databricks · pietern · Oct 16, 2025 · Oct 15, 2025
diff --git a/...orkflow_with_ai_parse_document/.gitignore → ...rib/job_with_ai_parse_document/.gitignore b/...orkflow_with_ai_parse_document/.gitignore → ...rib/job_with_ai_parse_document/.gitignore
diff --git a/...workflow_with_ai_parse_document/README.md → contrib/job_with_ai_parse_document/README.md b/...workflow_with_ai_parse_document/README.md → contrib/job_with_ai_parse_document/README.md
@@ -96,7 +96,7 @@ Uses `ai_parse_document` to extract text, tables, and metadata from PDFs/images:
 
 Extracts clean concatenated text using `transform()`:
 - Reads from previous task's table via streaming
-- Handles both parser v1.0 and v2.0 formats
+- Extracts text from parsed document elements
 - Uses `transform()` for efficient text extraction
 - Includes error handling for failed parses
 

diff --git a/...arse_document/assets/document_summary.png → ...arse_document/assets/document_summary.png b/...arse_document/assets/document_summary.png → ...arse_document/assets/document_summary.png
diff --git a/...se_document/assets/figure_description.png → ...se_document/assets/figure_description.png b/...se_document/assets/figure_description.png → ...se_document/assets/figure_description.png
diff --git a/..._document/assets/page1_bounding_boxes.png → ..._document/assets/page1_bounding_boxes.png b/..._document/assets/page1_bounding_boxes.png → ..._document/assets/page1_bounding_boxes.png
diff --git a/...e_document/assets/page1_elements_list.png → ...e_document/assets/page1_elements_list.png b/...e_document/assets/page1_elements_list.png → ...e_document/assets/page1_elements_list.png
diff --git a/..._document/assets/page2_contents_table.png → ..._document/assets/page2_contents_table.png b/..._document/assets/page2_contents_table.png → ..._document/assets/page2_contents_table.png
diff --git a/...arse_document/assets/table_extraction.png → ...arse_document/assets/table_extraction.png b/...arse_document/assets/table_extraction.png → ...arse_document/assets/table_extraction.png
diff --git a/...low_with_ai_parse_document/databricks.yml → ...job_with_ai_parse_document/databricks.yml b/...low_with_ai_parse_document/databricks.yml → ...job_with_ai_parse_document/databricks.yml
diff --git a/...t/resources/ai_parse_document_job.job.yml → ...t/resources/ai_parse_document_job.job.yml b/...t/resources/ai_parse_document_job.job.yml → ...t/resources/ai_parse_document_job.job.yml
diff --git a/...ions/ai_parse_document -- debug output.py → ...ions/ai_parse_document -- debug output.py b/...ions/ai_parse_document -- debug output.py → ...ions/ai_parse_document -- debug output.py
diff --git a/...src/transformations/01_parse_documents.py → ...src/transformations/01_parse_documents.py b/...src/transformations/01_parse_documents.py → ...src/transformations/01_parse_documents.py
diff --git a/...nt/src/transformations/02_extract_text.py → ...nt/src/transformations/02_extract_text.py b/...nt/src/transformations/02_extract_text.py → ...nt/src/transformations/02_extract_text.py
@@ -47,11 +47,7 @@
                 "\n\n",
                 expr("""
                 transform(
-                    CASE
-                        WHEN try_cast(parsed:metadata:version AS STRING) = '1.0'
-                        THEN try_cast(parsed:document:pages AS ARRAY<VARIANT>)
-                        ELSE try_cast(parsed:document:elements AS ARRAY<VARIANT>)
-                    END,
+                    try_cast(parsed:document:elements AS ARRAY<VARIANT>),
                     element -> try_cast(element:content AS STRING)
                 )
             """),

diff --git a/...sformations/03_extract_structured_data.py → ...sformations/03_extract_structured_data.py b/...sformations/03_extract_structured_data.py → ...sformations/03_extract_structured_data.py