2022-05-19 14:45:09 <KipIngram   > Oh.  A day or two ago I posted some rambling about letting compilation flow from one block to the next, seamlessly, as a prelude to eventually having a file system.  Then later I said maybe that was unnecessary - I could just implement --> instead.
2022-05-19 14:45:36 <KipIngram   > Well, I think maybe I do want the first thing - I'd actually like for a word to be able to straddle a pair of blocks, if that's how things fell.
2022-05-19 14:45:44 <KipIngram   > --> wouldn't support that.
2022-05-19 14:46:19 <KipIngram   > What I'm doing now is checking to see if >IN has reached 4096; if it has, I return a zero, so it sort of forces a null word into the input stream - terminates the LOAD.
2022-05-19 14:46:41 <KipIngram   > But it would be easy to replace that with BLK 1+! >IN OFF.
2022-05-19 14:47:04 <KipIngram   > I think that's all I'd need to do, actually.
2022-05-19 14:47:21 <KipIngram   > Then eventually I'd have to actually have a null character in a block, to end the LOAD.
2022-05-19 14:48:22 <KipIngram   > So, if I do that LOAD will just keep loading, however many blocks, until it reaches a null termination.
2022-05-19 14:53:15 <KipIngram   > I'll have to move that end of block check - right now I'm fetching the next character (which may be beyond the end of the block) and replacing it with null if it was - I'll want to move it before the BLK @ BLOCK that appears in that code.
2022-05-19 16:37:07 <KipIngram   > Ok, that seems to work just fine.