6/30 - Another weekly meeting day. We've been having trouble getting the Firefox doxygen outputs to combine into one file (the output comes with a combine XSLT stylesheet and an index XML file to do this automatically) because the XSLT processor we are using is java, and java keeps running out of heap space to run so many files at one time. We even tried running it on the server with 8G's of space, but it still ran out of space. So we're giving up on Firefox for our project and are going to focus on Apache, since this project is considerably smaller. Our hope for this week is to glean the inheritance constraints from the Apache system through the doxygen output. We shall see. Our other task is to start writing the script to mine the developers' mailing list for Apache and try to get the data into the database.
7/2 - So for the past two days, I've been feverishly working at converting a script written entirely in Perl (a language I'm barely familiar with) into Ruby (another language I'm slightly more familiar with). I swear, my eyes have crossed at least 6 times in the last three days. It's been a tough week for them, for sure. As for the doxygen output, the only constraints I can find that would be easily accessible are the includes and includedby tags, which tells which classes/header files are included by this module, and which other modules include this module. However, this is not what Sunny wants, so we are stuck as to how to find the constraints otherwise. He wants function calls so as to know the exact function that is being inherited, but the only way to do this would be to parse through all of the output's codelines (basically, the source code) and find all of the function calls. I really, really do not want to do that. |
