// ============================================================================ // compiler.zc: the resident zc compiler. it lives in the genome next to the // runtime, compiles source straight to vm words in memory, and is visible to // the programs it compiles: every function and global here can be called or // patched by name, and parse words (immediate fns) run inside it. // ============================================================================ // ---- compiler data region (fixed addresses, 0xA0100 .. 0xBE100) ---- const IHN = 8192; const R_IH = 0xA0100; // intern hash table const R_POOL = 0xA2100; // interned strings: [symbol head] chars 0 const R_POOLEND = 0xAC000; const NS = 4096; // symbol table, one array per field const R_SYM = 0xAC100; const sname = 0xAC100, skind = 0xAD100, sval = 0xAE100, ssig = 0xAF100; const sflag = 0xB0100, sfix = 0xB1100, snext = 0xB2100; const NL = 512; // locals, one array per field const lname = 0xB3100, lslot = 0xB3300, lhome = 0xB3500, lstate = 0xB3700; const lowner = 0xB3900, lused = 0xB3B00, larr = 0xB3D00, lown = 0xB3F00; // lown: has owned an object const NF = 8192; // patch nodes: a site and a link const fsite = 0xB4100, fnext = 0xB6100; const R_SNAP = 0xB8100; // analyzer snapshots const R_QBUF = 0xBA100; // token buffer for quote and grab const R_SRC = 0xBC100; // input source stack, 12 words per frame const R_SEQ = 0xBC500; // operator code sequences const bname = 0xBCD00, bprec = 0xBCD40, bseq = 0xBCD80, blen = 0xBCDC0; // binary operators const uname = 0xBCE00, useq = 0xBCE20, ulen = 0xBCE40; // unary operators const aname = 0xBCE60, abin = 0xBCEA0; // compound assignment const kops = 0xBCF00; // every operator spelling the lexer knows const R_CTX = 0xBD100; // saved compiler contexts for nested compiles const R_LOOP = 0xBD900; // loop stack, 6 words per loop const R_ASM = 0xBDD00; // asm block labels // ---- token kinds, symbol kinds and flags ---- const T_EOF = 0, T_NUM = 1, T_ID = 2, T_STR = 3, T_OP = 4; const K_FN = 1, K_GVAR = 2, K_GARR = 3, K_CONST = 4, K_KW = 5; const F_DEF = 1, F_SIG = 2, F_UNTYPED = 4, F_STMT = 8, F_EXPR = 16, F_TOP = 32; // variable states for the lifetime analyzer const S_PLAIN = 0, S_M = 1, S_U = 2, S_D = 3, S_DEAD = 4, S_CHILD = 5, S_BORROW = 6; // expression classes: plain word, a new object, or a bare variable (C_VAR + index) const C_PLAIN = 0, C_FRESH = 1, C_VAR = 2; // lvalue kinds const LV_NONE = 0, LV_LOCAL = 1, LV_GLOBAL = 2, LV_ELEM = 3, LV_OTHER = 4; // opcodes const O_Q = 0, O_LD = 1, O_ST = 2, O_PICK = 3, O_DROP = 4, O_SWAP = 5, O_CALL = 6, O_JMP = 7; const O_JZ = 8, O_RET = 9, O_ADD = 13, O_SUB = 14; const SEQREF = -2; // in an operator sequence: the next word names a symbol // ---- state ---- var pooltop, nsym, ubase, vislo, nfixu, fixfree; var nbin, nun, nasg, nkop, seqtop; var mnem = "q@!$_~cjzr,?.+-*&|^><"; var check_on = 1; // set to 0 to switch the analyzer off for what follows var tk, tv, tglue, tline, line = 1, ch, chkind, chptr, qdone, isp = -1; var tbuf[1100]; var here, hdelta, hlimit, ptop; var nloc, lbase, fdepth, fslots, fchk, fret, live, bxdepth, epichain, fframe, infn; var lsp, loopbase, ctxsp, snaptop; var lv, lvk, lvi, kw_expr, xcls; // interned spellings the parser compares against var s_lp, s_rp, s_lb, s_rb, s_lk, s_rk, s_comma, s_semi, s_eq, s_arrow, s_inc, s_dec; var s_star, s_amp, s_colon, s_hash, s_minus, s_andand, s_oror; var s_fn, s_var, s_const, s_else, s_as, s_unchecked, s_immediate, s_m, s_d, s_u; var s_start, s_main, s_free, s_own, s_qnew; // ---- errors ---- fn die(msg) { puts("zc: line "); putn(tline); puts(": "); puts(msg); putc(10); halt(); return 0; } fn dien(name, msg) { puts("zc: line "); putn(tline); puts(": "); puts(name); puts(": "); puts(msg); putc(10); halt(); return 0; } // analyzer errors only count in checked, reachable code fn cerr(msg) { if (fchk && live) die(msg); return 0; } fn cerrn(i, msg) { if (fchk && live) dien(lname[i], msg); return 0; } // ---- interning ---- fn hashs(s) { var h = 0; while (*s) { h = h * 31 + *s; s++; } return h & (IHN - 1); } fn intern(s) { var h = hashs(s), p, n; while (1) { p = R_IH[h]; if (!p) break; if (streq(p, s)) return p; h = (h + 1) & (IHN - 1); } n = strlen(s); if (pooltop + n + 2 >= R_POOLEND) die("string pool full"); pooltop[0] = 0; p = pooltop + 1; wcopy(p, s, n + 1); pooltop = pooltop + n + 2; R_IH[h] = p; return p; } // ---- symbols ---- // each interned name keeps the index (+1) of its newest symbol in p[-1], and // each symbol links to the one it shadows, so lookup is one step. fn addsym(name, kind, val) { var i = nsym; if (nsym >= NS) die("symbol table full"); nsym++; sname[i] = name; skind[i] = kind; sval[i] = val; ssig[i] = 0; sflag[i] = F_DEF; sfix[i] = 0; snext[i] = name[-1]; name[-1] = i + 1; return i; } // while an image builds, names below vislo are hidden, except keywords fn lookup(name) { var i = name[-1] - 1; while (i >= 0) { if (i >= vislo || skind[i] == K_KW) return i; i = snext[i] - 1; } return -1; } fn popsyms(to) { while (nsym > to) { nsym--; sname[nsym][-1] = snext[nsym]; } return 0; } fn lookups(s) { return lookup(intern(s)); } // patch nodes fn fixnew(site, link) { var i = fixfree; if (i) fixfree = fnext[i]; else { nfixu++; i = nfixu; if (i >= NF) die("too many forward references"); } fsite[i] = site; fnext[i] = link; return i; } // patch every lit in a chain to point at v, and recycle the nodes fn fixall(head, v) { var n; while (head) { setlit(fsite[head], v); n = fnext[head]; fnext[head] = fixfree; fixfree = head; head = n; } return 0; } // ---- code emission ---- fn emit(w) { if (here >= hlimit) die("program too big"); *(here + hdelta) = w; here++; return 0; } fn lit(v) { emit(v + v + 1); return 0; } fn op(k) { emit(k + k); return 0; } fn setlit(at, v) { *(at + hdelta) = v + v + 1; return 0; } fn jmp(t) { lit(t); op(O_JMP); return 0; } // a lit that will be patched later; returns its address fn hole() { var a = here; lit(MW - 1); return a; } fn fwdfn(name) { var s = addsym(name, K_FN, 0); sflag[s] = 0; return s; } fn fnref(s) { if (sflag[s] & F_DEF) { lit(sval[s]); return 0; } sfix[s] = fixnew(here, sfix[s]); lit(MW - 1); // a jump to MW-1 halts if it ever runs unpatched return 0; } fn callname(name) { var s = lookup(name); if (s < 0) s = fwdfn(name); fnref(s); op(O_CALL); return 0; } fn laddr(k) { lit(ZFP); op(O_LD); if (k) { lit(k); op(O_ADD); } return 0; } fn strlit(s) { var j = hole(), a; op(O_JMP); a = here; while (*s) { emit(*s); s++; } emit(0); setlit(j, here); lit(a); return 0; } // ---- input sources ---- // frame: 0 kind (0 stdin, 1 token list, 2 string) 1 object 2 index 3 end // 4 transparent 5..7 saved token 8..10 saved char state 11 owned fn nextch() { if (chkind == 0) { if (qdone || eof()) { ch = -1; return 0; } ch = getc(); if (ch == 96) { qdone = 1; ch = -1; return 0; } // a backtick ends the source } else { ch = *chptr; if (ch) chptr++; else ch = -1; } if (ch == 10) line++; return 0; } fn src_push(kind, obj, trans, owned) { var f; isp++; if (isp >= 32) die("sources nested too deeply"); f = R_SRC + isp * 12; f[0] = kind; f[1] = obj; f[2] = 0; f[3] = 0; f[4] = trans; f[11] = owned; f[5] = tk; f[6] = tv; f[7] = tglue; f[8] = ch; f[9] = chkind; f[10] = chptr; if (kind == 1) f[3] = obj[0]; if (kind == 0) { chkind = 0; qdone = 0; nextch(); } if (kind == 2) { chkind = 2; chptr = obj; nextch(); } return 0; } fn src_pop() { var f = R_SRC + isp * 12; tk = f[5]; tv = f[6]; tglue = f[7]; ch = f[8]; chkind = f[9]; chptr = f[10]; isp--; if (f[11]) free(f[1]); return 0; } fn isopch(c) { var s = "+-*/%=<>!&|^~?:.@#$"; while (*s) { if (*s == c) return 1; s++; } return 0; } // is buf[0..n) followed by c the start of some operator the lexer knows? fn opprefix(n, c) { var i = 0, k, j; while (i < nkop) { k = kops[i]; j = 0; while (j < n && k[j] == tbuf[j]) j++; if (j == n && k[n] == c) return 1; i++; } return 0; } fn isdig(c) { return c >= '0' && c <= '9'; } fn isal(c) { return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || c == '_'; } fn hexv(c) { if (isdig(c)) return c - '0'; if (c >= 'a' && c <= 'f') return c - 'a' + 10; if (c >= 'A' && c <= 'F') return c - 'A' + 10; return -1; } fn escch() { var c = ch; nextch(); if (c != 92) return c; c = ch; nextch(); if (c == 'n') return 10; if (c == 't') return 9; if (c == 'r') return 13; if (c == '0') return 0; if (c == 'e') return 27; return c; } fn oplex(n) { while (ch >= 0 && isopch(ch) && opprefix(n, ch)) { tbuf[n] = ch; n++; nextch(); } tbuf[n] = 0; tk = T_OP; tv = intern(tbuf); return 0; } fn lex() { var n = 0, v, d; tglue = 1; while (1) { while (ch == 32 || ch == 9 || ch == 10 || ch == 13) { nextch(); tglue = 0; } if (ch != '/') break; nextch(); if (ch == '/') { while (ch >= 0 && ch != 10) nextch(); tglue = 0; continue; } if (ch == '*') { nextch(); while (ch >= 0) { if (ch == '*') { nextch(); if (ch == '/') break; } else nextch(); } nextch(); tglue = 0; continue; } tline = line; tbuf[0] = '/'; oplex(1); return 0; } tline = line; if (ch < 0) { tk = T_EOF; tv = 0; return 0; } if (isdig(ch)) { v = 0; if (ch == '0') { nextch(); if (ch == 'x' || ch == 'X') { nextch(); while ((d = hexv(ch)) >= 0) { v = v * 16 + d; nextch(); } tk = T_NUM; tv = v; return 0; } } while (isdig(ch)) { v = v * 10 + ch - '0'; nextch(); } tk = T_NUM; tv = v; return 0; } if (isal(ch)) { while (isal(ch) || isdig(ch)) { if (n < 1000) { tbuf[n] = ch; n++; } nextch(); } tbuf[n] = 0; tk = T_ID; tv = intern(tbuf); return 0; } if (ch == 39) { nextch(); tv = escch(); if (ch != 39) die("bad character literal"); nextch(); tk = T_NUM; return 0; } if (ch == 34) { nextch(); while (ch >= 0 && ch != 34) { v = escch(); if (n < 1000) { tbuf[n] = v; n++; } } nextch(); tbuf[n] = 0; tk = T_STR; tv = intern(tbuf); return 0; } tbuf[0] = ch; nextch(); if (isopch(tbuf[0])) return oplex(1); tbuf[1] = 0; tk = T_OP; tv = intern(tbuf); return 0; } fn next() { var f, q, i; f = R_SRC + isp * 12; if (f[0] != 1) { lex(); return 0; } i = f[2]; if (i < f[3]) { q = f[1]; tk = q[1 + i + i]; tv = q[2 + i + i]; tglue = 0; f[2] = i + 1; return 0; } if (!f[4]) { tk = T_EOF; tv = 0; return 0; } src_pop(); // a fed list ran out: continue with the token that followed it return 0; } fn is_op(s) { return tk == T_OP && tv == s; } fn is_id(s) { return tk == T_ID && tv == s; } fn accept(s) { if ((tk == T_OP || tk == T_ID) && tv == s) { next(); return 1; } return 0; } fn expect(s) { if (!accept(s)) { puts("zc: line "); putn(tline); puts(": expected "); puts(s); putc(10); halt(); } return 0; } fn name() { var n = tv; if (tk != T_ID) die("expected a name"); next(); return n; } // ---- operator tables ---- // a sequence is written like asm: one-char mnemonics, numbers, and names // (resolved when the operator is used, so image builds stay self-contained) fn seqasm(s) { var start = seqtop, n, neg, v, k, c; while (1) { while (*s == 32) s++; if (!*s) break; n = 0; while (s[n] && s[n] != 32) { tbuf[n] = s[n]; n++; } tbuf[n] = 0; s = s + n; neg = tbuf[0] == '-' && isdig(tbuf[1]); if (isdig(tbuf[neg])) { v = 0; k = neg; while (tbuf[k]) { v = v * 10 + tbuf[k] - '0'; k++; } if (neg) v = 0 - v; R_SEQ[seqtop] = v + v + 1; seqtop++; } else if (n == 1 && (c = mnemidx(tbuf[0])) >= 0) { R_SEQ[seqtop] = c + c; seqtop++; } else { R_SEQ[seqtop] = SEQREF; R_SEQ[seqtop + 1] = intern(tbuf); seqtop = seqtop + 2; } } return start; } fn mnemidx(c) { var i = 0; while (mnem[i]) { if (mnem[i] == c) return i; i++; } return -1; } fn emitseq(a, n) { var e = a + n, s; while (a < e) { if (R_SEQ[a] == SEQREF) { s = lookup(R_SEQ[a + 1]); if (s < 0) s = fwdfn(R_SEQ[a + 1]); if (skind[s] == K_FN) fnref(s); else lit(sval[s]); a = a + 2; } else { emit(R_SEQ[a]); a++; } } return 0; } fn addkop(p) { var i = 0; while (i < nkop) { if (kops[i] == p) return 0; i++; } kops[nkop] = p; nkop++; return 0; } fn binfind(p) { var i = nbin; while (i) { i--; if (bname[i] == p) return i; } return -1; } fn unfind(p) { var i = nun; while (i) { i--; if (uname[i] == p) return i; } return -1; } fn asgfind(p) { var i = nasg; while (i) { i--; if (aname[i] == p) return i; } return -1; } // add or replace a binary operator. code is a sequence string, or 0 for the // built-in && (prec 2) and || (prec 1) behaviour, chosen by -1 / -2 in prec. fn add_binop(nm, prec, code) { var p = intern(nm), i = binfind(p), a, n; if (i < 0) { i = nbin; nbin++; } bname[i] = p; bprec[i] = prec; if (prec < 0) { bprec[i] = 0 - prec; blen[i] = prec; bseq[i] = 0; addkop(p); return i; } bseq[i] = seqasm(code); blen[i] = seqtop - bseq[i]; addkop(p); // and its compound assignment, x op= y n = strlen(nm); wcopy(tbuf + 1000, nm, n); tbuf[1000 + n] = '='; tbuf[1001 + n] = 0; a = intern(tbuf + 1000); if (prec >= 8 || prec == 5 || prec == 4 || prec == 3) { if (asgfind(a) < 0) { aname[nasg] = a; nasg++; } abin[asgfind(a)] = i; addkop(a); } return i; } fn add_unop(nm, code) { var p = intern(nm), i = unfind(p); if (i < 0) { i = nun; nun++; } uname[i] = p; useq[i] = seqasm(code); ulen[i] = seqtop - useq[i]; addkop(p); return i; } fn add_kw(nm, handler, flags) { var s = addsym(intern(nm), K_KW, handler); sflag[s] = F_DEF | flags; return s; } // ---- locals and the analyzer's view of them ---- fn findloc(n) { var i = nloc; while (i > lbase) { i--; if (lname[i] == n) return i; } return -1; } fn newloc(n) { var i = nloc; if (nloc >= NL) die("too many locals"); nloc++; lname[i] = n; lslot[i] = fslots; lhome[i] = fdepth; lstate[i] = S_PLAIN; lowner[i] = 0; lused[i] = 0; larr[i] = 0; lown[i] = 0; return i; } fn lroot(i) { while (lstate[i] == S_CHILD) i = lowner[i]; return i; } fn tracked(i) { return lstate[i] != S_PLAIN; } // a use that doesn't give the object away fn touch(i) { var r; lused[i] = 1; if (!fchk) return 0; r = lroot(i); lused[r] = 1; if (lstate[r] == S_DEAD) { cerrn(i, "used after it was given away"); return 0; } // a d object that shows up in a deeper block is given to that block if (lstate[r] == S_D && lhome[r] < fdepth) { lhome[r] = fdepth; lstate[r] = S_M; lown[r] = 1; } return 0; } fn plainuse(c) { if (c == C_FRESH) cerr("a new object can't be used as a plain value; bind it to a name first"); if (c >= C_VAR) touch(c - C_VAR); return C_PLAIN; } fn freevar(i) { laddr(lslot[i]); op(O_LD); callname(s_free); op(O_DROP); return 0; } // leaving every block deeper than or equal to dep: free what is owned, and // complain about what should have left already. final means the block really // ends here rather than an early exit. fn exit_vars(dep, final) { var i = nloc, st; if (!fchk || !live) return 0; while (i > lbase) { i--; if (lhome[i] < dep) continue; st = lstate[i]; if (st == S_M) freevar(i); else if (st == S_U) cerrn(i, "u object must leave this block (block value, assignment to the enclosing block, or return)"); else if (st == S_D) cerrn(i, "d object was never given away"); } return 0; } // an m object has to be used somewhere in the block that owns it, on any path fn unused_check(from) { while (from < nloc) { if (fchk && lown[from] && !lused[from]) dien(lname[from], "m object is never used"); from++; } return 0; } fn block_close(save) { var i = lbase; unused_check(save); exit_vars(fdepth, 1); while (i < save) { if (lhome[i] >= fdepth) { lstate[i] = S_DEAD; lhome[i] = fdepth - 1; } i++; } nloc = save; return 0; } // snapshots: [lbase] [nloc] then state, owner, home per local fn snap_push() { var b = snaptop, i = lbase, p; p = R_SNAP + b; p[0] = lbase; p[1] = nloc; p = p + 2; while (i < nloc) { p[0] = lstate[i]; p[1] = lowner[i]; p[2] = lhome[i]; p = p + 3; i++; } snaptop = p - R_SNAP; if (snaptop >= 8000) die("nesting too deep for the analyzer"); return b; } fn snap_pop(b) { snaptop = b; return 0; } fn snap_restore(b) { var p = R_SNAP + b, i = p[0], e = p[1]; if (e > nloc) e = nloc; p = p + 2; while (i < e) { lstate[i] = p[0]; lowner[i] = p[1]; lhome[i] = p[2]; p = p + 3; i++; } return 0; } fn empty(s) { return s == S_PLAIN || s == S_DEAD; } // compare the current states with a snapshot: -1 if they agree, else the // first local that differs. plain and dead agree and merge to dead. fn snap_cmp(b) { var p = R_SNAP + b, i = p[0], e = p[1], s; if (e > nloc) e = nloc; p = p + 2; while (i < e) { s = lstate[i]; if (empty(s) && empty(p[0])) { if (p[0] == S_DEAD) lstate[i] = S_DEAD; } else if (s != p[0] || lhome[i] != p[2] || (s == S_CHILD && lowner[i] != p[1])) return i; p = p + 3; i++; } return -1; } fn statename(s) { if (s == S_PLAIN) return "empty"; if (s == S_M) return "m"; if (s == S_U) return "u"; if (s == S_D) return "d"; if (s == S_DEAD) return "given away"; if (s == S_CHILD) return "owned by a container"; return "borrowed"; } fn merge_err(i, sa, what) { if (!fchk) return 0; puts("zc: line "); putn(tline); puts(": "); puts(lname[i]); puts(": "); puts(what); puts(" (one path leaves it "); puts(statename(sa)); puts(", another "); puts(statename(lstate[i])); puts(")\n"); halt(); return 0; } // after an if: the current state is the else path, sa is the then path fn merge(sa, livea) { var i, p; if (!livea) return 0; if (!live) { snap_restore(sa); live = 1; return 0; } i = snap_cmp(sa); if (i >= 0) { p = R_SNAP + sa + 2 + 3 * (i - R_SNAP[sa]); merge_err(i, p[0], "ends the branches of an if in different states"); } return 0; } fn loopcheck(s0) { var i, p; if (!fchk || !live) return 0; i = snap_cmp(s0); if (i >= 0) { p = R_SNAP + s0 + 2 + 3 * (i - R_SNAP[s0]); merge_err(i, p[0], "changes state inside a loop; each pass has to leave it as it found it"); } return 0; } // ---- analyzer rules for the places values flow ---- // recast: x as u / x as d / x as m fn recast(c, t) { var i, st; if (!fchk) return c; if (c < C_VAR) { cerr("only a named owned object can be recast"); return c; } i = c - C_VAR; st = lstate[i]; lused[i] = 1; if (st == S_CHILD) cerrn(i, "belongs to a container; recast the container instead"); else if (st == S_BORROW) cerrn(i, "is borrowed and can't be recast"); else if (st == S_PLAIN) cerrn(i, "isn't an owned object"); else if (st == S_DEAD) cerrn(i, "used after it was given away"); else if (t == S_M && st != S_M) cerrn(i, "can't be recast back to m"); else lstate[i] = t; return c; } // an argument for a parameter tagged t (0 plain, 1 m, 2 d) fn argcheck(c, t) { var i, st; if (!fchk) return 0; if (c == C_FRESH) { if (t != 2) cerr("a new object can only go straight to a d parameter; bind it to a name first"); return 0; } if (c < C_VAR || !tracked(c - C_VAR)) return 0; i = c - C_VAR; st = lstate[i]; if (lstate[lroot(i)] == S_DEAD) { cerrn(i, "used after it was given away"); return 0; } if (t == 0) { cerrn(i, "is an owned object; it can only be passed to an m or d parameter"); return 0; } if (t == 1) { if (st == S_D) { cerrn(i, "is d; pass it to a d parameter (lending it would lose it)"); return 0; } touch(i); return 0; } if (st == S_D) { lstate[i] = S_DEAD; lused[i] = 1; return 0; } if (st == S_M) cerrn(i, "is m; write 'as d' to give it away"); else if (st == S_U) cerrn(i, "is u; write 'as d' to give it away instead of returning it"); else if (st == S_CHILD) cerrn(i, "belongs to a container and can't be given away alone"); else cerrn(i, "is borrowed and can't be given away"); return 0; } fn emitown(p) { laddr(lslot[p]); op(O_LD); callname(s_own); return 0; } // a store of class rc into an lvalue of kind k (local or container index i) fn assigncheck(k, i, rc) { var st, j, r, js; if (!fchk) return C_PLAIN; j = -1; if (rc >= C_VAR && tracked(rc - C_VAR)) j = rc - C_VAR; if (j >= 0 && lstate[lroot(j)] == S_DEAD) { cerrn(j, "used after it was given away"); return C_PLAIN; } if (k == LV_LOCAL) { st = lstate[i]; if (j < 0 && rc != C_FRESH) { if (!empty(st)) cerrn(i, "still owns an object; overwriting it would leak it"); else lstate[i] = S_PLAIN; return C_PLAIN; } if (!empty(st)) { cerrn(i, "still owns an object; overwriting it would leak it"); return C_PLAIN; } if (rc == C_FRESH) { lstate[i] = S_M; lused[i] = 0; lown[i] = 1; return C_PLAIN; } js = lstate[j]; if (js == S_U && lhome[i] == lhome[j] - 1) { // handing a u object up one block lstate[i] = S_M; lused[i] = 0; lown[i] = 1; lstate[j] = S_DEAD; lused[j] = 1; return C_PLAIN; } if (js == S_U) cerrn(j, "is u; it can only be handed to a variable of the enclosing block, or returned"); else cerrn(j, "is an owned object and can't be copied into another variable"); return C_PLAIN; } if (k == LV_ELEM) { r = lroot(i); st = lstate[r]; touch(i); if (rc == C_FRESH) { emitown(i); return C_PLAIN; } if (j < 0) return C_PLAIN; js = lstate[j]; if (js == S_CHILD) cerrn(j, "already belongs to a container"); else if (js == S_BORROW) cerrn(j, "is borrowed and can't be stored"); else if (st == S_BORROW) { if (js != S_D) cerrn(j, "must be d to be stored into a borrowed object"); else { lstate[j] = S_DEAD; lused[j] = 1; emitown(i); } } else if (js != st) cerrn(j, "its tag doesn't match its container's (both have to be m, u or d)"); else if (lhome[j] < lhome[r]) cerrn(j, "lives in an outer block than the container it's stored into"); else { lstate[j] = S_CHILD; lowner[j] = i; lused[j] = 1; emitown(i); } return C_PLAIN; } // a global, or memory the analyzer doesn't track: the collector takes it if (j >= 0) { if (lstate[j] == S_D) { lstate[j] = S_DEAD; lused[j] = 1; } else cerrn(j, "must be d to be stored in a global or in untracked memory (the collector takes it over)"); } return C_PLAIN; } fn retcheck(c) { var i; if (!fchk) return 0; i = -1; if (c >= C_VAR && tracked(c - C_VAR)) i = c - C_VAR; if (i >= 0 && lstate[lroot(i)] == S_DEAD) { cerrn(i, "used after it was given away"); return 0; } if (fret) { if (i >= 0) { if (lstate[i] == S_U) { lstate[i] = S_DEAD; lused[i] = 1; } else cerrn(i, "has to be u to be returned; write 'return x as u'"); } return 0; } if (c == C_FRESH) cerr("returns a new object; declare the function with -> u"); else if (i >= 0) cerrn(i, "is owned and can't be returned from a function without -> u"); return 0; } // the last expression of a block expression fn handup(c) { var i; if (!fchk) return C_PLAIN; if (c == C_FRESH) return C_FRESH; if (c < C_VAR || !tracked(c - C_VAR)) return C_PLAIN; i = c - C_VAR; if (lstate[i] == S_U && lhome[i] == fdepth) { lstate[i] = S_DEAD; lused[i] = 1; return C_FRESH; } if (lstate[i] == S_U) cerrn(i, "only a u object of this block can be its value"); else cerrn(i, "has to be u to be handed up as the block's value"); return C_PLAIN; } fn discard(c) { if (c == C_FRESH) cerr("a new object (or a u result) can't be thrown away"); else if (c >= C_VAR) touch(c - C_VAR); return 0; } // ---- expressions ---- fn rval() { if (lv) { op(O_LD); lv = 0; } return 0; } fn kwcall(s, e) { var f = sval[s], old = kw_expr, c; kw_expr = e; xcls = C_PLAIN; f(); kw_expr = old; c = xcls; lv = 0; lvk = LV_NONE; return c; } fn stmtonly() { if (kw_expr) die("this can't be used inside an expression"); return 0; } fn call(s) { var np = 0, tags = 0, ret = 0, c; if (sflag[s] & F_SIG) { tags = ssig[s] >> 6; ret = ssig[s] & 1; } else sflag[s] = sflag[s] | F_UNTYPED; expect(s_lp); if (!is_op(s_rp)) { while (1) { c = expr(); rval(); argcheck(c, (tags >> (np + np)) & 3); np++; if (!accept(s_comma)) break; } } expect(s_rp); if ((sflag[s] & F_SIG) && np != ((ssig[s] >> 1) & 31)) dien(sname[s], "wrong number of arguments"); fnref(s); op(O_CALL); lv = 0; lvk = LV_NONE; if (ret) return C_FRESH; return C_PLAIN; } fn primary() { var n, i, s, c; lv = 0; lvk = LV_NONE; if (tk == T_NUM) { lit(tv); next(); return C_PLAIN; } if (tk == T_STR) { strlit(tv); next(); return C_PLAIN; } if (tk == T_OP) { if (tv == s_lp) { next(); c = expr(); expect(s_rp); return c; } if (tv == s_lb) return blockexpr(); die("expected an expression"); } if (tk != T_ID) die("expected an expression"); n = tv; i = findloc(n); if (i >= 0) { next(); if (!fchk && tracked(i)) dien(n, "is tracked by the analyzer and can't be used in unchecked code"); laddr(lslot[i]); if (larr[i]) return C_PLAIN; lv = 1; lvk = LV_LOCAL; lvi = i; if (tracked(i)) return C_VAR + i; return C_PLAIN; } s = lookup(n); if (s >= 0 && skind[s] == K_KW) { if (!(sflag[s] & F_EXPR)) dien(n, "can't be used inside an expression"); return kwcall(s, 1); } next(); if (s >= 0 && skind[s] == K_CONST) { lit(sval[s]); return C_PLAIN; } if (s >= 0 && skind[s] == K_GVAR) { lit(sval[s]); lv = 1; lvk = LV_GLOBAL; return C_PLAIN; } if (s >= 0 && skind[s] == K_GARR) { lit(sval[s]); return C_PLAIN; } if (s < 0) s = fwdfn(n); if (is_op(s_lp)) return call(s); fnref(s); return C_PLAIN; } // ++ and -- on the address at the top of the stack fn incdec(post, d) { if (!lv) die("++ and -- need something to assign to"); if (lvk == LV_LOCAL && fchk && tracked(lvi)) cerrn(lvi, "owns an object and can't be incremented"); lv = 0; lvk = LV_NONE; lit(0); op(O_PICK); op(O_LD); if (post) { lit(0); op(O_PICK); lit(d); op(O_ADD); lit(2); op(O_PICK); op(O_ST); } else { lit(d); op(O_ADD); lit(0); op(O_PICK); lit(2); op(O_PICK); op(O_ST); } op(O_SWAP); op(O_DROP); return C_PLAIN; } fn postfix() { var c = primary(), b, n, t; while (1) { if (tk == T_ID && tv == s_as) { next(); if (tk != T_ID) die("expected u, d or m after as"); t = S_M; if (tv == s_u) t = S_U; else if (tv == s_d) t = S_D; else if (tv != s_m) die("expected u, d or m after as"); next(); rval(); c = recast(c, t); lvk = LV_NONE; continue; } if (tk != T_OP) return c; if (tv == s_lk) { next(); rval(); b = -1; if (c == C_FRESH) cerr("index a new object only after binding it to a name"); else if (c >= C_VAR) { touch(c - C_VAR); b = c - C_VAR; } c = expr(); rval(); plainuse(c); expect(s_rk); op(O_ADD); lv = 1; if (b >= 0) { lvk = LV_ELEM; lvi = b; } else lvk = LV_OTHER; c = C_PLAIN; continue; } if (tv == s_lp) { // call through a value rval(); plainuse(c); next(); n = 0; if (!is_op(s_rp)) { while (1) { c = expr(); rval(); argcheck(c, 0); n++; if (!accept(s_comma)) break; } } expect(s_rp); lit(n); op(O_PICK); op(O_CALL); op(O_SWAP); op(O_DROP); lv = 0; lvk = LV_NONE; c = C_PLAIN; continue; } if (tv == s_inc) { next(); c = incdec(1, 1); continue; } if (tv == s_dec) { next(); c = incdec(1, -1); continue; } return c; } return c; } fn unary() { var c, u; if (tk == T_OP) { if (tv == s_star) { next(); c = unary(); rval(); plainuse(c); lv = 1; lvk = LV_OTHER; return C_PLAIN; } if (tv == s_amp) { next(); c = unary(); if (!lv) die("& needs something addressable"); if (c >= C_VAR) touch(c - C_VAR); lv = 0; lvk = LV_NONE; return C_PLAIN; } if (tv == s_inc) { next(); unary(); return incdec(0, 1); } if (tv == s_dec) { next(); unary(); return incdec(0, -1); } u = unfind(tv); if (u >= 0) { next(); c = unary(); rval(); plainuse(c); emitseq(useq[u], ulen[u]); lv = 0; lvk = LV_NONE; return C_PLAIN; } } return postfix(); } fn bin(minp) { var c = unary(), b, p, a, j; while (tk == T_OP) { b = binfind(tv); if (b < 0) return c; p = bprec[b]; if (p < minp) return c; next(); rval(); plainuse(c); if (blen[b] == -1) { // && : 0 if the left side is 0, else the right side as 0 or 1 a = hole(); op(O_JZ); c = bin(p + 1); rval(); plainuse(c); emitseq(useq[1], ulen[1]); // !! normalizes j = hole(); op(O_JMP); setlit(a, here); lit(0); setlit(j, here); } else if (blen[b] == -2) { // || : 1 if the left side isn't 0, else the right side as 0 or 1 a = hole(); op(O_JZ); lit(1); j = hole(); op(O_JMP); setlit(a, here); c = bin(p + 1); rval(); plainuse(c); emitseq(useq[1], ulen[1]); setlit(j, here); } else { c = bin(p + 1); rval(); plainuse(c); emitseq(bseq[b], blen[b]); } lv = 0; lvk = LV_NONE; c = C_PLAIN; } return c; } // [addr v] -> [v], storing v at addr fn storekeep() { op(O_SWAP); lit(1); op(O_PICK); op(O_SWAP); op(O_ST); return 0; } fn expr() { var c = bin(1), k, i, a, rc; if (tk != T_OP) return c; if (tv == s_eq) { if (!lv) die("this can't be assigned to"); k = lvk; i = lvi; next(); rc = expr(); rval(); storekeep(); lv = 0; lvk = LV_NONE; return assigncheck(k, i, rc); } a = asgfind(tv); if (a >= 0) { if (!lv) die("this can't be assigned to"); if (lvk == LV_LOCAL && fchk && tracked(lvi)) cerrn(lvi, "owns an object; compound assignment would lose it"); next(); lit(0); op(O_PICK); op(O_LD); lv = 0; rc = expr(); rval(); plainuse(rc); emitseq(bseq[abin[a]], blen[abin[a]]); storekeep(); lv = 0; lvk = LV_NONE; return C_PLAIN; } return c; } // does the current token start a statement rather than an expression? fn stmtstart() { var s; if (tk == T_OP) return tv == s_semi || tv == s_lb; if (tk != T_ID || findloc(tv) >= 0) return 0; s = lookup(tv); return s >= 0 && skind[s] == K_KW && !(sflag[s] & F_EXPR); } // { stmts; tail } as an expression: its value is the tail, or 0 fn blockexpr() { var save = nloc, c = C_PLAIN, tail = 0; next(); fdepth++; bxdepth++; while (!is_op(s_rb)) { if (tk == T_EOF) die("missing }"); if (stmtstart()) { stmt(); continue; } c = expr(); rval(); if (accept(s_semi)) { op(O_DROP); discard(c); c = C_PLAIN; continue; } if (!is_op(s_rb)) die("expected ; or } after an expression"); tail = 1; } next(); if (tail) c = handup(c); else lit(0); block_close(save); fdepth--; bxdepth--; lv = 0; lvk = LV_NONE; return c; } // ---- constant expressions ---- fn cprim() { var v, s; if (tk == T_NUM) { v = tv; next(); return v; } if (accept(s_minus)) return 0 - cprim(); if (tk == T_OP && tv == uname[2]) { next(); return ~cprim(); } if (accept(s_lp)) { v = cexpr(); expect(s_rp); return v; } if (tk == T_ID) { s = lookup(tv); if (s >= 0 && (skind[s] == K_CONST || skind[s] == K_GARR || (skind[s] == K_FN && (sflag[s] & F_DEF)) || skind[s] == K_GVAR)) { next(); return sval[s]; } } die("expected a constant"); return 0; } fn cexpr() { var v = cprim(), o; while (tk == T_OP) { o = tv; if (o == bname[3]) { next(); v = v + cprim(); } else if (o == bname[4]) { next(); v = v - cprim(); } else if (o == bname[0]) { next(); v = v * cprim(); } else if (o == bname[15]) { next(); v = v | cprim(); } else if (o == bname[13]) { next(); v = v & cprim(); } else if (o == bname[5]) { next(); v = v << cprim(); } else if (o == bname[6]) { next(); v = v >> cprim(); } else return v; } return v; } // ---- statements ---- fn stmt() { var s, c; if (tk == T_ID && findloc(tv) < 0) { s = lookup(tv); if (s >= 0 && skind[s] == K_KW && (sflag[s] & F_STMT)) { kwcall(s, 0); return 0; } } if (tk == T_OP) { if (tv == s_semi) { next(); return 0; } if (tv == s_lb) { block(); return 0; } } c = expr(); rval(); op(O_DROP); expect(s_semi); discard(c); return 0; } fn block() { var save = nloc; expect(s_lb); fdepth++; while (!is_op(s_rb)) { if (tk == T_EOF) die("missing }"); stmt(); } next(); block_close(save); fdepth--; return 0; } // the body of if/while/for: its own block, even without braces fn substmt() { var save = nloc; if (is_op(s_lb)) return block(); fdepth++; stmt(); block_close(save); fdepth--; return 0; } fn kw_var() { var n, i, c, k; stmtonly(); next(); while (1) { n = name(); i = newloc(n); if (accept(s_lk)) { k = cexpr(); expect(s_rk); larr[i] = 1; fslots = fslots + k; } else fslots++; if (accept(s_eq)) { c = expr(); rval(); laddr(lslot[i]); op(O_ST); assigncheck(LV_LOCAL, i, c); } if (!accept(s_comma)) break; } expect(s_semi); return 0; } fn kw_if() { var a, b, c, s0, sa, live0, livea; stmtonly(); next(); expect(s_lp); c = expr(); rval(); plainuse(c); expect(s_rp); a = hole(); op(O_JZ); live0 = live; s0 = snap_push(); substmt(); livea = live; sa = snap_push(); if (is_id(s_else)) { next(); b = hole(); op(O_JMP); setlit(a, here); snap_restore(s0); live = live0; substmt(); setlit(b, here); } else { setlit(a, here); snap_restore(s0); live = live0; } merge(sa, livea); snap_pop(s0); return 0; } // loop frame: 0 continue target 1 break chain 2 depth 3 snapshot 4 block-expression depth fn loop_push(cont, s0) { var f = R_LOOP + lsp * 6; if (lsp >= 60) die("loops nested too deeply"); f[0] = cont; f[1] = 0; f[2] = fdepth; f[3] = s0; f[4] = bxdepth; lsp++; return 0; } fn loop_pop() { lsp--; fixall(R_LOOP[lsp * 6 + 1], here); return 0; } fn kw_while() { var c0 = here, a, s0, c; stmtonly(); next(); s0 = snap_push(); expect(s_lp); c = expr(); rval(); plainuse(c); expect(s_rp); a = hole(); op(O_JZ); loop_push(c0, s0); substmt(); loopcheck(s0); jmp(c0); setlit(a, here); loop_pop(); snap_restore(s0); live = 1; snap_pop(s0); return 0; } fn kw_for() { var save = nloc, s0, c0, a, b, d, c; stmtonly(); next(); expect(s_lp); fdepth++; if (is_id(s_var)) kw_var(); else { if (!is_op(s_semi)) { c = expr(); rval(); op(O_DROP); discard(c); } expect(s_semi); } s0 = snap_push(); c0 = here; if (!is_op(s_semi)) { c = expr(); rval(); plainuse(c); } else lit(1); expect(s_semi); a = hole(); op(O_JZ); b = hole(); op(O_JMP); d = here; if (!is_op(s_rp)) { c = expr(); rval(); op(O_DROP); discard(c); } expect(s_rp); jmp(c0); setlit(b, here); loop_push(d, s0); substmt(); loopcheck(s0); jmp(d); setlit(a, here); loop_pop(); snap_restore(s0); live = 1; snap_pop(s0); block_close(save); fdepth--; return 0; } fn loopexit(f) { if (lsp <= loopbase) die("break or continue outside a loop"); if (f[4] != bxdepth) die("break and continue can't leave a block expression"); loopcheck(f[3]); exit_vars(f[2] + 1, 0); return 0; } fn kw_break() { var f = R_LOOP + (lsp - 1) * 6; stmtonly(); next(); expect(s_semi); loopexit(f); f[1] = fixnew(here, f[1]); lit(MW - 1); op(O_JMP); live = 0; return 0; } fn kw_continue() { var f = R_LOOP + (lsp - 1) * 6; stmtonly(); next(); expect(s_semi); loopexit(f); jmp(f[0]); live = 0; return 0; } fn kw_return() { var c; stmtonly(); next(); if (!infn) die("return outside a function"); if (bxdepth) die("return can't leave a block expression"); if (is_op(s_semi)) { lit(0); c = C_PLAIN; } else { c = expr(); rval(); } expect(s_semi); retcheck(c); exit_vars(1, 0); epichain = fixnew(here, epichain); lit(MW - 1); op(O_JMP); live = 0; return 0; } fn kw_unchecked() { var old = fchk; stmtonly(); next(); fchk = 0; substmt(); fchk = old; return 0; } // ---- asm blocks: raw vm words, one per token ---- // numbers push, -n pushes a negative number, #n emits a raw word, one-char // tokens from the mnemonic table are opcodes, names push the address of a // variable or function, :name defines a label for this block. fn asmblock() { var nl = 0, nr = 0, i, k, c, s, ln = R_ASM, la = R_ASM + 64, rn = R_ASM + 128, ra = R_ASM + 384; expect(s_lb); while (!is_op(s_rb)) { if (tk == T_EOF) die("unterminated asm block"); if (tk == T_NUM) { lit(tv); next(); continue; } if (tk == T_STR) { strlit(tv); next(); continue; } if (is_op(s_hash)) { next(); if (tk != T_NUM) die("# needs a number"); emit(tv); next(); continue; } if (is_op(s_minus)) { next(); if (tk == T_NUM && tglue) { lit(0 - tv); next(); } else op(O_SUB); continue; } if (is_op(s_colon)) { next(); if (nl >= 64) die("too many asm labels"); ln[nl] = name(); la[nl] = here; nl++; continue; } if (tv[0] && !tv[1] && (c = mnemidx(tv[0])) >= 0) { op(c); next(); continue; } if (tk != T_ID) die("bad token in asm"); i = findloc(tv); if (i >= 0) { lused[i] = 1; laddr(lslot[i]); next(); continue; } k = -1; i = 0; while (i < nl) { if (ln[i] == tv) k = i; i++; } if (k >= 0) { lit(la[k]); next(); continue; } s = lookup(tv); if (s >= 0 && skind[s] != K_FN && skind[s] != K_KW) { lit(sval[s]); next(); continue; } if (s >= 0 && skind[s] == K_FN) { fnref(s); next(); continue; } if (nr >= 250) die("too many asm label references"); rn[nr] = tv; ra[nr] = hole(); nr++; next(); // a label further down this block } next(); i = 0; while (i < nr) { k = -1; c = 0; while (c < nl) { if (ln[c] == rn[i]) k = c; c++; } if (k < 0) dien(rn[i], "unknown name in asm"); setlit(ra[i], la[k]); i++; } return 0; } fn kw_asm() { next(); asmblock(); xcls = C_PLAIN; return 0; } // ---- quote: code as data ---- // a token list is an object: [n] then kind, value for each token. names, // strings and operators are interned pointers. fn grab() -> u { var n = 0, depth = 0, q; if (!is_op(s_lb)) die("expected {"); next(); while (1) { if (tk == T_EOF) die("unterminated quote"); if (is_op(s_lb)) depth++; if (is_op(s_rb)) { if (!depth) break; depth--; } if (n >= 4000) die("quote too long"); R_QBUF[n + n] = tk; R_QBUF[n + n + 1] = tv; n++; next(); } next(); q = new(n + n + 1); q[0] = n; wcopy(q + 1, R_QBUF, n + n); return q as u; } unchecked fn qnew(src) -> u { var n = src[0] * 2 + 1, q = new(n); wcopy(q, src, n); return q; } fn qdup(m q) -> u { var n = q[0] * 2 + 1, r = new(n); wcopy(r, q, n); return r as u; } fn wcopyto(m q, n) { var i = 0; while (i < n) { emit(q[i]); i++; } return 0; } fn kw_quote() { var q, j, a, n; next(); q = grab(); j = hole(); op(O_JMP); a = here; n = q[0] * 2 + 1; wcopyto(q, n); free(q as d); setlit(j, here); lit(a); callname(s_qnew); xcls = C_FRESH; return 0; } // tokens of a string, as a token list fn tokens(s) -> u { var n = 0, q; src_push(2, s, 0, 0); next(); while (tk != T_EOF) { if (n >= 4000) die("too many tokens"); R_QBUF[n + n] = tk; R_QBUF[n + n + 1] = tv; n++; next(); } src_pop(); q = new(n + n + 1); q[0] = n; wcopy(q + 1, R_QBUF, n + n); return q as u; } // feed a token list into the input: the parser reads it next, then carries // on where it was. the list is freed once it has been read. unchecked fn feed(d q) { src_push(1, q, 1, 1); next(); return 0; } // ---- nested compiles ---- // compile() and load() can run while another function is being compiled (from // a parse word). the function context is saved, the new code is emitted in // place with a jump around it, and then everything is put back. fn ctx_push() { var f = R_CTX + ctxsp * 24; if (ctxsp >= 30) die("compiles nested too deeply"); ctxsp++; f[0] = nloc; f[1] = lbase; f[2] = fdepth; f[3] = fslots; f[4] = fchk; f[5] = fret; f[6] = live; f[7] = bxdepth; f[8] = epichain; f[9] = fframe; f[10] = infn; f[11] = lsp; f[12] = loopbase; f[13] = lv; f[14] = lvk; f[15] = lvi; f[16] = kw_expr; f[17] = xcls; f[18] = snaptop; return 0; } fn ctx_pop() { var f; ctxsp--; f = R_CTX + ctxsp * 24; nloc = f[0]; lbase = f[1]; fdepth = f[2]; fslots = f[3]; fchk = f[4]; fret = f[5]; live = f[6]; bxdepth = f[7]; epichain = f[8]; fframe = f[9]; infn = f[10]; lsp = f[11]; loopbase = f[12]; lv = f[13]; lvk = f[14]; lvi = f[15]; kw_expr = f[16]; xcls = f[17]; snaptop = f[18]; return 0; } // ---- functions and the top level ---- fn fn_begin(chk, ret) { lbase = nloc; fdepth = 1; fslots = 0; fchk = chk && check_on; fret = ret; live = 1; bxdepth = 0; epichain = 0; infn = 1; loopbase = lsp; lv = 0; lvk = LV_NONE; return 0; } fn prologue() { lit(ZFP); op(O_LD); fframe = hole(); op(O_SUB); lit(ZFP); op(O_ST); return 0; } fn epilogue() { lit(0); fixall(epichain, here); lit(ZFP); op(O_LD); lit(fslots); op(O_ADD); lit(ZFP); op(O_ST); op(O_RET); setlit(fframe, fslots); infn = 0; nloc = lbase; return 0; } fn body() { expect(s_lb); while (!is_op(s_rb)) { if (tk == T_EOF) die("missing } at the end of a function"); stmt(); } next(); unused_check(lbase); if (fchk && live) { exit_vars(1, 1); if (fret) cerr("a -> u function has to end with a return"); } return 0; } fn fndef(chk) { var n, s, np = 0, tags = 0, t, ret = 0, sig, i, pn[24], pt[24]; next(); n = name(); s = lookup(n); if (s < 0 || s < ubase || skind[s] != K_FN || (sflag[s] & F_DEF)) { s = addsym(n, K_FN, 0); sflag[s] = 0; } expect(s_lp); if (!is_op(s_rp)) { while (1) { t = 0; if (tk == T_ID && (tv == s_m || tv == s_d)) { t = 1; if (tv == s_d) t = 2; pn[np] = tv; next(); if (tk == T_ID) pn[np] = name(); else t = 0; // a parameter called m or d } else pn[np] = name(); if (np >= 12 && t) die("only the first 12 parameters can be tagged"); pt[np] = t; if (np < 12) tags = tags + (t << (np + np)); np++; if (np >= 24) die("too many parameters"); if (!accept(s_comma)) break; } } expect(s_rp); if (accept(s_arrow)) { if (!is_id(s_u)) die("the only return tag is -> u"); next(); ret = 1; } sig = ret + np + np + tags * 64; if (sflag[s] & F_SIG) { if (ssig[s] != sig) dien(n, "doesn't match its earlier declaration"); } else if ((sflag[s] & F_UNTYPED) && (tags || ret)) dien(n, "was called before its tags were declared; put a prototype above the first call"); ssig[s] = sig; sflag[s] = sflag[s] | F_SIG; if (accept(s_semi)) return s; if (sflag[s] & F_DEF) dien(n, "is defined twice"); sval[s] = here; sflag[s] = sflag[s] | F_DEF; fixall(sfix[s], here); sfix[s] = 0; fn_begin(chk, ret); prologue(); i = 0; while (i < np) { t = newloc(pn[i]); fslots++; if (fchk && pt[i] == 1) lstate[t] = S_BORROW; if (fchk && pt[i] == 2) { lstate[t] = S_M; lown[t] = 1; } i++; } i = np; while (i) { i--; laddr(i); op(O_ST); } body(); epilogue(); return s; } fn gvar() { var n, a, k; next(); while (1) { n = name(); if (accept(s_lk)) { k = cexpr(); expect(s_rk); addsym(n, K_GARR, here); while (k) { emit(0); k--; } } else { a = here; addsym(n, K_GVAR, a); emit(0); if (accept(s_eq)) { if (tk == T_STR) { *(a + hdelta) = here; k = tv; while (*k) { emit(*k); k++; } emit(0); next(); } else *(a + hdelta) = cexpr(); } } if (!accept(s_comma)) break; } expect(s_semi); return 0; } fn gconst() { var n; next(); while (1) { n = name(); expect(s_eq); addsym(n, K_CONST, cexpr()); if (!accept(s_comma)) break; } expect(s_semi); return 0; } fn topitem() { var chk = 1, imm = 0, s, k; if (is_id(s_unchecked)) { next(); if (accept(s_semi)) { check_on = 0; return 0; } // unchecked; switches the analyzer off from here on chk = 0; } if (is_id(s_immediate)) { next(); imm = 1; } if (is_id(s_fn)) { s = fndef(chk); if (imm) { if (hdelta) die("parse words can't be defined while building an image"); k = addsym(sname[s], K_KW, sval[s]); sflag[k] = F_DEF | F_STMT | F_EXPR | F_TOP; } return 0; } if (!chk || imm) die("expected fn"); if (is_id(s_var)) return gvar(); if (is_id(s_const)) return gconst(); if (tk == T_ID) { s = lookup(tv); if (s >= 0 && skind[s] == K_KW && (sflag[s] & F_TOP)) { kwcall(s, 0); return 0; } } die("expected fn, var or const"); return 0; } fn unit_top() { while (tk != T_EOF) topitem(); return 0; } fn unit_end(from) { var i = from; while (i < nsym) { if (skind[i] == K_FN && !(sflag[i] & F_DEF) && sfix[i]) dien(sname[i], "is used but never defined"); i++; } return 0; } // compile a token list as the body of a new function, or (top) as top-level // declarations. returns the entry address of the function. unchecked fn compile_list(m q, top) { var hop = -1, entry = 0, from = nsym; ctx_push(); if (infn) { hop = hole(); op(O_JMP); } src_push(1, q, 0, 0); next(); if (top) unit_top(); else { entry = here; fn_begin(1, 0); prologue(); while (tk != T_EOF) stmt(); unused_check(lbase); if (fchk && live) exit_vars(1, 1); epilogue(); } src_pop(); unit_end(from); if (hop >= 0) setlit(hop, here); if (here > ptop) ptop = here; ctx_pop(); return entry; } fn compile(m q) { return compile_list(q, 0); } fn load(m q) { compile_list(q, 1); return 0; } fn eval(m q) { var f = compile_list(q, 0); return f(); } fn evals(s) { var q = tokens(s), r = eval(q); free(q as d); return r; } // ---- symbol blob ---- // an image ends with its own symbol table so the genome knows its names: // kind, value, signature, name chars, 0 ... then a 0 kind. fn blob_emit(from) { var i = from, p; while (i < nsym) { if (skind[i] >= K_FN && skind[i] <= K_CONST) { emit(skind[i]); emit(sval[i]); emit(ssig[i]); p = sname[i]; while (*p) { emit(*p); p++; } emit(0); } i++; } emit(0); return 0; } fn blob_load(p) { var s, n; while (*p) { n = intern(p + 3); s = addsym(n, p[0], p[1]); ssig[s] = p[2]; sflag[s] = F_DEF | F_SIG; p = p + 3 + strlen(p + 3) + 1; } return 0; } // compile the rest of stdin as a genome image that will run at org, writing // it to dst. only the source's own names are visible while it compiles. // returns the image length in words. fn build_image(org, dst) { var sh = here, sd = hdelta, sl = hlimit, sub = ubase, svl = vislo, sns = nsym, n; ubase = nsym; vislo = nsym; here = org; hdelta = dst - org; hlimit = org + 0x30000; lit(FTOP); lit(ZFP); op(O_ST); fnref(fwdfn(s_start)); op(O_JMP); emit(0); emit(0); src_push(0, 0, 0, 0); next(); unit_top(); unit_end(ubase); src_pop(); *(dst + 5) = here; blob_emit(ubase); *(dst + 6) = here; n = here - org; popsyms(sns); ubase = sub; vislo = svl; here = sh; hdelta = sd; hlimit = sl; return n; } // ---- setup ---- fn gc_roots() { markrange(0, ptop); if (!hdelta && here > ptop) markrange(ptop, here); // globals of a unit still compiling markrange(GORG, GORG[6]); markrange(R_SRC, R_SRC + 400); markrange(*ZFP, FTOP); markrange(DSLO, DSHI); return 0; } fn zc_init() { pooltop = R_POOL; wfill(R_IH, 0, IHN); nsym = 0; nfixu = 0; fixfree = 0; nbin = 0; nun = 0; nasg = 0; nkop = 0; seqtop = 0; lsp = 0; ctxsp = 0; snaptop = 0; nloc = 0; lbase = 0; s_lp = intern("("); s_rp = intern(")"); s_lb = intern("{"); s_rb = intern("}"); s_lk = intern("["); s_rk = intern("]"); s_comma = intern(","); s_semi = intern(";"); s_eq = intern("="); s_arrow = intern("->"); s_inc = intern("++"); s_dec = intern("--"); s_star = intern("*"); s_amp = intern("&"); s_colon = intern(":"); s_hash = intern("#"); s_minus = intern("-"); s_andand = intern("&&"); s_oror = intern("||"); s_fn = intern("fn"); s_var = intern("var"); s_const = intern("const"); s_else = intern("else"); s_as = intern("as"); s_unchecked = intern("unchecked"); s_immediate = intern("immediate"); s_m = intern("m"); s_d = intern("d"); s_u = intern("u"); s_start = intern("start"); s_main = intern("main"); s_free = intern("free"); s_own = intern("own"); s_qnew = intern("qnew"); return 0; } // operators and keywords. these are plain tables in memory: a program can // add operators, change precedences or replace keywords with its own. fn zc_tables() { add_unop("!", "0 $ 0 ~ - | 31 > 1 +"); // index 0 add_unop("!!", "0 $ 0 ~ - | 31 > 0 ~ -"); // index 1: normalize to 0 or 1 add_unop("~", "-1 ^"); // index 2 add_unop("-", "0 ~ -"); add_binop("*", 10, "*"); // the first few indexes are used by cexpr add_binop("/", 10, "zdiv c"); add_binop("%", 10, "zmod c"); add_binop("+", 9, "+"); // 3 add_binop("-", 9, "-"); // 4 add_binop("<<", 8, "zshl c"); // 5 add_binop(">>", 8, ">"); // 6 add_binop("<", 7, "<"); add_binop(">", 7, "~ <"); add_binop("<=", 7, "~ < 1 ^"); add_binop(">=", 7, "< 1 ^"); add_binop("==", 6, "- 0 $ 0 ~ - | 31 > 1 +"); add_binop("!=", 6, "- 0 $ 0 ~ - | 31 > 0 ~ -"); add_binop("&", 5, "&"); // 13 add_binop("^", 4, "^"); add_binop("|", 3, "|"); // 15 add_binop("&&", -1, 0); add_binop("||", -2, 0); addkop(s_eq); addkop(s_arrow); addkop(s_inc); addkop(s_dec); add_kw("var", kw_var, F_STMT); add_kw("if", kw_if, F_STMT); add_kw("while", kw_while, F_STMT); add_kw("for", kw_for, F_STMT); add_kw("break", kw_break, F_STMT); add_kw("continue", kw_continue, F_STMT); add_kw("return", kw_return, F_STMT); add_kw("unchecked", kw_unchecked, F_STMT); add_kw("asm", kw_asm, F_STMT | F_EXPR); add_kw("quote", kw_quote, F_EXPR); return 0; } fn start() { var s, f; heap_init(); zc_init(); blob_load(GORG[5]); zc_tables(); here = 0; hdelta = 0; hlimit = HEAP_LO; ptop = 0; ubase = nsym; vislo = 0; src_push(0, 0, 0, 0); next(); unit_top(); unit_end(ubase); if (here > ptop) ptop = here; s = lookup(s_main); if (s >= 0 && skind[s] == K_FN && (sflag[s] & F_DEF)) { f = sval[s]; f(); } halt(); return 0; }