On Fri, Feb 4, 2011 at 4:45 PM, Bill Campbell <[email protected]> wrote:
> On Fri, Feb 04, 2011, Jeff Chimene wrote: > >On 02/04/2011 04:13 PM, Bill Campbell wrote: > ... > >>> The current OO <==> Docbook interchange technique uses styles to > >>> indicate how to interchange OO text and Docbook. That isn't to say that > >>> using bookmarks is wrong, just that I haven't tried a stylesheet that > >>> uses that technique. As part of your OO installation, you should have a > >>> Docbook filter. That filter is tailored to Docbook 4, not 5. > >> > >> I tried importing a manually created DB XML file into OO. The original > >> document had an <author> section, which seems to have been translated > into > >> OO field variables, one for surname, the surname was in a variable named > >> "authorinfo.author_0.surname=Spooner", but the firstname showed up in an > >> unnamed variable. > > > >That sounds like an imperfect import filter. Did you try the built-in > >Docbook filter? > > Good question Jeff. > > The one I used for my first test is the one that's bundled with > the Latest&Greatest version of NeoOffice, OpenOffice.org with a > native Mac interface. I just tried importing the same file using > the current version of OpenOffice.org which has a slightly > different "docbooktosoffheadings.xsl" file, but it still had the > same variable names including the missing name for <firstname>. > > I just did a "diff" on these files from NeoOffice and > OpenOffice.org, and the only changes were in the Copyright and > Revision header text. There were no changes in the working part > of the files. > > >> My current procedure is to process the docbook.xml file generated by OO > >> through a python script using elementtree to clean up the file. One > thing > >> that was causing serious DTD problems was that the informaltable header > >> entries had the text wrapped in <title> tags. > >> > >> I have quite a few python scripts I've written over the years to do > things > >> like de-Microsoft html files, removing all their font and style cruft to > >> produce clean html. With a reasonable knowledge of regular expressions > >> this isn't too difficult. > >> > > >Python is interesting, but that's XSLT kind of processing. > > I know python, but have only a rudimentary knowledge of XSLT now. > I've done a first-reading of a couple of books on it, but have > never done anything serious. > > >Interesting you mention WP. That's the only WYSIWYG word processor I've > >seen that has the horsepower to handle importing/exporting markup > languages. > > > >Anyway: do you want to round-trip these documents? > > Not really. I'll be happy to be able to do original typing in OO > or a variant, the use vim for the Real Work(tm). > > FWIW, I've tried getting my head and fingers around emacs several > times in the last 20 years or so, but they seem terminally > infected with almost 30 years of vi(m) use. > > Bill > -- > INTERNET: [email protected] Bill Campbell; Celestial Software LLC > URL: http://www.celestial.com/ PO Box 820; 6641 E. Mercer Way > Voice: (206) 236-1676 Mercer Island, WA 98040-0820 > Fax: (206) 232-9186 Skype: jwccsllc (206) 855-5792 > > Memoirs -- Bill Clinton is getting $12 million for his memoirs, and his > wife Hillary got $8 million for hers. That's $20 million for memories from > two people who for eight years repeatedly testified they couldn't remember > anything. >
