I can't repair corrupt .Docx files, are they recoverable at all?

[saarinen]

Hi, I've been trying to recover these files for days and I have gotten absolutely nowhere.
First of all, these are files from late last year and earlier this year my Macbook Pro crashed. I was able to recover most of the documents with Stellar Phoenix however some of them seem too corrupt to repair.
The main problem: When I try to open these .DOCX files on Win 7 in word, text edit or open office all I get is a bunch of symbols eg. "]≥D @]≥< @]≥4 @]≥, @]≥$ @]≥ @]≥ @]≥"
Before it opens in Microsoft Word (2011) the 'Convert File' window pops up and I proceed to the option 'recover text from any file'.
The document then opens in compatibility mode and with a compatibility check it displays the following errors:
"1. Word 6.0/95 compatibility options are set
2. This document displays unicode characters that might not display correctly in earlier versions of office."
When I change the compatibility preferences there is no difference to the text and the second, unicode error persists.
From here when I try to 'Save as' it automatically tries to save it as a '.Doc' file, not '.Docx'.
I have tried a few different things such as putting the files on to a Windows computer, zipping and unzipping to view the . XML content which didn't work but I am an amateur Windows user so I could have gone wrong. I have tried converting it to an RTF, no luck. And I have tried downloading an XML editor but it is beyond me.
Any suggestions? How do you know if the file is too corrupt?
Any help would be really appreciated!

I don't know hot to attach file in this post.(

[peterb]

Doc files are quite complex, and have an internal structure similar to a directory, with sub directories containing the data and metadata.

If you examine the file with a hex editor, you might be able to recover the text portion of the file. If you have access to a Linux machine, BLESS is a suitable Hex editor. For Windows machines, there is WinHex (there is a free evaluation version IIRC) so that should enable you to see if there is anything recoverable.

[smargh]

They might be Cryptolocker'd ?

docx/xlsx etc files are compressed XML. If you open the file in a hex editor, the first letters should be "PK"

If the file is filled with what seem to be random characters, then it might well be encrypted with something like Cryptolocker/Cryptowall. If there seems to be some non-random parts the file, which would point to some kind of structure, then it's more likely a corrupt file.

The least-worst scenario could potentially be simply recovering the plain text portions from the XML.

[BobF64]

docx/xlsx etc files are compressed XML. If you open the file in a hex editor, the first letters should be "PK"

Because its more than just "compressed XML" its a Zip file.

Try changing the extension from .docx to .zip and seeing if 7-Zip or something can open or extract anything from it.

Of course, if as suggested, its been encrypted some how, or really is corrupt, that probably wont do much.