[Hint] How did I fix my ODT file

[acknak]

It's very hard to do anything but guess without having the actual (.odt) file to look at. I'm sure if you can attach it here, or upload it somewhere else and give us a link, that someone can take a look and see if anything can be recovered.

You can use the "Upload Attachment" link (below the message entry area after you click "POST REPLY"). [Forum] How to attach a document here

[garbledtext]

Hi acknak.

Thanks for your quick reply. I will attach the troubled document.

I read on another forum that images in OOo are by default linked rather than embedded when pasted into a document. Not realizing this, I deleted the image that I had pasted into this document off the hard drive and I think that's what caused this file to go hay-wire.

[acknak]

Sorry, I don't know of anything that can be done with it. This type of damage is new to me; it looks like some binary data was saved in the document archive in place of the document content. Very strange.

There is an intact image in the file, some kind of process flowchart. That can be recovered easily, but I presume that you still have the original image as a separate file.

[RoryOF]

I could get nothing from it. I also looked to see if content.xml was (incorrectly ) some form of archive, but again, nothing.

[EsTr3ToR]

Hi,

Could you guys check on this file?
Tried several ways, none seems to be working.
I would be very thankfull
http://www.mediafire.com/?b53evbgbn8zxz3g

[acknak]

Try this one: uk op 6_recovered.ods

Let me know when you have it and I'll remove the link+file.

[EsTr3ToR]

Works! Thanks a bunch!
What was the problem? Invalid characters or style errors?

[acknak]

Yes; there was a single invalid character in the content.xml.

All I had to do was pass the content.xml file through an xml syntax checker ("xmllint") which found that In one attribute "office:value-type="float", the dash character was replaced with a Ctrl+A (0x01) character.

Once I replaced the invalid character with the hyphen, and replaced the content.xml in the archive, the document opened fine.

[omega_x]

Could anyone help me repair this file?
I already tried to delete all data between "<office:automatic-styles>" and "</office:automatic-styles>" without any results.
A lot of thanks in advance.

Edit: Hagar (Moderator): file removed.

[Hagar Delest]

I've sent you a PM with the recovered file. Since there are some personal data inside, I've removed it from your post.

I think that the file has been cropped, that is the end of the content.xml file has not been correctly written. So by adding the tags that close the document, I've saved what was before the "cut". Don't know how much I've recovered.

[omega_x]

Thank you very much!!

[Hagar Delest]

Just for the record, do you have suffered a heavy data loss or just some bits?

[omega_x]

No, the data is pretty much intact.

[abe]

i am having tough time recovering attached file. I did file recovery using Recuva after accidental mess up deleting files. i have few about 30 files like this state. If somebody can help on this and guide how i might recover rest of the file using same pattern would be much appreciated

[acknak]

Sorry, but the file you attached does not contain the document's data. It seems to contain random information from other files on the system.

There's nothing there to recover.

[mac1]

Similar but different . Did a search ref above if this is wrong place let me know and I'll change
I don't get the above error code.
When I try to open the hour glass shows up , have it left run several hours with no joy, the task manager reports running ( no non_ responding report). When I terminate the process with the task manger then this file shows up .~lock.simplification of meshes.odt# Have deleted all my OO programs via the add and remove route, including those left behind form previous installs, cleaned the registry, ran virus check and then installed latest OO 3.3 ( The problem file is a 3.2 version). The results are the same. Have not tried any fixes ref above. Should I try the above or should I attach file. It is large 1.5 MB??
Ran the validator and it showed 11 warnings and this error;
This file is NOT valid

Result details:

upload:///simplification of meshes.odt/META-INF/manifest.xml[2,88]:Error:element "manifest:manifest" is missing "version" attribute
upload:///simplification of meshes.odt:Info:Generator: OpenOffice.org/3.2$Win32 OpenOffice.org_project/320m18$Build-9502
I don't have a manifest .xml but have a mainfest.rdf.
I extracted the files by rename of my file from xxxx.odt to xxxx.zip and then extracted and then changed back for the above test

[acknak]

It's probably simplest to make the file available so others can try it--if the file works for others, then we'll know it's not a problem with the file. The validator sometimes complains about things that OOo doesn't care about.

You can use the "Upload Attachment" link (below the message entry area after you click "POST REPLY"). [Forum] How to attach a document here

Note: large documents have to be shared through some other means; there are suggestions in the link above.

[mac1]

Up date;
ajut001 determined what to do but I did change the flow a little:
1) Unzip the xxxxxx.zip ( the original xxxxx.odt OO with extension change) . I use Zip Genius for this. ;
2) Open the xml file using the xml NotePad 2007 I downloaded form Microsoft. The program has a tree capability so search for the target to delete is not required because it appears in the top level of the hierarchy;
2) Delete target and save file;
4) Select this save file and the others extracted from the xxxxxx.zip and use Zip Genius to make a archive;
5) Save archive and change file name back to xxxxxxx.odt
File now opens OK. It appears the number of jpg's I had in th file were causing the problem. File must be re- formatted and pictures re-inserted.
How does one avoid this goat rope in the future
I was very close to submitting this as a WIKI and these problems are not acceptable.

[chariteb]

Hello,

Can someone fix my file, 6 months of work is in this file.

It's a spreadsheet and the error is:

Error in subdocument content.xml on position 2,528190 (row,column).

I've tried every tip in this forum, still Ooo refuses to read.

http://www.mediafire.com/?7oz2hf5f3skf5hn

Regards,

Boudewijn

[acknak]

Try this one: Dagboeken 2011_recovered.ods

Do remember to make regular backup copies of important documents. Every time you open a document you're taking a chance that it will be corrupted or lost. Your only real protection is to make backup copies regularly, using different file names.

[chariteb]

Thanks it worked,

and I will start immediate with a daily backup system. I had but didn't installed it on my new file server yet.

I don't know why but I opened this file in Libre Office 3.4.0 did some work in it saved it, got some error, saved it again without any warning now so I assumed it was ok. A few days later found this error.

What was wrong with the file because I could not find the "real" error. The validator http://tools.services.openoffice.org/odfvalidator/ of Ooo gave me some errors in some columns but still couldn't open it after fixing most of them.

Which tool you used to edit the xml file, notepad++ was terribly slow editing such a big file.

Regards,
Boudewijn

[Hagar Delest]

Don't know for acknak but personally, I use XML Copy Editor (both on Windows and GNU/Linux) and it works fine, it can check the XML structure directly so you can undo modifications if you introduce errors.

[acknak]

Any software that I've ever tried for working with ODF's xml has been very unpleasant--either too slow or somehow unable to cope with OOo's "all in one line" xml, or an xml file that OOo has made a hash of.

I generally use two command-line utilities and a plain text editor. Emacs is my text editor; xmlindent is handy to re-format OOo's all-in-one-line xml, and xmllint tries to check an xml file for errors.

In this case, the problem turned out to be very easy to deal with.

Xmllint gave me this output:

$ xmlindent -i2 content.xml | xmllint -noout -
-:4725: parser error : Attribute table:end-cell-address redefined
-:4725: parser error : Attribute table:end-x redefined
-:4725: parser error : Attribute table:end-y redefined
-:4725: parser error : Attribute draw:z-index redefined

Line 4725 looks like this:

<table:table-cell table:end-cell-address="&apos;Kasboek per dag&apos;.J42" table:end-x="0.926cm" table:end-y="0.417cm" draw:z-index="0" table:end-cell-address="&apos;Kasboek per dag&apos;.J40" table:end-x="0.476cm" table:end-y="0.365cm" draw:z-index="0" table:style-name="ce251" office:value-type="string" table:number-columns-spanned="6" table:number-rows-spanned="2">

You can see that four of the attributes appear twice in the same xml element. That's not allowed, so the xml is not valid and won't be loaded.

I used a text editor to remove the first set of attribute values, leaving only the second set, shown in blue. I then updated the content.xml file in your document with the modified one and that opened without errors in Calc.

[n1120ec]

I don't know if you guys are still helping out people with file errors, I've tried all the things listed and at wits end right now. Wondering if anyone can help out with a file. I've posted it http://www.mediafire.com/file/4ou69opcq ... orynew.ods

Thank you for the help, I used ALZIP to unzip the file so I will upload that content.xml http://www.mediafire.com/?qd24fb2u18d98p6

[floris v]

There's a problem with your file: I couldn't extract content.xml so I downloaded a demo of ZipRepair Pro to see if I could repair it and it reports that the filesize of the archive seems to be incorrect. To be sure, a compression of 95% seems a bit much. I'm afraid that part of the file got truncated and that content.xml can't be recovered.

[Hagar Delest]

I managed to extract the content.xml file (but having the CRC error). Managed to get a part of it in vi but no way to make it displayed in Gedit or XML Copy Editor so not sure it can be recovered.

[n1120ec]

Do you guys think if I use Sysinfo tools for Calc recovery to fix, if it would work???? And thank you guys for all the help

[TheGurkha]

Do you guys think if I use Sysinfo tools for Calc recovery to fix, if it would work? And thank you guys for all the help

Sadly no.

[acknak]

The text in the content.xml file is very badly garbled. I very much doubt that any usable data can be recovered from it, by any tool. The only solution is a backup file.

[Bill]

Maybe you can recover a previous version of the file:

Previous versions of files: frequently asked questions