Since CSV (comma-separated-values) files are supported by almost every statistics and/or math tool, I sticked to use the format for all my data collections.

Yesterday I wanted to start using the Eclipse extension PyDev, because I was tired of using the pdb module. Even more sophisticated debugging modules like pudb did not satisfy my need for intuitive and comfortable python debugging. Even though I heard of the PyDev extension quite a while ago, I never gave it a try, as I was reluctant using the Eclipse Framework for Python.

To parse a gzipped xml file using the minidom parser in Python, one has two options: Either hand over the file object pointing to the xml file or hand over the full content as a string. As I thought it would be the more powerful variant in terms of efficiency, I chose to use the file object, given by gzip.open(). This works for the SAX parser, but fails for the minidom parser somehow. This is for sure a bug, but it persists over many versions and many operating systems (tried Linux and Mac).