diff --git a/js/i18n.d.ts b/js/i18n.d.ts new file mode 100644 index 0000000..4eccda4 --- /dev/null +++ b/js/i18n.d.ts @@ -0,0 +1,2 @@ +export function t(key: string): string; +export function initializeI18n(): void; diff --git a/js/i18n.js b/js/i18n.js index 09f9365..9542849 100644 --- a/js/i18n.js +++ b/js/i18n.js @@ -8,6 +8,7 @@ import { imageResizeLocales } from "./locales/image-resize.js"; import { imageCompressorLocales } from "./locales/image-compressor.js"; import { imageMetadataLocales } from "./locales/image-metadata.js"; import { imageToTextLocales } from "./locales/image-to-text.js"; +import { pdfToTextLocales } from "./locales/pdf-to-text.js"; import { privacyHubLocales } from "./locales/privacy-hub.js"; import { metadataUxLocales } from "./locales/metadata-ux.js"; @@ -15,13 +16,14 @@ const STORAGE_KEY = "secure-tools-language"; const baseTranslations = { en, ko, ja, es, de, fr }; export const translations = Object.fromEntries(Object.entries(baseTranslations).map(([language, catalog]) => [language, { ...catalog, - metadata: { ...catalog.metadata, imageResize: imageResizeLocales[language].metadata, imageCompressor: imageCompressorLocales[language].metadata, imageMetadata: imageMetadataLocales[language].metadata, imageToText: imageToTextLocales[language].metadata, privacyCategory: privacyHubLocales[language].metadata }, - tools: { ...catalog.tools, imageMetadata: imageMetadataLocales[language].toolName, imageToText: imageToTextLocales[language].toolName, categoryDescriptions: { ...catalog.tools.categoryDescriptions, privacy: privacyHubLocales[language].categoryDescription } }, - categories: { ...catalog.categories, image: { ...catalog.categories.image, metadata: imageMetadataLocales[language].categoryDescription, toText: imageToTextLocales[language].categoryDescription } }, + metadata: { ...catalog.metadata, imageResize: imageResizeLocales[language].metadata, imageCompressor: imageCompressorLocales[language].metadata, imageMetadata: imageMetadataLocales[language].metadata, imageToText: imageToTextLocales[language].metadata, pdfToText: pdfToTextLocales[language].metadata, privacyCategory: privacyHubLocales[language].metadata }, + tools: { ...catalog.tools, imageMetadata: imageMetadataLocales[language].toolName, imageToText: imageToTextLocales[language].toolName, pdfToText: pdfToTextLocales[language].toolName, categoryDescriptions: { ...catalog.tools.categoryDescriptions, privacy: privacyHubLocales[language].categoryDescription } }, + categories: { ...catalog.categories, pdf: { ...catalog.categories.pdf, toText: pdfToTextLocales[language].categoryDescription }, image: { ...catalog.categories.image, metadata: imageMetadataLocales[language].categoryDescription, toText: imageToTextLocales[language].categoryDescription }, scan: { ...catalog.categories.scan, pdfToText: pdfToTextLocales[language].categoryDescription } }, imageResize: imageResizeLocales[language].copy, imageCompressor: imageCompressorLocales[language].copy, imageMetadata: { ...imageMetadataLocales[language].copy, source: { ...imageMetadataLocales[language].copy.source, ...metadataUxLocales[language].image.source }, inspector: { ...imageMetadataLocales[language].copy.inspector, ...metadataUxLocales[language].image.inspector }, clean: { ...imageMetadataLocales[language].copy.clean, ...metadataUxLocales[language].image.clean }, policy: metadataUxLocales[language].image.policy }, imageToText: imageToTextLocales[language].copy, + pdfToText: pdfToTextLocales[language].copy, pdfMetadata: { ...catalog.pdfMetadata, source: { ...catalog.pdfMetadata.source, ...metadataUxLocales[language].pdf.source }, inspector: { ...catalog.pdfMetadata.inspector, ...metadataUxLocales[language].pdf.inspector }, actions: { ...catalog.pdfMetadata.actions, ...metadataUxLocales[language].pdf.actions }, custom: metadataUxLocales[language].pdf.custom, errors: { ...catalog.pdfMetadata.errors, ...metadataUxLocales[language].pdf.errors } }, privacyHub: privacyHubLocales[language].copy, }])); diff --git a/js/locales/pdf-to-text.js b/js/locales/pdf-to-text.js new file mode 100644 index 0000000..2e7caa7 --- /dev/null +++ b/js/locales/pdf-to-text.js @@ -0,0 +1,65 @@ +const copy = { + eyebrow: "Local PDF OCR", title: "PDF to Text", description: "Recognize editable text from every page or a selected page range without uploading your PDF.", + drop: { title: "Add one PDF", description: "Drop a PDF here or use the picker.", choose: "Choose PDF", localTitle: "Processed locally.", localBody: "The PDF, rendered pages, and recognized text stay on this device.", privacyLink: "How privacy works" }, + source: { title: "Source PDF", empty: "No PDF selected yet.", selectedLabel: "Selected source PDF", meta: "{pages} pages · {size}", replace: "Replace PDF", remove: "Remove PDF" }, + warning: { large: "Large PDFs can take a while. Keep this tab open while local OCR runs." }, + settings: { title: "Recognition", pages: "Pages", all: "All pages", selected: "Selected pages", range: "Page range", rangePlaceholder: "1, 3-5", rangeHelp: "Use commas and ranges, such as 1, 3-5.", language: "Text language", english: "English", korean: "Korean", combined: "English + Korean", recognize: "Recognize text", cancel: "Cancel recognition", progress: "Recognition progress" }, + result: { title: "Recognized text", description: "Review and edit each page before copying or downloading it.", page: "Page {page}", pageLabel: "Editable recognized text for page {page}", copyPage: "Copy page", copyAll: "Copy all", download: "Download TXT" }, + status: { preparing: "Preparing the PDF locally…", ready: "PDF ready. Choose pages and a language to begin.", cancelled: "Recognition cancelled. The PDF remains ready.", success: "Recognition complete. You can edit every page.", copied: "All recognized text copied.", pageCopied: "Page {page} copied.", downloaded: "Downloaded {name}." }, + progress: { starting: "Starting local recognition…", loading: "Loading the PDF locally…", "rendering-page": "Rendering page {page} of {total}…", "recognizing-page": "Recognizing page {page} of {total}… {percent}%", complete: "Finishing recognition…" }, + errors: { oneFile: "Choose exactly one PDF.", UNSUPPORTED_PDF: "Choose a valid PDF file.", UNREADABLE_PDF: "This PDF could not be read.", ENCRYPTED_PDF: "Password-protected PDFs are not supported.", PDF_HAS_NO_PAGES: "This PDF has no pages.", PAGE_RANGE_REQUIRED: "Enter at least one page.", PAGE_RANGE_INVALID: "Enter pages as numbers and ranges, such as 1, 3-5.", PAGE_RANGE_REVERSED: "A page range cannot run backwards.", PAGE_OUT_OF_RANGE: "One or more pages are outside this PDF.", PDF_OCR_PAGE_DUPLICATE: "Choose each page only once.", PDF_OCR_RENDER_DIMENSION_EXCEEDED: "A page is too large to render safely.", PDF_OCR_RENDER_PIXELS_EXCEEDED: "A page is too large to render safely.", PDF_OCR_FAILED: "Recognition failed. Try again or choose another PDF.", PDF_LIBRARY_UNAVAILABLE: "The local PDF reader is unavailable.", copy: "The text could not be copied. Select it manually and copy it.", download: "The text file could not be downloaded." }, +}; +const localized = { + en: { title: "PDF to Text", description: copy.description, toolName: "PDF to Text OCR", category: "Extract editable text from PDF pages with local English and Korean OCR." }, + ko: { title: "PDF → 텍스트", description: "PDF를 업로드하지 않고 전체 또는 선택한 페이지에서 편집 가능한 텍스트를 인식합니다.", toolName: "PDF 텍스트 OCR", category: "로컬 영어·한국어 OCR로 PDF 페이지에서 편집 가능한 텍스트를 추출합니다." }, + ja: { title: "PDFからテキスト", description: "PDFをアップロードせず、全ページまたは選択ページから編集可能なテキストを認識します。", toolName: "PDFテキストOCR", category: "ローカルの英語・韓国語OCRでPDFページから編集可能なテキストを抽出します。" }, + es: { title: "PDF a texto", description: "Reconoce texto editable de todas las páginas o de una selección sin subir el PDF.", toolName: "OCR de PDF a texto", category: "Extrae texto editable de páginas PDF con OCR local en inglés y coreano." }, + de: { title: "PDF zu Text", description: "Erkennt bearbeitbaren Text auf allen oder ausgewählten Seiten, ohne die PDF hochzuladen.", toolName: "PDF-zu-Text-OCR", category: "Extrahiert bearbeitbaren Text aus PDF-Seiten mit lokaler englischer und koreanischer OCR." }, + fr: { title: "PDF en texte", description: "Reconnaît du texte modifiable dans toutes les pages ou une sélection sans téléverser le PDF.", toolName: "OCR PDF en texte", category: "Extrait du texte modifiable des pages PDF par OCR local anglais et coréen." }, +}; +const overrides = { + ko: { + eyebrow: "로컬 PDF OCR", drop: { title: "PDF 한 개 추가", description: "PDF를 놓거나 파일 선택기를 사용하세요.", choose: "PDF 선택", localTitle: "로컬에서 처리됩니다.", localBody: "PDF, 렌더링한 페이지, 인식된 텍스트는 이 기기에만 남습니다.", privacyLink: "개인정보 보호 방식" }, + source: { title: "원본 PDF", empty: "선택한 PDF가 없습니다.", selectedLabel: "선택한 원본 PDF", meta: "{pages}페이지 · {size}", replace: "PDF 교체", remove: "PDF 제거" }, warning: { large: "큰 PDF는 시간이 걸릴 수 있습니다. 로컬 OCR이 실행되는 동안 이 탭을 열어 두세요." }, + settings: { title: "텍스트 인식", pages: "페이지", all: "모든 페이지", selected: "선택한 페이지", range: "페이지 범위", rangePlaceholder: "1, 3-5", rangeHelp: "1, 3-5처럼 쉼표와 범위를 사용하세요.", language: "텍스트 언어", english: "영어", korean: "한국어", combined: "영어 + 한국어", recognize: "텍스트 인식", cancel: "인식 취소", progress: "인식 진행률" }, + result: { title: "인식된 텍스트", description: "복사하거나 다운로드하기 전에 각 페이지를 확인하고 편집하세요.", page: "{page}페이지", pageLabel: "{page}페이지의 편집 가능한 인식 텍스트", copyPage: "페이지 복사", copyAll: "모두 복사", download: "TXT 다운로드" }, + status: { preparing: "PDF를 로컬에서 준비하는 중…", ready: "PDF가 준비되었습니다. 페이지와 언어를 선택하세요.", cancelled: "인식을 취소했습니다. PDF는 준비 상태로 유지됩니다.", success: "인식이 완료되었습니다. 각 페이지를 편집할 수 있습니다.", copied: "인식된 텍스트를 모두 복사했습니다.", pageCopied: "{page}페이지를 복사했습니다.", downloaded: "{name}을 다운로드했습니다." }, + progress: { starting: "로컬 인식을 시작하는 중…", loading: "PDF를 로컬에서 불러오는 중…", "rendering-page": "{total}페이지 중 {page}페이지를 렌더링하는 중…", "recognizing-page": "{total}페이지 중 {page}페이지를 인식하는 중… {percent}%", complete: "인식을 마무리하는 중…" }, + }, + ja: { + eyebrow: "ローカルPDF OCR", drop: { title: "PDFを1件追加", description: "PDFをドロップするか、ファイル選択を使用します。", choose: "PDFを選択", localTitle: "ローカル処理。", localBody: "PDF、描画ページ、認識テキストは端末外へ送信されません。", privacyLink: "プライバシーの仕組み" }, + source: { title: "元のPDF", empty: "PDFが未選択です。", selectedLabel: "選択した元のPDF", meta: "{pages}ページ · {size}", replace: "PDFを置換", remove: "PDFを削除" }, warning: { large: "大きなPDFは時間がかかることがあります。ローカルOCR中はこのタブを開いたままにしてください。" }, + settings: { title: "文字認識", pages: "ページ", all: "全ページ", selected: "選択したページ", range: "ページ範囲", rangePlaceholder: "1, 3-5", rangeHelp: "1, 3-5のようにカンマと範囲を使用します。", language: "テキスト言語", english: "英語", korean: "韓国語", combined: "英語 + 韓国語", recognize: "テキストを認識", cancel: "認識をキャンセル", progress: "認識の進行状況" }, + result: { title: "認識したテキスト", description: "コピーまたはダウンロード前に各ページを確認・編集できます。", page: "ページ{page}", pageLabel: "ページ{page}の編集可能な認識テキスト", copyPage: "ページをコピー", copyAll: "すべてコピー", download: "TXTをダウンロード" }, + status: { preparing: "PDFをローカルで準備中…", ready: "PDFの準備ができました。ページと言語を選択してください。", cancelled: "認識をキャンセルしました。PDFは準備済みです。", success: "認識が完了しました。各ページを編集できます。", copied: "認識テキストをすべてコピーしました。", pageCopied: "ページ{page}をコピーしました。", downloaded: "{name}をダウンロードしました。" }, + progress: { starting: "ローカル認識を開始中…", loading: "PDFをローカルで読み込み中…", "rendering-page": "{total}ページ中{page}ページを描画中…", "recognizing-page": "{total}ページ中{page}ページを認識中… {percent}%", complete: "認識を完了中…" }, + }, + es: { + eyebrow: "OCR local de PDF", drop: { title: "Añadir un PDF", description: "Suelta un PDF aquí o usa el selector.", choose: "Elegir PDF", localTitle: "Procesamiento local.", localBody: "El PDF, las páginas renderizadas y el texto reconocido permanecen en este dispositivo.", privacyLink: "Cómo funciona la privacidad" }, + source: { title: "PDF de origen", empty: "No hay ningún PDF seleccionado.", selectedLabel: "PDF de origen seleccionado", meta: "{pages} páginas · {size}", replace: "Reemplazar PDF", remove: "Quitar PDF" }, warning: { large: "Los PDF grandes pueden tardar. Mantén esta pestaña abierta mientras se ejecuta el OCR local." }, + settings: { title: "Reconocimiento", pages: "Páginas", all: "Todas las páginas", selected: "Páginas seleccionadas", range: "Rango de páginas", rangePlaceholder: "1, 3-5", rangeHelp: "Usa comas y rangos, por ejemplo 1, 3-5.", language: "Idioma del texto", english: "Inglés", korean: "Coreano", combined: "Inglés + coreano", recognize: "Reconocer texto", cancel: "Cancelar reconocimiento", progress: "Progreso del reconocimiento" }, + result: { title: "Texto reconocido", description: "Revisa y edita cada página antes de copiarla o descargarla.", page: "Página {page}", pageLabel: "Texto reconocido editable de la página {page}", copyPage: "Copiar página", copyAll: "Copiar todo", download: "Descargar TXT" }, + status: { preparing: "Preparando el PDF localmente…", ready: "PDF listo. Elige páginas e idioma para comenzar.", cancelled: "Reconocimiento cancelado. El PDF sigue listo.", success: "Reconocimiento terminado. Puedes editar cada página.", copied: "Se copió todo el texto reconocido.", pageCopied: "Página {page} copiada.", downloaded: "Se descargó {name}." }, + progress: { starting: "Iniciando el reconocimiento local…", loading: "Cargando el PDF localmente…", "rendering-page": "Renderizando página {page} de {total}…", "recognizing-page": "Reconociendo página {page} de {total}… {percent}%", complete: "Finalizando el reconocimiento…" }, + }, + de: { + eyebrow: "Lokale PDF-OCR", drop: { title: "Eine PDF hinzufügen", description: "PDF hier ablegen oder Dateiauswahl verwenden.", choose: "PDF auswählen", localTitle: "Lokale Verarbeitung.", localBody: "PDF, gerenderte Seiten und erkannter Text bleiben auf diesem Gerät.", privacyLink: "So funktioniert der Datenschutz" }, + source: { title: "Quell-PDF", empty: "Keine PDF ausgewählt.", selectedLabel: "Ausgewählte Quell-PDF", meta: "{pages} Seiten · {size}", replace: "PDF ersetzen", remove: "PDF entfernen" }, warning: { large: "Große PDFs können länger dauern. Lassen Sie diesen Tab während der lokalen OCR geöffnet." }, + settings: { title: "Erkennung", pages: "Seiten", all: "Alle Seiten", selected: "Ausgewählte Seiten", range: "Seitenbereich", rangePlaceholder: "1, 3-5", rangeHelp: "Kommas und Bereiche verwenden, zum Beispiel 1, 3-5.", language: "Textsprache", english: "Englisch", korean: "Koreanisch", combined: "Englisch + Koreanisch", recognize: "Text erkennen", cancel: "Erkennung abbrechen", progress: "Erkennungsfortschritt" }, + result: { title: "Erkannter Text", description: "Jede Seite vor dem Kopieren oder Herunterladen prüfen und bearbeiten.", page: "Seite {page}", pageLabel: "Bearbeitbarer erkannter Text für Seite {page}", copyPage: "Seite kopieren", copyAll: "Alles kopieren", download: "TXT herunterladen" }, + status: { preparing: "PDF wird lokal vorbereitet…", ready: "PDF bereit. Seiten und Sprache auswählen.", cancelled: "Erkennung abgebrochen. Die PDF bleibt bereit.", success: "Erkennung abgeschlossen. Jede Seite kann bearbeitet werden.", copied: "Gesamten erkannten Text kopiert.", pageCopied: "Seite {page} kopiert.", downloaded: "{name} wurde heruntergeladen." }, + progress: { starting: "Lokale Erkennung wird gestartet…", loading: "PDF wird lokal geladen…", "rendering-page": "Seite {page} von {total} wird gerendert…", "recognizing-page": "Seite {page} von {total} wird erkannt… {percent}%", complete: "Erkennung wird abgeschlossen…" }, + }, + fr: { + eyebrow: "OCR PDF local", drop: { title: "Ajouter un PDF", description: "Déposez un PDF ici ou utilisez le sélecteur.", choose: "Choisir un PDF", localTitle: "Traitement local.", localBody: "Le PDF, les pages rendues et le texte reconnu restent sur cet appareil.", privacyLink: "Fonctionnement de la confidentialité" }, + source: { title: "PDF source", empty: "Aucun PDF sélectionné.", selectedLabel: "PDF source sélectionné", meta: "{pages} pages · {size}", replace: "Remplacer le PDF", remove: "Retirer le PDF" }, warning: { large: "Les PDF volumineux peuvent prendre du temps. Gardez cet onglet ouvert pendant l’OCR local." }, + settings: { title: "Reconnaissance", pages: "Pages", all: "Toutes les pages", selected: "Pages sélectionnées", range: "Plage de pages", rangePlaceholder: "1, 3-5", rangeHelp: "Utilisez des virgules et des plages, par exemple 1, 3-5.", language: "Langue du texte", english: "Anglais", korean: "Coréen", combined: "Anglais + coréen", recognize: "Reconnaître le texte", cancel: "Annuler la reconnaissance", progress: "Progression de la reconnaissance" }, + result: { title: "Texte reconnu", description: "Vérifiez et modifiez chaque page avant de la copier ou de la télécharger.", page: "Page {page}", pageLabel: "Texte reconnu modifiable de la page {page}", copyPage: "Copier la page", copyAll: "Tout copier", download: "Télécharger le TXT" }, + status: { preparing: "Préparation locale du PDF…", ready: "PDF prêt. Choisissez les pages et la langue.", cancelled: "Reconnaissance annulée. Le PDF reste prêt.", success: "Reconnaissance terminée. Chaque page est modifiable.", copied: "Tout le texte reconnu a été copié.", pageCopied: "Page {page} copiée.", downloaded: "{name} téléchargé." }, + progress: { starting: "Démarrage de la reconnaissance locale…", loading: "Chargement local du PDF…", "rendering-page": "Rendu de la page {page} sur {total}…", "recognizing-page": "Reconnaissance de la page {page} sur {total}… {percent}%", complete: "Finalisation de la reconnaissance…" }, + }, +}; +export const pdfToTextLocales = Object.fromEntries(Object.entries(localized).map(([language, value]) => [language, { + metadata: { title: `${value.toolName} — Secure Tools`, description: value.description }, toolName: value.toolName, + categoryDescription: value.category, copy: Object.fromEntries(Object.entries({ ...copy, title: value.title, description: value.description }).map(([key, item]) => [key, item && typeof item === "object" ? { ...item, ...(overrides[language]?.[key] || {}) } : (overrides[language]?.[key] || item)])), +}])); diff --git a/scripts/site-routes.mjs b/scripts/site-routes.mjs index 4465863..025ffe9 100644 --- a/scripts/site-routes.mjs +++ b/scripts/site-routes.mjs @@ -11,6 +11,7 @@ export const canonicalPages = [ { source: "tools/pdf/organize/index.html", route: "/pdf/organize/" }, { source: "tools/pdf/to-images/index.html", route: "/pdf/to-images/" }, { source: "tools/pdf/metadata/index.html", route: "/pdf/metadata/" }, + { source: "tools/pdf/to-text/index.html", route: "/pdf/to-text/", legacy: false }, { source: "tools/image/index.html", route: "/image/" }, { source: "tools/image/converter/index.html", route: "/image/converter/" }, { source: "tools/image/resize/index.html", route: "/image/resize/" }, @@ -23,7 +24,7 @@ export const canonicalPages = [ export const legacyRedirects = [ ...canonicalPages - .filter(({ source }) => source.startsWith("tools/")) + .filter(({ source, legacy }) => source.startsWith("tools/") && legacy !== false) .map(({ route }) => ({ from: `/tools${route}`, to: route })), { from: "/tools/privacy/", to: "/privacy/" }, { from: "/tools/image-to-pdf/", to: "/pdf/images-to-pdf/" }, diff --git a/scripts/typescript-modules.mjs b/scripts/typescript-modules.mjs index 6951c96..2205fbb 100644 --- a/scripts/typescript-modules.mjs +++ b/scripts/typescript-modules.mjs @@ -1,4 +1,19 @@ export const compiledBrowserModules = Object.freeze([ + Object.freeze({ + source: "tools/pdf/to-text/app.ts", + compiled: "tools/pdf/to-text/app.js", + public: "pdf/to-text/app.js", + }), + Object.freeze({ + source: "tools/pdf/to-text/controller.ts", + compiled: "tools/pdf/to-text/controller.js", + public: "pdf/to-text/controller.js", + }), + Object.freeze({ + source: "tools/pdf/to-text/output.ts", + compiled: "tools/pdf/to-text/output.js", + public: "pdf/to-text/output.js", + }), Object.freeze({ source: "tools/shared/pdf-ocr.ts", compiled: "tools/shared/pdf-ocr.js", diff --git a/sitemap.xml b/sitemap.xml index 574798e..e0fc784 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -10,6 +10,7 @@ https://tools.securetools.app/pdf/organize/ https://tools.securetools.app/pdf/to-images/ https://tools.securetools.app/pdf/metadata/ + https://tools.securetools.app/pdf/to-text/ https://tools.securetools.app/image/ https://tools.securetools.app/image/converter/ https://tools.securetools.app/image/resize/ diff --git a/tests/browser/pdf-to-text-smoke.html b/tests/browser/pdf-to-text-smoke.html new file mode 100644 index 0000000..06e6d9a --- /dev/null +++ b/tests/browser/pdf-to-text-smoke.html @@ -0,0 +1,3 @@ + +PDF to Text browser smoke +

PDF to Text browser smoke

Running…

    diff --git a/tests/browser/pdf-to-text-smoke.js b/tests/browser/pdf-to-text-smoke.js new file mode 100644 index 0000000..ee89b71 --- /dev/null +++ b/tests/browser/pdf-to-text-smoke.js @@ -0,0 +1,59 @@ +import { createPdfToTextController } from "/pdf/to-text/controller.js"; +import { copyText, downloadPdfText, formatPdfOcrText } from "/pdf/to-text/output.js"; +import { translations } from "/js/i18n.js"; + +const status = document.querySelector("#status"); +const results = document.querySelector("#results"); +const checks = []; +const assert = (condition, message) => { if (!condition) throw new Error(message); checks.push(message); }; +const file = (name) => new File(["%PDF-1.7"], name, { type: "application/pdf" }); +const show = () => { results.replaceChildren(...checks.map((message) => Object.assign(document.createElement("li"), { textContent: message }))); }; + +try { + const happy = createPdfToTextController({ + inspect: async () => ({ pageCount: 2 }), + service: { + async recognizeDocument(request) { + const pages = [{ pageNumber: 1, text: "first" }, { pageNumber: 2, text: "second" }]; + pages.forEach((page, index) => { request.onProgress?.({ phase: "recognizing-page", pageNumber: page.pageNumber, pageIndex: index + 1, pageCount: 2, documentPageCount: 2, pageProgress: 1, overallProgress: (index + 1) / 2, ocrStage: "recognizing" }); request.onPageResult?.(page); }); + return { pageCount: 2, selectedPageNumbers: [1, 2], pages, largeDocument: false }; + }, async cancel() { return false; }, async dispose() {}, + }, + }); + await happy.select(file("multi.pdf")); await happy.recognize("eng", { mode: "all" }); + assert(happy.getState().phase === "success", "happy path reaches success"); + assert(happy.getState().pages.map(({ pageNumber }) => pageNumber).join(",") === "1,2", "multi-page order is preserved"); + + let release; + const cancelled = createPdfToTextController({ inspect: async () => ({ pageCount: 1 }), service: { + async recognizeDocument() { await new Promise((resolve) => { release = resolve; }); throw Object.assign(new Error("cancelled"), { code: "PDF_OCR_CANCELLED" }); }, + async cancel() { release?.(); return true; }, async dispose() {}, + } }); + await cancelled.select(file("cancel.pdf")); const job = cancelled.recognize("eng", { mode: "all" }); await cancelled.cancel(); await job; + assert(cancelled.getState().phase === "cancelled" && cancelled.getState().pages.length === 0, "cancellation suppresses stale results"); + + let attempts = 0; + const retry = createPdfToTextController({ inspect: async () => ({ pageCount: 1 }), service: { + async recognizeDocument() { attempts += 1; if (attempts === 1) throw Object.assign(new Error("failed"), { code: "PDF_OCR_FAILED" }); return { pages: [{ pageNumber: 1, text: "recovered" }] }; }, + async cancel() { return false; }, async dispose() {}, + } }); + await retry.select(file("retry.pdf")); await retry.recognize("eng", { mode: "all" }); await retry.recognize("eng", { mode: "all" }); + assert(retry.getState().phase === "success", "retry recovers after a controlled failure"); + + let releaseOld; + const replacement = createPdfToTextController({ service: { async recognizeDocument() {}, async cancel() { return false; }, async dispose() {} }, inspect: async (source) => source.name === "old.pdf" ? new Promise((resolve) => { releaseOld = () => resolve({ pageCount: 9 }); }) : { pageCount: 2 } }); + const oldJob = replacement.select(file("old.pdf")); await replacement.select(file("new.pdf")); releaseOld(); await oldJob; + assert(replacement.getState().source.file.name === "new.pdf", "source replacement rejects stale preparation"); + + let clipboard = ""; + const text = formatPdfOcrText(happy.getState().pages); + await copyText(text, { navigatorObject: { clipboard: { writeText: async (value) => { clipboard = value; } } } }); + assert(clipboard.includes("--- Page 1 ---") && clipboard.includes("--- Page 2 ---"), "copy-all preserves page boundaries"); + let blob; const anchor = { click() { this.clicked = true; }, remove() {} }; + downloadPdfText(happy.getState().pages, "multi.pdf", { documentObject: { body: { append() {} }, createElement: () => anchor }, urlObject: { createObjectURL(value) { blob = value; return "blob:test"; }, revokeObjectURL() {} }, schedule() {} }); + assert(anchor.download === "multi.txt" && (await blob.text()) === text, "TXT export preserves filename and UTF-8 content"); + assert(["en", "ko", "ja", "es", "de", "fr"].every((language) => translations[language].pdfToText.title && translations[language].pdfToText.result.copyAll), "all six locale catalogs load without raw keys"); + show(); status.textContent = `${checks.length} browser checks passed.`; document.body.dataset.state = "passed"; +} catch (error) { + status.textContent = error instanceof Error ? error.message : String(error); document.body.dataset.state = "failed"; console.error(error); +} diff --git a/tests/category-availability.test.mjs b/tests/category-availability.test.mjs index 2cdc228..141281f 100644 --- a/tests/category-availability.test.mjs +++ b/tests/category-availability.test.mjs @@ -84,8 +84,8 @@ assert.match(privacyHtml, /data-i18n="privacyHub\.pdfDescription"/); assertPublicRoutesExist("/privacy/", privacyRoutes); const pdfList = categoryList(read("tools/pdf/index.html")); -assert.equal(linkedRoutes(pdfList).length, 6, "Every PDF production card must remain linked"); -assert.equal((pdfList.match(/status--available/g) || []).length, 6); +assert.equal(linkedRoutes(pdfList).length, 7, "Every PDF production card must remain linked"); +assert.equal((pdfList.match(/status--available/g) || []).length, 7); for (const category of ["pdf", "image", "scan"]) { const html = read(`tools/${category}/index.html`); diff --git a/tests/i18n-quality.test.mjs b/tests/i18n-quality.test.mjs index 39f5a2f..baa84d0 100644 --- a/tests/i18n-quality.test.mjs +++ b/tests/i18n-quality.test.mjs @@ -46,7 +46,7 @@ function placeholders(value) { function testCatalogParityAndQuality() { assert.deepEqual([...Object.keys(translations)], [...languageNames.keys()]); const english = flatten(translations.en); - assert.equal(english.size, 817); + assert.equal(english.size, 887); for (const [language, catalog] of Object.entries(translations)) { const flattened = flatten(catalog); @@ -78,7 +78,7 @@ function testResolutionDetectionAndPersistence() { function testSelectorsAndDocumentTranslation() { const pages = canonicalPages.map(({ source }) => path.join(root, source)); - assert.equal(pages.length, 18, "Every canonical page comes from the route manifest"); + assert.equal(pages.length, 19, "Every canonical page comes from the route manifest"); for (const file of pages) { const html = fs.readFileSync(file, "utf8"); const select = html.match(/]*data-language-select[^>]*>([\s\S]*?)<\/select>/)?.[1]; diff --git a/tests/ocr-foundation.test.mjs b/tests/ocr-foundation.test.mjs index b11f2bd..9ef4817 100644 --- a/tests/ocr-foundation.test.mjs +++ b/tests/ocr-foundation.test.mjs @@ -210,7 +210,7 @@ const publicOcrReferences = listAbsoluteFiles(path.join(root, "tools")) .filter((file) => file.endsWith(".html")) .map((file) => path.relative(root, file).replaceAll("\\", "/")) .filter((relative) => read(relative).includes("assets/vendor/tesseract")); -assert.deepEqual(publicOcrReferences, ["tools/image/to-text/index.html"], "OCR runtime must stay lazy to its public route"); +assert.deepEqual(publicOcrReferences, ["tools/image/to-text/index.html", "tools/pdf/to-text/index.html"], "OCR runtime must stay lazy to public OCR routes"); for (const required of ["engine/tesseract.min.js", "worker/worker.min.js", "lang/eng.traineddata.gz", "lang/kor.traineddata.gz"]) { assert.ok(manifest.assets[required], `missing ${required}`); } diff --git a/tests/pdf-to-text.test.mjs b/tests/pdf-to-text.test.mjs new file mode 100644 index 0000000..762b6e6 --- /dev/null +++ b/tests/pdf-to-text.test.mjs @@ -0,0 +1,98 @@ +import assert from "node:assert/strict"; +import fs from "node:fs"; +import path from "node:path"; +import { fileURLToPath } from "node:url"; +import { createPdfToTextController } from "../.ts-build/tools/pdf/to-text/controller.js"; +import { copyText, downloadPdfText, formatPdfOcrText, pdfTextFilename } from "../.ts-build/tools/pdf/to-text/output.js"; +import { canonicalPages, legacyRedirects } from "../scripts/site-routes.mjs"; +import { translations } from "../js/i18n.js"; + +const root = path.resolve(path.dirname(fileURLToPath(import.meta.url)), ".."); +const read = (file) => fs.readFileSync(path.join(root, file), "utf8"); +const html = read("tools/pdf/to-text/index.html"); + +assert.equal(canonicalPages.length, 19); +assert.equal(legacyRedirects.length, 17); +assert.ok(canonicalPages.some(({ route }) => route === "/pdf/to-text/")); +assert.ok(!legacyRedirects.some(({ from }) => from === "/tools/pdf/to-text/")); +assert.match(html, /data-page="pdfToText"/); +assert.match(html, /href="https:\/\/tools\.securetools\.app\/pdf\/to-text\/"/); +assert.match(html, /id="pages-all"[\s\S]*id="pages-selected"[\s\S]*id="page-range"/); +assert.match(html, /value="eng"[\s\S]*value="kor"[\s\S]*value="eng\+kor"/); +assert.match(html, /connect-src 'none'/); +assert.doesNotMatch(html, /https?:\/\/(?!tools\.securetools\.app|github\.com)/); +for (const language of ["en", "ko", "ja", "es", "de", "fr"]) { + assert.ok(translations[language].metadata.pdfToText.title); + assert.ok(translations[language].pdfToText.result.copyAll); + assert.ok(translations[language].categories.pdf.toText); +} + +assert.equal(formatPdfOcrText([{ pageNumber: 1, text: "One" }, { pageNumber: 3, text: "Three\n" }]), "--- Page 1 ---\nOne\n\n--- Page 3 ---\nThree\n"); +assert.equal(pdfTextFilename("private/report.pdf"), "private_report.txt"); + +let resolveFirst; +const first = new Promise((resolve) => { resolveFirst = resolve; }); +let request; +const states = []; +const service = { + async recognizeDocument(value) { request = value; await first; value.onPageResult?.({ pageNumber: 1, text: "stale" }); return { pages: [{ pageNumber: 1, text: "stale" }] }; }, + async cancel() { resolveFirst?.(); return true; }, + async dispose() {}, +}; +const controller = createPdfToTextController({ service, inspect: async () => ({ pageCount: 2 }), onChange: (state) => states.push(state.phase) }); +const file = Object.assign(new Blob(["%PDF"], { type: "application/pdf" }), { name: "sample.pdf", lastModified: 0 }); +await controller.select(file); +const running = controller.recognize("eng", { mode: "all" }); +await controller.cancel(); +await running; +assert.equal(controller.getState().phase, "cancelled"); +assert.deepEqual(controller.getState().pages, []); +assert.equal(request.selection.mode, "all"); +assert.ok(states.includes("recognizing")); + +let copied = ""; +await copyText(formatPdfOcrText([{ pageNumber: 2, text: "editable" }]), { navigatorObject: { clipboard: { writeText: async (text) => { copied = text; } } } }); +assert.equal(copied, "--- Page 2 ---\neditable\n"); +let downloaded; +const anchor = { click() { this.clicked = true; }, remove() {} }; +downloadPdfText([{ pageNumber: 2, text: "editable" }], "unsafe:name.pdf", { + documentObject: { body: { append() {} }, createElement: () => anchor }, + urlObject: { createObjectURL(blob) { downloaded = blob; return "blob:test"; }, revokeObjectURL() {} }, schedule() {}, +}); +assert.equal(anchor.download, "unsafe_name.txt"); +assert.equal(anchor.clicked, true); +assert.equal(await downloaded.text(), "--- Page 2 ---\neditable\n"); + +let attempts = 0; +const retryController = createPdfToTextController({ + inspect: async () => ({ pageCount: 1 }), + service: { + async recognizeDocument(value) { + attempts += 1; + if (attempts === 1) throw Object.assign(new Error("failed"), { code: "PDF_OCR_FAILED" }); + const page = { pageNumber: 1, text: "retry worked" }; value.onPageResult?.(page); + return { pages: [page] }; + }, + async cancel() { return false; }, async dispose() {}, + }, +}); +await retryController.select(file); +await retryController.recognize("eng", { mode: "all" }); +assert.equal(retryController.getState().phase, "error"); +await retryController.recognize("eng", { mode: "all" }); +assert.equal(retryController.getState().phase, "success"); +assert.equal(retryController.getState().pages[0].text, "retry worked"); + +let releaseOld; +const replacementController = createPdfToTextController({ + service: { async recognizeDocument() { throw new Error("unused"); }, async cancel() { return false; }, async dispose() {} }, + inspect: async (source) => source.name === "old.pdf" ? new Promise((resolve) => { releaseOld = () => resolve({ pageCount: 9 }); }) : { pageCount: 2 }, +}); +const oldFile = Object.assign(new Blob(["%PDF"], { type: "application/pdf" }), { name: "old.pdf", lastModified: 0 }); +const newFile = Object.assign(new Blob(["%PDF"], { type: "application/pdf" }), { name: "new.pdf", lastModified: 0 }); +const oldSelection = replacementController.select(oldFile); +await replacementController.select(newFile); +releaseOld(); await oldSelection; +assert.equal(replacementController.getState().source.file.name, "new.pdf"); +assert.equal(replacementController.getState().source.pageCount, 2); +console.log("PDF to Text route, localization, output, and stale-job cancellation checks passed."); diff --git a/tests/run-all.mjs b/tests/run-all.mjs index 5d4b8cb..09cf70f 100644 --- a/tests/run-all.mjs +++ b/tests/run-all.mjs @@ -32,6 +32,7 @@ for (const test of [ "tests/image-metadata.test.mjs", "tests/ocr-foundation.test.mjs", "tests/pdf-ocr-foundation.test.mjs", + "tests/pdf-to-text.test.mjs", "tests/image-to-text.test.mjs", "tests/typescript-foundation.test.mjs", "tests/category-availability.test.mjs", diff --git a/tests/seo-foundation.test.mjs b/tests/seo-foundation.test.mjs index 6501b86..beb4c61 100644 --- a/tests/seo-foundation.test.mjs +++ b/tests/seo-foundation.test.mjs @@ -15,7 +15,7 @@ const indexableRoutes = new Map(canonicalPages.map(({ source, route }) => [sourc const excludedRoutes = ["404.html", "tools/image-to-pdf/index.html"]; const allHtmlRoutes = [...indexableRoutes.keys(), ...excludedRoutes]; -assert.equal(indexableRoutes.size, 18, "all canonical pages come from the route manifest"); +assert.equal(indexableRoutes.size, 19, "all canonical pages come from the route manifest"); const shareImagePath = "assets/images/og-image.png"; const shareImageUrl = `${origin}/${shareImagePath}`; const iconLinks = new Map([ diff --git a/tests/serve-ocr-smoke.mjs b/tests/serve-ocr-smoke.mjs index b1d098a..ac86357 100644 --- a/tests/serve-ocr-smoke.mjs +++ b/tests/serve-ocr-smoke.mjs @@ -52,5 +52,6 @@ const server = http.createServer((request, response) => { server.listen(port, "127.0.0.1", () => { console.log(`OCR browser smoke: http://127.0.0.1:${port}/tests/browser/ocr-smoke.html`); console.log(`PDF OCR browser smoke: http://127.0.0.1:${port}/tests/browser/pdf-ocr-smoke.html`); + console.log(`PDF to Text browser smoke: http://127.0.0.1:${port}/tests/browser/pdf-to-text-smoke.html`); console.log(`Image to Text UI QA: http://127.0.0.1:${port}/image/to-text/`); }); diff --git a/tests/typescript-foundation.test.mjs b/tests/typescript-foundation.test.mjs index 7180ac7..92ec1ba 100644 --- a/tests/typescript-foundation.test.mjs +++ b/tests/typescript-foundation.test.mjs @@ -21,7 +21,7 @@ assert.equal(typeConfig.compilerOptions.noEmit, true); assert.equal(typeConfig.compilerOptions.sourceMap, false); assert.equal(buildConfig.compilerOptions.outDir, ".ts-build"); -assert.equal(compiledBrowserModules.length, 4, "the compiled TypeScript boundary remains intentionally bounded"); +assert.equal(compiledBrowserModules.length, 7, "the compiled TypeScript boundary remains intentionally bounded"); assert.equal(new Set(compiledBrowserModules.map((module) => module.public)).size, compiledBrowserModules.length); for (const module of compiledBrowserModules) { assert.ok(fs.existsSync(path.join(root, module.source)), `${module.source} exists`); diff --git a/tests/url-namespace.test.mjs b/tests/url-namespace.test.mjs index c2865bd..47cc2c9 100644 --- a/tests/url-namespace.test.mjs +++ b/tests/url-namespace.test.mjs @@ -8,7 +8,7 @@ import { canonicalPages, legacyRedirects, productionOrigin, redirectStatus } fro const root = path.resolve(path.dirname(fileURLToPath(import.meta.url)), ".."); const routes = new Set(canonicalPages.map(({ route }) => route)); -assert.equal(canonicalPages.length, 18); +assert.equal(canonicalPages.length, 19); assert.equal(legacyRedirects.length, 17); assert.equal(redirectStatus, 308); assert.ok(routes.has("/image/to-text/")); diff --git a/tools/pdf/index.html b/tools/pdf/index.html index 2d527ae..64bdaf7 100644 --- a/tools/pdf/index.html +++ b/tools/pdf/index.html @@ -31,5 +31,6 @@
  1. Available

    Organize PDF

    Preview, reorder, rotate, and remove pages from one PDF.

  2. Available

    PDF to Images

    Convert ordered PDF pages to PNG, JPEG, or WebP images.

  3. Available

    PDF Metadata

    Inspect common document metadata and create a verified cleaned copy.

    4. +
  4. Available

    PDF to Text OCR

    Extract editable text from PDF pages with local English and Korean OCR.

    5. Available tools process file contents locally in browser memory.

    diff --git a/tools/pdf/split/pdf.d.ts b/tools/pdf/split/pdf.d.ts new file mode 100644 index 0000000..92c98af --- /dev/null +++ b/tools/pdf/split/pdf.d.ts @@ -0,0 +1 @@ +export function parsePageSelection(value: unknown, pageCount: number): number[]; diff --git a/tools/pdf/to-text/app.ts b/tools/pdf/to-text/app.ts new file mode 100644 index 0000000..b963259 --- /dev/null +++ b/tools/pdf/to-text/app.ts @@ -0,0 +1,125 @@ +import { t } from "../../../js/i18n.js"; +import { formatBytes } from "../../shared/file.js"; +import { inspectPdf } from "../../shared/pdf.js"; +import { PDF_OCR_LARGE_DOCUMENT_THRESHOLD } from "../../shared/pdf-ocr.js"; +import { parsePageSelection } from "../split/pdf.js"; +import { createPdfToTextController } from "./controller.js"; +import type { PdfToTextState } from "./controller.js"; +import { copyText, downloadPdfText, formatPdfOcrText, pdfTextFilename } from "./output.js"; + +declare global { interface Window { PDFLib?: { PDFDocument?: unknown } } } +type ElementMap = Record; +const byId = (id: string): T => { + const value = document.getElementById(id); + if (!value) throw new Error(`Missing #${id}`); + return value as T; +}; +const elements: ElementMap & { + input: HTMLInputElement; language: HTMLSelectElement; range: HTMLInputElement; + progress: HTMLProgressElement; all: HTMLInputElement; selected: HTMLInputElement; +} = { + input: byId("file-input"), language: byId("ocr-language"), + range: byId("page-range"), progress: byId("ocr-progress"), + all: byId("pages-all"), selected: byId("pages-selected"), + drop: byId("drop-zone"), sourceEmpty: byId("source-empty"), sourceCard: byId("source-card"), + sourceName: byId("source-name"), sourceMeta: byId("source-meta"), remove: byId("remove-source"), + replace: byId("replace-source"), warning: byId("large-warning"), rangePanel: byId("range-panel"), + rangeError: byId("range-error"), recognize: byId("recognize"), cancel: byId("cancel"), + status: byId("tool-status"), result: byId("result-panel"), pages: byId("result-pages"), + copyAll: byId("copy-all"), download: byId("download-result"), +}; + +const template = (key: string, values: Record = {}) => Object.entries(values).reduce( + (text, [name, value]) => text.replaceAll(`{${name}}`, String(value)), t(key), +); +const codeOf = (error: unknown) => error instanceof Error && "code" in error ? String(error.code) : "PDF_OCR_FAILED"; +let latest: PdfToTextState; + +function statusKey(state: PdfToTextState): string { + if (state.phase === "preparing") return "pdfToText.status.preparing"; + if (state.phase === "ready") return "pdfToText.status.ready"; + if (state.phase === "cancelled") return "pdfToText.status.cancelled"; + if (state.phase === "success") return "pdfToText.status.success"; + if (state.phase === "error") return `pdfToText.errors.${codeOf(state.error)}`; + return ""; +} + +function renderPages(state: PdfToTextState): void { + elements.pages.replaceChildren(); + state.pages.forEach((page) => { + const article = document.createElement("article"); article.className = "page-result surface"; + const header = document.createElement("div"); header.className = "page-result__header"; + const heading = document.createElement("h3"); heading.textContent = template("pdfToText.result.page", { page: page.pageNumber }); + const button = document.createElement("button"); button.className = "button button--secondary"; button.type = "button"; button.textContent = t("pdfToText.result.copyPage"); + const label = document.createElement("label"); const id = `page-text-${page.pageNumber}`; label.htmlFor = id; label.className = "visually-hidden"; label.textContent = template("pdfToText.result.pageLabel", { page: page.pageNumber }); + const field = document.createElement("textarea"); field.id = id; field.rows = 9; field.spellcheck = true; field.value = page.text; + field.addEventListener("input", () => controller.updatePageText(page.pageNumber, field.value)); + button.addEventListener("click", async () => { try { await copyText(field.value); showTransient("pdfToText.status.pageCopied", { page: page.pageNumber }); } catch { showError("pdfToText.errors.copy"); } }); + header.append(heading, button); article.append(header, label, field); elements.pages.append(article); + }); +} + +function progressText(state: PdfToTextState): string { + const progress = state.progress; + if (!progress) return t("pdfToText.progress.starting"); + if (progress.phase === "loading-document") return t("pdfToText.progress.loading"); + if (progress.phase === "complete") return t("pdfToText.progress.complete"); + return template(`pdfToText.progress.${progress.phase}`, { page: progress.pageNumber || 1, total: progress.pageCount || state.source?.pageCount || 1, percent: Math.round((progress.pageProgress || 0) * 100) }); +} + +function render(state: PdfToTextState): void { + latest = state; + const hasSource = Boolean(state.source); const busy = state.phase === "preparing" || state.phase === "recognizing"; + elements.sourceEmpty.hidden = hasSource; + elements.sourceCard.hidden = !hasSource; + elements.warning.hidden = !state.source || state.source.pageCount < PDF_OCR_LARGE_DOCUMENT_THRESHOLD; + if (state.source) { elements.sourceName.textContent = state.source.file.name; elements.sourceMeta.textContent = template("pdfToText.source.meta", { pages: state.source.pageCount, size: formatBytes(state.source.file.size) }); } + elements.input.toggleAttribute("disabled", busy); elements.language.toggleAttribute("disabled", busy); + elements.all.toggleAttribute("disabled", busy); elements.selected.toggleAttribute("disabled", busy); elements.range.toggleAttribute("disabled", busy || !elements.selected.checked); + elements.remove.toggleAttribute("disabled", busy); elements.replace.toggleAttribute("disabled", busy); + elements.recognize.toggleAttribute("disabled", !hasSource || busy); elements.recognize.hidden = state.phase === "recognizing"; + elements.cancel.hidden = state.phase !== "recognizing"; + elements.progress.hidden = state.phase !== "recognizing"; + if (state.phase === "recognizing") { + const value = state.progress?.overallProgress; + if (value === null || value === undefined) elements.progress.removeAttribute("value"); else elements.progress.value = value; + elements.status.textContent = progressText(state); + } else { elements.status.textContent = statusKey(state) ? t(statusKey(state)) : ""; } + elements.status.dataset.tone = state.phase === "error" ? "error" : state.phase === "success" ? "success" : state.phase === "cancelled" ? "warning" : ""; + elements.result.hidden = state.pages.length === 0; + renderPages(state); +} + +const controller = createPdfToTextController({ + inspect: (file) => inspectPdf(file, window.PDFLib?.PDFDocument), + onChange: render, +}); + +function selectedPages() { + if (elements.all.checked) return { mode: "all" as const }; + const pageNumbers = parsePageSelection(elements.range.value, latest.source?.pageCount || 0).map((page) => page + 1); + if (new Set(pageNumbers).size !== pageNumbers.length) throw Object.assign(new Error("PDF_OCR_PAGE_DUPLICATE"), { code: "PDF_OCR_PAGE_DUPLICATE" }); + return { mode: "selected" as const, pageNumbers }; +} +function clearRangeError(): void { elements.range.removeAttribute("aria-invalid"); elements.rangeError.textContent = ""; } +function showError(key: string): void { elements.status.textContent = t(key); elements.status.dataset.tone = "error"; } +function showTransient(key: string, values: Record = {}): void { elements.status.textContent = template(key, values); elements.status.dataset.tone = "success"; } +async function choose(files: FileList | File[]): Promise { + if (files.length !== 1) { showError("pdfToText.errors.oneFile"); return; } + clearRangeError(); await controller.select(files[0]); elements.input.value = ""; +} + +elements.input.addEventListener("change", () => { if (elements.input.files) void choose(elements.input.files); }); +elements.replace.addEventListener("click", () => elements.input.click()); elements.remove.addEventListener("click", () => void controller.reset()); +elements.cancel.addEventListener("click", () => void controller.cancel()); +elements.recognize.addEventListener("click", () => { clearRangeError(); try { void controller.recognize(elements.language.value as "eng" | "kor" | "eng+kor", selectedPages()); } catch (error) { elements.range.setAttribute("aria-invalid", "true"); elements.rangeError.textContent = t(`pdfToText.errors.${codeOf(error)}`); } }); +[elements.language, elements.all, elements.selected].forEach((element) => element.addEventListener("change", () => { elements.rangePanel.hidden = !elements.selected.checked; clearRangeError(); controller.invalidateResults(); })); +elements.range.addEventListener("input", () => { clearRangeError(); controller.invalidateResults(); }); +elements.copyAll.addEventListener("click", async () => { try { await copyText(formatPdfOcrText(latest.pages)); showTransient("pdfToText.status.copied"); } catch { showError("pdfToText.errors.copy"); } }); +elements.download.addEventListener("click", () => { try { downloadPdfText(latest.pages, latest.source?.file.name); showTransient("pdfToText.status.downloaded", { name: pdfTextFilename(latest.source?.file.name) }); } catch { showError("pdfToText.errors.download"); } }); +for (const type of ["dragenter", "dragover"]) elements.drop.addEventListener(type, (event) => { event.preventDefault(); elements.drop.dataset.dragging = "true"; }); +for (const type of ["dragleave", "drop"]) elements.drop.addEventListener(type, (event) => { event.preventDefault(); delete elements.drop.dataset.dragging; }); +elements.drop.addEventListener("drop", (event) => { const data = (event as DragEvent).dataTransfer; if (data) void choose(data.files); }); +document.addEventListener("securetools:languagechange", () => render(latest)); +window.addEventListener("beforeunload", () => void controller.dispose()); +render(controller.getState()); diff --git a/tools/pdf/to-text/controller.ts b/tools/pdf/to-text/controller.ts new file mode 100644 index 0000000..b64eebf --- /dev/null +++ b/tools/pdf/to-text/controller.ts @@ -0,0 +1,89 @@ +import { createPdfOcrService, readPdfOcrSource } from "../../shared/pdf-ocr.js"; +import type { PdfOcrPageResult, PdfOcrPageSelection, PdfOcrProgress, PdfOcrService } from "../../shared/pdf-ocr.js"; + +export type PdfToTextPhase = "empty" | "preparing" | "ready" | "recognizing" | "success" | "error" | "cancelled"; + +export interface PdfToTextSource { file: File; bytes: ArrayBuffer; pageCount: number } +export interface PdfToTextState { + phase: PdfToTextPhase; + source: PdfToTextSource | null; + pages: readonly PdfOcrPageResult[]; + progress: PdfOcrProgress | null; + error: unknown; +} +export interface PdfToTextControllerOptions { + service?: PdfOcrService; + inspect(file: File): Promise<{ pageCount: number }>; + onChange?(state: PdfToTextState): void; +} + +export function createPdfToTextController(options: PdfToTextControllerOptions) { + const service = options.service || createPdfOcrService(); + let generation = 0; + let state: PdfToTextState = { phase: "empty", source: null, pages: [], progress: null, error: null }; + const publish = (patch: Partial) => { + state = Object.freeze({ ...state, ...patch }); + options.onChange?.(state); + }; + const current = (token: number) => token === generation; + + async function select(file: File): Promise { + const token = ++generation; + await service.cancel(); + publish({ phase: "preparing", source: null, pages: [], progress: null, error: null }); + try { + const [bytes, details] = await Promise.all([readPdfOcrSource(file), options.inspect(file)]); + if (!current(token)) return; + publish({ phase: "ready", source: { file, bytes, pageCount: details.pageCount }, error: null }); + } catch (error) { + if (!current(token)) return; + publish({ phase: "error", source: null, error }); + } + } + + async function recognize(language: "eng" | "kor" | "eng+kor", selection: PdfOcrPageSelection): Promise { + if (!state.source || state.phase === "recognizing") return; + const token = ++generation; + const source = state.source; + publish({ phase: "recognizing", pages: [], progress: null, error: null }); + try { + const result = await service.recognizeDocument({ + sourceBytes: source.bytes, + language, + selection, + onProgress(progress) { if (current(token)) publish({ progress }); }, + onPageResult(page) { if (current(token)) publish({ pages: [...state.pages, page] }); }, + }); + if (!current(token)) return; + publish({ phase: "success", pages: result.pages, error: null }); + } catch (error) { + if (!current(token)) return; + const code = error instanceof Error && "code" in error ? String(error.code) : ""; + publish({ phase: code === "PDF_OCR_CANCELLED" ? "cancelled" : "error", progress: null, error }); + } + } + + async function cancel(): Promise { + if (state.phase !== "recognizing") return; + generation += 1; + await service.cancel(); + publish({ phase: "cancelled", progress: null, error: null }); + } + + async function reset(): Promise { + generation += 1; + await service.cancel(); + publish({ phase: "empty", source: null, pages: [], progress: null, error: null }); + } + + function invalidateResults(): void { + if (!state.source || state.phase === "recognizing") return; + publish({ phase: "ready", pages: [], progress: null, error: null }); + } + + function updatePageText(pageNumber: number, text: string): void { + publish({ pages: state.pages.map((page) => page.pageNumber === pageNumber ? { ...page, text } : page) }); + } + + return Object.freeze({ select, recognize, cancel, reset, invalidateResults, updatePageText, getState: () => state, dispose: () => service.dispose() }); +} diff --git a/tools/pdf/to-text/index.html b/tools/pdf/to-text/index.html new file mode 100644 index 0000000..2e4c1ce --- /dev/null +++ b/tools/pdf/to-text/index.html @@ -0,0 +1,16 @@ + + + + +PDF to Text OCR — Secure Tools + + +
    Secure Tools
    + +

    Local PDF OCR

    PDF to Text

    Recognize editable text from every page or a selected page range without uploading your PDF.

    +

    Add one PDF

    Drop a PDF here or use the picker.

    Processed locally. The PDF, rendered pages, and recognized text stay on this device. How privacy works

    +

    Source PDF

    No PDF selected yet.

    +
    +
    + + diff --git a/tools/pdf/to-text/output.ts b/tools/pdf/to-text/output.ts new file mode 100644 index 0000000..9244a74 --- /dev/null +++ b/tools/pdf/to-text/output.ts @@ -0,0 +1,23 @@ +import { downloadBlob } from "../../shared/save.js"; +import type { DownloadEnvironment } from "../../shared/save.js"; +import type { CopyEnvironment } from "../../image/to-text/output.js"; +import { copyText } from "../../image/to-text/output.js"; +import type { PdfOcrPageResult } from "../../shared/pdf-ocr.js"; + +const ILLEGAL_FILENAME_CHARACTERS = /[\\/:*?"<>|\u0000-\u001f]+/g; + +export function formatPdfOcrText(pages: readonly PdfOcrPageResult[]): string { + return pages.map(({ pageNumber, text }) => `--- Page ${pageNumber} ---\n${String(text).trimEnd()}`).join("\n\n") + (pages.length ? "\n" : ""); +} + +export function pdfTextFilename(sourceName: unknown): string { + const base = String(sourceName || "").replace(/\.pdf$/i, "").trim().replace(ILLEGAL_FILENAME_CHARACTERS, "_").replace(/[. ]+$/g, ""); + return `${base || "recognized-document"}.txt`; +} + +export { copyText }; +export function downloadPdfText(pages: readonly PdfOcrPageResult[], sourceName: unknown, environment?: DownloadEnvironment): void { + const blob = new Blob([formatPdfOcrText(pages)], { type: "text/plain;charset=utf-8" }); + downloadBlob(blob, pdfTextFilename(sourceName), environment); +} +export type { CopyEnvironment }; diff --git a/tools/pdf/to-text/tool.css b/tools/pdf/to-text/tool.css new file mode 100644 index 0000000..886495e --- /dev/null +++ b/tools/pdf/to-text/tool.css @@ -0,0 +1,15 @@ +.mode-list { grid-template-columns: 1fr; } +[hidden] { display: none !important; } +.mode-list legend { margin-bottom: var(--space-2); color: var(--text-secondary); font-size: .78rem; font-weight: var(--font-weight-bold); } +.source-actions, .result-actions { display: flex; flex-wrap: wrap; gap: var(--space-2); } +.tool-warning { margin: var(--space-4) 0 0; padding: var(--space-3) var(--space-4); border: 1px solid var(--tool-warning); border-radius: var(--radius-sm); background: var(--tool-warning-soft); color: var(--tool-warning); } +.result-summary { display: flex; align-items: center; justify-content: space-between; gap: var(--space-4); padding: var(--space-5); } +.result-summary h2 { margin-bottom: var(--space-2); } +.result-summary p { margin: 0; color: var(--text-secondary); } +.result-pages { display: grid; gap: var(--space-4); margin-top: var(--space-4); } +.page-result { padding: var(--space-5); } +.page-result__header { display: flex; align-items: center; justify-content: space-between; gap: var(--space-3); margin-bottom: var(--space-3); } +.page-result h3 { margin: 0; font-size: 1.1rem; } +.page-result textarea { width: 100%; min-height: 12rem; resize: vertical; padding: var(--space-3); border: 1px solid var(--border); border-radius: var(--radius-sm); background: var(--bg-primary); color: var(--text-primary); font: inherit; line-height: 1.55; } +.clipboard-fallback { position: fixed; left: -9999px; } +@media (max-width: 42rem) { .result-summary, .page-result__header { align-items: stretch; flex-direction: column; } .result-actions .button, .page-result__header .button { width: 100%; } } diff --git a/tools/scan/index.html b/tools/scan/index.html index 2830f61..eefaf9c 100644 --- a/tools/scan/index.html +++ b/tools/scan/index.html @@ -21,4 +21,4 @@ Scan and OCR tools — Secure Tools
    Secure Tools
    -

    Tool category

    Scan & OCR tools

    Turn scans into useful documents while keeping source material on your device.

    Available tools process file contents locally in browser memory.

    +

    Tool category

    Scan & OCR tools

    Turn scans into useful documents while keeping source material on your device.

    Available tools process file contents locally in browser memory.

    diff --git a/tools/shared/file.d.ts b/tools/shared/file.d.ts new file mode 100644 index 0000000..9a1807f --- /dev/null +++ b/tools/shared/file.d.ts @@ -0,0 +1,2 @@ +export function formatBytes(bytes: unknown): string; +export function sanitizePdfFilename(value: unknown, fallback?: string): string; diff --git a/tools/shared/pdf.d.ts b/tools/shared/pdf.d.ts index 18e3307..4ca016b 100644 --- a/tools/shared/pdf.d.ts +++ b/tools/shared/pdf.d.ts @@ -3,3 +3,4 @@ export interface PdfFileLike extends Blob { } export function isSupportedPdf(file: unknown): boolean; +export function inspectPdf(file: PdfFileLike, PDFDocument: unknown): Promise<{ pageCount: number }>; diff --git a/tsconfig.build.json b/tsconfig.build.json index 072290c..4521e33 100644 --- a/tsconfig.build.json +++ b/tsconfig.build.json @@ -9,6 +9,9 @@ "tools/image/to-text/controller.ts", "tools/image/to-text/output.ts", "tools/image/to-text/preview.ts", + "tools/pdf/to-text/app.ts", + "tools/pdf/to-text/controller.ts", + "tools/pdf/to-text/output.ts", "tools/shared/pdf-ocr.ts", "tools/shared/*.d.ts" ] diff --git a/tsconfig.json b/tsconfig.json index 1f8c68d..77c1eff 100644 --- a/tsconfig.json +++ b/tsconfig.json @@ -16,7 +16,10 @@ }, "include": [ "tools/image/to-text/*.ts", + "tools/pdf/to-text/*.ts", "tools/shared/pdf-ocr.ts", - "tools/shared/*.d.ts" + "tools/shared/*.d.ts", + "tools/pdf/split/*.d.ts", + "js/*.d.ts" ] }