Dylan's Journal

I have been working on article about how I wrote Lua code to generate C code to help wrapping object-oriented C (glib-style) and exposing it to Lua, however I have realized that as neat as this sounds, it is terribly difficult to maintain. So let’s pretend it never happened.

Last night was leap-ahead night in the US, thanks to Congress changing when DST accures. Almost all my computer-based clocks updated properly, except for my handspring visor. The poor old thing is incapable of such updates, since all its programming is in ROM.

I will have to set its clock forward manually, and later set it back manually. As it thinks DST should happen in about three weeks. Fun stuff.

Today was a remarkably productive. Resuming the work I started last night on my remind2pda script, I figured out why it was crashing my handspring visor. Apparently it doesn’t you can’t have repeating entries with a frequency of 0! Oops.

The second problem was not my fault, but the documentation of Palm::Datebook. The perldoc states:

The number of the week on which the event occurs. 0 means the first week of the month, 1 means the second week of the month, and so forth. A value of 5 means that the event occurs on the last week of the month

However, that is a lie! A value of 4 means the last week of the month, and a value of 5 causes a strange bug that makes all events appear on the 00th day of the next month.

I also got tired of parsing devtodo’s XML files, so I rewrote devtodo in perl, and using YAML to store todo lists in. More on this later.

Spoony and I tried to get cables to sync our phones using bitpim I was able to find a cable for my phone (a nokia 2810i or something), but not for spoony’s LG VX3400. However, the phone cable I got doesn’t seem to be acting like a proper USB device.