[Solved] This is not a WinWord6 file, General input/output

[auname]

I was working on a file, then there was a power outrage, the computer switched off.
When the power was back, I tried to open the file, at first, it said "This is not a WinWord6 file", and then "General input/output error".
I had auto recovery switched on in the options but not backup.
Read through the similar topics and it did not help me.
What can I do?

[jrkrideau]

What format was the file in? Was it an original odt file or converted from Word or some other word processor?

[auname]

I saved it as word97-2003

[John_Ha]

1. Search the forum with winword or winword6 for other posts.
2. Upload the file somewhere so it can be looked at.

[auname]

I already searched, did what they said and it did not help.
File:
https://files.fm/u/vmppvpks

[erbsenzahl]

I tried to open on LibreOffice 6.0.3.2 - no success (only some cryptic characters)

Then tried to convert on zamzar.com - this is their reply:

Hello from Zamzar !
There was a problem converting your file - We attempted to perform the conversion you requested from doc to odt but unfortunately the conversion failed - our support team have been notified and are looking into the problem.

Your file seems to be corrupted, probably a specialist's efforts may have more success...

[John_Ha]

The file is corrupted. I did the following:

1. Opened the .doc file with a Hex Editor (HxD)
2. Copied out all the text.
3. Pasted it into a .odt file.

fairy.doc viewed in HxD

The .odt file is 159 pages long ... but every character is separated by a " full stop " character. I cannot seem to remove these " full stop " characters with Find and Replace. They appear as NUL characters in NotePad++ ... and I cannot remove them with Find and Replace there either.

The full file is too big to upload here so I have uploaded it here. I have also uploaded the first page as first page.odt here to see if anyone can work out how to remove the " full stop " characters. If they can then doing the same to the full file should recover all the text.

See [Tutorial] Differences between Writer and MS Word files for why you should always work in and save files as .odt and never use .doc files.

You could also try recovering previous deleted versions of the file using a file un-deleter like Recuva.

You could also try to undelete any temporary files AOO may have written while you were editing the file. Unfortunately, as you were saving as a .doc file, and not saving as a .odt file, this will not work unless you have AutoSave or Create a backup copy enabled. See [Tutorial] How to find and un-delete Writer temporary files.

[John_Ha]

See fairy text no full stops.odt where I have got rid of the full stops leaving just the text. It needs re-formatting but at least you don't have to rewrite it.

I used NotePad++ to remove all the " 00 " which took many seconds as there were 500,000 or so. AOO was also able to do it. I then pasted the result back into HxD which has a View > Visible columns > Text ..., which displays only the text. I then copied the text into a .odt file.

One thing which failed was to create a word6 .doc file with one word and copy all the text from fairy.doc into that new .doc file keeping the binary stuff before and after the text component. AOO said the resulting .doc file was corrupted.

Showing that a problem has been solved helps others searching so, if your problem is now solved, please view your first post in this thread and click the Edit button (top right in the post) and add [Solved] in front of the subject.

[auname]

See fairy text no full stops.odt where I have got rid of the full stops leaving just the text. It needs re-formatting but at least you don't have to rewrite it.

I used NotePad++ to remove all the " 00 " which took many seconds as there were 500,000 or so. AOO was also able to do it. I then pasted the result back into HxD which has a View > Visible columns > Text ..., which displays only the text. I then copied the text into a .odt file.

Clipboard02.gif

One thing which failed was to create a word6 .doc file with one word and copy all the text from fairy.doc into that new .doc file keeping the binary stuff before and after the text component. AOO said the resulting .doc file was corrupted.

Showing that a problem has been solved helps others searching so, if your problem is now solved, please view your first post in this thread and click the Edit button (top right in the post) and add [Solved] in front of the subject.

I love you,John_Ha.

[John_Ha]

Several things.

1. Be sure to make backup copies in future. Plug in a USB memory stick and make daily? weekly? backup copies to the USB stick. Google for free backup programs.

2. Stop saving as a .doc file and always save as a .odt file.

3. Tools > Options > Load/Save > General. Tick Always create a backup copy. Tick Save AutoRecovery information every ..., and choose a value - say 10 minutes. See [Tutorial] How to find and un-delete Writer temporary files for a discussion of the options.

[auname]

I save every day when I am finished writing. This time, the power switched off in the middle of the day and I lost several last pages.
Activated backup copy already.

[MrProgrammer]

I saved it as word97-2003

Bad idea. Always save as ODT.

I had auto recovery switched on in the options but not backup.

You should enable that option, if you didn't yet. You may also want to investigate software which makes periodic copies of files on external devices. I don't use Windoze and can't suggest anything. Perhaps Windoze even can do that without additional software. Remember that all hard drives eventually fail.

but every character is separated by a " full stop " character.

No. Viewed as ASCII, the characters are separated by NUL (X'00'), not a "full stop" (period). Your editor displays the NUL as a period.

I changed the file extension from DOC to TXT and I opened the file with a text editor to remove the first bunch of lines so OpenOffice could not recognize it as a Microsnot Word Document. Then I opened the TXT file with OpenOffice and specified character set Unicode. There are several ways to encode Unicode characters in a file. The DOC file may be using UTF-16/UCS-2. After the initial paragraphs there is other data in the file which represents formatting, but you can delete that and quicky recover most of the text.

[John_Ha]

I googled repair doc file and came across lots of useful stuff.

I downloaded GetData's Repair my Word but it failed.

I then went to Recovery Toolbox which offers an on-line recovery service for .doc files. I uploaded fairy.doc and it seemed to recover the entire file. It reported it recovered 217 pages and the preview showed the text was double spaced.

When I pressed Next it asked for $10 so I left . I think you would be well advised to use the site as it will save hours of work.

I could not identify the cause of the corruption in the file but the last 1/3 of the .doc file was full of NUL characters and the content after the text I recovered and before the NULS appeared seemed to be garbage.

[Zizi64]

Activated backup copy already.

The built-in backup feature can create one backup copy per document, and the backup file will be rewrite again and again. If something go wrong in your file, the wrong version will overwrite the previous saved version.

Save more than one backup copies manually - with timestamped file name. Or use this extension:
viewtopic.php?f=47&t=86742

[John_Ha]

The DOC file may be using UTF-16/UCS-2.

I have been told that a .doc file is effectively a dump of the memory contents.

... a Microsnot ...