removed utils.parallelLimit

well, that was a pointless endeavor.
i wasn't thinking clearly.

also updated all scripts that previously used them, to use a combo of
for-loop and setInterval to print progress.
This commit is contained in:
Bobby Wibowo 2020-12-27 19:44:10 +07:00
parent ed27ad18b5
commit 0a19b025a0
No known key found for this signature in database
GPG Key ID: 51C3A1E1E22D26CF
4 changed files with 39 additions and 63 deletions

View File

@ -176,36 +176,6 @@ self.stripIndents = string => {
return result
}
self.parallelLimit = (promiseFactories, limit, progressCallback) => {
// https://stackoverflow.com/a/40377750
// CC BY-SA 3.0
const hasProgressCallback = typeof progressCallback === 'function'
const total = promiseFactories.length
const result = []
let cnt = 0
function chain (promiseFactories) {
if (!promiseFactories.length) return
const i = cnt++ // preserve order in result
if (hasProgressCallback) {
progressCallback({ done: cnt, total })
}
return promiseFactories.shift().then((res) => {
result[i] = res // save result
return chain(promiseFactories) // append next promise
})
}
const arrChains = []
while (limit-- > 0 && promiseFactories.length > 0) {
// create "limit" chains which run in parallel
arrChains.push(chain(promiseFactories))
}
// return when all arrChains are finished
return Promise.all(arrChains).then(() => result)
}
self.authorize = async (req, res) => {
// TODO: Improve usage of this function by the other APIs
const token = req.headers.token

View File

@ -38,7 +38,7 @@ const self = {
const mode = parseInt(args[0]) || 0
const dryrun = mode === 0
console.log('Querying uploads\u2026')
console.log('Querying and mapping uploads\u2026')
const uploads = await self.getFiles(paths.uploads)
console.log(`Uploads: ${uploads.length}`)

View File

@ -19,53 +19,57 @@ const db = require('knex')(config.database)
Rebuild file hashes.
Usage:
node ${location} <mode=0|1|2> [parallel]
node ${location} <mode=0|1|2>
mode:
0 = Dry run (recalculate hashes, print them, but do NOT store to DB).
1 = Recalculate hashes and store to DB.
2 = Verbose (recalculate hashes, print them, and store to DB).
parallel:
Amount of uploads to hash in parallel (not to be confused with multi-threading).
`).trim())
}
const dryrun = mode === 0
const verbose = [0, 2].includes(mode)
const parallel = Math.max(parseInt(args[1]), 1) || 1
console.log(`Parallel: ${parallel}`)
console.log('Querying uploads\u2026')
const hrstart = process.hrtime()
const uploads = await db.table('files')
.select('id', 'name', 'hash')
console.log(`Uploads : ${uploads.length}`)
let lastProgressOut
await utils.parallelLimit(uploads.map(upload => {
return new Promise((resolve, reject) => {
let done = 0
const printProgress = () => {
console.log(`PROGRESS: ${done}/${uploads.length}`)
if (done >= uploads.length) clearInterval(progressInterval)
}
const progressInterval = setInterval(printProgress, 1000)
printProgress()
for (const upload of uploads) {
await new Promise((resolve, reject) => {
fs.createReadStream(path.join(paths.uploads, upload.name))
.on('error', () => reject)
.on('error', reject)
.pipe(blake3.createHash())
.on('data', async hasher => {
const hash = hasher.toString('hex')
.on('error', reject)
.on('data', async source => {
const hash = source.toString('hex')
if (verbose) console.log(`${upload.name}: ${hash}`)
if (!dryrun && upload.hash !== hash) {
await db.table('files')
.update('hash', hash)
.where('id', upload.id)
}
done++
resolve()
})
}).catch(error => {
console.log(`${upload.name}: ${error.toString()}`)
})
}), parallel, progress => {
const now = Date.now()
if (!lastProgressOut || (now - lastProgressOut >= 1000) || progress.done === progress.total) {
console.log(`Progress: ${progress.done}/${progress.total}`)
lastProgressOut = now
}
})
}
clearInterval(progressInterval)
printProgress()
const hrend = process.hrtime(hrstart)
console.log(`Done in : ${(hrend[0] + (hrend[1] / 1e9)).toFixed(4)}s`)

View File

@ -31,7 +31,6 @@ const self = {
const force = parseInt(args[1]) || 0
const verbose = parseInt(args[2]) || 0
const cfcache = parseInt(args[3]) || 0
const parallel = Math.max(parseInt(args[1]), 1) || 1
if (![1, 2, 3].includes(self.mode) ||
![0, 1].includes(force) ||
@ -43,17 +42,15 @@ const self = {
Generate thumbnails.
Usage:
node ${location} <mode=1|2|3> [force=0|1] [verbose=0|1] [cfcache=0|1] [parallel]
node ${location} <mode=1|2|3> [force=0|1] [verbose=0|1] [cfcache=0|1]
mode : 1 = images only, 2 = videos only, 3 = both images and videos
force : 0 = no force (default), 1 = overwrite existing thumbnails
verbose : 0 = only print missing thumbs (default), 1 = print all, 2 = print nothing
cfcache : 0 = do not clear cloudflare cache (default), 1 = clear cloudflare cache
parallel: amount of thumbs to generate in parallel (not to be confused with multi-threading).
`).trim())
}
console.log(`Parallel: ${parallel}`)
console.log('Looking through existing thumbnails\u2026')
const uploads = await db.table('files')
.select('id', 'name')
@ -69,8 +66,16 @@ const self = {
let error = 0
let exists = 0
let skipped = 0
let lastProgressOut
await utils.parallelLimit(uploads.map(async upload => {
const printProgress = () => {
const done = succeeded.length + error + exists + skipped
console.log(`PROGRESS: ${done}/${uploads.length}`)
if (done >= uploads.length) clearInterval(progressInterval)
}
const progressInterval = setInterval(printProgress, 1000)
printProgress()
for (const upload of uploads) {
const extname = utils.extname(upload.name)
const basename = upload.name.slice(0, -extname.length)
@ -92,13 +97,10 @@ const self = {
}
generated ? succeeded.push({ upload, extname }) : error++
}
}), parallel, progress => {
const now = Date.now()
if (!lastProgressOut || (now - lastProgressOut >= 1000) || progress.done === progress.total) {
console.log(`Progress: ${progress.done}/${progress.total}`)
lastProgressOut = now
}
})
}
clearInterval(progressInterval)
printProgress()
console.log(utils.stripIndents(`
---