Outstanding issues which may or may not get fixed/done someday

Good thing the python scripts are examples, meant to spark
improvement by others or development of entirely new packages;
you know you could write something better if you just sat down and
spent some time, so go do it!

sqlfilter could read INSERT statements that span multiple
lines as long as each line is started by a tupe with possibly leading
whitespace; this would let it process output from sql2txt

timing tests with importDump.php to compare loading of content

pretty sure some of these programs don't provide version information

bash script to use sqlfilter on sql table X from mw version Y
to write it in format for version Z (earlier version) by dropping
columns as needed

speeding up C code?  compare to speed of equiv python or perl scripts?

do something about media import

script to get all pages in category or all its subcategories etc

maybe convert ooooold (2001 through 2003) dumps to a more modern
format?

in wikicontent2sql.py, we should be able to get away with not using
a stub file and generating text len on the fly, all other data we need
is in the content file downloaded; also we should be able to use
regular content dumps and not just those produced by Special:Export.
