Word & PowerPoint File Repair (DOCX, PPTX, DOC, PPT)

About the Word & PowerPoint File Repair (DOCX, PPTX, DOC, PPT)

A DOCX or PPTX file is a ZIP archive of XML parts — the text, the styles, the pictures, each slide — tied together by links (ECMA-376 Office Open XML). When one part is cut off by a broken download, holds a character XML does not allow, or points at a picture that is gone, Word says it found unreadable content or that the file cannot be opened because there are problems with the contents, and PowerPoint that it found a problem with content.

This page rebuilds the file around what survived: it finds every part even when the end of the archive is missing, repairs each XML part, keeps every complete paragraph, table and slide, takes out the references to what is gone, and writes a file that opens. When the structure is beyond saving — or the file is an old DOC or PPT — its text rescue keeps every readable paragraph in order, with its headings, tables and pictures, as a new Word document or plain text. You see the report and the start of the text before anything is saved, and nothing is uploaded.

Common uses

• Documents and presentations that will not open after a broken download, an interrupted save or a sync error.
• Files Word or PowerPoint only open after their own repair, which drops too much.
• Documents recovered by a deep scan of a disk image and marked partly damaged.
• Old DOC and PPT files that newer programs call corrupt.
• Getting the pictures out of a document or presentation that will not open.

What the repair does to a DOCX or PPTX

• Finds the parts. The archive is read from the front, so its parts are found even when its end is missing (see ZIP Archive Repair for how).
• Repairs the XML. Characters XML forbids are removed, elements left open closed. A part cut short keeps every complete paragraph or table — directly in the body, a table cell, a header, a footer or a note; never one half-written — and a slide every complete shape.
• Takes out references to what is gone. A picture whose image did not survive, a header whose part is gone, footnote and comment marks without their part, list numbering without its definitions, a slide without its layout. Each is listed in the report.
• Keeps a presentation’s chain whole. A slide needs its layout, a layout its master, a master a theme; a master that lost its theme gets a plain one.
• Ties it back together. Links to missing parts are taken out, and the list of content types is written again.

The text rescue

The text rescue reads the document as text, in order: every paragraph that could be read, with its heading level where the styles survived (or where the style’s name says so), list items, tables and pictures; text boxes are read once, even though Word stores them twice. For a presentation it reads each slide in the order of the deck — its title, its text and tables, its pictures — and the speaker notes.

It is saved as a new Word document with Word’s own heading styles, so the navigation pane works, or as plain text. It has none of the damaged file’s structure, so it always opens; it does not keep fonts, colours, page layout or tracked changes.

Old DOC and PPT files

Word 97–2003 and PowerPoint 97–2003 files are compound files, a small file system inside one file (MS-CFB). In a .doc, the header at the start of the text stream points to a piece table that says where each piece of the text is stored and how (MS-DOC); in a .ppt, the slides’ text is in text records (MS-PPT). The page follows both; field codes are left out and their results kept. When the structure is too damaged, it scans the file for readable text instead, and says so. Complete pictures stored in the file are found and kept too.

What the repair is measured against

The automated tests use documents written by Microsoft Word and by other programs, and break them: cut short in the middle of the main text, with a damaged picture, without the footnotes part, with a text box Word stores twice; presentations without a slide layout or without their theme; old .doc files with their header or their text index overwritten. Every repaired file must pass an independent XML check and read back in another program’s reader, and every text rescue must read back as a Word document with its headings, lists, tables and pictures.

No success rate is published for other people’s files: what comes back depends on how much of the file reached the disk.

Limitations

• Missing data stays missing: text after the point where a part was cut off, and pictures whose files are damaged, cannot be recreated.
• Formatting, tracked changes, charts, SmartArt, embedded objects and macros come back only when their own parts survived whole; the text rescue keeps text, headings, lists, tables and pictures, not formatting or page layout.
• Old DOC and PPT files are rescued as text and pictures, not rebuilt in their own format.
• Password-protected documents are encrypted: they cannot be repaired without their password, and this page never tries to guess one.
• Damaged PDFs go to PDF Repair, workbooks to Excel File Repair, archives to ZIP Archive Repair.
• Files of up to 400 MB: the repair works on the whole file in memory.
• Word, PowerPoint and Microsoft are trademarks of Microsoft; this page is not made by or affiliated with Microsoft.

Privacy

The document is read by your browser on your device; nothing from it is sent anywhere, and the page works offline once it has loaded.