find:: get rid of nested function (it's a gcc-ism)

[oweals/busybox.git] / editors / awk.c

diff --git a/editors/awk.c b/editors/awk.c
index 71abca215a4bb3b3942f8cad35c8207cc3fee517..0b573a065c18bfd237886ee563f661c5199acc78 100644 (file)
--- a/editors/awk.c
+++ b/editors/awk.c
@@ -155,7 +155,7 @@ typedef struct tsplitter_s {
 
 /* simple token classes */
 /* Order and hex values are very important!!!  See next_token() */
-#define        TC_SEQSTART      1                              /* ( */
+#define        TC_SEQSTART     1                       /* ( */
 #define        TC_SEQTERM      (1 << 1)                /* ) */
 #define        TC_REGEXP       (1 << 2)                /* /.../ */
 #define        TC_OUTRDR       (1 << 3)                /* | > >> */
@@ -696,6 +696,10 @@ static char nextchar(char **s)
        pps = *s;
        if (c == '\\')
                c = bb_process_escape_sequence((const char**)s);
+       /* Example awk statement:
+        * s = "abc\"def"
+        * we must treat \" as "
+        */
        if (c == '\\' && *s == pps) { /* unrecognized \z? */
                c = *(*s); /* yes, fetch z */
                if (c)
@@ -704,6 +708,15 @@ static char nextchar(char **s)
        return c;
 }
 
+/* TODO: merge with strcpy_and_process_escape_sequences()?
+ */
+static void unescape_string_in_place(char *s1)
+{
+       char *s = s1;
+       while ((*s1 = nextchar(&s)) != '\0')
+               s1++;
+}
+
 static ALWAYS_INLINE int isalnum_(int c)
 {
        return (isalnum(c) || c == '_');
@@ -1799,6 +1812,18 @@ static void handle_special(var *v)
                is_f0_split = FALSE;
 
        } else if (v == intvar[FS]) {
+               /*
+                * The POSIX-2008 standard says that changing FS should have no effect on the
+                * current input line, but only on the next one. The language is:
+                *
+                * > Before the first reference to a field in the record is evaluated, the record
+                * > shall be split into fields, according to the rules in Regular Expressions,
+                * > using the value of FS that was current at the time the record was read.
+                *
+                * So, split up current line before assignment to FS:
+                */
+               split_f0();
+
                mk_splitter(getvar_s(v), &fsplitter);
 
        } else if (v == intvar[RS]) {
@@ -2636,7 +2661,8 @@ static var *evaluate(node *op, var *res)
                        var *vbeg, *v;
                        const char *sv_progname;
 
-                       if (!op->r.f->body.first)
+                       /* The body might be empty, still has to eval the args */
+                       if (!op->r.n->info)
                                syntax_error(EMSG_UNDEF_FUNC);
 
                        vbeg = v = nvalloc(op->r.f->nargs + 1);
@@ -2992,7 +3018,7 @@ static int awk_exit(int r)
  * otherwise return 0 */
 static int is_assignment(const char *expr)
 {
-       char *exprc, *val, *s, *s1;
+       char *exprc, *val;
 
        if (!isalnum_(*expr) || (val = strchr(expr, '=')) == NULL) {
                return FALSE;
@@ -3002,10 +3028,7 @@ static int is_assignment(const char *expr)
        val = exprc + (val - expr);
        *val++ = '\0';
 
-       s = s1 = val;
-       while ((*s1 = nextchar(&s)) != '\0')
-               s1++;
-
+       unescape_string_in_place(val);
        setvar_u(newvar(exprc), val);
        free(exprc);
        return TRUE;
@@ -3118,8 +3141,10 @@ int awk_main(int argc, char **argv)
        opt = getopt32(argv, "F:v:f:W:", &opt_F, &list_v, &list_f, NULL);
        argv += optind;
        argc -= optind;
-       if (opt & 0x1)
-               setvar_s(intvar[FS], opt_F); // -F
+       if (opt & 0x1) { /* -F */
+               unescape_string_in_place(opt_F);
+               setvar_s(intvar[FS], opt_F);
+       }
        while (list_v) { /* -v */
                if (!is_assignment(llist_pop(&list_v)))
                        bb_show_usage();