feat: collapse redundant orchestration layers and retire pdf_table_extractor shim (RFC #19)

.gitignore

@@ -246,3 +246,6 @@ tmp/
 *.tgz
 *.rar
 *.7z
+
+# GSD planning artifacts
+.planning/

packages/parser-core/pyproject.toml

@@ -93,6 +93,7 @@ markers = [
 ]
 filterwarnings = [
     "ignore:TestResult has no addDuration method:RuntimeWarning",
+    "ignore::DeprecationWarning:bankstatements_core.pdf_table_extractor",
 ]
 
 [tool.coverage.run]

packages/parser-core/src/bankstatements_core/extraction/pdf_extractor.py

@@ -144,7 +144,9 @@ def _extract_page(self, page: Any, page_num: int) -> list[dict] | None:
         page_rows = self._row_builder.build_rows(words)
 
         if self.page_validation_enabled:
-            from bankstatements_core.pdf_table_extractor import validate_page_structure
+            from bankstatements_core.extraction.validation_facade import (
+                validate_page_structure,
+            )
 
             if not validate_page_structure(page_rows, self.columns):
                 logger.info(
@@ -153,7 +155,9 @@ def _extract_page(self, page: Any, page_num: int) -> list[dict] | None:
                 )
                 return None
 
-        from bankstatements_core.pdf_table_extractor import merge_continuation_lines
+        from bankstatements_core.extraction.validation_facade import (
+            merge_continuation_lines,
+        )
 
         return merge_continuation_lines(page_rows, self.columns)
 
@@ -184,7 +188,9 @@ def _determine_boundaries_and_extract(
             all_words = initial_area.extract_words(use_text_flow=True)
 
             if self.header_check_enabled:
-                from bankstatements_core.pdf_table_extractor import detect_table_headers
+                from bankstatements_core.extraction.validation_facade import (
+                    detect_table_headers,
+                )
 
                 header_top = (
                     header_check_top_y
@@ -198,7 +204,7 @@ def _determine_boundaries_and_extract(
                     logger.info(f"Page {page_num}: No table headers detected, skipping")
                     return None
 
-            from bankstatements_core.pdf_table_extractor import (
+            from bankstatements_core.extraction.extraction_facade import (
                 detect_table_end_boundary_smart,
             )
 
@@ -225,7 +231,9 @@ def _determine_boundaries_and_extract(
         words = table_area.extract_words(use_text_flow=True)
 
         if self.header_check_enabled:
-            from bankstatements_core.pdf_table_extractor import detect_table_headers
+            from bankstatements_core.extraction.validation_facade import (
+                detect_table_headers,
+            )
 
             header_top = (
                 header_check_top_y

packages/parser-core/src/bankstatements_core/facades/processing_facade.py

@@ -11,8 +11,8 @@
 from typing import TYPE_CHECKING, Any
 
 from bankstatements_core.config.app_config import AppConfig, ConfigurationError
+from bankstatements_core.config.column_config import get_columns_config
 from bankstatements_core.entitlements import EntitlementError, Entitlements
-from bankstatements_core.pdf_table_extractor import get_columns_config
 
 if TYPE_CHECKING:
     from bankstatements_core.processor import BankStatementProcessor

packages/parser-core/src/bankstatements_core/pdf_table_extractor.py

@@ -12,9 +12,20 @@
 from __future__ import annotations
 
 import logging
+import warnings
 
 import pdfplumber  # noqa: F401 - used by extraction module
 
+warnings.warn(
+    "bankstatements_core.pdf_table_extractor is a backward-compatibility shim "
+    "and will be removed in a future version. "
+    "Import directly from bankstatements_core.extraction.extraction_facade, "
+    "bankstatements_core.extraction.validation_facade, or "
+    "bankstatements_core.extraction.row_classification_facade instead.",
+    DeprecationWarning,
+    stacklevel=2,
+)
+
 logger = logging.getLogger(__name__)
 
 # Re-export column configuration (backward compatibility)