fix: exclude __words__ keyword index from corruption detection and getStats()

The __words__ keyword index stores 50-5000 entries per entity (one per
word), which inflated avg entries/entity well above the corruption
threshold of 100. This caused:

1. validateConsistency() to falsely detect corruption on every startup,
   triggering unnecessary clearAllIndexData() + rebuild() cycles
2. getStats() to log false "Metadata index may be corrupted" warnings
   and report inflated totalEntries/totalIds stats

Both methods now skip __words__ when counting, so stats and health
checks reflect metadata fields only (noun, type, createdAt, etc.).
Keyword search is unaffected since the __words__ field index itself
is not modified.
This commit is contained in:
David Snelling 2026-01-27 15:38:21 -08:00
parent 32dbdcec61
commit 364360d447
128 changed files with 5637 additions and 5682 deletions

View file

@ -21,7 +21,7 @@ let brainyInstance: Brainy | null = null
const getBrainy = async (): Promise<Brainy> => {
if (!brainyInstance) {
brainyInstance = new Brainy()
await brainyInstance.init() // v5.0.1: Initialize brain (VFS auto-initialized here!)
await brainyInstance.init() // Initialize brain (VFS auto-initialized here!)
}
return brainyInstance
}
@ -52,8 +52,8 @@ export const vfsCommands = {
const spinner = ora('Reading file...').start()
try {
const brain = await getBrainy() // v5.0.1: Await async getBrainy
// v5.0.1: VFS auto-initialized, no need for vfs.init()
const brain = await getBrainy() // Await async getBrainy
// VFS auto-initialized, no need for vfs.init()
const buffer = await brain.vfs.readFile(path, {
encoding: options.encoding as any
})