Format error in sub-document context.xml

[KR88]

Hello

I've tried to help a friend who has a ODS.document and gets the line :
Read Error. Format error discovered in the file in sub-document content.xml at 2,95735(row,col) when he opens it.

He has got Libreoffice but when I open it with OpenOffice I get a text that the file is damaged. If I repare it with Openoffice I get an repared empty file. Is the file completely lost or can it be fixed ?

Thanks

[FJCC]

If you post the document here, someone can take a look at the internal structure and possibly fix it. To post a document, click Post Reply and look for the Attachments tab just below the box where you type a response.

[RoryOF]

You should post the original damaged document, not a version after any attempt to repair it.

[KR88]

According to my friend its some private personal stuff that is in the file (I dont have a clue whats in there) and I dont think he would want it to be all over the internet. Is there a way to send it private to someone ?

Does that happen a lot that you have this message when you open an ods.file ? I have never had it personally but I have around 75 different files on a usb and I don't want to lose 10 years of work

[FJCC]

I will send you a private message with my email address.
It is not frequent that documents get corrupted, but it can happen. Always back up documents. And a usb is particularly likely to get lost or damaged.

[FJCC]

I got the document but I can't extract the content.xml file. The ods file does open as a zip archive but the tools I have on WIndows and Linux Mint say "An error occurred while extracting the file" when I try to extract content.xml. I can successfully extract other files from the archive, e.g. meta.xml.
If someone else would like to try, please send a PM to KR88.
If I try to open the ods file in LibreOffice (the document started in LibreOffice), I do get a message about an error at 2,98000 (approx), so there is something in the content.xml.

[lader]

I think that a RAR or ZIP program can repair a damaged zip archive with a certain parameter turned on (it is indicated in the help of the given program).

[FJCC]

I managed to open content.xml and it's a mess. It seems to be only style definitions. I couldn't find any user data, though I was simply scrolling through, so I easily could have missed something. Near the end of the file, it is badly damaged. For example

Code: Select all

<style:style style:namepable-cell-properties style:text-align-source="value-type" style:repeat-content="false" style:rotation-align="none"/>
<style:paragraph-properties fo:margin-left="0cm"/>
<style:text-properties style:font-name="Calib8r1w -properties style:font-name=" Calibri" fo:font-size="8pt" style:font-size-asian="8pt" stylet-snt-style-name="Default" styiyle-name="Defaultyle:paragraph-properties fo:text-e=" Defa-style-name="N119">
	<style:table-cell-properties se-complex="8pt"/>
</style:style>
<style:styl-name="Deyle:foU"2ahaph-properties fo:text-align="center" forties l-thaph-properties fo:text-align="center" forties l-thaph-properties fo:text-align="center" forties l-thaph-properties fo:text-align="center" forties l-thaph-properties fo:text-align="center" forties l-thaph-pta-styare-21-/>
</style:style>

[RoryOF]

Once or twice on Forum we have come across content.xml being merely style definitions. I think it is probably caused by over-hasty close down of the computer, before the file has been fully generated and written to disk.

An outside chance: there might be a deleted work file on the disk drive. Worth running Recuva or PhotoRec as soon as may be to see if there is such a file, but don't have high expectations.