Using OpenOffice History at the Command Line
7 April 2008
I wanted to edit an OpenOffice document that I wrote a couple of months ago but I completely forgot the filename.
In OpenOffice, under the file menu, it shows you ten recent documents. I wondered if it stored more than the last ten, what if we want to see the last fifty or so?
The good news is that it does store them. There is an XML file in your home directory called Common.xcu where your file history information is stored.
Therefore I wrote a quick Python script to print out all the stored openoffice history.
Grab the file from here and save it or rename it to oohistory.py. Make it executable:
chmod +x oohistory.py
The basic usage is as follows:
./oohistory.py
This should print out your history. The default output is URLs, one URL per line. They are URLs because the files may not have been edited locally but over http or ftp or sftp or whatever.
However, if you want just the local files, you can use:
./oohistory.py -f
This will give it in a format suitable to piping to another utility.
Say, for example, we were going away for the weekend in a rush, but we are workaholics and want to take our recent office files with us, just in case. So if we wanted to copy all our recent files to a USB flash disk (mounted at /media/disk/ ), then we could go:
./oohistory.py -f | xargs -n1 -i cp {} /media/disk/Documents/
One slight issue with using Office documents in shell commands is that in the Unix world, you expect filenames to not have spaces in, whereas in a graphical program, a non-techincal user might have given their document a filename with spaces in. The moral of this story is to never put spaces in your filenames.
Getting the Path right
Until this point I have assumed you are using the same operating system as me, which might not be true. The script needs to know the location of Common.xcu which varies according to your operating system.
One of the catchphrases of the Gentoo handbook is "Maybe it just works?", i.e. just try it and see what happens.
If it does not work then we need to set the location of Common.xcu.
Firstly, using whatever search tool you like, you need to locate your user's copy of Common.xcu, on a Linux or Unix-like system you normally type:
locate Common.xcu
Now you know the location, there are two ways to set the location of Common.xcu. The first is to just add it as an argument:
./oohistory.py \
~/.openoffice.org2/user/registry/data/org/openoffice/Office/Common.xcu
The second way is to edit the PATH variable at the top of the file.
One of the nice things about Python is that it took me longer to document the script in this post than to write the script itself!
If anyone knows an easy portable way that the script can find Common.xcu, please let me know. (A quick way - I do not want to reimplement the whole of OpenOffice).



1 Lingerance says...
As I recall (never done this) xargs can take input with null terminators allowing you to have spaces in the file names, alternatively you could modify the python script to call a function on each file, or a count of files, much like find -exec will do (presuming you aren't using the busybox crippled one).
Posted at 6:21 p.m. on April 7, 2008
2 Steen says...
I don't know if the formatting comes through right (edit: it doesn't). If not, just add indentation after each ':'
Without the necessary research (it late, i'm lazy), I suspect that os and glob just works in a uniform fashion on all sane platforms
#!/usr/bin/env python import os, glob
fn = "Common.xcu" dr = "/"
print "searching for %s in dir %s and below"%( fn, dr)
for dir, subdir, files in os.walk( dr ): Â for file in files: Â if glob.fnmatch.fnmatch( file, fn ): Â print "%s/%s"%(dir, file)
Posted at 11 p.m. on April 7, 2008
3 Steen says...
I don't know if the formatting comes through right (edit: it doesn't). If not, just add indentation after each ':'
Without the necessary research (it late, i'm lazy), I suspect that os and glob just works in a uniform fashion on all sane platforms
#!/usr/bin/env python import os, glob
fn = "Common.xcu" dr = "/"
print "searching for %s in dir %s and below"%( fn, dr)
for dir, subdir, files in os.walk( dr ): Â for file in files: Â if glob.fnmatch.fnmatch( file, fn ): Â print "%s/%s"%(dir, file)
Posted at 11:11 p.m. on April 7, 2008