informatica-python 1.2.0__tar.gz → 1.2.1__tar.gz

{informatica_python-1.2.0 → informatica_python-1.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: informatica-python
-Version: 1.2.0
+Version: 1.2.1
 Summary: Convert Informatica PowerCenter workflow XML to Python/PySpark code
 License-Expression: MIT
 Requires-Python: >=3.8

{informatica_python-1.2.0 → informatica_python-1.2.1}/informatica_python/generators/mapping_gen.py

@@ -278,7 +278,7 @@ def _generate_transformation(lines, tx, connector_graph, source_dfs, transform_m
     elif tx_type == "sorter":
         _gen_sorter_transform(lines, tx, tx_safe, input_df, source_dfs)
     elif tx_type in ("joiner",):
-        _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_dfs)
+        _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_dfs, connector_graph)
     elif tx_type in ("lookup procedure", "lookup"):
         _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs)
     elif tx_type == "router":
@@ -410,7 +410,7 @@ def _gen_sorter_transform(lines, tx, tx_safe, input_df, source_dfs):
     source_dfs[tx.name] = f"df_{tx_safe}"
 
 
-def _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_dfs):
+def _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_dfs, connector_graph=None):
     join_type = "inner"
     join_condition = ""
     for attr in tx.attributes:
@@ -436,10 +436,31 @@ def _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_df
 
     left_keys, right_keys = parse_join_condition(join_condition)
 
+    master_src = None
+    detail_src = None
+    input_conns = connector_graph.get("to", {}).get(tx.name, []) if connector_graph else []
+    for conn in input_conns:
+        to_field = conn.to_field
+        if to_field in master_fields:
+            master_src = conn.from_instance
+        elif to_field in detail_fields:
+            detail_src = conn.from_instance
+
     src_list = list(input_sources)
-    if len(src_list) >= 2:
-        df_master = source_dfs.get(src_list[0], f"df_{_safe_name(src_list[0])}")
-        df_detail = source_dfs.get(src_list[1], f"df_{_safe_name(src_list[1])}")
+    if not master_src and not detail_src and len(src_list) >= 2:
+        master_src = src_list[0]
+        detail_src = src_list[1]
+    elif not master_src and len(src_list) >= 1:
+        master_src = src_list[0]
+    if not detail_src:
+        for s in src_list:
+            if s != master_src:
+                detail_src = s
+                break
+
+    if master_src and detail_src:
+        df_master = source_dfs.get(master_src, f"df_{_safe_name(master_src)}")
+        df_detail = source_dfs.get(detail_src, f"df_{_safe_name(detail_src)}")
 
         lines.append(f"    # Join ({join_type}): {join_condition or 'auto'}")
         if left_keys and right_keys:
@@ -451,9 +472,7 @@ def _gen_joiner_transform(lines, tx, tx_safe, input_df, input_sources, source_df
             lines.append(f"        suffixes=('', '_master')")
             lines.append(f"    )")
         else:
-            common_cols = []
-            if master_fields and detail_fields:
-                common_cols = [f for f in detail_fields if f in master_fields]
+            common_cols = [f for f in detail_fields if f in master_fields]
             if common_cols:
                 lines.append(f"    df_{tx_safe} = {df_detail}.merge(")
                 lines.append(f"        {df_master},")
@@ -539,9 +558,13 @@ def _gen_lookup_transform(lines, tx, tx_safe, input_df, source_dfs):
 
         drop_cols = [k for k in lookup_keys if k not in input_keys]
         if drop_cols:
-            lines.append(f"    lkp_drop = [c for c in {drop_cols} if c in df_{tx_safe}.columns and c + '_lkp' not in df_{tx_safe}.columns]")
+            lines.append(f"    _lkp_drop = [c for c in {drop_cols} if c in df_{tx_safe}.columns]")
+            lines.append(f"    if _lkp_drop:")
+            lines.append(f"        df_{tx_safe} = df_{tx_safe}.drop(columns=_lkp_drop)")
 
         for rf in all_output_fields:
+            lines.append(f"    if '{rf.name}' not in df_{tx_safe}.columns:")
+            lines.append(f"        df_{tx_safe}['{rf.name}'] = None")
             if rf.default_value:
                 lines.append(f"    df_{tx_safe}['{rf.name}'] = df_{tx_safe}['{rf.name}'].fillna({repr(rf.default_value)})")
     else:

{informatica_python-1.2.0 → informatica_python-1.2.1}/informatica_python/utils/expression_converter.py

@@ -95,11 +95,11 @@ INFA_FUNC_MAP = {
 }
 
 
-AGG_FUNC_NAMES = {
-    "SUM", "COUNT", "AVG", "MAX", "MIN", "MEDIAN",
-    "STDDEV", "VARIANCE", "PERCENTILE", "FIRST", "LAST",
-    "MOVINGAVG", "MOVINGSUM", "CUME",
-}
+AGG_FUNC_NAMES = [
+    "MOVINGAVG", "MOVINGSUM", "PERCENTILE", "VARIANCE",
+    "STDDEV", "MEDIAN", "COUNT", "FIRST", "LAST",
+    "CUME", "SUM", "AVG", "MAX", "MIN",
+]
 
 
 def convert_expression(expr):
@@ -131,6 +131,8 @@ def convert_expression(expr):
 
     converted = re.sub(r'<>', '!=', converted)
 
+    converted = re.sub(r'(?<![<>!])=(?!=)', '==', converted)
+
     converted = re.sub(r':LKP\.(\w+)\(', r'lookup_func("\1", ', converted)
 
     converted = re.sub(r'\$\$(\w+)', r'get_variable("\1")', converted)
@@ -202,8 +204,11 @@ def parse_aggregate_expression(expr):
     cleaned = expr.strip()
 
     for func_name in AGG_FUNC_NAMES:
-        pattern = re.compile(r'\b' + func_name + r'\s*\(\s*([^)]*)\s*\)', re.IGNORECASE)
-        match = pattern.search(cleaned)
+        pattern = re.compile(
+            r'^\s*' + func_name + r'\s*\(\s*([A-Za-z_][A-Za-z0-9_]*|\*)\s*\)\s*$',
+            re.IGNORECASE
+        )
+        match = pattern.match(cleaned)
         if match:
             col = match.group(1).strip()
             return func_name.lower(), col

{informatica_python-1.2.0 → informatica_python-1.2.1}/informatica_python.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: informatica-python
-Version: 1.2.0
+Version: 1.2.1
 Summary: Convert Informatica PowerCenter workflow XML to Python/PySpark code
 License-Expression: MIT
 Requires-Python: >=3.8

{informatica_python-1.2.0 → informatica_python-1.2.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "informatica-python"
-version = "1.2.0"
+version = "1.2.1"
 description = "Convert Informatica PowerCenter workflow XML to Python/PySpark code"
 readme = "README.md"
 license = "MIT"

{informatica_python-1.2.0 → informatica_python-1.2.1}/tests/test_converter.py

@@ -239,6 +239,8 @@ def test_expression_converter_expanded():
 
     result = convert_expression("IIF(STATUS = 'A', 'Active', 'Inactive')")
     assert "iif_expr" in result
+    assert "==" in result, f"Expected == in result, got: {result}"
+    assert "= =" not in result
 
     result = convert_expression("DECODE(TYPE, 1, 'One', 2, 'Two', 'Other')")
     assert "decode_expr" in result
@@ -301,6 +303,17 @@ def test_expression_converter_expanded():
     result = convert_expression("STATUS <> 'X'")
     assert "!=" in result
 
+    result = convert_expression("AMOUNT >= 100")
+    assert ">=" in result
+    assert ">==" not in result
+
+    result = convert_expression("AMOUNT <= 100")
+    assert "<=" in result
+    assert "<==" not in result
+
+    result = convert_expression("SUM(A)/COUNT(*)")
+    assert "sum_val" in result or "count_val" in result
+
     result = convert_expression("$$MY_VARIABLE")
     assert 'get_variable("MY_VARIABLE")' in result
 
@@ -405,6 +418,13 @@ def test_parse_aggregate_expression():
     assert func is None
     assert col is None
 
+    func, col = parse_aggregate_expression("SUM(A)/COUNT(*)")
+    assert func is None, f"Compound expression should not match, got func={func}"
+    assert col is None
+
+    func, col = parse_aggregate_expression("AVG(A+B)")
+    assert func is None, f"Expression with operators should not match, got func={func}"
+
     print("PASS: test_parse_aggregate_expression")
 
 
@@ -459,7 +479,7 @@ def test_generated_joiner_code():
     lines = []
     source_dfs = {"SRC_CUST": "df_src_cust", "SRC_ORDER": "df_src_order"}
     input_sources = {"SRC_CUST", "SRC_ORDER"}
-    _gen_joiner_transform(lines, tx, "jnr_cust_order", "df_src_cust", input_sources, source_dfs)
+    _gen_joiner_transform(lines, tx, "jnr_cust_order", "df_src_cust", input_sources, source_dfs, connector_graph=None)
     code = "\n".join(lines)
 
     assert "merge" in code