Rule/command for Word Perfect (x-fmt/44)

Skip to first unread message

Nov 3, 2021, 1:20:46 PM11/3/21
to archivematica
Hi all,

I didn't find this topic anywhere, at least not with a recent enough date, but if I missed one that already answers this, please feel free to redirect.

I'm wondering if anyone has a rule/command written to normalize Word Perfect files (specifically x-fmt/44) to ODF/ODT files for preservation. I realize that the search is ongoing for an "officially approved" normalization strategy for these, but I'm dealing with the results of a late '90s/early 2000s database, so I need to get them converted.

Has anybody had any success doing something similar or have an idea of a proper command for this? Similarly, anyone have any luck adapting the Office to pdf command to work with Word Press files?

I assume Libre Office is the tool to use, but my command writing skills aren't great, so any help that anyone can provide would be much appreciated.



Nov 3, 2021, 1:20:54 PM11/3/21
Message has been deleted


Nov 10, 2021, 11:14:12 AM11/10/21

Grant Hurley

Nov 25, 2021, 4:02:03 PM11/25/21
to archivematica
Hi Jarad,

I've tested with the below command to convert WordPerfect to ODT using LibreOffice installed in Ubuntu 18.04. The source of this was an Archivematica tech forum post by Tessa Walsh from a few years ago, though, so I can take no credit! -

The basic steps are:
- Install Libreoffice on the VM where Archivematica is installed - sudo apt install libreoffice
- Install Microsoft true type coreutils fonts package here too - I have found this is helpful in making sure LibreOffice has access to some of the fonts you would commonly find in doc files - though I can't guarantee it will always work - see comment below. I haven't tested whether this package benefits WordPerfect conversion specifically but it might be worth checking out. sudo apt-get install msttcorefonts
- Add LibreOffice as a new tool under 'Format policy registry' > 'Tools' in the Preservation Planning tab
- Create a normalization command for 'Transcoding to odt with LibreOffice' using the below information
- Create a normalization rule using this command for each WordPerfect version that is required - it depends if your source files are all in one version or not. 

One thing is that with LibreOffice is that you're in great shape if you're only concerned about getting the content from the source file. Presentation is another matter. General layout like paragraph breaks and indents and things like emphasis and underlining should be retained in the ODT output. But for any office doc-type conversions, LibreOffice needs the same fonts also installed if you want the same presentation as the original document. As far as I understand, in the case of WordPerfect 5.1, original fonts were selected by users in relation to the output to be sent to their printer rather than being part of the software per se - not sure what this looked like in subsequent versions of WordPerfect. So you might not be able to perfectly replicate what the original printed version looked like without somehow also emulating a dot matrix printer! But in any case, it's a good idea to do some QA testing whenever you are using LibreOffice in this context.


for i in `seq 1 10`; do libreoffice --headless --invisible --convert-to odt --outdir "%outputDirectory%" "%fileFullName%" 
if [ -f "%outputDirectory%%fileName%.odt" ]; 
    then mv "%outputDirectory%%fileName%.odt" "%outputDirectory%%prefix%%fileName%%postfix%.odt" 

Script type: Bash script
Output file format: ODF 1.0 generic
Output location: %outputDirectory%%prefix%%fileName%%postfix%.odt
Command usage: Normalization
Verification command: Standard verification command (non zero filesize)

Hope this helps!


Nov 26, 2021, 2:13:30 PM11/26/21
to archivematica

Hi Grant,

Thanks for this. I won't be able to implement/test immediately as I'm using a cloud-based setup for which I don't have access to the back end / OS to install anything, myself but I'll see what I can get done. Failing that, I may just opt for manual normalization. At any rate, thanks for the info!

Reply all
Reply to author
0 new messages