v3.17.0

v3.17.0

This release advances the listing-key-first dataset work and closes another substantial block of metadata gaps.

Highlights

• Added collision-safe core exports:
  • data/core_listings.csv
  • data/core_aliases.csv
  • matching JSON, Parquet, and SQLite tables
• Kept tickers.csv as the legacy compatibility layer while making listing_key the canonical core identity.
• Extended deterministic ETF-category backfills across KRX, SSE, SZSE, B3, and parts of NYSE ARCA.
• Added core_listings.csv support to the EODHD ISIN backfill path and accepted new Canada (TSX/TSXV) ISIN fills.

Current snapshot

• Primary tickers: 61,941
• Core listings: 54,173
• Listings: 71,092
• ISIN coverage: 56,175 (90.7%)
• Sector/category coverage: 50,898 (82.2%)
• ETF category coverage: 14,244
• Core primary rows missing ISIN: 4,535
• Legacy primary ticker collision rows: 1

Net improvements in this cycle

• missing_isin_primary: 4,578 -> 4,535
• missing_etf_category: 2,030 -> 1,360
• sector/category coverage: 50,268 -> 50,898
• ETF category coverage: 13,574 -> 14,244

Validation

• Release validation passed
• Test suite: 854 passed