Sangeetha-Grantha

Metadata	Value
Status	Completed
Version	1.0.0
Last Updated	2026-09-10
Author	Sangeetha Grantha Team
Document Type	Evidence record

Guru Guha Blog Source Adapter

---

[!NOTE] Historical evidence: results, counts, commands, and observations below belong to the original work described here. The editorial update date is not a new test or corpus verification. For present behavior, use current feature map.

Purpose

Fix the guru-guha.blogspot.com extraction pipeline end-to-end so that all 26+ krithis imported from this source have correct structural sections (PALLAVI, ANUPALLAVI, CHARANAM) and clean lyrics — no metadata pollution, no navigation cruft, no URLs in lyric text.

Root Causes

1. False metadata boundary: Structure parser’s “Meaning” pattern matched “Meaning of Kriti” navigation links at offset 48, truncating the entire lyric window
2. Navigation table pollution: Language-switch <table> elements extracted as lyric content
3. URL noise in lyrics: Fragment links rendered with full URLs
4. Stale sections not upgraded: Re-extraction didn’t replace all-OTHER sections from bad extractions
5. Retry counter not reset: retry() reset status to PENDING but left attempts at max

Implementation Details

Python Extraction Worker Fixes

• Strip navigation <table> elements with language-switch links
• Skip navigation boilerplate links (Meaning of Kriti, Notation, etc.)
• Omit URLs for fragment/same-page links in lyric content
• Negative lookaheads on MEANING/NOTES boundary patterns
• Minimum-content guard (skip early boundaries < 200 chars)

Backend Fixes

• LyricVariantPersistenceService: Upgrade stale all-OTHER sections on re-extraction; backfill includes MAPPED and IN_REVIEW statuses
• ImportRoutes: POST /re-extract endpoint with typed ReExtractResponse
• KrithiLyricRepository: deleteAllVariants() for clearing before re-extraction
• ExtractionQueueRepository: Reset attempts = 0 in retry() and retryAllFailed()

Code Changes

File	Change
tools/krithi-extract-enrich-worker/src/html_extractor.py	Navigation table stripping, link filtering
tools/krithi-extract-enrich-worker/src/structure_parser.py	Metadata boundary hardening, minimum-content guard
tools/krithi-extract-enrich-worker/src/worker.py	URL-based composer inference fallback
tools/krithi-extract-enrich-worker/tests/test_html_extractor.py	Guru-guha blog tests
modules/backend/api/.../services/LyricVariantPersistenceService.kt	Section upgrade, multi-status backfill
modules/backend/api/.../routes/ImportRoutes.kt	Re-extract endpoint, typed response DTOs
modules/backend/dal/.../repositories/KrithiLyricRepository.kt	deleteAllVariants()
modules/backend/dal/.../repositories/ExtractionQueueRepository.kt	Reset attempts in retry()
database/migrations/42__reset_guru_guha_extraction_attempts.sql	Reset extraction attempts for guru-guha entries
database/migrations/43__reset_guru_guha_to_pending.sql	Reset guru-guha queue entries to PENDING

Results

• All 26/26 guru-guha krithis have lyric variants and correct structural sections
• Clean lyrics with no navigation cruft or URL pollution
• Re-extraction pipeline fully functional with proper retry semantics

Ref: application_documentation/10-implementations/track-097-guru-guha-blog-source-adapter.md

---

Section index · Documentation home · Feature status