Skip to content

Sync-in · Sovereign platform for file storage, sharing, synchronization, and collaboration.

feat(backend:files): add pdf ocr indexing

johavend37c531313d4
diff --git a/.gitignore b/.gitignore
index 39fe4586..f058f190 100755
--- a/.gitignore
+++ b/.gitignore
@@ -15,5 +15,5 @@ frontend/tmp
frontend/out-tsc
frontend/bazel-out
/frontend/src/assets/pdfjs/
-circular-dependencies.png!
+circular-dependencies.png
backend/src/applications/files/assets/ocr-languages/*.traineddata
\ No newline at end of file
diff --git a/backend/package.json b/backend/package.json
index 9b4ed9f1..38d85d09 100755
--- a/backend/package.json
+++ b/backend/package.json
@@ -80,6 +80,7 @@
"sharp": "^0.34.4",
"socket.io": "^4.8.1",
"tar": "^7.5.10",
+ "tesseract.js": "^7.0.0",
"text-to-svg": "^3.1.5",
"time2fa": "^1.4.2",
"unpdf": "^1.3.2",
diff --git a/backend/src/applications/files/files.config.ts b/backend/src/applications/files/files.config.ts
index f6f7feae..bf2e99fc 100644
--- a/backend/src/applications/files/files.config.ts
+++ b/backend/src/applications/files/files.config.ts
@@ -1,8 +1,33 @@
import { Type } from 'class-transformer'
-import { IsBoolean, IsInt, IsNotEmpty, IsNotEmptyObject, IsString, ValidateNested } from 'class-validator'
+import { ArrayNotEmpty, IsArray, IsBoolean, IsInt, IsNotEmpty, IsNotEmptyObject, IsString, ValidateIf, ValidateNested } from 'class-validator'
import { CollaboraOnlineConfig } from './modules/collabora-online/collabora-online.config'
import { OnlyOfficeConfig } from './modules/only-office/only-office.config'
+export class FilesContentIndexingOCRConfig {
+ @IsBoolean()
+ enabled: boolean = true
+
+ @ValidateIf((o: FilesContentIndexingOCRConfig) => o.enabled)
+ @ArrayNotEmpty()
+ @IsArray()
+ @IsString({ each: true })
+ languages: string[] = ['eng']
+
+ @IsBoolean()
+ offline: boolean = false
+}
+
+export class FilesContentIndexingConfig {
+ @IsBoolean()
+ enabled: boolean = true
+
+ @ValidateIf((o: FilesContentIndexingConfig) => o.enabled)
+ @IsNotEmptyObject()
+ @ValidateNested()
+ @Type(() => FilesContentIndexingOCRConfig)
+ ocr: FilesContentIndexingOCRConfig = new FilesContentIndexingOCRConfig()
+}
+
export class FilesConfig {
@IsNotEmpty()
@IsString()
@@ -23,8 +48,10 @@ export class FilesConfig {
@IsInt()
maxUploadSize: number = 5368709120 // 5 GB
- @IsBoolean()
- contentIndexing: boolean = true
+ @IsNotEmptyObject()
+ @ValidateNested()
+ @Type(() => FilesContentIndexingConfig)
+ contentIndexing: FilesContentIndexingConfig = new FilesContentIndexingConfig()
@IsBoolean()
showHiddenFiles: boolean = false
diff --git a/backend/src/applications/files/services/files-content-manager.service.ts b/backend/src/applications/files/services/files-content-manager.service.ts
index ba092a68..c0af33df 100644
--- a/backend/src/applications/files/services/files-content-manager.service.ts
+++ b/backend/src/applications/files/services/files-content-manager.service.ts
@@ -7,6 +7,7 @@ import { FileIndexContext, FileParseContext } from '../interfaces/file-parse-ind
import { FilesIndexer } from '../models/files-indexer'
import { FileContent } from '../schemas/file-content.interface'
import { docTextify } from '../utils/doc-textify/doc-textify'
+import { PdfOCRWorkerManager } from '../utils/doc-textify/utils/pdf-ocr'
import { getMimeType } from '../utils/files'
import { FilesParser } from './files-parser.service'
@@ -14,6 +15,7 @@ import { FilesParser } from './files-parser.service'
export class FilesContentManager {
private readonly maxDocumentSize = 150 * 1_000_000
private readonly logger = new Logger(FilesContentManager.name)
+ private pdfOcrWorkerManager: PdfOCRWorkerManager | null = null
constructor(
private readonly filesIndexer: FilesIndexer,
@@ -21,28 +23,39 @@ export class FilesContentManager {
) {}
async parseAndIndexAllFiles(): Promise<void> {
- const indexSuffixes: string[] = []
- for await (const [id, type, paths] of this.filesParser.allPaths()) {
- let indexSuffix: string
- switch (type) {
- case 'user':
- indexSuffix = `${userIndexPrefix}${id}`
- break
- case 'space':
- indexSuffix = `${spaceIndexPrefix}${id}`
- break
- case 'share':
- indexSuffix = `${shareIndexPrefix}${id}`
- }
- try {
- await this.indexFiles(indexSuffix, paths)
- } catch (e) {
- this.logger.error({ tag: this.parseAndIndexAllFiles.name, msg: `${e}` })
+ this.pdfOcrWorkerManager = PdfOCRWorkerManager.getInstance(this.logger)
+ try {
+ await this.pdfOcrWorkerManager.start()
+ } catch (e) {
+ this.logger.warn({ tag: this.parseAndIndexAllFiles.name, msg: `unable to initialize OCR worker: ${e}` })
+ }
+ try {
+ const indexSuffixes: string[] = []
+ for await (const [id, type, paths] of this.filesParser.allPaths()) {
+ let indexSuffix: string
+ switch (type) {
+ case 'user':
+ indexSuffix = `${userIndexPrefix}${id}`
+ break
+ case 'space':
+ indexSuffix = `${spaceIndexPrefix}${id}`
+ break
+ case 'share':
+ indexSuffix = `${shareIndexPrefix}${id}`
+ }
+ try {
+ await this.indexFiles(indexSuffix, paths)
+ } catch (e) {
+ this.logger.error({ tag: this.parseAndIndexAllFiles.name, msg: `${e}` })
+ }
+ indexSuffixes.push(indexSuffix)
}
- indexSuffixes.push(indexSuffix)
+ // clean up old tables
+ await this.filesIndexer.cleanIndexes(indexSuffixes)
+ } finally {
+ await this.pdfOcrWorkerManager?.stop()
+ this.pdfOcrWorkerManager = null
}
- // clean up old tables
- await this.filesIndexer.cleanIndexes(indexSuffixes)
}
private async indexFiles(indexSuffix: string, paths: FileParseContext[]): Promise<void> {
@@ -117,7 +130,7 @@ export class FilesContentManager {
yield* this.parseFiles(realPath, context)
continue
}
- const fileContent = await this.analyzeFile(realPath, context)
+ const fileContent = await this.analyzeFile(realPath, context, false)
if (fileContent !== null) {
yield fileContent
}
@@ -177,7 +190,11 @@ export class FilesContentManager {
try {
const content = await docTextify(
rPath,
- { newlineDelimiter: ' ', minCharsToExtract: 10 },
+ {
+ newlineDelimiter: ' ',
+ minCharsToExtract: 10,
+ ocrWorker: this.pdfOcrWorkerManager?.worker
+ },
{
extension: extension,
verified: true
diff --git a/backend/src/applications/files/services/files-search-manager.service.ts b/backend/src/applications/files/services/files-search-manager.service.ts
index c4fd72b3..8a7e4865 100644
--- a/backend/src/applications/files/services/files-search-manager.service.ts
+++ b/backend/src/applications/files/services/files-search-manager.service.ts
@@ -28,7 +28,7 @@ export class FilesSearchManager {
async search(user: UserModel, search: SearchFilesDto): Promise<FileContent[]> {
const [spaceIds, shareIds] = await Promise.all([this.spacesQueries.spaceIds(user.id), this.sharesQueries.shareIds(user.id, +user.isAdmin)])
if (search.fullText) {
- if (!configuration.applications.files.contentIndexing) {
+ if (!configuration.applications.files.contentIndexing.enabled) {
throw new HttpException('Full-text search is disabled', HttpStatus.BAD_REQUEST)
}
return await this.searchFullText(user.id, spaceIds, shareIds, search.content, search.limit)
diff --git a/backend/src/applications/files/utils/doc-textify/adapters/pdf.ts b/backend/src/applications/files/utils/doc-textify/adapters/pdf.ts
index 16f39753..370231c2 100644
--- a/backend/src/applications/files/utils/doc-textify/adapters/pdf.ts
+++ b/backend/src/applications/files/utils/doc-textify/adapters/pdf.ts
@@ -1,63 +1,176 @@
import { readFile } from 'node:fs/promises'
-import { getDocumentProxy } from 'unpdf'
+import sharp from 'sharp'
+import { extractImages, getDocumentProxy } from 'unpdf'
import type { PDFDocumentProxy } from 'unpdf/pdfjs'
-import type { DocTextifyOptions } from '../interfaces/doc-textify.interfaces'
+import type { DocTextifyOCRWorkerLike, DocTextifyOptions } from '../interfaces/doc-textify.interfaces'
+
+const ignorePdfBadFormat = new Set([0x0000, 0x0001])
-// Type guard to filter true text items
interface TextItem {
str: string
transform: [number, number, number, number, number, number]
}
-function isTextItem(item: any): item is TextItem {
- return typeof item.str === 'string' && Array.isArray(item.transform)
+
+type ExtractedImages = Awaited<ReturnType<typeof extractImages>>
+
+function isTextItem(item: unknown): item is TextItem {
+ return (
+ typeof item === 'object' &&
+ item !== null &&
+ 'str' in item &&
+ 'transform' in item &&
+ typeof (item as TextItem).str === 'string' &&
+ Array.isArray((item as TextItem).transform)
+ )
}
-const ignorePdfBadFormat = new Set([0x0000, 0x0001])
+function shouldOcrImage(image: { width: number; height: number; channels: number }): boolean {
+ const { width, height } = image
+
+ // too small for OCR
+ if (width < 32 || height < 32) return false
+
+ // not enough pixels
+ if (width * height < 4000) return false
+
+ const minSide = Math.min(width, height)
+ const maxSide = Math.max(width, height)
+ const ratio = maxSide / minSide
+
+ // too thin overall
+ if (minSide < 12) return false
+
+ // reject extreme shapes
+ return ratio <= 8
+}
+
+function extractText(items: unknown[], newlineDelimiter: string): string {
+ const fragments: string[] = []
+ let lastY: number | undefined = undefined
+
+ for (const item of items) {
+ if (!isTextItem(item)) continue
+ const currentY = item.transform[5]
+ if (lastY !== undefined && currentY !== lastY) {
+ fragments.push(newlineDelimiter)
+ }
+ fragments.push(item.str)
+ lastY = currentY
+ }
+
+ return fragments.join('')
+}
+
+async function extractTextFromImages(images: ExtractedImages, options: DocTextifyOptions): Promise<string[]> {
+ const contents: string[] = []
+ const worker: DocTextifyOCRWorkerLike | undefined = options.ocrWorker
+
+ if (!worker) return contents
+
+ for (const image of images) {
+ if (!shouldOcrImage(image)) {
+ continue
+ }
+
+ let imageBuffer: Buffer | undefined
+
+ try {
+ let imageProcessor = sharp(image.data, {
+ raw: {
+ width: image.width,
+ height: image.height,
+ channels: image.channels
+ }
+ })
+
+ if (image.channels === 4) {
+ imageProcessor = imageProcessor.flatten({ background: '#ffffff' })
+ }
+
+ imageBuffer = await imageProcessor
+ .resize({
+ width: 1800,
+ height: 1800,
+ fit: 'inside',
+ withoutEnlargement: true,
+ fastShrinkOnLoad: true
+ })
+ .grayscale()
+ .jpeg({ quality: 85 })
+ // .png({ compressionLevel: 0, adaptiveFiltering: false })
+ .toBuffer()
+ const result = await worker.recognize(imageBuffer)
+ const content = result.data?.text?.trim() || ''
+
+ if (content.length > 0) {
+ contents.push(content)
+ }
+ } catch {
+ // ignore OCR errors for a single image
+ } finally {
+ imageBuffer = undefined
+ }
+ }
+
+ return contents
+}
-/** Parse PDF files */
export async function parsePdf(filePath: string, options: DocTextifyOptions): Promise<string> {
let doc: PDFDocumentProxy | undefined
const buffer = await readFile(filePath)
+ const canUseOcr = !!options.ocrWorker
try {
- // Load the document, allowing system fonts as fallback
doc = await getDocumentProxy(new Uint8Array(buffer.buffer, buffer.byteOffset, buffer.byteLength), {
disableFontFace: true,
verbosity: 0
})
- const fragments: string[] = []
- let lastY: number | undefined = undefined
+ const contents: string[] = []
for (let pageNum = 1; pageNum <= doc.numPages; pageNum++) {
const page = await doc.getPage(pageNum)
- const { items } = await page.getTextContent()
-
- for (const item of items) {
- // Skip non-text items
- if (!isTextItem(item)) continue
- const currentY = item.transform[5]
- if (lastY !== undefined && currentY !== lastY) {
- fragments.push(options.newlineDelimiter)
+ try {
+ let pageText = ''
+ try {
+ const { items } = await page.getTextContent()
+ pageText = extractText(items, options.newlineDelimiter).trim()
+ } catch {
+ // ignore text extraction error and fallback to image OCR when possible
}
-
- fragments.push(item.str)
- lastY = currentY
+ const pageHasText = pageText.length > 1
+ if (pageHasText) {
+ contents.push(pageText)
+ continue
+ }
+ if (!canUseOcr) {
+ continue
+ }
+ let images: ExtractedImages = []
+ try {
+ images = await extractImages(doc, pageNum)
+ } catch {
+ // ignore image extraction error for this page
+ }
+ const pageHasImages = images.some(shouldOcrImage)
+ if (!pageHasImages) {
+ continue
+ }
+ const ocrContents = await extractTextFromImages(images, options)
+ if (ocrContents.length > 0) {
+ contents.push(...ocrContents)
+ }
+ } finally {
+ page.cleanup()
}
- page.cleanup()
}
- const content = fragments.join('')
- if (ignorePdfBadFormat.has(content.charCodeAt(0))) {
+ const content = contents.join(options.newlineDelimiter)
+ if (content.length > 0 && ignorePdfBadFormat.has(content.charCodeAt(0))) {
return ''
}
return content
- } catch (e) {
- if (options.outputErrorToConsole) {
- console.error('Error parsing PDF:', e)
- }
- throw e
} finally {
- doc?.destroy().catch((e: Error) => console.error(e))
+ await doc?.destroy().catch(() => undefined)
}
}
diff --git a/backend/src/applications/files/utils/doc-textify/doc-textify.ts b/backend/src/applications/files/utils/doc-textify/doc-textify.ts
index 473c7ae5..1699ecf6 100644
--- a/backend/src/applications/files/utils/doc-textify/doc-textify.ts
+++ b/backend/src/applications/files/utils/doc-textify/doc-textify.ts
@@ -19,7 +19,6 @@ export async function docTextify(
options = {
newlineDelimiter: '\n',
minCharsToExtract: 10,
- outputErrorToConsole: false,
...options
}
@@ -33,32 +32,25 @@ export async function docTextify(
const ext = fileProperties?.extension || path.extname(filePath).slice(1).toLowerCase()
- try {
- switch (ext) {
- case 'docx':
- return cleanContent(await parseWord(filePath, options), options)
- case 'pptx':
- return cleanContent(await parsePowerPoint(filePath, options), options)
- case 'xlsx':
- return cleanContent(await parseExcel(filePath, options), options)
- case 'odt':
- case 'odp':
- case 'ods':
- return cleanContent(await parseOpenOffice(filePath, options), options)
- case 'pdf':
- return cleanContent(await parsePdf(filePath, options), options)
- case 'txt':
- return cleanContent(await parseText(filePath), options)
- case 'html':
- case 'htm':
- return cleanContent(await parseHtml(filePath), options)
- default:
- throw new Error('currently only supports docx, pptx, xlsx, odt, odp, ods, pdf files')
- }
- } catch (e) {
- if (options.outputErrorToConsole) {
- console.error(e.message)
- }
- throw e
+ switch (ext) {
+ case 'docx':
+ return cleanContent(await parseWord(filePath, options), options)
+ case 'pptx':
+ return cleanContent(await parsePowerPoint(filePath, options), options)
+ case 'xlsx':
+ return cleanContent(await parseExcel(filePath, options), options)
+ case 'odt':
+ case 'odp':
+ case 'ods':
+ return cleanContent(await parseOpenOffice(filePath, options), options)
+ case 'pdf':
+ return cleanContent(await parsePdf(filePath, options), options)
+ case 'txt':
+ return cleanContent(await parseText(filePath), options)
+ case 'html':
+ case 'htm':
+ return cleanContent(await parseHtml(filePath), options)
+ default:
+ throw new Error('currently only supports docx, pptx, xlsx, odt, odp, ods, pdf files')
}
}
diff --git a/backend/src/applications/files/utils/doc-textify/interfaces/doc-textify.interfaces.ts b/backend/src/applications/files/utils/doc-textify/interfaces/doc-textify.interfaces.ts
index 9a298f83..d5182424 100644
--- a/backend/src/applications/files/utils/doc-textify/interfaces/doc-textify.interfaces.ts
+++ b/backend/src/applications/files/utils/doc-textify/interfaces/doc-textify.interfaces.ts
@@ -1,5 +1,9 @@
+export interface DocTextifyOCRWorkerLike {
+ recognize: (image: Buffer) => Promise<{ data?: { text?: string } }>
+}
+
export interface DocTextifyOptions {
- outputErrorToConsole?: boolean
newlineDelimiter: string
minCharsToExtract: number
+ ocrWorker?: DocTextifyOCRWorkerLike | null
}
diff --git a/backend/src/applications/files/utils/doc-textify/utils/pdf-ocr.ts b/backend/src/applications/files/utils/doc-textify/utils/pdf-ocr.ts
new file mode 100644
index 00000000..5f088f1a
--- /dev/null
+++ b/backend/src/applications/files/utils/doc-textify/utils/pdf-ocr.ts
@@ -0,0 +1,132 @@
+import type { Logger } from '@nestjs/common'
+import { mkdir } from 'node:fs/promises'
+import path from 'node:path'
+import Tesseract from 'tesseract.js'
+import { configuration } from '../../../../../configuration/config.environment'
+
+export type PdfOCRWorker = Awaited<ReturnType<typeof Tesseract.createWorker>>
+type PdfOCRWorkerOptions = NonNullable<Parameters<typeof Tesseract.createWorker>[2]>
+
+export class PdfOCRWorkerManager {
+ private static instance: PdfOCRWorkerManager
+ public worker: PdfOCRWorker | null
+ private readonly ocrLanguagesPath = path.resolve(__dirname, '../../../assets/ocr-languages')
+ private readonly ocrUserDefinedDpi = '300'
+ private logger: Logger
+ private workerInitializationPromise: Promise<PdfOCRWorker | null> | null
+
+ private constructor(logger: Logger) {
+ this.logger = logger
+ this.worker = null
+ this.workerInitializationPromise = null
+ }
+
+ static getInstance(logger: Logger): PdfOCRWorkerManager {
+ if (!PdfOCRWorkerManager.instance) {
+ PdfOCRWorkerManager.instance = new PdfOCRWorkerManager(logger)
+ }
+ return PdfOCRWorkerManager.instance
+ }
+
+ async start(): Promise<PdfOCRWorker | null> {
+ if (this.worker) {
+ return this.worker
+ }
+ if (this.workerInitializationPromise) {
+ return this.workerInitializationPromise
+ }
+ this.logger.verbose({ tag: this.constructor.name, msg: 'Starting OCR' })
+ this.workerInitializationPromise = this.createWorkerFromConfiguration()
+ .then((worker) => {
+ this.worker = worker
+ return worker
+ })
+ .finally(() => {
+ this.workerInitializationPromise = null
+ })
+ return this.workerInitializationPromise
+ }
+
+ async stop(): Promise<void> {
+ this.logger.verbose({ tag: this.constructor.name, msg: 'Stopping OCR' })
+ if (this.workerInitializationPromise) {
+ await this.workerInitializationPromise.catch((e) =>
+ this.logger.error({
+ tag: this.constructor.name,
+ msg: `Worker promise: ${e}`
+ })
+ )
+ }
+ const worker = this.worker
+ this.worker = null
+ if (!worker) {
+ return
+ }
+ await worker.terminate().catch((e) =>
+ this.logger.error({
+ tag: this.constructor.name,
+ msg: `Worker stop: ${e}`
+ })
+ )
+ }
+
+ private async createConfiguredWorker(languages: string[], options: PdfOCRWorkerOptions): Promise<PdfOCRWorker> {
+ const worker = await Tesseract.createWorker(languages, Tesseract.OEM.LSTM_ONLY, {
+ ...options,
+ errorHandler: (err: unknown) =>
+ this.logger.error({
+ tag: this.constructor.name,
+ msg: `Tesseract worker error: ${err}`
+ }),
+ logger: () => {
+ // intentionally disabled
+ }
+ })
+ await worker.setParameters({
+ user_defined_dpi: this.ocrUserDefinedDpi
+ })
+ return worker
+ }
+
+ private async createWorkerFromConfiguration(): Promise<PdfOCRWorker | null> {
+ const ocrOptions = configuration.applications.files.contentIndexing.ocr
+ if (!ocrOptions.enabled) {
+ return null
+ }
+ try {
+ await mkdir(this.ocrLanguagesPath, { recursive: true })
+ } catch (e) {
+ this.logger.error({
+ tag: this.constructor.name,
+ msg: `unable to create languages directory: ${e}`
+ })
+ }
+ if (ocrOptions.offline) {
+ const offlineWorkerOptions = {
+ langPath: this.ocrLanguagesPath,
+ cacheMethod: 'none' as const
+ }
+ try {
+ // First try for gzipped files: <lang>.traineddata.gz
+ return await this.createConfiguredWorker(ocrOptions.languages, {
+ ...offlineWorkerOptions,
+ gzip: true
+ })
+ } catch (error) {
+ this.logger.warn({
+ tag: this.constructor.name,
+ msg: `unable to load offline OCR languages as .traineddata, retrying with .traineddata.gz: ${error}`
+ })
+ // Fallback for non-gz files: <lang>.traineddata
+ return this.createConfiguredWorker(ocrOptions.languages, {
+ ...offlineWorkerOptions,
+ gzip: false
+ })
+ }
+ }
+ return this.createConfiguredWorker(ocrOptions.languages, {
+ cachePath: this.ocrLanguagesPath,
+ cacheMethod: 'write' as const
+ })
+ }
+}
diff --git a/backend/src/configuration/config.environment.ts b/backend/src/configuration/config.environment.ts
index c85b6057..1fe8e776 100755
--- a/backend/src/configuration/config.environment.ts
+++ b/backend/src/configuration/config.environment.ts
@@ -1,4 +1,5 @@
import { join } from 'node:path'
+import { FilesContentIndexingConfig } from '../applications/files/files.config'
import { AuthTokenAccessConfig, AuthTokenRefreshConfig } from '../authentication/auth.config'
import { ACCESS_KEY, CSRF_KEY, TWO_FA_VERIFY_EXPIRATION, WS_KEY } from '../authentication/constants/auth'
import { TOKEN_TYPE } from '../authentication/interfaces/token.interface'
@@ -41,6 +42,18 @@ function loadConfiguration(): GlobalConfig {
config.applications.files.usersPath = join(config.applications.files.dataPath, 'users')
config.applications.files.spacesPath = join(config.applications.files.dataPath, 'spaces')
config.applications.files.tmpPath = join(config.applications.files.dataPath, 'tmp')
+ // DEPRECATIONS
+ // applications.files.contentIndexing → applications.files.contentIndexing.enabled
+ if (typeof config.applications.files.contentIndexing === 'boolean') {
+ const filesContentIndexingConfig = new FilesContentIndexingConfig()
+ filesContentIndexingConfig.enabled = config.applications.files.contentIndexing
+ config.applications.files.contentIndexing = filesContentIndexingConfig
+ console.warn(
+ '[DEPRECATED][CONFIGURATION] applications.files.contentIndexing is deprecated and will be removed in a future version. ' +
+ 'Please use applications.files.contentIndexing.enabled instead.'
+ )
+ }
+
return transformAndValidate(
GlobalConfig,
config,
diff --git a/environment/environment.dist.yaml b/environment/environment.dist.yaml
index ebdf81bd..78a4ad6b 100755
--- a/environment/environment.dist.yaml
+++ b/environment/environment.dist.yaml
@@ -297,9 +297,23 @@ applications:
dataPath: /home/sync-in
# default: 5368709120 (5 GB)
maxUploadSize: 5368709120
- # Enable indexing of file contents for search (disabling this turns off full-text search)
- # default: true
- contentIndexing: true
+ contentIndexing:
+ # Enable indexing of file contents for search (disabling this turns off full-text search)
+ # default: true
+ enabled: true
+ ocr:
+ # Enable OCR on PDF
+ # default: true
+ enabled: true
+ # OCR languages used by tesseract.js
+ # Supports ISO 639-2/T three-letter codes: 'eng', 'spa', 'fra', 'deu', etc.
+ # examples: `[eng,fra]`, `[fra]`
+ # default: [eng]
+ languages: [eng]
+ # Offline mode: do not download OCR languages, only use local language files from the built-in OCR folder
+ # To download languages, use this scheme: https://cdn.jsdelivr.net/npm/@tesseract.js-data/<lang>@1.0.0/4.0.0_best_int/<lang>.traineddata.gz
+ # default: false
+ offline: false
# Show files starting with a dot in the file explorer
# default: false
showHiddenFiles: false
diff --git a/package-lock.json b/package-lock.json
index 705aa1a3..0e78db0b 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -93,6 +93,7 @@
"sharp": "^0.34.4",
"socket.io": "^4.8.1",
"tar": "^7.5.10",
+ "tesseract.js": "^7.0.0",
"text-to-svg": "^3.1.5",
"time2fa": "^1.4.2",
"unpdf": "^1.3.2",
@@ -13292,6 +13293,12 @@
"readable-stream": "^3.4.0"
}
},
+ "node_modules/bmp-js": {
+ "version": "0.1.0",
+ "resolved": "https://registry.npmjs.org/bmp-js/-/bmp-js-0.1.0.tgz",
+ "integrity": "sha512-vHdS19CnY3hwiNdkaqk93DvjVLfbEcI8mys4UjuWrlX1haDmroo8o4xCzh4wD6DGV6HxRCyauwhHRqMTfERtjw==",
+ "license": "MIT"
+ },
"node_modules/body-parser": {
"version": "2.2.2",
"resolved": "https://registry.npmjs.org/body-parser/-/body-parser-2.2.2.tgz",
@@ -18007,6 +18014,12 @@
"url": "https://opencollective.com/express"
}
},
+ "node_modules/idb-keyval": {
+ "version": "6.2.2",
+ "resolved": "https://registry.npmjs.org/idb-keyval/-/idb-keyval-6.2.2.tgz",
+ "integrity": "sha512-yjD9nARJ/jb1g+CvD0tlhUHOrJ9Sy0P8T9MF3YaLlHnSRpwPfpTX0XIvpmw3gAJUmEu3FiICLBDPXVwyEvrleg==",
+ "license": "Apache-2.0"
+ },
"node_modules/ieee754": {
"version": "1.2.1",
"resolved": "https://registry.npmjs.org/ieee754/-/ieee754-1.2.1.tgz",
@@ -18383,7 +18396,6 @@
"version": "1.2.4",
"resolved": "https://registry.npmjs.org/is-url/-/is-url-1.2.4.tgz",
"integrity": "sha512-ITvGim8FhRiYe4IQ5uHSkj7pVaPDrCTkNd3yq3cV7iZAcJdHTUMPMEHcqSOy9xZ9qFenQCvi+2wjH9a1nXqHww==",
- "dev": true,
"license": "MIT"
},
"node_modules/is-url-superb": {
@@ -22595,6 +22607,26 @@
"lodash": "^4.17.21"
}
},
+ "node_modules/node-fetch": {
+ "version": "2.7.0",
+ "resolved": "https://registry.npmjs.org/node-fetch/-/node-fetch-2.7.0.tgz",
+ "integrity": "sha512-c4FRfUm/dbcWZ7U+1Wq0AwCyFL+3nt2bEw05wfxSz+DWpWsitgmSgYmy2dQdWyKC1694ELPqMs/YzUSNozLt8A==",
+ "license": "MIT",
+ "dependencies": {
+ "whatwg-url": "^5.0.0"
+ },
+ "engines": {
+ "node": "4.x || >=6.0.0"
+ },
+ "peerDependencies": {
+ "encoding": "^0.1.0"
+ },
+ "peerDependenciesMeta": {
+ "encoding": {
+ "optional": true
+ }
+ }
+ },
"node_modules/node-gyp": {
"version": "12.2.0",
"resolved": "https://registry.npmjs.org/node-gyp/-/node-gyp-12.2.0.tgz",
@@ -23055,6 +23087,15 @@
"url": "https://github.com/sponsors/sindresorhus"
}
},
+ "node_modules/opencollective-postinstall": {
+ "version": "2.0.3",
+ "resolved": "https://registry.npmjs.org/opencollective-postinstall/-/opencollective-postinstall-2.0.3.tgz",
+ "integrity": "sha512-8AV/sCtuzUeTo8gQK5qDZzARrulB3egtLzFgteqB2tcT4Mw7B8Kt7JcDHmltjz6FOAHsvTevk70gZEbhM4ZS9Q==",
+ "license": "MIT",
+ "bin": {
+ "opencollective-postinstall": "index.js"
+ }
+ },
"node_modules/openid-client": {
"version": "6.8.2",
"resolved": "https://registry.npmjs.org/openid-client/-/openid-client-6.8.2.tgz",
@@ -24556,6 +24597,12 @@
"integrity": "sha512-urBwgfrvVP/eAyXx4hluJivBKzuEbSQs9rKWCrCkbSxNv8mxPcUZKeuoF3Uy4mJl3Lwprp6yy5/39VWigZ4K6Q==",
"license": "Apache-2.0"
},
+ "node_modules/regenerator-runtime": {
+ "version": "0.13.11",
+ "resolved": "https://registry.npmjs.org/regenerator-runtime/-/regenerator-runtime-0.13.11.tgz",
+ "integrity": "sha512-kY1AZVr2Ra+t+piVaJ4gxaFaReZVH40AKNo7UCX6W+dEwBo/2oZJzqfuN1qLq1oL45o56cPaTXELwrTh8Fpggg==",
+ "license": "MIT"
+ },
"node_modules/require-directory": {
"version": "2.1.1",
"resolved": "https://registry.npmjs.org/require-directory/-/require-directory-2.1.1.tgz",
@@ -26299,6 +26346,30 @@
"dev": true,
"license": "MIT"
},
+ "node_modules/tesseract.js": {
+ "version": "7.0.0",
+ "resolved": "https://registry.npmjs.org/tesseract.js/-/tesseract.js-7.0.0.tgz",
+ "integrity": "sha512-exPBkd+z+wM1BuMkx/Bjv43OeLBxhL5kKWsz/9JY+DXcXdiBjiAch0V49QR3oAJqCaL5qURE0vx9Eo+G5YE7mA==",
+ "hasInstallScript": true,
+ "license": "Apache-2.0",
+ "dependencies": {
+ "bmp-js": "^0.1.0",
+ "idb-keyval": "^6.2.0",
+ "is-url": "^1.2.4",
+ "node-fetch": "^2.6.9",
+ "opencollective-postinstall": "^2.0.3",
+ "regenerator-runtime": "^0.13.3",
+ "tesseract.js-core": "^7.0.0",
+ "wasm-feature-detect": "^1.8.0",
+ "zlibjs": "^0.3.1"
+ }
+ },
+ "node_modules/tesseract.js-core": {
+ "version": "7.0.0",
+ "resolved": "https://registry.npmjs.org/tesseract.js-core/-/tesseract.js-core-7.0.0.tgz",
+ "integrity": "sha512-WnNH518NzmbSq9zgTPeoF8c+xmilS8rFIl1YKbk/ptuuc7p6cLNELNuPAzcmsYw450ca6bLa8j3t0VAtq435Vw==",
+ "license": "Apache-2.0"
+ },
"node_modules/test-exclude": {
"version": "6.0.0",
"resolved": "https://registry.npmjs.org/test-exclude/-/test-exclude-6.0.0.tgz",
@@ -26525,6 +26596,12 @@
"url": "https://github.com/sponsors/Borewit"
}
},
+ "node_modules/tr46": {
+ "version": "0.0.3",
+ "resolved": "https://registry.npmjs.org/tr46/-/tr46-0.0.3.tgz",
+ "integrity": "sha512-N3WMsuqV66lT30CrXNbEjx4GEwlow3v6rr4mCcv6prnfwhS01rkgyFdjPNBYd9br7LpXV1+Emh01fHnq2Gdgrw==",
+ "license": "MIT"
+ },
"node_modules/trim-newlines": {
"version": "3.0.1",
"resolved": "https://registry.npmjs.org/trim-newlines/-/trim-newlines-3.0.1.tgz",
@@ -27430,6 +27507,12 @@
"makeerror": "1.0.12"
}
},
+ "node_modules/wasm-feature-detect": {
+ "version": "1.8.0",
+ "resolved": "https://registry.npmjs.org/wasm-feature-detect/-/wasm-feature-detect-1.8.0.tgz",
+ "integrity": "sha512-zksaLKM2fVlnB5jQQDqKXXwYHLQUVH9es+5TOOHwGOVJOCeRBCiPjwSg+3tN2AdTCzjgli4jijCH290kXb/zWQ==",
+ "license": "Apache-2.0"
+ },
"node_modules/watchpack": {
"version": "2.4.4",
"resolved": "https://registry.npmjs.org/watchpack/-/watchpack-2.4.4.tgz",
@@ -27475,6 +27558,12 @@
"license": "MIT",
"optional": true
},
+ "node_modules/webidl-conversions": {
+ "version": "3.0.1",
+ "resolved": "https://registry.npmjs.org/webidl-conversions/-/webidl-conversions-3.0.1.tgz",
+ "integrity": "sha512-2JAn3z8AR6rjK8Sm8orRC0h/bcl/DqL7tRPdGZ4I1CjdF+EaMLmYxBHyXuKL849eucPFhvBoxMsflfOb8kxaeQ==",
+ "license": "BSD-2-Clause"
+ },
"node_modules/webpack": {
"version": "5.105.4",
"resolved": "https://registry.npmjs.org/webpack/-/webpack-5.105.4.tgz",
@@ -27651,6 +27740,16 @@
"node": ">=10.13.0"
}
},
+ "node_modules/whatwg-url": {
+ "version": "5.0.0",
+ "resolved": "https://registry.npmjs.org/whatwg-url/-/whatwg-url-5.0.0.tgz",
+ "integrity": "sha512-saE57nupxk6v3HY35+jzBwYa0rKSy0XR8JSxZPwgLr7ys0IBzhGviA1/TUGJLmSVqs8pb9AnvICXEuOHLprYTw==",
+ "license": "MIT",
+ "dependencies": {
+ "tr46": "~0.0.3",
+ "webidl-conversions": "^3.0.0"
+ }
+ },
"node_modules/which": {
"version": "2.0.2",
"resolved": "https://registry.npmjs.org/which/-/which-2.0.2.tgz",
@@ -28032,6 +28131,15 @@
"node": "^12.22.0 || ^14.17.0 || >=16.0.0"
}
},
+ "node_modules/zlibjs": {
+ "version": "0.3.1",
+ "resolved": "https://registry.npmjs.org/zlibjs/-/zlibjs-0.3.1.tgz",
+ "integrity": "sha512-+J9RrgTKOmlxFSDHo0pI1xM6BLVUv+o0ZT9ANtCxGkjIVCCUdx9alUF8Gm+dGLKbkkkidWIHFDZHDMpfITt4+w==",
+ "license": "MIT",
+ "engines": {
+ "node": "*"
+ }
+ },
"node_modules/zod": {
"version": "4.1.13",
"resolved": "https://registry.npmjs.org/zod/-/zod-4.1.13.tgz",