commit cold helper scripts
This commit is contained in:
localhost
parent
cee4ce06f2
commit
f938346008
|
|
@ -0,0 +1,262 @@
|
|||
import { getVideo } from '@/utils/metadata'
|
||||
import { db } from '@/utils/database'
|
||||
import * as fs from 'node:fs'
|
||||
|
||||
type FileEntry = {
|
||||
id: string
|
||||
size: number
|
||||
line: string
|
||||
}
|
||||
|
||||
type VideoStatus = {
|
||||
id: string
|
||||
size: number
|
||||
title: string | null
|
||||
stillUp: boolean
|
||||
}
|
||||
|
||||
const INPUT_PATH = 'ratios3.txt'
|
||||
const OUTPUT_PATH = 'colds3.txt'
|
||||
const CONCURRENCY = 2
|
||||
const checkOnlyChannelId = false
|
||||
|
||||
function parseOutput(text: string): FileEntry[] {
|
||||
const entries: FileEntry[] = []
|
||||
|
||||
for (const line of text.split('\n')) {
|
||||
const trimmed = line.trim()
|
||||
if (!trimmed) continue
|
||||
|
||||
const ratioEntry = parseRatioEntry(trimmed, line)
|
||||
if (ratioEntry) {
|
||||
entries.push(ratioEntry)
|
||||
continue
|
||||
}
|
||||
|
||||
const lsEntry = parseLsEntry(trimmed, line)
|
||||
if (lsEntry) {
|
||||
entries.push(lsEntry)
|
||||
}
|
||||
}
|
||||
|
||||
return entries
|
||||
}
|
||||
|
||||
function parseLsEntry(trimmed: string, line: string): FileEntry | null {
|
||||
const match = trimmed.match(
|
||||
/^[^\s]+\s+\d+\s+\S+\s+\S+\s+(\d+)\s+[A-Za-z]{3}\s+\d{1,2}\s+(?:\d{4}|\d{2}:\d{2})\s+(.+)$/
|
||||
)
|
||||
if (!match) return null
|
||||
|
||||
const [, sizeText, filename] = match
|
||||
const size = Number(sizeText)
|
||||
if (!Number.isFinite(size)) return null
|
||||
|
||||
const idMatch = filename.match(/^([^.]+)\.\w+$/)
|
||||
const id = idMatch?.[1]
|
||||
if (!id) return null
|
||||
|
||||
return { id, size, line }
|
||||
}
|
||||
|
||||
function parseRatioEntry(trimmed: string, line: string): FileEntry | null {
|
||||
const columns = trimmed.split('\t')
|
||||
if (columns.length < 7) return null
|
||||
|
||||
const [rankText, , , sizeText, , , filename] = columns
|
||||
if (!/^\d+$/.test(rankText)) return null
|
||||
|
||||
const size = Number(sizeText)
|
||||
if (!Number.isFinite(size)) return null
|
||||
|
||||
const idMatch = filename.match(/^([^.]+)\.\w+$/)
|
||||
const id = idMatch?.[1]
|
||||
if (!id) return null
|
||||
|
||||
return { id, size, line }
|
||||
}
|
||||
|
||||
function formatBytes(bytes: number): string {
|
||||
const units = ['B', 'KB', 'MB', 'GB', 'TB']
|
||||
let value = bytes
|
||||
let unitIndex = 0
|
||||
|
||||
while (value >= 1024 && unitIndex < units.length - 1) {
|
||||
value /= 1024
|
||||
unitIndex += 1
|
||||
}
|
||||
|
||||
return `${value.toFixed(value >= 100 || unitIndex === 0 ? 0 : 1)} ${units[unitIndex]}`
|
||||
}
|
||||
|
||||
async function mapConcurrent<T, R>(
|
||||
items: T[],
|
||||
concurrency: number,
|
||||
mapper: (item: T, index: number) => Promise<R>,
|
||||
): Promise<R[]> {
|
||||
const results = new Array<R>(items.length)
|
||||
let nextIndex = 0
|
||||
|
||||
async function worker() {
|
||||
while (nextIndex < items.length) {
|
||||
const currentIndex = nextIndex
|
||||
nextIndex += 1
|
||||
results[currentIndex] = await mapper(items[currentIndex], currentIndex)
|
||||
}
|
||||
}
|
||||
|
||||
await Promise.all(
|
||||
Array.from({ length: Math.min(concurrency, items.length) }, () => worker()),
|
||||
)
|
||||
|
||||
return results
|
||||
}
|
||||
|
||||
async function getVideoWithRetry(id: string, retries = 3, delay = 1000): Promise<any> {
|
||||
for (let attempt = 1; attempt <= retries; attempt++) {
|
||||
try {
|
||||
const json = await getVideo(id)
|
||||
if (json && !json.error) {
|
||||
return json
|
||||
}
|
||||
console.warn(`[Retry ${attempt}/${retries}] getVideo(${id}) returned empty or error:`, json)
|
||||
} catch (e: any) {
|
||||
console.error(`[Retry ${attempt}/${retries}] getVideo(${id}) threw:`, e)
|
||||
}
|
||||
if (attempt < retries) {
|
||||
await new Promise((resolve) => setTimeout(resolve, delay * attempt))
|
||||
}
|
||||
}
|
||||
return null
|
||||
}
|
||||
|
||||
async function getPageViewsWithRetry(id: string, retries = 3, delay = 1000): Promise<number> {
|
||||
for (let attempt = 1; attempt <= retries; attempt++) {
|
||||
try {
|
||||
const response = await fetch(`https://a.gloe.net/api/stats/preservetube.com/top-stats?period=all&date=2026-04-29&filters=%7B%22page%22%3A%22%2Fwatch%2F${id}%22%7D&with_imported=true&comparison=previous_period&compare_from=undefined&compare_to=undefined&match_day_of_week=true`, {
|
||||
headers: {
|
||||
'cookie': 'logged_in=true; _plausible_key=SFMyNTY.g3QAAAAFbQAAAAtfY3NyZl90b2tlbm0AAAAYN0drdE52TGVzejBCSHVBdmlTdWQzTXdTbQAAAA9jdXJyZW50X3VzZXJfaWRhAW0AAAAJbGFzdF9zZWVuYmnxNDttAAAACmxvZ2luX2Rlc3RkAANuaWxtAAAAEnNlc3Npb25fdGltZW91dF9hdGJqA6pE.q1ENNi4aJLNquuSIo_1cca054Tj9lWPmalOAnMNuK8A'
|
||||
}
|
||||
})
|
||||
if (response.ok) {
|
||||
const aJson = await response.json() as any
|
||||
const topStats = aJson?.['top_stats']
|
||||
if (Array.isArray(topStats)) {
|
||||
const pageViewsObj = topStats.find((s: any) => s.name == 'Total pageviews')
|
||||
if (pageViewsObj && typeof pageViewsObj.value === 'number') {
|
||||
return pageViewsObj.value
|
||||
}
|
||||
}
|
||||
}
|
||||
console.warn(`[Retry ${attempt}/${retries}] Failed to fetch page views for ${id}`)
|
||||
} catch (e: any) {
|
||||
console.error(`[Retry ${attempt}/${retries}] Fetch page views for ${id} threw:`, e)
|
||||
}
|
||||
if (attempt < retries) {
|
||||
await new Promise((resolve) => setTimeout(resolve, delay * attempt))
|
||||
}
|
||||
}
|
||||
return 0
|
||||
}
|
||||
|
||||
async function main() {
|
||||
const text = await Bun.file(INPUT_PATH).text()
|
||||
const files = parseOutput(text)
|
||||
|
||||
if (files.length === 0) {
|
||||
throw new Error(`No video entries could be parsed from ${INPUT_PATH}`)
|
||||
}
|
||||
|
||||
await Bun.write(OUTPUT_PATH, '')
|
||||
|
||||
const statuses = (await mapConcurrent(files, CONCURRENCY, async (file) => {
|
||||
const isAlreadyCold = await db.selectFrom('files')
|
||||
.where('videoId', '=', file.id)
|
||||
.selectAll()
|
||||
.executeTakeFirst()
|
||||
|
||||
if (isAlreadyCold) return {
|
||||
id: file.id,
|
||||
size: file.size,
|
||||
title: 'already cold',
|
||||
stillUp: false,
|
||||
}
|
||||
|
||||
const archivedMetadata = await db.selectFrom('videos')
|
||||
.where('id', '=', file.id)
|
||||
.selectAll()
|
||||
.executeTakeFirst()
|
||||
|
||||
if (!archivedMetadata || archivedMetadata.deletion_stage) {
|
||||
console.log(`${file.id} on disk, but not in database`)
|
||||
fs.appendFileSync('missing.txt', `${file.line}\n`)
|
||||
return false
|
||||
}
|
||||
|
||||
const channelList = ['UCMnULQ6F6kLDAHxofDWIbrw', 'UCWEoAwbfSQtgB-_OUN-g7gg']
|
||||
if (archivedMetadata?.channelId && channelList.includes(archivedMetadata.channelId)) {
|
||||
console.log(file.id, archivedMetadata?.channelId, 'added based on channel id')
|
||||
fs.appendFileSync(OUTPUT_PATH, `${file.line}\n`)
|
||||
return {
|
||||
id: file.id,
|
||||
size: file.size,
|
||||
title: archivedMetadata.title,
|
||||
stillUp: true
|
||||
}
|
||||
}
|
||||
|
||||
if (checkOnlyChannelId) return {
|
||||
id: file.id,
|
||||
size: file.size,
|
||||
title: archivedMetadata.title,
|
||||
stillUp: true
|
||||
}
|
||||
|
||||
const json = await getVideoWithRetry(file.id) as { videoDetails?: { title?: string } } | null
|
||||
console.log(json)
|
||||
const title = json?.videoDetails?.title ?? null
|
||||
const stillUp = Boolean(title)
|
||||
|
||||
const pageViews = await getPageViewsWithRetry(file.id)
|
||||
|
||||
console.log(file.id, archivedMetadata?.channelId, archivedMetadata?.title, stillUp, pageViews)
|
||||
|
||||
if ((stillUp && pageViews < 5) || (!stillUp && pageViews < 2)) {
|
||||
fs.appendFileSync(OUTPUT_PATH, `${file.line}\n`)
|
||||
}
|
||||
|
||||
const status: VideoStatus = {
|
||||
id: file.id,
|
||||
size: file.size,
|
||||
title,
|
||||
stillUp,
|
||||
}
|
||||
|
||||
return status
|
||||
})).filter(s => s != false)
|
||||
|
||||
const stillUpCount = statuses.filter((status) => status.stillUp).length
|
||||
const gone = statuses
|
||||
.filter((status) => !status.stillUp)
|
||||
.sort((a, b) => b.size - a.size)
|
||||
|
||||
console.log(`Parsed ${files.length} files from ${INPUT_PATH}`)
|
||||
console.log(`Checked ${files.length} videos`)
|
||||
console.log(`Still up: ${stillUpCount}`)
|
||||
console.log(`Gone: ${gone.length}`)
|
||||
console.log(`Wrote ${OUTPUT_PATH}`)
|
||||
|
||||
if (gone.length === 0) {
|
||||
return
|
||||
}
|
||||
|
||||
console.log('')
|
||||
console.log('Gone videos:')
|
||||
console.log('')
|
||||
|
||||
for (const video of gone) {
|
||||
console.log(`${formatBytes(video.size)}\t${video.id}`)
|
||||
}
|
||||
}
|
||||
|
||||
await main()
|
||||
|
|
@ -0,0 +1,200 @@
|
|||
import { readdir } from 'node:fs/promises'
|
||||
|
||||
type FfprobeStream = {
|
||||
codec_name?: string
|
||||
codec_type?: string
|
||||
duration?: string
|
||||
width?: number
|
||||
height?: number
|
||||
avg_frame_rate?: string
|
||||
r_frame_rate?: string
|
||||
}
|
||||
|
||||
type FfprobeFormat = {
|
||||
duration?: string
|
||||
}
|
||||
|
||||
type FfprobeOutput = {
|
||||
streams?: FfprobeStream[]
|
||||
format?: FfprobeFormat
|
||||
}
|
||||
|
||||
function printUsage() {
|
||||
console.error('Usage: bun run generateColdRow.ts --input-folder=/mnt/cold --prefix=cold/ --suffix=.webm')
|
||||
console.error('Example: bun run generateColdRow.ts --input-folder=./cold --prefix=cold/ --suffix=.webm')
|
||||
}
|
||||
|
||||
function formatResolution(width?: number, height?: number): string {
|
||||
if (!width || !height) return 'unknown'
|
||||
return `${width}x${height}`
|
||||
}
|
||||
|
||||
function parseDuration(durationText?: string): number {
|
||||
if (!durationText) return 0
|
||||
const duration = Number(durationText)
|
||||
return Number.isFinite(duration) ? duration : 0
|
||||
}
|
||||
|
||||
function parseFps(rate?: string): number {
|
||||
if (!rate) return 0
|
||||
const [numeratorText, denominatorText] = rate.split('/')
|
||||
const numerator = Number(numeratorText)
|
||||
const denominator = denominatorText ? Number(denominatorText) : 1
|
||||
|
||||
if (!Number.isFinite(numerator) || !Number.isFinite(denominator) || denominator === 0) {
|
||||
return 0
|
||||
}
|
||||
|
||||
return Number((numerator / denominator).toFixed(3))
|
||||
}
|
||||
|
||||
async function hashFile(path: string): Promise<string> {
|
||||
const proc = Bun.spawn(['sha256sum', '--', path], {
|
||||
stdout: 'pipe',
|
||||
stderr: 'pipe',
|
||||
})
|
||||
|
||||
const [stdout, stderr, exitCode] = await Promise.all([
|
||||
new Response(proc.stdout).text(),
|
||||
new Response(proc.stderr).text(),
|
||||
proc.exited,
|
||||
])
|
||||
|
||||
if (exitCode !== 0) {
|
||||
throw new Error(stderr.trim() || 'sha256sum failed')
|
||||
}
|
||||
|
||||
const hash = stdout.trim().split(/\s+/)[0]
|
||||
if (!hash) throw new Error('sha256sum did not return a hash')
|
||||
return hash
|
||||
}
|
||||
|
||||
async function getVideoMetadata(path: string): Promise<FfprobeOutput> {
|
||||
const proc = Bun.spawn([
|
||||
'ffprobe',
|
||||
'-v',
|
||||
'error',
|
||||
'-print_format',
|
||||
'json',
|
||||
'-show_format',
|
||||
'-show_streams',
|
||||
path,
|
||||
], {
|
||||
stdout: 'pipe',
|
||||
stderr: 'pipe',
|
||||
})
|
||||
|
||||
const [stdout, stderr, exitCode] = await Promise.all([
|
||||
new Response(proc.stdout).text(),
|
||||
new Response(proc.stderr).text(),
|
||||
proc.exited,
|
||||
])
|
||||
|
||||
if (exitCode !== 0) {
|
||||
throw new Error(stderr.trim() || 'ffprobe failed')
|
||||
}
|
||||
|
||||
return JSON.parse(stdout) as FfprobeOutput
|
||||
}
|
||||
|
||||
function normalizeDirectory(path: string): string {
|
||||
return path.endsWith('/') ? path : `${path}/`
|
||||
}
|
||||
|
||||
function escapeRegExp(text: string): string {
|
||||
return text.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')
|
||||
}
|
||||
|
||||
function readFlagValue(arg: string, longName: string, shortName: string): string | undefined {
|
||||
if (arg.startsWith(`--${longName}=`)) {
|
||||
return arg.slice(longName.length + 3)
|
||||
}
|
||||
|
||||
if (arg.startsWith(`-${shortName}=`)) {
|
||||
return arg.slice(shortName.length + 2)
|
||||
}
|
||||
}
|
||||
|
||||
function parseArgs(args: string[]): { inputFolder: string, suffix: string } {
|
||||
let inputFolder: string | undefined
|
||||
let prefix: string | undefined
|
||||
let suffix: string | undefined
|
||||
|
||||
for (const arg of args) {
|
||||
inputFolder ??= readFlagValue(arg, 'input-folder', 'input-folder')
|
||||
inputFolder ??= readFlagValue(arg, 'input', 'input')
|
||||
prefix ??= readFlagValue(arg, 'prefix', 'prefix')
|
||||
suffix ??= readFlagValue(arg, 'suffix', 'suffix')
|
||||
}
|
||||
|
||||
if (!prefix || !suffix) {
|
||||
printUsage()
|
||||
process.exit(1)
|
||||
}
|
||||
|
||||
return {
|
||||
inputFolder: normalizeDirectory(inputFolder ?? prefix),
|
||||
suffix,
|
||||
}
|
||||
}
|
||||
|
||||
async function readInputIds(inputFolder: string, suffix: string): Promise<string[]> {
|
||||
const entries = await readdir(inputFolder, { withFileTypes: true })
|
||||
const suffixPattern = new RegExp(`${escapeRegExp(suffix)}$`)
|
||||
|
||||
return entries
|
||||
.filter(entry => entry.isFile() && entry.name.endsWith(suffix))
|
||||
.map(entry => entry.name.replace(suffixPattern, ''))
|
||||
.sort()
|
||||
}
|
||||
|
||||
async function generateRow(videoId: string, inputFolder: string, suffix: string) {
|
||||
const filePath = `${inputFolder}${videoId}${suffix}`
|
||||
const file = Bun.file(filePath)
|
||||
|
||||
if (!(await file.exists())) {
|
||||
throw new Error(`File not found for ${videoId}: ${filePath}`)
|
||||
}
|
||||
|
||||
const filename = filePath.split('/').pop()
|
||||
if (!filename) {
|
||||
throw new Error(`Invalid file path for ${videoId}: ${filePath}`)
|
||||
}
|
||||
|
||||
const [hash, metadata] = await Promise.all([
|
||||
hashFile(filePath),
|
||||
getVideoMetadata(filePath),
|
||||
])
|
||||
|
||||
const videoStream = metadata.streams?.find(stream => stream.codec_type === 'video')
|
||||
const audioStream = metadata.streams?.find(stream => stream.codec_type === 'audio')
|
||||
|
||||
return {
|
||||
videoId,
|
||||
filename,
|
||||
hash,
|
||||
hash_algorithm: 'sha256',
|
||||
size_bytes: file.size,
|
||||
duration_seconds: parseDuration(videoStream?.duration ?? metadata.format?.duration),
|
||||
video_codec: videoStream?.codec_name ?? 'unknown',
|
||||
audio_codec: audioStream?.codec_name ?? 'unknown',
|
||||
resolution: formatResolution(videoStream?.width, videoStream?.height),
|
||||
fps: parseFps(videoStream?.avg_frame_rate ?? videoStream?.r_frame_rate),
|
||||
}
|
||||
}
|
||||
|
||||
async function main() {
|
||||
const { inputFolder, suffix } = parseArgs(Bun.argv.slice(2))
|
||||
const ids = await readInputIds(inputFolder, suffix)
|
||||
if (ids.length === 0) {
|
||||
console.error(`No files ending in ${suffix} found in ${inputFolder}`)
|
||||
printUsage()
|
||||
process.exit(1)
|
||||
}
|
||||
|
||||
for (const videoId of ids) {
|
||||
console.log(await generateRow(videoId, inputFolder, suffix))
|
||||
}
|
||||
}
|
||||
|
||||
await main()
|
||||
|
|
@ -0,0 +1,179 @@
|
|||
import { db } from '@/utils/database'
|
||||
import type { NewFile } from '@/types'
|
||||
|
||||
const DEFAULT_INPUT_PATH = './coldRows.txt'
|
||||
|
||||
function chunkArray<T>(array: T[], size: number): T[][] {
|
||||
const result: T[][] = []
|
||||
for (let i = 0; i < array.length; i += size) {
|
||||
result.push(array.slice(i, i + size))
|
||||
}
|
||||
return result
|
||||
}
|
||||
|
||||
function getLineAndColumn(text: string, index: number): { line: number; column: number; context: string } {
|
||||
const lines = text.slice(0, index).split('\n')
|
||||
const line = lines.length
|
||||
const column = lines[lines.length - 1].length + 1
|
||||
|
||||
const allLines = text.split('\n')
|
||||
const startLineIdx = Math.max(0, line - 3)
|
||||
const endLineIdx = Math.min(allLines.length - 1, line + 2)
|
||||
|
||||
const contextLines = allLines.slice(startLineIdx, endLineIdx + 1).map((l, i) => {
|
||||
const currentLineNum = startLineIdx + i + 1
|
||||
const prefix = currentLineNum === line ? '-> ' : ' '
|
||||
return `${prefix}${String(currentLineNum).padStart(4, ' ')} | ${l}`
|
||||
})
|
||||
|
||||
return {
|
||||
line,
|
||||
column,
|
||||
context: contextLines.join('\n'),
|
||||
}
|
||||
}
|
||||
|
||||
function extractObjectBlocks(text: string): string[] {
|
||||
const blocks: string[] = []
|
||||
let depth = 0
|
||||
let startIndex = -1
|
||||
const openBraceIndices: number[] = []
|
||||
|
||||
for (let index = 0; index < text.length; index++) {
|
||||
const char = text[index]
|
||||
|
||||
if (char === '{') {
|
||||
if (depth === 0) {
|
||||
startIndex = index
|
||||
}
|
||||
depth += 1
|
||||
openBraceIndices.push(index)
|
||||
continue
|
||||
}
|
||||
|
||||
if (char !== '}') continue
|
||||
|
||||
depth -= 1
|
||||
openBraceIndices.pop()
|
||||
|
||||
if (depth < 0) {
|
||||
const { line, column, context } = getLineAndColumn(text, index)
|
||||
console.error(`Error: Unexpected closing brace at line ${line}, column ${column}:\n${context}`)
|
||||
throw new Error(`Unexpected closing brace at position ${index}`)
|
||||
}
|
||||
|
||||
if (depth === 0 && startIndex !== -1) {
|
||||
blocks.push(text.slice(startIndex, index + 1))
|
||||
startIndex = -1
|
||||
}
|
||||
}
|
||||
|
||||
if (depth !== 0) {
|
||||
console.error(`Error: Input contains an unterminated object literal (depth: ${depth}).`)
|
||||
console.error(`There are ${openBraceIndices.length} unclosed '{' brace(s).`)
|
||||
|
||||
const maxToPrint = Math.min(openBraceIndices.length, 3)
|
||||
for (let i = 0; i < maxToPrint; i++) {
|
||||
const braceIndex = openBraceIndices[i]
|
||||
const { line, column, context } = getLineAndColumn(text, braceIndex)
|
||||
console.error(`\nUnclosed '{' #${i + 1} at line ${line}, column ${column}:`)
|
||||
console.error(context)
|
||||
}
|
||||
|
||||
if (openBraceIndices.length > maxToPrint) {
|
||||
console.error(`\n... and ${openBraceIndices.length - maxToPrint} more unclosed brace(s).`)
|
||||
}
|
||||
|
||||
throw new Error('Input contains an unterminated object literal')
|
||||
}
|
||||
|
||||
return blocks
|
||||
}
|
||||
|
||||
function parseColdRow(block: string): NewFile {
|
||||
const parsed = Function(`"use strict"; return (${block})`)()
|
||||
|
||||
if (!parsed || typeof parsed !== 'object' || Array.isArray(parsed)) {
|
||||
throw new Error('Parsed row is not an object')
|
||||
}
|
||||
|
||||
return parsed as NewFile
|
||||
}
|
||||
|
||||
async function loadRows(inputPath: string): Promise<NewFile[]> {
|
||||
const text = await Bun.file(inputPath).text()
|
||||
const blocks = extractObjectBlocks(text)
|
||||
|
||||
if (blocks.length === 0) {
|
||||
throw new Error(`No row objects found in ${inputPath}`)
|
||||
}
|
||||
|
||||
return blocks.map(parseColdRow)
|
||||
}
|
||||
|
||||
async function main() {
|
||||
const args = Bun.argv.slice(2)
|
||||
const dryRun = args.includes('--dry-run')
|
||||
const inputPath = args.find(arg => arg !== '--dry-run') ?? DEFAULT_INPUT_PATH
|
||||
const rows = await loadRows(inputPath)
|
||||
const videoIds = [...new Set(rows.map(row => row.videoId))]
|
||||
|
||||
const uniqueHashes = [...new Set(rows.map(r => r.hash))]
|
||||
const existingFilesChunks = []
|
||||
for (const chunk of chunkArray(uniqueHashes, 1000)) {
|
||||
const filesChunk = await db.selectFrom('files')
|
||||
.where('hash', 'in', chunk)
|
||||
.selectAll()
|
||||
.execute()
|
||||
existingFilesChunks.push(filesChunk)
|
||||
}
|
||||
const existingFiles = existingFilesChunks.flat()
|
||||
|
||||
const rowsToInsert: NewFile[] = []
|
||||
for (const row of rows) {
|
||||
const matches = existingFiles.filter(f => f.hash === row.hash)
|
||||
const isDuplicate = matches.some(match => {
|
||||
return Object.keys(row).every(key => {
|
||||
const k = key as keyof NewFile
|
||||
return row[k] == (match as any)[k]
|
||||
})
|
||||
})
|
||||
|
||||
if (!isDuplicate) {
|
||||
rowsToInsert.push(row)
|
||||
}
|
||||
}
|
||||
|
||||
if (dryRun) {
|
||||
console.log(`Parsed ${rows.length} row(s) from ${inputPath}`)
|
||||
console.log(`Would insert ${rowsToInsert.length} new row(s) and skip ${rows.length - rowsToInsert.length} duplicate(s)`)
|
||||
console.log(JSON.stringify(rowsToInsert, null, 2))
|
||||
console.log(`Would set deletion_stage to cold_storage for ${videoIds.length} video(s)`)
|
||||
return
|
||||
}
|
||||
|
||||
if (rowsToInsert.length > 0) {
|
||||
for (const chunk of chunkArray(rowsToInsert, 100)) {
|
||||
await db.insertInto('files')
|
||||
.values(chunk)
|
||||
.execute()
|
||||
}
|
||||
}
|
||||
|
||||
for (const chunk of chunkArray(videoIds, 1000)) {
|
||||
await db.updateTable('videos')
|
||||
.where('id', 'in', chunk)
|
||||
.set({ deletion_stage: 'cold_storage' })
|
||||
.execute()
|
||||
}
|
||||
|
||||
console.log(`Inserted ${rowsToInsert.length} row(s) from ${inputPath}`)
|
||||
console.log(`Skipped ${rows.length - rowsToInsert.length} duplicate row(s)`)
|
||||
console.log(`Set deletion_stage to cold_storage for ${videoIds.length} video(s)`)
|
||||
}
|
||||
|
||||
try {
|
||||
await main()
|
||||
} finally {
|
||||
await db.destroy()
|
||||
}
|
||||
|
|
@ -0,0 +1,151 @@
|
|||
import { readFile } from "node:fs/promises";
|
||||
import { db } from '@/utils/database'
|
||||
|
||||
const INPUT_FILE = "s3.txt";
|
||||
const AMSTERDAM_TIMEZONE = "Europe/Amsterdam";
|
||||
const MIN_SIZE_BYTES = 75 * 1024 ** 2; // 75 MiB
|
||||
|
||||
const coldFiles = await db.selectFrom('files')
|
||||
.select(['filename'])
|
||||
.execute()
|
||||
|
||||
const SIZE_MULTIPLIERS: Record<string, number> = {
|
||||
KiB: 1024,
|
||||
MiB: 1024 ** 2,
|
||||
GiB: 1024 ** 3,
|
||||
};
|
||||
|
||||
type ParsedEntry = {
|
||||
timestamp: string;
|
||||
filename: string;
|
||||
sizeText: string;
|
||||
sizeBytes: number;
|
||||
ageDays: number;
|
||||
ratio: number;
|
||||
};
|
||||
|
||||
const linePattern =
|
||||
/^\[(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}) (UTC|CET|CEST)\]\s+([0-9.]+)(KiB|MiB|GiB)\s+\S+\s+(.+)$/;
|
||||
|
||||
function parseTimestamp(dateTime: string, zone: string): Date {
|
||||
const [datePart, timePart] = dateTime.split(" ");
|
||||
const [year, month, day] = datePart.split("-").map(Number);
|
||||
const [hour, minute, second] = timePart.split(":").map(Number);
|
||||
|
||||
if (zone === "UTC") {
|
||||
return new Date(Date.UTC(year, month - 1, day, hour, minute, second));
|
||||
}
|
||||
|
||||
if (zone === "CET" || zone === "CEST") {
|
||||
const utcOffsetHours = zone === "CEST" ? 2 : 1;
|
||||
return new Date(Date.UTC(year, month - 1, day, hour - utcOffsetHours, minute, second));
|
||||
}
|
||||
|
||||
const utcGuess = new Date(Date.UTC(year, month - 1, day, hour, minute, second));
|
||||
const formatter = new Intl.DateTimeFormat("en-CA", {
|
||||
timeZone: AMSTERDAM_TIMEZONE,
|
||||
year: "numeric",
|
||||
month: "2-digit",
|
||||
day: "2-digit",
|
||||
hour: "2-digit",
|
||||
minute: "2-digit",
|
||||
second: "2-digit",
|
||||
hourCycle: "h23",
|
||||
});
|
||||
|
||||
const parts = formatter.formatToParts(utcGuess);
|
||||
const localYear = Number(parts.find((part) => part.type === "year")?.value);
|
||||
const localMonth = Number(parts.find((part) => part.type === "month")?.value);
|
||||
const localDay = Number(parts.find((part) => part.type === "day")?.value);
|
||||
const localHour = Number(parts.find((part) => part.type === "hour")?.value);
|
||||
const localMinute = Number(parts.find((part) => part.type === "minute")?.value);
|
||||
const localSecond = Number(parts.find((part) => part.type === "second")?.value);
|
||||
|
||||
const desiredUtcMs = Date.UTC(year, month - 1, day, hour, minute, second);
|
||||
const displayedUtcMs = Date.UTC(
|
||||
localYear,
|
||||
localMonth - 1,
|
||||
localDay,
|
||||
localHour,
|
||||
localMinute,
|
||||
localSecond,
|
||||
);
|
||||
|
||||
return new Date(utcGuess.getTime() - (displayedUtcMs - desiredUtcMs));
|
||||
}
|
||||
|
||||
function formatReferenceTime(date: Date): string {
|
||||
return new Intl.DateTimeFormat("sv-SE", {
|
||||
timeZone: AMSTERDAM_TIMEZONE,
|
||||
year: "numeric",
|
||||
month: "2-digit",
|
||||
day: "2-digit",
|
||||
hour: "2-digit",
|
||||
minute: "2-digit",
|
||||
second: "2-digit",
|
||||
hourCycle: "h23",
|
||||
timeZoneName: "short",
|
||||
}).format(date);
|
||||
}
|
||||
|
||||
async function main(): Promise<void> {
|
||||
const input = await readFile(INPUT_FILE, "utf8");
|
||||
const now = new Date();
|
||||
const entries: ParsedEntry[] = [];
|
||||
|
||||
for (const line of input.split(/\r?\n/)) {
|
||||
const match = line.match(linePattern);
|
||||
if (!match) {
|
||||
continue;
|
||||
}
|
||||
|
||||
const [, timestampWithoutZone, zone, sizeValueText, sizeUnit, filename] = match;
|
||||
const timestamp = `${timestampWithoutZone} ${zone}`;
|
||||
const sizeValue = Number(sizeValueText);
|
||||
const sizeBytes = sizeValue * SIZE_MULTIPLIERS[sizeUnit];
|
||||
const createdAt = parseTimestamp(timestampWithoutZone, zone);
|
||||
const ageDays = Math.max(now.getTime() - createdAt.getTime(), 0) / 86_400_000;
|
||||
|
||||
entries.push({
|
||||
timestamp,
|
||||
filename,
|
||||
sizeText: `${sizeValueText}${sizeUnit}`,
|
||||
sizeBytes,
|
||||
ageDays,
|
||||
ratio: sizeBytes * ageDays,
|
||||
});
|
||||
}
|
||||
|
||||
entries.sort(
|
||||
(left, right) =>
|
||||
right.ratio - left.ratio ||
|
||||
right.sizeBytes - left.sizeBytes ||
|
||||
right.ageDays - left.ageDays ||
|
||||
left.filename.localeCompare(right.filename),
|
||||
);
|
||||
|
||||
const lines = [
|
||||
`Reference time: ${formatReferenceTime(now)}`,
|
||||
"Ratio formula: size_bytes * age_days",
|
||||
"Sorted from biggest ratio to lowest ratio.",
|
||||
"",
|
||||
"Rank\tRatio\tAge(days)\tSize(bytes)\tSize\tTimestamp\tFilename",
|
||||
...entries
|
||||
.filter(e => e.sizeBytes >= MIN_SIZE_BYTES && !coldFiles.find(c => c.filename == e.filename))
|
||||
.map((entry, index) =>
|
||||
[
|
||||
String(index + 1),
|
||||
entry.ratio.toFixed(2),
|
||||
entry.ageDays.toFixed(2),
|
||||
String(Math.round(entry.sizeBytes)),
|
||||
entry.sizeText,
|
||||
entry.timestamp,
|
||||
entry.filename,
|
||||
].join("\t"),
|
||||
),
|
||||
];
|
||||
|
||||
console.log(lines.join("\n"));
|
||||
}
|
||||
|
||||
await main();
|
||||
|
|
@ -0,0 +1,37 @@
|
|||
import { readFile } from 'fs/promises'
|
||||
import { db } from '@/utils/database'
|
||||
|
||||
async function main() {
|
||||
const raw = await readFile('bucket.txt', 'utf-8')
|
||||
|
||||
const bucketFiles = new Set(
|
||||
raw
|
||||
.split('\n')
|
||||
.map((l) => l.trim())
|
||||
.filter(Boolean)
|
||||
.map((l) => l.split(/\s+/).pop()!)
|
||||
)
|
||||
|
||||
const dbRows = await db
|
||||
.selectFrom('files')
|
||||
.select('filename')
|
||||
.execute()
|
||||
|
||||
const dbFiles = new Set(dbRows.map((r) => r.filename))
|
||||
|
||||
const missingInDb = [...bucketFiles].filter((f) => !dbFiles.has(f))
|
||||
|
||||
for (const m of missingInDb) {
|
||||
const video = await db.selectFrom('videos')
|
||||
.selectAll()
|
||||
.where('id', '=', m.split('.')[0])
|
||||
.executeTakeFirstOrThrow()
|
||||
|
||||
console.log((new URL(video.source)).hostname, m)
|
||||
}
|
||||
}
|
||||
|
||||
main().catch((err) => {
|
||||
console.error(err)
|
||||
process.exit(1)
|
||||
})
|
||||
|
|
@ -0,0 +1,87 @@
|
|||
import { db } from '@/utils/database'
|
||||
import type { File } from '@/types'
|
||||
|
||||
async function main() {
|
||||
const args = Bun.argv.slice(2)
|
||||
const dryRun = args.includes('--dry-run')
|
||||
|
||||
console.log('Fetching all files to find duplicates...')
|
||||
const allFiles = await db.selectFrom('files').selectAll().execute()
|
||||
|
||||
// Group files by hash
|
||||
const filesByHash = new Map<string, File[]>()
|
||||
for (const file of allFiles) {
|
||||
const group = filesByHash.get(file.hash) ?? []
|
||||
group.push(file)
|
||||
filesByHash.set(file.hash, group)
|
||||
}
|
||||
|
||||
const uuidsToDelete: string[] = []
|
||||
let duplicateCount = 0
|
||||
|
||||
for (const [hash, group] of filesByHash.entries()) {
|
||||
if (group.length > 1) {
|
||||
// Find duplicates within the group by comparing properties
|
||||
const uniqueFiles: File[] = []
|
||||
|
||||
for (const file of group) {
|
||||
// Check if `file` is a duplicate of any in `uniqueFiles`
|
||||
const isDuplicate = uniqueFiles.some(uniqueFile => {
|
||||
return (
|
||||
file.videoId === uniqueFile.videoId &&
|
||||
file.filename === uniqueFile.filename &&
|
||||
file.hash_algorithm === uniqueFile.hash_algorithm &&
|
||||
file.size_bytes == uniqueFile.size_bytes &&
|
||||
file.duration_seconds == uniqueFile.duration_seconds &&
|
||||
file.video_codec === uniqueFile.video_codec &&
|
||||
file.audio_codec === uniqueFile.audio_codec &&
|
||||
file.resolution === uniqueFile.resolution &&
|
||||
file.fps == uniqueFile.fps
|
||||
)
|
||||
})
|
||||
|
||||
if (isDuplicate) {
|
||||
uuidsToDelete.push(file.uuid as string)
|
||||
duplicateCount++
|
||||
} else {
|
||||
uniqueFiles.push(file)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if (uuidsToDelete.length === 0) {
|
||||
console.log('No duplicate files found.')
|
||||
return
|
||||
}
|
||||
|
||||
console.log(`Found ${duplicateCount} duplicate file(s) across ${filesByHash.size} unique hash(es).`)
|
||||
|
||||
if (dryRun) {
|
||||
console.log(`\n[DRY RUN] Would delete the following ${uuidsToDelete.length} UUID(s):`)
|
||||
console.log(uuidsToDelete.join('\n'))
|
||||
console.log('\nRun without --dry-run to execute the deletion.')
|
||||
return
|
||||
}
|
||||
|
||||
console.log(`Deleting ${uuidsToDelete.length} duplicate file(s)...`)
|
||||
|
||||
// Chunk the deletions in case there are many
|
||||
const chunkSize = 1000
|
||||
for (let i = 0; i < uuidsToDelete.length; i += chunkSize) {
|
||||
const chunk = uuidsToDelete.slice(i, i + chunkSize)
|
||||
await db.deleteFrom('files')
|
||||
.where('uuid', 'in', chunk)
|
||||
.execute()
|
||||
}
|
||||
|
||||
console.log('Successfully deleted duplicate files.')
|
||||
}
|
||||
|
||||
try {
|
||||
await main()
|
||||
} catch (error) {
|
||||
console.error(error)
|
||||
} finally {
|
||||
await db.destroy()
|
||||
}
|
||||
|
|
@ -0,0 +1,69 @@
|
|||
import { db } from '@/utils/database'
|
||||
import * as fs from 'node:fs'
|
||||
|
||||
const target: string = 's0'
|
||||
let totalSize = 0
|
||||
|
||||
const bucket = (await Bun.file('bucket.txt').text()).split('\n')
|
||||
const alreadyRemoved = (await Bun.file('headRemoved.txt').text()).split('\n')
|
||||
const rmFiles = []
|
||||
|
||||
const videos = await db.selectFrom('videos')
|
||||
.where('source', 'like', `%${target}.archive.party%`)
|
||||
.where('deletion_stage', '=', 'cold_storage')
|
||||
.select(['id', 'source'])
|
||||
.execute()
|
||||
|
||||
const files = await db.selectFrom('files')
|
||||
.where('videoId', 'in', videos.map(v => v.id))
|
||||
.orderBy('files.size_bytes desc')
|
||||
.selectAll()
|
||||
.execute()
|
||||
|
||||
for (const f of files) {
|
||||
if (alreadyRemoved.includes(f.filename)) continue
|
||||
|
||||
let isFlagged = false
|
||||
const bucketSize = bucket.find(b => b.includes(f.filename))?.split(' ').filter(Boolean)[2]
|
||||
if (!bucketSize) {
|
||||
console.log('file not in bucket', f.filename)
|
||||
continue
|
||||
}
|
||||
|
||||
const video = videos.find(v => v.id == f.videoId)
|
||||
const headReq = await fetch(video!.source
|
||||
.replace('https://s0.archive.party', 'http://193.24.209.204:1337')
|
||||
.replace('https://minio.archive.party', 'http://5.83.147.250:9000')
|
||||
.replace('https://s2.archive.party', 'http://185.207.125.54:9000')
|
||||
.replace('https://s3.archive.party', 'http://193.24.209.233:8080'),
|
||||
{
|
||||
method: 'HEAD'
|
||||
})
|
||||
|
||||
if (headReq.status == 404) {
|
||||
fs.appendFileSync('headRemoved.txt', f.filename + '\n')
|
||||
continue
|
||||
}
|
||||
if (headReq.headers.get('Content-Length') != bucketSize) {
|
||||
console.log('head size not same for', f.filename)
|
||||
console.log([headReq.headers.get('Content-Length'), bucketSize])
|
||||
isFlagged = true
|
||||
}
|
||||
|
||||
if (f.size_bytes.toString() != bucketSize) {
|
||||
console.log('size not same for', f.filename)
|
||||
console.log(f.size_bytes, bucketSize)
|
||||
isFlagged = true
|
||||
}
|
||||
|
||||
if (isFlagged) continue
|
||||
console.log(f.size_bytes / 1e9, f.filename + '\n')
|
||||
totalSize += f.size_bytes / 1e9
|
||||
|
||||
if (f.size_bytes / 1e9 > 0) rmFiles.push(f.filename)
|
||||
if (f.size_bytes / 1e9 < 0) break;
|
||||
}
|
||||
|
||||
console.log(totalSize)
|
||||
if (target == 'minio' || target == 's2' || target == 's3') console.log(`./mc rm ${rmFiles.map(r => `local/preservetube/${r}`).join(' ')}`)
|
||||
if (target == 's0') console.log(`rm -- ${rmFiles.join(' ')}`)
|
||||
|
|
@ -0,0 +1,38 @@
|
|||
#!/usr/bin/env bash
|
||||
set -euo pipefail
|
||||
|
||||
input_file="${1:-cold.txt}"
|
||||
|
||||
cat "$input_file" | awk '
|
||||
function multiplier(unit) {
|
||||
if (unit == "KiB") return 1024
|
||||
if (unit == "MiB") return 1024 * 1024
|
||||
if (unit == "GiB") return 1024 * 1024 * 1024
|
||||
return 0
|
||||
}
|
||||
|
||||
function human(bytes) {
|
||||
if (bytes >= 1024 * 1024 * 1024) return sprintf("%.2fGiB", bytes / (1024 * 1024 * 1024))
|
||||
if (bytes >= 1024 * 1024) return sprintf("%.2fMiB", bytes / (1024 * 1024))
|
||||
if (bytes >= 1024) return sprintf("%.2fKiB", bytes / 1024)
|
||||
return sprintf("%dB", bytes)
|
||||
}
|
||||
|
||||
{
|
||||
size = $5
|
||||
if (size ~ /^[0-9.]+(KiB|MiB|GiB)$/) {
|
||||
value = size
|
||||
unit = size
|
||||
sub(/(KiB|MiB|GiB)$/, "", value)
|
||||
sub(/^[0-9.]+/, "", unit)
|
||||
total += value * multiplier(unit)
|
||||
parsed++
|
||||
}
|
||||
}
|
||||
|
||||
END {
|
||||
printf "Parsed lines: %d\n", parsed
|
||||
printf "Total bytes: %.0f\n", total
|
||||
printf "Total size: %s\n", human(total)
|
||||
}
|
||||
'
|
||||
|
|
@ -0,0 +1,200 @@
|
|||
import { readdir } from 'node:fs/promises'
|
||||
|
||||
type FfprobeStream = {
|
||||
codec_name?: string
|
||||
codec_type?: string
|
||||
duration?: string
|
||||
width?: number
|
||||
height?: number
|
||||
avg_frame_rate?: string
|
||||
r_frame_rate?: string
|
||||
}
|
||||
|
||||
type FfprobeFormat = {
|
||||
duration?: string
|
||||
}
|
||||
|
||||
type FfprobeOutput = {
|
||||
streams?: FfprobeStream[]
|
||||
format?: FfprobeFormat
|
||||
}
|
||||
|
||||
function printUsage() {
|
||||
console.error('Usage: bun run generateColdRow.ts --input-folder=/mnt/cold --prefix=cold/ --suffix=.webm')
|
||||
console.error('Example: bun run generateColdRow.ts --input-folder=./cold --prefix=cold/ --suffix=.webm')
|
||||
}
|
||||
|
||||
function formatResolution(width?: number, height?: number): string {
|
||||
if (!width || !height) return 'unknown'
|
||||
return `${width}x${height}`
|
||||
}
|
||||
|
||||
function parseDuration(durationText?: string): number {
|
||||
if (!durationText) return 0
|
||||
const duration = Number(durationText)
|
||||
return Number.isFinite(duration) ? duration : 0
|
||||
}
|
||||
|
||||
function parseFps(rate?: string): number {
|
||||
if (!rate) return 0
|
||||
const [numeratorText, denominatorText] = rate.split('/')
|
||||
const numerator = Number(numeratorText)
|
||||
const denominator = denominatorText ? Number(denominatorText) : 1
|
||||
|
||||
if (!Number.isFinite(numerator) || !Number.isFinite(denominator) || denominator === 0) {
|
||||
return 0
|
||||
}
|
||||
|
||||
return Number((numerator / denominator).toFixed(3))
|
||||
}
|
||||
|
||||
async function hashFile(path: string): Promise<string> {
|
||||
const proc = Bun.spawn(['sha256sum', '--', path], {
|
||||
stdout: 'pipe',
|
||||
stderr: 'pipe',
|
||||
})
|
||||
|
||||
const [stdout, stderr, exitCode] = await Promise.all([
|
||||
new Response(proc.stdout).text(),
|
||||
new Response(proc.stderr).text(),
|
||||
proc.exited,
|
||||
])
|
||||
|
||||
if (exitCode !== 0) {
|
||||
throw new Error(stderr.trim() || 'sha256sum failed')
|
||||
}
|
||||
|
||||
const hash = stdout.trim().split(/\s+/)[0]
|
||||
if (!hash) throw new Error('sha256sum did not return a hash')
|
||||
return hash
|
||||
}
|
||||
|
||||
async function getVideoMetadata(path: string): Promise<FfprobeOutput> {
|
||||
const proc = Bun.spawn([
|
||||
'ffprobe',
|
||||
'-v',
|
||||
'error',
|
||||
'-print_format',
|
||||
'json',
|
||||
'-show_format',
|
||||
'-show_streams',
|
||||
path,
|
||||
], {
|
||||
stdout: 'pipe',
|
||||
stderr: 'pipe',
|
||||
})
|
||||
|
||||
const [stdout, stderr, exitCode] = await Promise.all([
|
||||
new Response(proc.stdout).text(),
|
||||
new Response(proc.stderr).text(),
|
||||
proc.exited,
|
||||
])
|
||||
|
||||
if (exitCode !== 0) {
|
||||
throw new Error(stderr.trim() || 'ffprobe failed')
|
||||
}
|
||||
|
||||
return JSON.parse(stdout) as FfprobeOutput
|
||||
}
|
||||
|
||||
function normalizeDirectory(path: string): string {
|
||||
return path.endsWith('/') ? path : `${path}/`
|
||||
}
|
||||
|
||||
function escapeRegExp(text: string): string {
|
||||
return text.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')
|
||||
}
|
||||
|
||||
function readFlagValue(arg: string, longName: string, shortName: string): string | undefined {
|
||||
if (arg.startsWith(`--${longName}=`)) {
|
||||
return arg.slice(longName.length + 3)
|
||||
}
|
||||
|
||||
if (arg.startsWith(`-${shortName}=`)) {
|
||||
return arg.slice(shortName.length + 2)
|
||||
}
|
||||
}
|
||||
|
||||
function parseArgs(args: string[]): { inputFolder: string, suffix: string } {
|
||||
let inputFolder: string | undefined
|
||||
let prefix: string | undefined
|
||||
let suffix: string | undefined
|
||||
|
||||
for (const arg of args) {
|
||||
inputFolder ??= readFlagValue(arg, 'input-folder', 'input-folder')
|
||||
inputFolder ??= readFlagValue(arg, 'input', 'input')
|
||||
prefix ??= readFlagValue(arg, 'prefix', 'prefix')
|
||||
suffix ??= readFlagValue(arg, 'suffix', 'suffix')
|
||||
}
|
||||
|
||||
if (!prefix || !suffix) {
|
||||
printUsage()
|
||||
process.exit(1)
|
||||
}
|
||||
|
||||
return {
|
||||
inputFolder: normalizeDirectory(inputFolder ?? prefix),
|
||||
suffix,
|
||||
}
|
||||
}
|
||||
|
||||
async function readInputIds(inputFolder: string, suffix: string): Promise<string[]> {
|
||||
const entries = await readdir(inputFolder, { withFileTypes: true })
|
||||
const suffixPattern = new RegExp(`${escapeRegExp(suffix)}$`)
|
||||
|
||||
return entries
|
||||
.filter(entry => entry.isFile() && entry.name.endsWith(suffix))
|
||||
.map(entry => entry.name.replace(suffixPattern, ''))
|
||||
.sort()
|
||||
}
|
||||
|
||||
async function generateRow(videoId: string, inputFolder: string, suffix: string) {
|
||||
const filePath = `${inputFolder}${videoId}${suffix}`
|
||||
const file = Bun.file(filePath)
|
||||
|
||||
if (!(await file.exists())) {
|
||||
throw new Error(`File not found for ${videoId}: ${filePath}`)
|
||||
}
|
||||
|
||||
const filename = filePath.split('/').pop()
|
||||
if (!filename) {
|
||||
throw new Error(`Invalid file path for ${videoId}: ${filePath}`)
|
||||
}
|
||||
|
||||
const [hash, metadata] = await Promise.all([
|
||||
hashFile(filePath),
|
||||
getVideoMetadata(filePath),
|
||||
])
|
||||
|
||||
const videoStream = metadata.streams?.find(stream => stream.codec_type === 'video')
|
||||
const audioStream = metadata.streams?.find(stream => stream.codec_type === 'audio')
|
||||
|
||||
return {
|
||||
videoId,
|
||||
filename,
|
||||
hash,
|
||||
hash_algorithm: 'sha256',
|
||||
size_bytes: file.size,
|
||||
duration_seconds: parseDuration(videoStream?.duration ?? metadata.format?.duration),
|
||||
video_codec: videoStream?.codec_name ?? 'unknown',
|
||||
audio_codec: audioStream?.codec_name ?? 'unknown',
|
||||
resolution: formatResolution(videoStream?.width, videoStream?.height),
|
||||
fps: parseFps(videoStream?.avg_frame_rate ?? videoStream?.r_frame_rate),
|
||||
}
|
||||
}
|
||||
|
||||
async function main() {
|
||||
const { inputFolder, suffix } = parseArgs(Bun.argv.slice(2))
|
||||
const ids = await readInputIds(inputFolder, suffix)
|
||||
if (ids.length === 0) {
|
||||
console.error(`No files ending in ${suffix} found in ${inputFolder}`)
|
||||
printUsage()
|
||||
process.exit(1)
|
||||
}
|
||||
|
||||
for (const videoId of ids) {
|
||||
console.log(await generateRow(videoId, inputFolder, suffix))
|
||||
}
|
||||
}
|
||||
|
||||
await main()
|
||||
|
|
@ -0,0 +1,179 @@
|
|||
import { db } from '@/utils/database'
|
||||
import type { NewFile } from '@/types'
|
||||
|
||||
const DEFAULT_INPUT_PATH = './coldRows.txt'
|
||||
|
||||
function chunkArray<T>(array: T[], size: number): T[][] {
|
||||
const result: T[][] = []
|
||||
for (let i = 0; i < array.length; i += size) {
|
||||
result.push(array.slice(i, i + size))
|
||||
}
|
||||
return result
|
||||
}
|
||||
|
||||
function getLineAndColumn(text: string, index: number): { line: number; column: number; context: string } {
|
||||
const lines = text.slice(0, index).split('\n')
|
||||
const line = lines.length
|
||||
const column = lines[lines.length - 1].length + 1
|
||||
|
||||
const allLines = text.split('\n')
|
||||
const startLineIdx = Math.max(0, line - 3)
|
||||
const endLineIdx = Math.min(allLines.length - 1, line + 2)
|
||||
|
||||
const contextLines = allLines.slice(startLineIdx, endLineIdx + 1).map((l, i) => {
|
||||
const currentLineNum = startLineIdx + i + 1
|
||||
const prefix = currentLineNum === line ? '-> ' : ' '
|
||||
return `${prefix}${String(currentLineNum).padStart(4, ' ')} | ${l}`
|
||||
})
|
||||
|
||||
return {
|
||||
line,
|
||||
column,
|
||||
context: contextLines.join('\n'),
|
||||
}
|
||||
}
|
||||
|
||||
function extractObjectBlocks(text: string): string[] {
|
||||
const blocks: string[] = []
|
||||
let depth = 0
|
||||
let startIndex = -1
|
||||
const openBraceIndices: number[] = []
|
||||
|
||||
for (let index = 0; index < text.length; index++) {
|
||||
const char = text[index]
|
||||
|
||||
if (char === '{') {
|
||||
if (depth === 0) {
|
||||
startIndex = index
|
||||
}
|
||||
depth += 1
|
||||
openBraceIndices.push(index)
|
||||
continue
|
||||
}
|
||||
|
||||
if (char !== '}') continue
|
||||
|
||||
depth -= 1
|
||||
openBraceIndices.pop()
|
||||
|
||||
if (depth < 0) {
|
||||
const { line, column, context } = getLineAndColumn(text, index)
|
||||
console.error(`Error: Unexpected closing brace at line ${line}, column ${column}:\n${context}`)
|
||||
throw new Error(`Unexpected closing brace at position ${index}`)
|
||||
}
|
||||
|
||||
if (depth === 0 && startIndex !== -1) {
|
||||
blocks.push(text.slice(startIndex, index + 1))
|
||||
startIndex = -1
|
||||
}
|
||||
}
|
||||
|
||||
if (depth !== 0) {
|
||||
console.error(`Error: Input contains an unterminated object literal (depth: ${depth}).`)
|
||||
console.error(`There are ${openBraceIndices.length} unclosed '{' brace(s).`)
|
||||
|
||||
const maxToPrint = Math.min(openBraceIndices.length, 3)
|
||||
for (let i = 0; i < maxToPrint; i++) {
|
||||
const braceIndex = openBraceIndices[i]
|
||||
const { line, column, context } = getLineAndColumn(text, braceIndex)
|
||||
console.error(`\nUnclosed '{' #${i + 1} at line ${line}, column ${column}:`)
|
||||
console.error(context)
|
||||
}
|
||||
|
||||
if (openBraceIndices.length > maxToPrint) {
|
||||
console.error(`\n... and ${openBraceIndices.length - maxToPrint} more unclosed brace(s).`)
|
||||
}
|
||||
|
||||
throw new Error('Input contains an unterminated object literal')
|
||||
}
|
||||
|
||||
return blocks
|
||||
}
|
||||
|
||||
function parseColdRow(block: string): NewFile {
|
||||
const parsed = Function(`"use strict"; return (${block})`)()
|
||||
|
||||
if (!parsed || typeof parsed !== 'object' || Array.isArray(parsed)) {
|
||||
throw new Error('Parsed row is not an object')
|
||||
}
|
||||
|
||||
return parsed as NewFile
|
||||
}
|
||||
|
||||
async function loadRows(inputPath: string): Promise<NewFile[]> {
|
||||
const text = await Bun.file(inputPath).text()
|
||||
const blocks = extractObjectBlocks(text)
|
||||
|
||||
if (blocks.length === 0) {
|
||||
throw new Error(`No row objects found in ${inputPath}`)
|
||||
}
|
||||
|
||||
return blocks.map(parseColdRow)
|
||||
}
|
||||
|
||||
async function main() {
|
||||
const args = Bun.argv.slice(2)
|
||||
const dryRun = args.includes('--dry-run')
|
||||
const inputPath = args.find(arg => arg !== '--dry-run') ?? DEFAULT_INPUT_PATH
|
||||
const rows = await loadRows(inputPath)
|
||||
const videoIds = [...new Set(rows.map(row => row.videoId))]
|
||||
|
||||
const uniqueHashes = [...new Set(rows.map(r => r.hash))]
|
||||
const existingFilesChunks = []
|
||||
for (const chunk of chunkArray(uniqueHashes, 1000)) {
|
||||
const filesChunk = await db.selectFrom('files')
|
||||
.where('hash', 'in', chunk)
|
||||
.selectAll()
|
||||
.execute()
|
||||
existingFilesChunks.push(filesChunk)
|
||||
}
|
||||
const existingFiles = existingFilesChunks.flat()
|
||||
|
||||
const rowsToInsert: NewFile[] = []
|
||||
for (const row of rows) {
|
||||
const matches = existingFiles.filter(f => f.hash === row.hash)
|
||||
const isDuplicate = matches.some(match => {
|
||||
return Object.keys(row).every(key => {
|
||||
const k = key as keyof NewFile
|
||||
return row[k] == (match as any)[k]
|
||||
})
|
||||
})
|
||||
|
||||
if (!isDuplicate) {
|
||||
rowsToInsert.push(row)
|
||||
}
|
||||
}
|
||||
|
||||
if (dryRun) {
|
||||
console.log(`Parsed ${rows.length} row(s) from ${inputPath}`)
|
||||
console.log(`Would insert ${rowsToInsert.length} new row(s) and skip ${rows.length - rowsToInsert.length} duplicate(s)`)
|
||||
console.log(JSON.stringify(rowsToInsert, null, 2))
|
||||
console.log(`Would set deletion_stage to cold_storage for ${videoIds.length} video(s)`)
|
||||
return
|
||||
}
|
||||
|
||||
if (rowsToInsert.length > 0) {
|
||||
for (const chunk of chunkArray(rowsToInsert, 100)) {
|
||||
await db.insertInto('files')
|
||||
.values(chunk)
|
||||
.execute()
|
||||
}
|
||||
}
|
||||
|
||||
for (const chunk of chunkArray(videoIds, 1000)) {
|
||||
await db.updateTable('videos')
|
||||
.where('id', 'in', chunk)
|
||||
.set({ deletion_stage: 'cold_storage' })
|
||||
.execute()
|
||||
}
|
||||
|
||||
console.log(`Inserted ${rowsToInsert.length} row(s) from ${inputPath}`)
|
||||
console.log(`Skipped ${rows.length - rowsToInsert.length} duplicate row(s)`)
|
||||
console.log(`Set deletion_stage to cold_storage for ${videoIds.length} video(s)`)
|
||||
}
|
||||
|
||||
try {
|
||||
await main()
|
||||
} finally {
|
||||
await db.destroy()
|
||||
}
|
||||
Loading…
Reference in New Issue