This report is a bit late, sorry. The Evanescence concert trip in the beginning of the week and Midsummer in the end made last week a bit disorganized. I still got work done, though (about 22 working hours) and this is what I accomplished:
- Made the manimporter to create a basic wiki namespace hierarchy.
- Tried to figure out why I was getting weird exceptions from 4suite. The reason was that it generates some sort of DB for itself and when I moved directories around, I should have removed the DB and let it regenerate that.
- Tried to get to know doclifter source code a bit better. In total it’s something like a bit over 8000 lines of code, so it took a while to even figure out how to use it in my own code. These things took about one day.
- Then I spent one full day trying to figure out if I could use doclifter without temporary files, so that I could feed it the man source and get the XML data out of it as a Python string. I couldn’t figure that out and didn’t want to spend any more time on it. I made my manimporter so that it calls doclifter, which creates a temporary .xml file, then manimporter takes that file, reads the XML data, puts it into a wiki page and then removes the temporary file. Not the most beautiful solution, but it works.
- That’s how the two full working days got spent. Then I implemented authentication via XML-RPC multicall over the weekend, fixing a bug in Moin’s xmlrpc implementation so that I could get it to work in the first place. This took me about six hours.
I basically got done what I wanted to have completed at this point. The manimporter works when you feed it a man file from the file system. What I’d like to get done this week is some sort of CVS integration, so that you could point manimporter to a CVS repository and it would import the man files there. I think I need quite a bit of studying here, since I’ve never used CVS via Python. I’m guessing I might get some help from Dimitris or just by looking at his source code, since his project needs Fedora CVS handling in Python too.
Then something about the schedule. I have to admit that I am a bit behind on that. The biggest reason for it is that most of the things I’ve needed so far have been completely new to me. Moin’s source code, doclifter, XML-RPC etc. According to my original schedule I have two weeks left for this publication phase. During that time I should get CVS, sisterdiff and info importing done. It is possible and I’m ready to work some extra hours (I’m about one full working day behind schedule if you count the hours…). I also had a one week slip possibility in my schedule before this first milestone and if I decide to use that, it definitely should give me enough time to finish phase one. But then again that week would be off from the editing phase. We’ll see. Now back to work 😀
Leave a Reply