router_default_mask" in code or "~" in code, \
+                        "Default group should exclude rows matching other groups"
+                    break
+        finally:
+            shutil.rmtree(tmpdir)
+
+
+class TestLookupWarning(unittest.TestCase):
+    """Tests for lookup empty DataFrame warning."""
+
+    LOOKUP_XML = '''<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE POWERMART SYSTEM "powrmart.dtd">
+<POWERMART CREATION_DATE="01/01/2025" REPOSITORY_VERSION="1">
+<REPOSITORY NAME="repo" VERSION="1" CODEPAGE="UTF-8" DATABASETYPE="Oracle">
+<FOLDER NAME="TEST" OWNER="admin">
+  <SOURCE NAME="SRC" DATABASETYPE="Flat File" DBDNAME="SRC">
+    <FLATFILE DELIMITEDBY="COMMA" HEADERROWPRESENT="YES" PADBYTES="NO" ROWDELIMITER="\\n"/>
+    <SOURCEFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" NULLABLE="NOTNULL" KEYTYPE="PRIMARY KEY" FIELDNUMBER="1"/>
+  </SOURCE>
+  <TARGET NAME="TGT" DATABASETYPE="Flat File">
+    <TARGETFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" NULLABLE="NULL" KEYTYPE="NOT A KEY" FIELDNUMBER="1"/>
+  </TARGET>
+  <MAPPING NAME="m_lkp_test" ISVALID="YES">
+    <TRANSFORMATION NAME="SQ_SRC" TYPE="Source Qualifier" REUSABLE="NO">
+      <TRANSFORMFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" PORTTYPE="OUTPUT"/>
+    </TRANSFORMATION>
+    <TRANSFORMATION NAME="LKP_TEST" TYPE="Lookup Procedure" REUSABLE="NO">
+      <TRANSFORMFIELD NAME="ID" DATATYPE="integer" PRECISION="10" SCALE="0" PORTTYPE="INPUT"/>
+      <TRANSFORMFIELD NAME="RESULT" DATATYPE="string" PRECISION="100" SCALE="0" PORTTYPE="OUTPUT/RETURN"/>
+      <TABLEATTRIBUTE NAME="Lookup table name" VALUE="DIM_TABLE"/>
+      <TABLEATTRIBUTE NAME="Lookup condition" VALUE="ID = ID"/>
+    </TRANSFORMATION>
+    <INSTANCE NAME="SRC" TYPE="Source Definition" TRANSFORMATION_NAME="SRC"/>
+    <INSTANCE NAME="SQ_SRC" TYPE="Source Qualifier" TRANSFORMATION_NAME="SQ_SRC"/>
+    <INSTANCE NAME="LKP_TEST" TYPE="Lookup Procedure" TRANSFORMATION_NAME="LKP_TEST"/>
+    <INSTANCE NAME="TGT" TYPE="Target Definition" TRANSFORMATION_NAME="TGT"/>
+    <CONNECTOR FROMINSTANCE="SRC" FROMFIELD="ID" TOINSTANCE="SQ_SRC" TOFIELD="ID"/>
+    <CONNECTOR FROMINSTANCE="SQ_SRC" FROMFIELD="ID" TOINSTANCE="LKP_TEST" TOFIELD="ID"/>
+    <CONNECTOR FROMINSTANCE="LKP_TEST" FROMFIELD="RESULT" TOINSTANCE="TGT" TOFIELD="ID"/>
+  </MAPPING>
+  <CONFIG NAME="default_session_config"/>
+  <WORKFLOW NAME="wf_lkp_test" ISVALID="YES">
+    <TASK NAME="Start" REUSABLE="NO" TYPE="Start"/>
+    <SESSION NAME="s_m_lkp_test" ISVALID="YES" REUSABLE="NO" MAPPINGNAME="m_lkp_test">
+      <CONFIGREFERENCE REFOBJECTNAME="default_session_config" TYPE="Session config"/>
+    </SESSION>
+    <TASKINSTANCE NAME="Start" TASKNAME="Start" TASKTYPE="Start"/>
+    <TASKINSTANCE NAME="s_m_lkp_test" TASKNAME="s_m_lkp_test" TASKTYPE="Session"/>
+    <WORKFLOWLINK FROMTASK="Start" TOTASK="s_m_lkp_test"/>
+  </WORKFLOW>
+</FOLDER>
+</REPOSITORY>
+</POWERMART>'''
+
+    def test_lookup_with_table_reads_from_db(self):
+        converter = InformaticaConverter()
+        tmpdir = tempfile.mkdtemp()
+        try:
+            converter.convert_string(self.LOOKUP_XML, output_dir=tmpdir)
+            for fn in os.listdir(tmpdir):
+                if fn.startswith("mapping_") and fn.endswith(".py"):
+                    with open(os.path.join(tmpdir, fn)) as f:
+                        code = f.read()
+                    assert "read_from_db" in code, "Lookup with table should use read_from_db"
+                    assert "DIM_TABLE" in code
+                    break
+        finally:
+            shutil.rmtree(tmpdir)