Recover a damaged file
Drop a file that will not open. webdave saves what is left of it and finds the files hidden inside it. Your browser opens it on your device: nothing is uploaded.
What it can save
| ZIP, DOCX, XLSX, PPTX | Rebuilt from their parts when the table of contents is missing or cut |
| Word | The text of a cut document, read on its own |
| Repaired by qpdf, or its readable pages copied out | |
| JPEG, PNG | Cut pictures closed; a PNG’s whole rows kept |
| Text | UTF-8 read as Windows-1252 put right, NULs taken out |
| Inside any file | JPEG, PNG, GIF, PDF, ZIP and Office files, MP4, MOV, WAV, WebP |
| Upload | None: recovered on your device |
How to recover a damaged file
Drop the file in the box above. A ZIP, Word, Excel or PowerPoint file is rebuilt from its parts, a PDF is repaired with qpdf, a cut JPEG or PNG keeps the part that arrived, and garbled text is put right. Then every file inside it is listed, to open or save. Nothing is uploaded.
- Drop the file. A file that will not open, stops short, or came from a memory card or a backup as one big file.
- Let it look. Recover tries what fits the file’s type, then reads every byte for pictures, PDFs, Office files, ZIPs, video and sound inside it.
- Save what it found. Each result says what was recovered. Open it here to check it, then Save it.
Questions
Drop it here. A DOCX or XLSX is a ZIP of parts; when the list of parts at its end is missing or cut, webdave reads each part from its own header, closes a part that stops short, and writes the file again. For Word, the text it holds is also given on its own.
Often. qpdf, a PDF library, rebuilds the cross-reference table that tells readers where each object is, the part most often broken. When that fails, the pages pdf-lib can still read are copied into a new PDF.
The grey part was never in the file: the download or copy stopped there. webdave keeps what is there and closes the file so every program shows it. A cut PNG becomes the rows that arrived.
Files can sit inside other files: pictures in a memory card dump, a PDF glued to the end of another file, a ZIP in a .dat. webdave reads every byte for the start of a known format and follows it to its end.
The text was UTF-8 but was read as Windows-1252 and saved again. Recover turns those pairs back into the letters they were.
No. Everything runs in your browser, on your device. Nothing is sent anywhere.