# PP Data: #1293 + #1306 done (11 Oct 2026)

## #1293 tapes refreshed (analyst1)
- /opt/ppcopy/r6weth.db, r6nat.db, r6usd.db, swapamt.db rebuilt 11 Oct 00:10-00:25 UTC from the chainlogs tape: tape_upto_block 85,407,614 (about 11 Oct 00:10 UTC; the chain tape is now at head, b083-b085 done). Format unchanged (ev, meta, names, kv).
- Cron on research-box: daily2.sh at 02:10, 08:10, 14:10, 20:10 UTC (r6weth, r6nat, r6usd, names fetch+attach, swapamt, then pool_paths). So the tapes are at most about 6 h behind the chain, not hourly. The hooks=0 forward tape you asked for is the same r6weth/r6nat/r6usd ev table (kind 1 swap, 10 liquidity); no separate hourly file.
- Example forward OOS cut: select pool,t,a,q,px from ev where pool in (select pool from meta where lower(hooks)='0x0000000000000000000000000000000000000000') and t>=strftime('%s','2026-10-01').

## #1306 pool_paths (analyst2)
- /opt/sw/radar/pool_paths.jsonl (research-box) and http://159.223.88.168/hx-c9ois/fw/pool_paths.jsonl: 31,042 pools (was 30,031), last t_first 11 Oct ~00:00 UTC, refreshed every 6 h by the same cron (build at /opt/crew/work/ppdata/pp2/build.sh, copy of ppfeeds' pathbuild with the extra fields; ppfeeds' pathbuild.py itself untouched).
- New keys on every row: t300k_s, t1m_s, t3m_s (seconds after t_first of the first print with cap >= level, null = never) and depth_usd_t300k, depth_usd_t1m, depth_usd_t3m (virtual depth at that print, same formula as depth_usd_5m).
- Counts: t300k 7.2k pools, t1m 3.9k, t3m 2.1k; 3,166 pools jump 300k to 1M in under 30 s (a first-print jump or thin pool: filter on depth).
- Existing keys unchanged for pools in both versions (max_cap differs >1% for 50 of 28,945 only, late prints). cap_* now null only when the age is past the tape end.
- Caveat: glitch rule is the old one (cap > $20B dropped, 20x one-print spikes dropped); t*_s are mark prints, not tradable.