Finish processing logs

[ctengel]

7. look through attempt files /logs.... a. look at url and find urls that aren't in DB summary (either vid URL, playlist URL, etc) b. deeper analysis of data...reconstruct playlists, etc

8. worker logs a. quieter b. success dl vs failed

9. flask logs anything???

then we need to somehow dump all this in a way that it can be stored by higher level maths

[ctengel]

ability to combine this with other log/retries (see journal)

[ctengel]

• users
• playlists
• failures

[ctengel]

Stick with flat files now for output - DB later

[ctengel]

Next steps:

1. Limit to -8 0 +8 seconds - 10 second resolution
2. output useful info from readlogs.py (or skip to 3-5?)
   • output error videos (retry)
   • output error playlist (retry/refresh - w/ ignore errors? #16 )
   • output success playlist (refresh)
   • track successful retries at later time - i.e. don't include as failed if it succeeded later
3. Output CSV from readlogs.py
4. Somehow combine raw, CSV, or combined output of readlogs.py and readdb.py
5. Combined, sane way to have human verify and refeed to retryer refresher

[ctengel]

don't count an ignoreerrors #16 run as "successful"

[ctengel]

From #4:

Cause #15 and related tools to output backend for said file

i.e. output backends where relevant into retry/refresh files etc and definetely csvs