diff --git a/awk.h b/awk.h index 70097b9..ddf2466 100644 --- a/awk.h +++ b/awk.h @@ -97,9 +97,14 @@ extern Array *symtab; extern Cell *nrloc; /* NR */ extern Cell *fnrloc; /* FNR */ +extern Cell *fsloc; /* FS */ extern Cell *nfloc; /* NF */ +extern Cell *ofsloc; /* OFS */ +extern Cell *orsloc; /* ORS */ +extern Cell *rsloc; /* RS */ extern Cell *rstartloc; /* RSTART */ extern Cell *rlengthloc; /* RLENGTH */ +extern Cell *subseploc; /* SUBSEP */ /* Cell.tval values: */ #define NUM 01 /* number value is valid */ diff --git a/b.c b/b.c index 89a7841..a54a234 100644 --- a/b.c +++ b/b.c @@ -823,7 +823,15 @@ int relex(void) /* lexical analyzer for reparse */ if (cc->cc_name != NULL && prestr[1 + cc->cc_namelen] == ':' && prestr[2 + cc->cc_namelen] == ']') { prestr += cc->cc_namelen + 3; - for (i = 0; i < NCHARS; i++) { + /* + * BUG: We begin at 1, instead of 0, since we + * would otherwise prematurely terminate the + * string for classes like [[:cntrl:]]. This + * means that we can't match the NUL character, + * not without first adapting the entire + * program to track each string's length. + */ + for (i = 1; i < NCHARS; i++) { if (!adjbuf((char **) &buf, &bufsz, bp-buf+1, 100, (char **) &bp, "relex2")) FATAL("out of space for reg expr %.10s...", lastre); if (cc->cc_func(i)) { diff --git a/bugs-fixed/README b/bugs-fixed/README index 87dafe3..9c644f9 100644 --- a/bugs-fixed/README +++ b/bugs-fixed/README @@ -24,14 +24,31 @@ and also if CONVFMT changed. 7. unary-plus: Unary plus on a string constant returned the string. Instead, it should convert the value to numeric and give that value. -8. missing-precision: When using the format string "%*s", the precision +8. concat-assign-same: Concatenation previously evaluated both sides of the +expression before doing its work, which, since assign() evaluates to the cell +being assigned to, meant that expressions like "print (a = 1) (a = 2)" would +print "22" rather than "12". + +9. missing-precision: When using the format string "%*s", the precision argument was used without checking if it was present first. -9. fmt-overflow: The buffer used for OFMT/CONVFMT conversions was written +10. missing-precision: When using the format string "%*s", the precision +argument was used without checking if it was present first. + +11. fmt-overflow: The buffer used for OFMT/CONVFMT conversions was written to with sprintf(), which meant that some conversions could write past the end. -X. nf-self-assign: "NF = NF" wouldn't force the record to be rebuilt. +12. numeric-subsep, numeric-fs, numeric-output-seps, numerics-rs: If SUBSEP, +FS, RS, OFS, or ORS were set to a numeric value, then their string values +wouldn't always be generated before being needed. + +13. subsep-overflow: The length of SUBSEP needs to be rechecked after +calling execute(), in case SUBSEP itself has been changed. + +14. split-fs-from-array: If the third argument to split() comes from the +array passed as the second argument, then split() would previously read +from the freed memory and possibly produce incorrect results (depending +on the system's malloc()/free() behaviour.) -X. negative-nf: Setting NF to a negative value caused a segmentation fault. diff --git a/bugs-fixed/concat-assign-same.awk b/bugs-fixed/concat-assign-same.awk new file mode 100644 index 0000000..ed19f35 --- /dev/null +++ b/bugs-fixed/concat-assign-same.awk @@ -0,0 +1,4 @@ +BEGIN { + print (a = 1) (a = 2) (a = 3) (a = 4) (a = 5); + print (a = 1), (a = 2), (a = 3), (a = 4), (a = 5); +} diff --git a/bugs-fixed/concat-assign-same.bad b/bugs-fixed/concat-assign-same.bad new file mode 100644 index 0000000..294725b --- /dev/null +++ b/bugs-fixed/concat-assign-same.bad @@ -0,0 +1,2 @@ +22345 +1 2 3 4 5 diff --git a/bugs-fixed/concat-assign-same.ok b/bugs-fixed/concat-assign-same.ok new file mode 100644 index 0000000..4475052 --- /dev/null +++ b/bugs-fixed/concat-assign-same.ok @@ -0,0 +1,2 @@ +12345 +1 2 3 4 5 diff --git a/bugs-fixed/fs-overflow.awk b/bugs-fixed/fs-overflow.awk new file mode 100644 index 0000000..be10f5a --- /dev/null +++ b/bugs-fixed/fs-overflow.awk @@ -0,0 +1,13 @@ +function foo() { + a = ""; + for (i = 0; i < 10000; i++) { + a = a "c"; + } + return a; +} + +BEGIN { + FS = foo(); + $0="foo"; + print $1; +} diff --git a/bugs-fixed/numeric-fs.awk b/bugs-fixed/numeric-fs.awk new file mode 100644 index 0000000..01e438d --- /dev/null +++ b/bugs-fixed/numeric-fs.awk @@ -0,0 +1,5 @@ +BEGIN { + FS = 0; split("20202", a); print a[1]; + FS = 1; $0="31313"; print $1; + FS = 2; "echo 42424" | getline; print $1; +} diff --git a/bugs-fixed/numeric-fs.ok b/bugs-fixed/numeric-fs.ok new file mode 100644 index 0000000..dcf37cd --- /dev/null +++ b/bugs-fixed/numeric-fs.ok @@ -0,0 +1,3 @@ +2 +3 +4 diff --git a/bugs-fixed/numeric-output-seps.awk b/bugs-fixed/numeric-output-seps.awk new file mode 100644 index 0000000..daa0f72 --- /dev/null +++ b/bugs-fixed/numeric-output-seps.awk @@ -0,0 +1,8 @@ +BEGIN { + $0 = "a b c"; + OFS = 1; + ORS = 2; + NF = 2; + print; + print "d", "e"; +} diff --git a/bugs-fixed/numeric-output-seps.bad b/bugs-fixed/numeric-output-seps.bad new file mode 100644 index 0000000..95310f7 --- /dev/null +++ b/bugs-fixed/numeric-output-seps.bad @@ -0,0 +1,2 @@ +a b +d e diff --git a/bugs-fixed/numeric-output-seps.ok b/bugs-fixed/numeric-output-seps.ok new file mode 100644 index 0000000..de6b202 --- /dev/null +++ b/bugs-fixed/numeric-output-seps.ok @@ -0,0 +1 @@ +a1b2d1e2 \ No newline at end of file diff --git a/bugs-fixed/numeric-rs.awk b/bugs-fixed/numeric-rs.awk new file mode 100644 index 0000000..cc7a0a0 --- /dev/null +++ b/bugs-fixed/numeric-rs.awk @@ -0,0 +1,6 @@ +BEGIN { + RS = 1; + while ("echo a1b1c1d" | getline > 0) { + print $1; + } +} diff --git a/bugs-fixed/numeric-rs.bad b/bugs-fixed/numeric-rs.bad new file mode 100644 index 0000000..2027bc6 --- /dev/null +++ b/bugs-fixed/numeric-rs.bad @@ -0,0 +1 @@ +a1b1c1d diff --git a/bugs-fixed/numeric-rs.ok b/bugs-fixed/numeric-rs.ok new file mode 100644 index 0000000..d68dd40 --- /dev/null +++ b/bugs-fixed/numeric-rs.ok @@ -0,0 +1,4 @@ +a +b +c +d diff --git a/bugs-fixed/numeric-subsep.awk b/bugs-fixed/numeric-subsep.awk new file mode 100644 index 0000000..1252e4a --- /dev/null +++ b/bugs-fixed/numeric-subsep.awk @@ -0,0 +1,5 @@ +BEGIN { + SUBSEP = 123.456; + a["hello", "world"] = "foo"; + print a["hello" SUBSEP "world"]; +} diff --git a/bugs-fixed/numeric-subsep.bad b/bugs-fixed/numeric-subsep.bad new file mode 100644 index 0000000..8b13789 --- /dev/null +++ b/bugs-fixed/numeric-subsep.bad @@ -0,0 +1 @@ + diff --git a/bugs-fixed/numeric-subsep.ok b/bugs-fixed/numeric-subsep.ok new file mode 100644 index 0000000..257cc56 --- /dev/null +++ b/bugs-fixed/numeric-subsep.ok @@ -0,0 +1 @@ +foo diff --git a/bugs-fixed/split-fs-from-array.awk b/bugs-fixed/split-fs-from-array.awk new file mode 100644 index 0000000..fce1607 --- /dev/null +++ b/bugs-fixed/split-fs-from-array.awk @@ -0,0 +1,5 @@ +BEGIN { + a[1] = "elephantie" + a[2] = "e" + print split(a[1],a,a[2]), a[2], a[3], split(a[2],a,a[2]) +} diff --git a/bugs-fixed/split-fs-from-array.ok b/bugs-fixed/split-fs-from-array.ok new file mode 100644 index 0000000..9402b94 --- /dev/null +++ b/bugs-fixed/split-fs-from-array.ok @@ -0,0 +1 @@ +4 l phanti 2 diff --git a/bugs-fixed/subsep-overflow.awk b/bugs-fixed/subsep-overflow.awk new file mode 100644 index 0000000..66c7c24 --- /dev/null +++ b/bugs-fixed/subsep-overflow.awk @@ -0,0 +1,24 @@ +function foo(c, n) { + s = ""; + for (i = 0; i < n; i++) { + s = s c; + } + return s; +} + +BEGIN { + str1 = foo("a", 4500); + str2 = foo("b", 9000); + + a[(SUBSEP = str1), (SUBSEP = str2), "c"] = 1; + + for (k in a) { + print length(k); + } + + print (((SUBSEP = str1), (SUBSEP = str2), "c") in a); + print (((SUBSEP = str1) SUBSEP (SUBSEP = str2) SUBSEP "c") in a); + delete a[(SUBSEP = str1), (SUBSEP = str2), "c"]; + print (((SUBSEP = str1), (SUBSEP = str2), "c") in a); + print (((SUBSEP = str1) SUBSEP (SUBSEP = str2) SUBSEP "c") in a); +} diff --git a/bugs-fixed/subsep-overflow.ok b/bugs-fixed/subsep-overflow.ok new file mode 100644 index 0000000..ddbbd78 --- /dev/null +++ b/bugs-fixed/subsep-overflow.ok @@ -0,0 +1,5 @@ +27001 +1 +1 +0 +0 diff --git a/lib.c b/lib.c index 8078db0..f9cb6ca 100644 --- a/lib.c +++ b/lib.c @@ -59,7 +59,7 @@ void recinit(unsigned int n) { if ( (record = (char *) malloc(n)) == NULL || (fields = (char *) malloc(n+1)) == NULL - || (fldtab = (Cell **) malloc((nfields+1) * sizeof(Cell *))) == NULL + || (fldtab = (Cell **) malloc((nfields+2) * sizeof(Cell *))) == NULL || (fldtab[0] = (Cell *) malloc(sizeof(Cell))) == NULL ) FATAL("out of space for $0 and fields"); *fldtab[0] = dollar0; @@ -189,12 +189,13 @@ int readrec(char **pbuf, int *pbufsize, FILE *inf) /* read one record into buf * int sep, c; char *rr, *buf = *pbuf; int bufsize = *pbufsize; + char *rs = getsval(rsloc); - if (strlen(*FS) >= sizeof(inputFS)) + if (strlen(getsval(fsloc)) >= sizeof (inputFS)) FATAL("field separator %.10s... is too long", *FS); /*fflush(stdout); avoids some buffering problem but makes it 25% slower*/ strcpy(inputFS, *FS); /* for subsequent field splitting */ - if ((sep = **RS) == 0) { + if ((sep = *rs) == 0) { sep = '\n'; while ((c=getc(inf)) == '\n' && c != EOF) /* skip leading \n's */ ; @@ -208,7 +209,7 @@ int readrec(char **pbuf, int *pbufsize, FILE *inf) /* read one record into buf * FATAL("input record `%.30s...' too long", buf); *rr++ = c; } - if (**RS == sep || c == EOF) + if (*rs == sep || c == EOF) break; if ((c = getc(inf)) == '\n' || c == EOF) /* 2 in a row */ break; @@ -283,6 +284,8 @@ void fldbld(void) /* create fields from current record */ } fr = fields; i = 0; /* number of fields accumulated here */ + if (strlen(getsval(fsloc)) >= sizeof (inputFS)) + FATAL("field separator %.10s... is too long", *FS); strcpy(inputFS, *FS); if (strlen(inputFS) > 1) { /* it's a regular expression */ i = refldbld(r, inputFS); @@ -481,6 +484,7 @@ void recbld(void) /* create $0 from $1..$NF if necessary */ { int i; char *r, *p; + char *sep = getsval(ofsloc); if (donerec == 1) return; @@ -492,9 +496,9 @@ void recbld(void) /* create $0 from $1..$NF if necessary */ while ((*r = *p++) != 0) r++; if (i < *NF) { - if (!adjbuf(&record, &recsize, 2+strlen(*OFS)+r-record, recsize, &r, "recbld 2")) + if (!adjbuf(&record, &recsize, 2+strlen(sep)+r-record, recsize, &r, "recbld 2")) FATAL("created $0 `%.30s...' too long", record); - for (p = *OFS; (*r = *p++) != 0; ) + for (p = sep; (*r = *p++) != 0; ) r++; } } diff --git a/main.c b/main.c index a183c45..1c38a1e 100644 --- a/main.c +++ b/main.c @@ -88,7 +88,7 @@ int main(int argc, char *argv[]) exit(0); break; } - if (strncmp(argv[1], "--", 2) == 0) { /* explicit end of args */ + if (strcmp(argv[1], "--") == 0) { /* explicit end of args */ argc--; argv++; break; diff --git a/run.c b/run.c index 0704e47..ce30e93 100644 --- a/run.c +++ b/run.c @@ -462,7 +462,7 @@ Cell *array(Node **a, int n) /* a[0] is symtab, a[1] is list of subscripts */ Node *np; char *buf; int bufsz = recsize; - int nsub = strlen(*SUBSEP); + int nsub; if ((buf = (char *) malloc(bufsz)) == NULL) FATAL("out of memory in array"); @@ -472,6 +472,7 @@ Cell *array(Node **a, int n) /* a[0] is symtab, a[1] is list of subscripts */ for (np = a[1]; np; np = np->nnext) { y = execute(np); /* subscript */ s = getsval(y); + nsub = strlen(getsval(subseploc)); if (!adjbuf(&buf, &bufsz, strlen(buf)+strlen(s)+nsub+1, recsize, 0, "array")) FATAL("out of memory for %s[%s...]", x->nval, buf); strcat(buf, s); @@ -500,7 +501,7 @@ Cell *awkdelete(Node **a, int n) /* a[0] is symtab, a[1] is list of subscripts * Cell *x, *y; Node *np; char *s; - int nsub = strlen(*SUBSEP); + int nsub; x = execute(a[0]); /* Cell* for symbol table */ if (!isarr(x)) @@ -519,9 +520,10 @@ Cell *awkdelete(Node **a, int n) /* a[0] is symtab, a[1] is list of subscripts * for (np = a[1]; np; np = np->nnext) { y = execute(np); /* subscript */ s = getsval(y); + nsub = strlen(getsval(subseploc)); if (!adjbuf(&buf, &bufsz, strlen(buf)+strlen(s)+nsub+1, recsize, 0, "awkdelete")) FATAL("out of memory deleting %s[%s...]", x->nval, buf); - strcat(buf, s); + strcat(buf, s); if (np->nnext) strcat(buf, *SUBSEP); tempfree(y); @@ -540,7 +542,7 @@ Cell *intest(Node **a, int n) /* a[0] is index (list), a[1] is symtab */ char *buf; char *s; int bufsz = recsize; - int nsub = strlen(*SUBSEP); + int nsub; ap = execute(a[1]); /* array name */ if (!isarr(ap)) { @@ -558,6 +560,7 @@ Cell *intest(Node **a, int n) /* a[0] is index (list), a[1] is symtab */ for (p = a[0]; p; p = p->nnext) { x = execute(p); /* expr */ s = getsval(x); + nsub = strlen(getsval(subseploc)); if (!adjbuf(&buf, &bufsz, strlen(buf)+strlen(s)+nsub+1, recsize, 0, "intest")) FATAL("out of memory deleting %s[%s...]", x->nval, buf); strcat(buf, s); @@ -1178,25 +1181,26 @@ Cell *cat(Node **a, int q) /* a[0] cat a[1] */ { Cell *x, *y, *z; int n1, n2; - char *s; + char *s = NULL; + int ssz = 0; x = execute(a[0]); + n1 = strlen(getsval(x)); + adjbuf(&s, &ssz, n1 + 1, recsize, 0, "cat1"); + (void) strncpy(s, x->sval, ssz); + y = execute(a[1]); - getsval(x); - getsval(y); - n1 = strlen(x->sval); - n2 = strlen(y->sval); - s = (char *) malloc(n1 + n2 + 1); - if (s == NULL) - FATAL("out of space concatenating %.15s... and %.15s...", - x->sval, y->sval); - strcpy(s, x->sval); - strcpy(s+n1, y->sval); + n2 = strlen(getsval(y)); + adjbuf(&s, &ssz, n1 + n2 + 1, recsize, 0, "cat2"); + (void) strncpy(s + n1, y->sval, ssz - n1); + tempfree(x); tempfree(y); + z = gettemp(); z->sval = s; z->tval = STR; + return(z); } @@ -1243,18 +1247,20 @@ Cell *split(Node **a, int nnn) /* split(a[0], a[1], a[2]); a[3] is type */ { Cell *x = 0, *y, *ap; char *s, *origs; + char *fs, *origfs = NULL; int sep; - char *t, temp, num[50], *fs = 0; + char *t, temp, num[50]; int n, tempstat, arg3type; y = execute(a[0]); /* source string */ origs = s = strdup(getsval(y)); arg3type = ptoi(a[3]); if (a[2] == 0) /* fs string */ - fs = *FS; + fs = getsval(fsloc); else if (arg3type == STRING) { /* split(str,arr,"string") */ x = execute(a[2]); - fs = getsval(x); + origfs = fs = strdup(getsval(x)); + tempfree(x); } else if (arg3type == REGEXPR) fs = "(regexpr)"; /* split(str,arr,/regexpr/) */ else @@ -1369,9 +1375,7 @@ Cell *split(Node **a, int nnn) /* split(a[0], a[1], a[2]); a[3] is type */ tempfree(ap); tempfree(y); free(origs); - if (a[2] != 0 && arg3type == STRING) { - tempfree(x); - } + free(origfs); x = gettemp(); x->tval = NUM; x->fval = n; @@ -1633,9 +1637,9 @@ Cell *printstat(Node **a, int n) /* print a[0] */ fputs(getpssval(y), fp); tempfree(y); if (x->nnext == NULL) - fputs(*ORS, fp); + fputs(getsval(orsloc), fp); else - fputs(*OFS, fp); + fputs(getsval(ofsloc), fp); } if (a[1] != 0) fflush(fp); diff --git a/tran.c b/tran.c index 6775b01..d1dfe2b 100644 --- a/tran.c +++ b/tran.c @@ -55,10 +55,14 @@ Cell *fsloc; /* FS */ Cell *nrloc; /* NR */ Cell *nfloc; /* NF */ Cell *fnrloc; /* FNR */ +Cell *ofsloc; /* OFS */ +Cell *orsloc; /* ORS */ +Cell *rsloc; /* RS */ Array *ARGVtab; /* symbol table containing ARGV[...] */ Array *ENVtab; /* symbol table containing ENVIRON[...] */ Cell *rstartloc; /* RSTART */ Cell *rlengthloc; /* RLENGTH */ +Cell *subseploc; /* SUBSEP */ Cell *symtabloc; /* SYMTAB */ Cell *nullloc; /* a guaranteed empty cell */ @@ -88,9 +92,12 @@ void syminit(void) /* initialize symbol table with builtin vars */ fsloc = setsymtab("FS", " ", 0.0, STR|DONTFREE, symtab); FS = &fsloc->sval; - RS = &setsymtab("RS", "\n", 0.0, STR|DONTFREE, symtab)->sval; - OFS = &setsymtab("OFS", " ", 0.0, STR|DONTFREE, symtab)->sval; - ORS = &setsymtab("ORS", "\n", 0.0, STR|DONTFREE, symtab)->sval; + rsloc = setsymtab("RS", "\n", 0.0, STR|DONTFREE, symtab); + RS = &rsloc->sval; + ofsloc = setsymtab("OFS", " ", 0.0, STR|DONTFREE, symtab); + OFS = &ofsloc->sval; + orsloc = setsymtab("ORS", "\n", 0.0, STR|DONTFREE, symtab); + ORS = &orsloc->sval; OFMT = &setsymtab("OFMT", "%.6g", 0.0, STR|DONTFREE, symtab)->sval; CONVFMT = &setsymtab("CONVFMT", "%.6g", 0.0, STR|DONTFREE, symtab)->sval; FILENAME = &setsymtab("FILENAME", "", 0.0, STR|DONTFREE, symtab)->sval; @@ -100,7 +107,8 @@ void syminit(void) /* initialize symbol table with builtin vars */ NR = &nrloc->fval; fnrloc = setsymtab("FNR", "", 0.0, NUM, symtab); FNR = &fnrloc->fval; - SUBSEP = &setsymtab("SUBSEP", "\034", 0.0, STR|DONTFREE, symtab)->sval; + subseploc = setsymtab("SUBSEP", "\034", 0.0, STR|DONTFREE, symtab); + SUBSEP = &subseploc->sval; rstartloc = setsymtab("RSTART", "", 0.0, NUM, symtab); RSTART = &rstartloc->fval; rlengthloc = setsymtab("RLENGTH", "", 0.0, NUM, symtab); @@ -310,6 +318,9 @@ Awkfloat setfval(Cell *vp, Awkfloat f) /* set float val of a Cell */ } else if (isrec(vp)) { donefld = 0; /* mark $1... invalid */ donerec = 1; + } else if (vp == ofsloc) { + if (donerec == 0) + recbld(); } if (freeable(vp)) xfree(vp->sval); /* free any previous string */ @@ -351,7 +362,7 @@ char *setsval(Cell *vp, const char *s) /* set string val of a Cell */ } else if (isrec(vp)) { donefld = 0; /* mark $1... invalid */ donerec = 1; - } else if (&vp->sval == OFS) { + } else if (vp == ofsloc) { if (donerec == 0) recbld(); }