asciilifeform: ( there is not, notably, any possibility of a pill that dun force cpointerism ; as gnatv is at birth a cpointeristic turd )
asciilifeform: aaand if anyone can think of a less revoltingly-tasting pill, plz to also write in.
asciilifeform: if anyone knows otherwise, plox to write in.
asciilifeform: it relies , as is obvious , on the guts of gnat's main() initializer, that sets the turds . but in so far as i can tell , it is the same in all known gnats
asciilifeform: ^ used in place of , rather than in addition to, Ada.Command_Line
asciilifeform: it's like encountering a toilet with a reverse gear, is what this is.
asciilifeform: and hats off to the 1st reader here who immediately thought, reading the ads earlier, 'wtf does it mean and who ever heard of ~removing~ arguments from cmdline ?!'
asciilifeform: ( the 2ndarystack is used to keep around the lengths of this variable-length turd that got thrown on the primary. )
asciilifeform: if you return a String, that's 2ndarystacktronic . no escape.
asciilifeform: this is the , in mircea_popescuine parlance, 'hole where the night came in'
asciilifeform: it sits on the edge between outright wrecking and simple down's syndrome
asciilifeform: well i have nfi , but it entirely needlessly drags along the secondarystack thing
asciilifeform: dunno why i didn't think of it earlier
asciilifeform: just as easily it is possible to give it a 'here's an empty string of length L to take a shit into, for each cmdline param, and if the actual exceeds it, trigger constrainterror'
asciilifeform: and thereby force the use of secondary stack
asciilifeform: and in the process discovered that the original was ~maliciously~ stupid, there is NO reason why it HAD to return an unknown-length string value
asciilifeform: in other noose, asciilifeform wrote a sane commandline-getter for gnat
asciilifeform: the surviving hruschebas asciilifeform saw in bucharest, however, looked quite like this
asciilifeform: quite heavy compared to even, say, trb.
asciilifeform: only 'small' compared to gcc monster.
asciilifeform looked again at last-known tcc. it is not , after all, so small : ~54kLoc ; ~1.7MB !!! of sores.
asciilifeform: in other olds, asciilifeform found insinuations that gcc 4.7 was at some time built successfully using tcc. but no detail re how, in particular which tcc, where to get that
asciilifeform: phf: there is no amount of bending-over-backwards that is toomuch for these usefulidjits
asciilifeform: it isn't clear that it is even possible to carry out an intellectual exchange with such folx, who have 0 shared priors with actual people. the only conversation can then consist of the lime pit.
asciilifeform: i was abouttosay, what good does 'proof' when reply is inevitably 'but where is the proof of yer proof's proofiness'
