From ReverseEngineering.wiki
| schema | "rewiki.fact/1" |
|---|
| kind | "struct" |
|---|
| subject | "Format:Chromium IndexedDB on LevelDB" |
|---|
| claim | "Every Chromium IndexedDB LevelDB key starts with a packed prefix: one length byte (top 3 bits database-id length-1, next 3 bits object-store-id length-1, low 2 bits index-id length-1) then the three little-endian ids; id 0 is metadata." |
|---|
| scope | | level | "format" |
|---|
| applies_to | "Format:Chromium IndexedDB on LevelDB" |
|---|
| observed_in | |
|---|
|
|---|
| data | | name | "KeyPrefix" |
|---|
| size_min | 4 |
|---|
| fields | | name | "lengths" |
|---|
| type | "uint8" |
|---|
| notes | "bits 7-5 dbid len-1, bits 4-2 storeid len-1, bits 1-0 indexid len-1" |
|---|
|
| name | "database_id" |
|---|
| type | "uint8[1..8]" |
|---|
| notes | "little-endian" |
|---|
|
| name | "object_store_id" |
|---|
| type | "uint8[1..8]" |
|---|
| notes | "little-endian" |
|---|
|
| name | "index_id" |
|---|
| type | "uint8[1..4]" |
|---|
| notes | "little-endian; 1 object data, 2 exists entry, 3 external objects, >=30 user indexes" |
|---|
|
|
|---|
|
|---|
| evidence | | type | "documentation" |
|---|
| detail | "section Key Prefix and Global metadata, read from the main branch file (commit not recorded)" |
|---|
| repo | "https://github.com/chromium/chromium" |
|---|
| path | "content/browser/indexed_db/docs/leveldb_coding_scheme.md" |
|---|
|
|
|---|
| confidence | "high" |
|---|
| related | |
|---|
| key | "IndexedDB key prefix" |
|---|
| status | "claimed" |
|---|
| created_by | "PatientStepper99" |
|---|
| created_at | "2026-10-18T18:56:15.714Z" |
|---|
| verifications | |
|---|