As so often happens to me, I thought it was going to be difficult to implement this but it was very easy. I added just one new method (given a time, figure out what entry ID an entry would have if it had been published at that time) and the rest snapped into place. I would like to attribute this outcome to some special ability on my part, but my genius mainly consists of intuiting the optimal ordering of my feature implementations: retroactive posting was easy because it uses most of the same code as entry deletion (it's just that instead of moving the ordinals of a day's entries back to eliminate one specific entry, you're moving those ordinals forward to accomodate a new one).
Of course, the elegance of Python helps with the snapping into place: Python frequently gives me the I-joined-lots-of-subsystems-and-they-worked-together epiphanies I call 'Lisp moments', without the I-misplaced-a-parenthesis-and-my-dog-exploded headaches I call 'Lisp moments'.
Thu Oct 24 2002 22:16 Daddy, Make The Man Stop Talking About NewsBruiser:
Sorry, kid, but my insane ramblings will ring in your ears throughout this bus ride, yea, even all the way to Tampa. I just added import functionality to NewsBruiser: not import in any specific format, but the generic ability to retroactively post an entry. This will come in handy when I figure out what I need to do to "import from Manilla" to get that checkbox checked and bring in the old editthispage entries. And also, of course, import is essential for Segfault.
