src.dackar.RCA.doc_parsers.pdfParser

Attributes

PdfConverter
LOGGER
handler
_DOC_TYPE_PATTERNS
_EARLY_TEXT_PATTERNS
home

Functions

infer_doc_type(filename[, early_text])	Infer doc_type from filename first, then first ~500 chars of text.
_now_iso()
_clean_table_cell(x)
_clean_table_obj(t)
_estimate_text_extraction_quality(markdown_text, ...)	Lightweight extraction-quality heuristic for Stage 1 provenance.
get_file_name_no_extension(path)	Return filename without extension from a path.
_ensure_dir(path)
_parse_marker_image_key(image_key)	Marker images are commonly saved with names like '_page_12_Picture_3.jpeg'.
_normalize_table_markdown_to_json(md_text)	Convert Marker-produced Markdown tables into a normalized JSON structure:
_tables_from_pdfplumber(filename)	Extract tables using pdfplumber and normalize to JSON objects.
pdfParser([home_folder, red_filepath, ...])	Parse a PDF file into:

Module Contents

src.dackar.RCA.doc_parsers.pdfParser.PdfConverter = None

src.dackar.RCA.doc_parsers.pdfParser.LOGGER

src.dackar.RCA.doc_parsers.pdfParser.handler

src.dackar.RCA.doc_parsers.pdfParser._DOC_TYPE_PATTERNS

src.dackar.RCA.doc_parsers.pdfParser._EARLY_TEXT_PATTERNS

src.dackar.RCA.doc_parsers.pdfParser.infer_doc_type(filename, early_text='')

Infer doc_type from filename first, then first ~500 chars of text. Returns one of: CR | WO | SOP | ECA | OTHER

Parameters:

• filename (str)

• early_text (str)

Return type:

str

src.dackar.RCA.doc_parsers.pdfParser._now_iso()

Return type:

str

src.dackar.RCA.doc_parsers.pdfParser._clean_table_cell(x)

Parameters:

x (Any)

Return type:

str

src.dackar.RCA.doc_parsers.pdfParser._clean_table_obj(t)

Parameters:

t (Dict[str, Any])

Return type:

Dict[str, Any]

src.dackar.RCA.doc_parsers.pdfParser._estimate_text_extraction_quality(markdown_text, page_count)

Lightweight extraction-quality heuristic for Stage 1 provenance.

Parameters:

• markdown_text (str)

• page_count (Optional[int])

Return type:

Dict[str, Any]

src.dackar.RCA.doc_parsers.pdfParser.get_file_name_no_extension(path)

Return filename without extension from a path.

Parameters:

path (str)

Return type:

str

src.dackar.RCA.doc_parsers.pdfParser._ensure_dir(path)

Parameters:

path (str)

Return type:

None

src.dackar.RCA.doc_parsers.pdfParser._parse_marker_image_key(image_key)

Marker images are commonly saved with names like ‘_page_12_Picture_3.jpeg’. Extract page/index if present; otherwise return None values.

Parameters:

image_key (str)

Return type:

Dict[str, Any]

src.dackar.RCA.doc_parsers.pdfParser._normalize_table_markdown_to_json(md_text)

Convert Marker-produced Markdown tables into a normalized JSON structure: [

{ “columns”: […], “rows”: [[…], …], “caption”: None, “page”: None }

] Assumes tables are separated in MD by blank lines and follow pipe ‘|’ syntax.

Parameters:

md_text (str)

Return type:

List[Dict[str, Any]]

src.dackar.RCA.doc_parsers.pdfParser._tables_from_pdfplumber(filename)

Extract tables using pdfplumber and normalize to JSON objects.

Parameters:

filename (str)

Return type:

List[Dict[str, Any]]

src.dackar.RCA.doc_parsers.pdfParser.pdfParser(home_folder=None, red_filepath=None, destination_folder=None, text2markdown='marker', tableParser='marker', classification='internal', ingest_id=None, source_path=None)

Parse a PDF file into:

• text markdown

• figures (images)

• tables (normalized JSON)

Emit:

• document_index.json (paths + metadata)

• text.md

• tables.json

• figures/*.jpeg

• metadata.json (Marker-render metadata when available)

Returns:

• index dict mirroring document_index.json

Parameters:

• home_folder (Optional[str]) – Base folder for relative path red_filepath.

• red_filepath (Optional[str]) – File path relative to home_folder (or absolute if home_folder is None).

• destination_folder (Optional[str]) – Root destination for parsed outputs.

• text2markdown (str) – Only ‘marker’ supported for now.

• tableParser (str) – ‘marker’ or ‘pdfplumber’.

• classification (str) – Data classification tag.

• ingest_id (Optional[str])

• source_path (Optional[str])

Returns:

Document index with artifact paths and rich metadata.

Return type:

Dict[str, Any]

Raises:

• FileNotFoundError – If input or destination paths are invalid.

• ValueError – For unsupported options.

src.dackar.RCA.doc_parsers.pdfParser.home = None