Fix a corrupted or damaged PDF that won't open. The tool analyses the file, rebuilds the cross-reference table and object structure, recovers pages one by one, and can rasterize as a last resort — with a full recovery report and before/after preview. Everything runs in your browser; your file is never uploaded.
📤 Drag PDF file(s) here or click to choose
Scans the raw bytes and reports what looks broken — the header, cross-reference table, end marker, encryption, linearization and truncation — with a severity rating.
Reloads the document leniently and rebuilds the cross-reference table and object structure from scratch, fixing the most common corruption.
Deep mode copies each page into a fresh document one at a time, recovering the good pages and skipping the ones that are broken.
Detects and strips junk bytes before the %PDF header and other stray data that stop a file from opening.
If the structure is unparseable, it renders the pages with a more tolerant engine and rebuilds a new PDF from the images.
Recovers the title, author, subject, keywords and creation dates and carries them into the repaired file.
Shows a recovery percentage, a step-by-step repair log of exactly what was fixed, and a before/after page preview.
Repair several PDFs at once and download them individually or bundled into a single ZIP archive.
Analysis and repair run locally with pdf-lib, pdf.js and JSZip. Your file is never uploaded — no server, no account.
The ApneSoftware PDF Repair tool analyses and repairs corrupted, damaged or unopenable PDF files entirely inside your browser, without uploading the file anywhere. A PDF is a surprisingly intricate container: it holds a header that declares the format version, a body of numbered objects (pages, fonts, images, streams), a cross-reference table that records the exact byte position of every object, and a trailer that points to where reading should begin. When any of these pieces is damaged — because a download was interrupted, a disk had an error, an email attachment was mangled, a sync went wrong, or a program wrote the file incorrectly — the whole document can refuse to open, showing errors like “file is damaged and could not be repaired”, “cannot open”, or a blank grey screen. This tool exists to get those files open again by rebuilding the parts that a PDF reader relies on, using well-established, deterministic techniques rather than guesswork or AI.
The first thing the tool does is analyse the file. It scans the raw bytes and checks the things that most often go wrong: whether the %PDF- header is present and at the very start of the file or buried behind junk; whether the closing %%EOF marker exists; whether the startxref pointer and the cross-reference table it points to are valid; whether the file is encrypted or linearized (“fast web view”); and whether the object and stream markers are balanced or the file looks truncated. From this it produces a clear damage report with a severity rating, so before you repair anything you can see what is actually wrong. This diagnostic step, familiar from professional tools like Adobe Acrobat's Preflight, Foxit and Sejda, turns a frightening “it won't open” into a specific, understandable problem.
Repair itself works in layers, from gentlest to most aggressive, so the tool always keeps as much of your original document as possible. The first layer is structural rebuild: the file is loaded with a deliberately lenient parser that tolerates invalid objects and ignores encryption, and is then written out again from scratch. This single step rebuilds the cross-reference table, renumbers and repacks the objects, and drops any broken incremental-update layers — which alone fixes a large share of real-world corruption, because a bad or out-of-date cross-reference table is the single most common reason a PDF won't open. If junk bytes are found before the header, they are stripped first so the parser sees a clean file. Throughout, your metadata — title, author, subject, keywords and creation dates — is preserved so the repaired file is not stripped of its identity.
When a file is more badly damaged, the tool switches to deep, page-by-page recovery. Instead of trusting the document as a whole, it creates a brand-new, clean PDF and copies your pages into it one at a time. If a particular page is broken beyond recovery, that page is skipped and noted in the log, while every page that can be salvaged is kept. This is how you recover a fifty-page report where only one page was damaged: you get back forty-nine perfect pages instead of nothing at all. The tool reports exactly how many pages were recovered out of the total, expressed as a recovery percentage, so you know precisely how much of the document survived.
For the hardest cases — files whose internal structure is so broken that no structural parser can read them — there is a rasterize last resort. Here the tool uses a second, more tolerant rendering engine (the same technology that displays PDFs inside Firefox) to draw each page as an image, and then rebuilds a fresh PDF from those images. The result is not perfect: the text is no longer selectable because each page becomes a picture. But when the alternative is a file that will not open at all, recovering a faithful visual copy of every page is often exactly what people need — a recoverable invoice, contract, certificate or scan they can read, print and archive. You can turn this last resort off if you only want structural repair, and choose the image quality to balance clarity against file size.
Around this repair engine sits a workflow designed to feel like the professional desktop tools people trust, while remaining free and completely private. You can add several damaged PDFs and repair them as a batch, downloading them individually or as a single ZIP. A progress bar keeps you informed, a cancel button stops a long job, and after every repair a detailed recovery report shows a colour-coded log of each step that was taken — junk removed, structure rebuilt, pages recovered, metadata restored — alongside a before/after preview that renders the first page of the original next to the first page of the repaired file so you can confirm the result with your own eyes. Nothing about the process is a black box: you can see what was wrong, what was done, and what came out.
It is important to be honest about what any PDF repair tool can and cannot do, because false hope wastes time. Repair works by reconstructing structure from the data that is still present in the file. If the underlying bytes are intact but their organisation is broken — a corrupt cross-reference table, a bad trailer, junk around the header, a damaged object stream, stale incremental updates — the odds of a full recovery are excellent. If, however, the actual content is physically gone — a file that stopped downloading at 60% and is simply missing its second half, storage that overwrote part of the file, or a document that was encrypted with a key you do not have — then no tool, however sophisticated, can invent the missing information. In those cases this tool will still recover whatever remains readable (often a run of early pages) and tell you clearly what it could and could not do, which is far more useful than a silent failure. Setting expectations honestly is itself a feature.
Above all, this tool is private by design. Analysis, structural rebuild, page recovery and rasterizing all happen entirely on your own device using pdf-lib, pdf.js and JSZip; after the libraries have loaded you can even work offline. Nothing is uploaded to a server, nothing is stored, and no account or email is required. This matters especially for a repair tool, because the documents people most need to rescue are frequently the most sensitive and irreplaceable — contracts, financial statements, legal filings, medical records, certificates, dissertations and official paperwork — and uploading a damaged copy of exactly that material to an unknown online service, as many free repair sites quietly do, is a privacy risk on top of an already stressful situation. Here, that never happens: your file is opened only in your browser's memory, repaired locally, and discarded the moment you close the tab. Whether you are an office worker facing a corrupt attachment, a student whose thesis won't open the night before a deadline, or anyone staring at a “file is damaged” message, this tool gives you a fast, transparent, private, best-effort recovery with nothing to install.
ApneSoftware PDF Repair टूल खराब, क्षतिग्रस्त या न खुलने वाली PDF फ़ाइलों का विश्लेषण और मरम्मत पूरी तरह आपके ब्राउज़र के अंदर करता है, फ़ाइल को कहीं अपलोड किए बिना। PDF एक जटिल container है: इसमें एक header होता है जो प्रारूप संस्करण बताता है, क्रमांकित objects का body (पेज, फ़ॉन्ट, इमेज, streams), एक cross-reference table जो हर object की सटीक byte स्थिति दर्ज करती है, और एक trailer जो बताता है कि पढ़ना कहाँ से शुरू हो। जब इनमें से कोई हिस्सा खराब हो जाता है — download बीच में रुकने, disk त्रुटि, ईमेल attachment खराब होने, गलत sync, या किसी प्रोग्राम द्वारा गलत तरीके से फ़ाइल लिखने से — तो पूरा दस्तावेज़ खुलने से मना कर देता है और “file is damaged” जैसी त्रुटियाँ दिखाता है। यह टूल उन फ़ाइलों को फिर से खोलने के लिए है, स्थापित, निश्चयात्मक तकनीकों से — अनुमान या AI से नहीं।
टूल सबसे पहले फ़ाइल का विश्लेषण करता है। यह raw bytes स्कैन करके उन चीज़ों की जाँच करता है जो अक्सर खराब होती हैं: क्या %PDF- header मौजूद है और बिल्कुल शुरुआत में है या junk के पीछे छिपा है; क्या समापन %%EOF marker है; क्या startxref pointer और जिस cross-reference table की ओर वह इशारा करता है वह मान्य है; क्या फ़ाइल encrypted या linearized है; और क्या object व stream markers संतुलित हैं या फ़ाइल truncated लगती है। इससे यह एक स्पष्ट damage report बनाता है जिसमें severity rating होती है, ताकि मरम्मत से पहले आप देख सकें कि असल में क्या गलत है।
मरम्मत परतों में काम करती है, सबसे हल्की से सबसे सख्त तक, ताकि टूल आपके मूल दस्तावेज़ का जितना संभव हो उतना बचाए रखे। पहली परत है structural rebuild: फ़ाइल को एक जानबूझकर सहनशील parser से लोड किया जाता है जो अमान्य objects को सहन करता है और encryption को अनदेखा करता है, फिर उसे नए सिरे से दोबारा लिखा जाता है। यह एक चरण cross-reference table दोबारा बनाता है, objects को फिर से क्रमांकित व पैक करता है, और किसी भी टूटे incremental-update layer को हटा देता है — जो अकेले ही वास्तविक जीवन की बहुत सी corruption ठीक कर देता है। यदि header से पहले junk bytes मिलते हैं तो उन्हें पहले हटा दिया जाता है। पूरे समय आपका metadata — title, author, subject, keywords और dates — संरक्षित रहता है।
जब फ़ाइल ज़्यादा खराब हो, तो टूल deep, page-by-page recovery पर चला जाता है। पूरे दस्तावेज़ पर भरोसा करने के बजाय, यह एक बिल्कुल नई, साफ़ PDF बनाता है और आपके पेजों को एक-एक करके उसमें कॉपी करता है। यदि कोई पेज पूरी तरह खराब है, तो उसे छोड़कर log में दर्ज कर दिया जाता है, जबकि हर बचाया जा सकने वाला पेज रखा जाता है। इसी तरह आप पचास-पेज की रिपोर्ट में से, जहाँ सिर्फ़ एक पेज खराब था, उनतालीस पेज वापस पाते हैं — कुछ भी न पाने के बजाय। टूल बताता है कि कुल में से कितने पेज बचाए गए, recovery percentage के रूप में।
सबसे कठिन मामलों के लिए — जिनकी internal structure इतनी टूटी है कि कोई structural parser पढ़ ही नहीं सकता — एक rasterize अंतिम उपाय है। यहाँ टूल एक दूसरा, अधिक सहनशील rendering engine (वही तकनीक जो Firefox में PDF दिखाती है) उपयोग करके हर पेज को एक छवि के रूप में बनाता है, और फिर उन छवियों से एक नई PDF बनाता है। नतीजा परफ़ेक्ट नहीं है: text अब चयन‑योग्य नहीं रहता क्योंकि हर पेज एक तस्वीर बन जाता है। पर जब विकल्प एक बिल्कुल न खुलने वाली फ़ाइल हो, तो हर पेज की विश्वसनीय दृश्य प्रति पाना अक्सर वही होता है जो लोगों को चाहिए। आप इस उपाय को बंद कर सकते हैं और image quality चुन सकते हैं।
इस मरम्मत इंजन के चारों ओर एक ऐसा workflow है जो पेशेवर desktop टूल जैसा महसूस कराता है, फिर भी मुफ़्त और पूरी तरह निजी है। आप कई खराब PDF जोड़कर batch में मरम्मत कर सकते हैं और अलग-अलग या एक ZIP के रूप में डाउनलोड कर सकते हैं। एक प्रगति पट्टी सूचित रखती है, एक कैंसल बटन लंबे काम को रोकता है, और हर मरम्मत के बाद एक विस्तृत recovery report हर चरण का रंग-कोडित log दिखाती है — junk हटाया, structure दोबारा बना, पेज बचाए, metadata बहाल — साथ में एक before/after preview। प्रक्रिया black box नहीं है: आप देख सकते हैं कि क्या गलत था, क्या किया गया, और क्या निकला।
यह ईमानदार होना ज़रूरी है कि कोई भी PDF repair टूल क्या कर सकता है और क्या नहीं। मरम्मत फ़ाइल में अभी भी मौजूद data से structure को फिर से बनाकर काम करती है। यदि अंतर्निहित bytes सही हैं पर उनका संगठन टूटा है — खराब cross-reference table, बुरा trailer, header के आसपास junk, क्षतिग्रस्त object stream — तो पूर्ण recovery की संभावना बहुत अच्छी है। पर यदि असल content भौतिक रूप से चला गया है — 60% पर रुकी download, अधिलेखित storage, या ऐसी key से encrypted दस्तावेज़ जो आपके पास नहीं — तो कोई भी टूल गायब जानकारी नहीं बना सकता। ऐसे मामलों में यह टूल जो भी पढ़ने योग्य बचा है वह बचाएगा और स्पष्ट रूप से बताएगा कि क्या हो सका और क्या नहीं।
सबसे बढ़कर, यह टूल डिज़ाइन से ही निजी है। विश्लेषण, structural rebuild, page recovery और rasterizing सब पूरी तरह आपके अपने डिवाइस पर pdf-lib, pdf.js और JSZip से होते हैं; लाइब्रेरी लोड होने के बाद आप ऑफ़लाइन भी काम कर सकते हैं। कुछ भी सर्वर पर अपलोड नहीं होता, कुछ भी सेव नहीं होता, और किसी अकाउंट की ज़रूरत नहीं। यह किसी repair टूल के लिए विशेष रूप से महत्वपूर्ण है, क्योंकि जिन दस्तावेज़ों को लोग सबसे ज़्यादा बचाना चाहते हैं वे अक्सर सबसे संवेदनशील और अपूरणीय होते हैं — अनुबंध, वित्तीय विवरण, कानूनी दस्तावेज़, मेडिकल रिकॉर्ड, प्रमाणपत्र और शोध-प्रबंध। यहाँ वह कभी नहीं होता: आपकी फ़ाइल केवल आपके ब्राउज़र की memory में खुलती है, स्थानीय रूप से मरम्मत होती है, और tab बंद करते ही हट जाती है।
This tool analyses and repairs your PDF entirely inside your browser using pdf-lib, pdf.js and JSZip. Your file is never uploaded to any server, never stored, and never transmitted anywhere. No account, email or signup is required, and once the libraries have loaded you can work offline. Because the PDFs people most need to repair — contracts, statements, legal filings, medical records, certificates and dissertations — are often sensitive and irreplaceable, this local-only design keeps a damaged copy of exactly that material under your sole control from upload to download.
Works in all modern browsers — Google Chrome, Mozilla Firefox, Microsoft Edge, Safari, Brave and Opera — on Windows, macOS, Linux, Android and iOS. It relies only on JavaScript and the standard File, Blob and Canvas APIs. On phones and tablets the layout stacks into a single column with touch-friendly controls and the before/after previews stack vertically. Repairing very large PDFs, and rasterizing, use more memory and time on a phone than on a desktop.
Structurally corrupt PDFs — broken cross-reference tables, bad trailers, junk before the header, damaged object streams and stale incremental updates — are the ideal case and usually repair fully. Partly damaged PDFs recover page by page in Deep mode. Unparseable PDFs can often be recovered visually by rasterizing. Encrypted PDFs are loaded with encryption ignored where possible; if a file needs a password, remove it first with our Unlock PDF tool. Truncated files (missing their end) recover whatever pages remain readable.
Rebuilds the cross-reference (xref) table and trailer; renumbers and repacks objects and object streams; strips junk bytes before the %PDF header; drops broken incremental-update layers; recovers pages one by one, skipping unrecoverable ones; recovers metadata (title, author, subject, keywords, dates); and, as a last resort, rasterizes pages with a tolerant renderer to rebuild a readable file. It reports a recovery percentage and a full log of every action taken.
Repair reconstructs structure from the data still in the file; it cannot restore content that is physically missing. A file truncated mid-download, storage that overwrote part of the file, or a document encrypted with a key you do not have cannot be fully recovered — though the tool will still salvage whatever remains readable. Rasterizing recovers pages as images, so their text is no longer selectable or searchable. Interactive elements such as form fields, some annotations, bookmarks and JavaScript may be simplified or dropped during a deep rebuild. Very large files use more memory, especially when rasterizing on a phone. This is a best-effort tool, not a guarantee.
“Could not parse the PDF structure.” Turn on Rasterize as last resort and try Deep mode — this recovers pages visually even when the structure is unreadable. Repaired file still won't open: the file may be truncated or overwritten; check the recovery percentage and log to see how much survived. Some pages are missing: those pages were unrecoverable and were skipped — the log lists them. Text is not selectable after repair: the file was rasterized as a last resort, turning pages into images; try again with rasterizing off if you need the text. “This PDF is encrypted.” Remove the password first with our Unlock PDF tool, then repair. Repair is slow: large files and rasterizing take longer and use more memory — use Cancel to stop.