v0.2.0 — Reference File Support

What's new

graphify-sf can now detect and index non-Salesforce reference files that co-exist in your SFDX repo — READMEs, specs, architecture docs, spreadsheets, and PDFs appear as nodes in the knowledge graph alongside your SF metadata.

Reference file support

• Documents (.md, .mdx, .txt, .rst, .html) — headings extracted as sub-nodes; Salesforce component name mentions create references edges back to SF nodes
• PDFs (.pdf) — text extracted via pypdf; SF name mentions detected
• Spreadsheets (.xlsx) — structural nodes: workbook → sheet → named table → column headers; content converted to markdown sidecar
• Word documents (.docx) — converted to markdown sidecar via python-docx, then processed as document
• Images (.png, .jpg, .jpeg, .gif, .webp, .svg) — metadata node, no text extraction

Other changes

• Office file sidecar conversion to graphify-sf-out/converted/ (stable SHA-256 filename)
• New optional extra: pip install graphify-sf[docs] installs pypdf, python-docx, openpyxl
• Graceful degradation: missing optional libraries skip files with a warning, never crash
• detect() returns new doc_files key; detect_incremental() tracks doc file changes
• 23 new tests; 203/203 passing

Installation

pip install graphify-sf==0.2.0

# With document support (PDF, docx, xlsx):
pip install "graphify-sf[docs]==0.2.0"

Full changelog

See CHANGELOG.md for complete history.