aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--ChangeLog32
-rw-r--r--array.c15
-rw-r--r--awk.h6
-rw-r--r--awkgram.c1
-rw-r--r--awkgram.y1
-rw-r--r--bool.notes53
-rw-r--r--builtin.c42
-rw-r--r--eval.c1
-rw-r--r--field.c2
-rw-r--r--gawkapi.h47
-rw-r--r--main.c1
-rw-r--r--node.c22
-rw-r--r--test/ChangeLog5
-rw-r--r--test/dumpvars.ok2
-rw-r--r--test/functab5.ok1
-rw-r--r--test/id.ok1
-rw-r--r--test/intest.awk4
-rw-r--r--test/symtab11.ok1
-rw-r--r--test/symtab8.ok2
19 files changed, 210 insertions, 29 deletions
diff --git a/ChangeLog b/ChangeLog
index 89767c0f..5f2794d7 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,35 @@
+2021-03-20 Arnold D. Robbins <arnold@skeeve.com>
+
+ * array.c (do_sort_up_value_type): Add logic for handling bools.
+
+2021-03-08 Arnold D. Robbins <arnold@skeeve.com>
+
+ * awk.h (warn_bool, do_bool): Add function declarations.
+ * awkgram.y (tokentab): Add entry for "bool" builtin.
+ * builtin.c (warn_bool): New function.
+ (do_index, do_substr, do_toupper, do_tolower, do_match,
+ do_length): Call it.
+ (do_sub): If first arg to sub/gsub is bool, fatal error.
+ (do_bool): New function.
+ * field.c (do_split, do_patsplit): Call warn_bool.
+ * main.c (load_procinfo_bools): Removed function and call.
+
+2021-03-05 Arnold D. Robbins <arnold@skeeve.com>
+
+ Start on a bool type for gawk.
+
+ * awk.h (BOOL): New flag value.
+ (make_bool_node): Add declaration of new function.
+ (bool_val): Check for BOOL along with NUMBER.
+ * builtin.c (do_typeof): Add support for BOOL.
+ * eval.c (flags2str): Ditto.
+ * gawkapi.h (awk_val_type): Add AWK_BOOL.
+ (awk_value_t): Add awk_bool_t element to the union and macro for it.
+ (struct gawk_api): Update the table of request/return values.
+ * main.c (load_procinfo_bools): New function.
+ (load_procinfo): Call it.
+ * node.c (make_bool_node): New function.
+
2021-02-13 Arnold D. Robbins <arnold@skeeve.com>
* io.c (nextfile): Use the value of ARGC directly in the for
diff --git a/array.c b/array.c
index 50efddb2..b4ce86bc 100644
--- a/array.c
+++ b/array.c
@@ -1209,11 +1209,24 @@ do_sort_up_value_type(const void *p1, const void *p2)
(void) fixtype(n1);
(void) fixtype(n2);
+ /* 3a. Bools first */
+ if ((n1->flags & BOOL) != 0 && (n2->flags & BOOL) != 0) {
+ return cmp_numbers(n1, n2);
+ }
+
+ /* 3b. Numbers next */
if ((n1->flags & NUMBER) != 0 && (n2->flags & NUMBER) != 0) {
return cmp_numbers(n1, n2);
}
- /* 3. All numbers are less than all strings. This is aribitrary. */
+ /* 3c. Bools before everything else */
+ if ((n1->flags & BOOL) != 0 && (n2->flags & BOOL) == 0) {
+ return -1;
+ } else if ((n1->flags & BOOL) == 0 && (n2->flags & BOOL) != 0) {
+ return 1;
+ }
+
+ /* 3d. All numbers are less than all strings. This is aribitrary. */
if ((n1->flags & NUMBER) != 0 && (n2->flags & STRING) != 0) {
return -1;
} else if ((n1->flags & STRING) != 0 && (n2->flags & NUMBER) != 0) {
diff --git a/awk.h b/awk.h
index c9eec663..f4fa4d4d 100644
--- a/awk.h
+++ b/awk.h
@@ -463,6 +463,7 @@ typedef struct exp_node {
XARRAY = 0x10000,
NUMCONSTSTR = 0x20000, /* have string value for numeric constant */
REGEX = 0x40000, /* this is a typed regex */
+ BOOL = 0x80000, /* this is a boolean value */
} flags;
long valref;
} NODE;
@@ -1454,6 +1455,7 @@ extern bool is_identchar(int c);
extern NODE *make_regnode(NODETYPE type, NODE *exp);
extern bool validate_qualified_name(char *token);
/* builtin.c */
+extern void warn_bool(const char *func, int argnum, NODE *n);
extern double double_to_int(double d);
extern NODE *do_exp(int nargs);
extern NODE *do_fflush(int nargs);
@@ -1503,6 +1505,7 @@ extern int strncasecmpmbs(const unsigned char *,
const unsigned char *, size_t);
extern int sanitize_exit_status(int status);
extern void check_symtab_functab(NODE *dest, const char *fname, const char *msg);
+extern NODE *do_bool(int nargs);
/* debug.c */
extern void init_debug(void);
extern int debug_prog(INSTRUCTION *pc);
@@ -1714,6 +1717,7 @@ extern NODE *r_force_number(NODE *n);
extern NODE *r_format_val(const char *format, int index, NODE *s);
extern NODE *r_dupnode(NODE *n);
extern NODE *make_str_node(const char *s, size_t len, int flags);
+extern NODE *make_bool_node(bool value);
extern NODE *make_typed_regex(const char *re, size_t len);
extern void *more_blocks(int id);
extern int parse_escape(const char **string_ptr);
@@ -1995,7 +1999,7 @@ static inline bool
boolval(NODE *t)
{
(void) fixtype(t);
- if ((t->flags & NUMBER) != 0)
+ if ((t->flags & (BOOL|NUMBER)) != 0)
return ! is_zero(t);
return (t->stlen > 0);
}
diff --git a/awkgram.c b/awkgram.c
index 9c2700a3..40c3cbd2 100644
--- a/awkgram.c
+++ b/awkgram.c
@@ -4779,6 +4779,7 @@ static const struct token tokentab[] = {
{"asorti", Op_builtin, LEX_BUILTIN, GAWKX|A(1)|A(2)|A(3), do_asorti, 0},
{"atan2", Op_builtin, LEX_BUILTIN, NOT_OLD|A(2), do_atan2, MPF(atan2)},
{"bindtextdomain", Op_builtin, LEX_BUILTIN, GAWKX|A(1)|A(2), do_bindtextdomain, 0},
+{"bool", Op_builtin, LEX_BUILTIN, GAWKX|A(1), do_bool, 0},
{"break", Op_K_break, LEX_BREAK, 0, 0, 0},
{"case", Op_K_case, LEX_CASE, GAWKX, 0, 0},
{"close", Op_builtin, LEX_BUILTIN, NOT_OLD|A(1)|A(2), do_close, 0},
diff --git a/awkgram.y b/awkgram.y
index 6dfbe7e0..3fe0272b 100644
--- a/awkgram.y
+++ b/awkgram.y
@@ -2277,6 +2277,7 @@ static const struct token tokentab[] = {
{"asorti", Op_builtin, LEX_BUILTIN, GAWKX|A(1)|A(2)|A(3), do_asorti, 0},
{"atan2", Op_builtin, LEX_BUILTIN, NOT_OLD|A(2), do_atan2, MPF(atan2)},
{"bindtextdomain", Op_builtin, LEX_BUILTIN, GAWKX|A(1)|A(2), do_bindtextdomain, 0},
+{"bool", Op_builtin, LEX_BUILTIN, GAWKX|A(1), do_bool, 0},
{"break", Op_K_break, LEX_BREAK, 0, 0, 0},
{"case", Op_K_case, LEX_CASE, GAWKX, 0, 0},
{"close", Op_builtin, LEX_BUILTIN, NOT_OLD|A(1)|A(2), do_close, 0},
diff --git a/bool.notes b/bool.notes
new file mode 100644
index 00000000..8e24c2b5
--- /dev/null
+++ b/bool.notes
@@ -0,0 +1,53 @@
+Thu Mar 18 21:05:29 IST 2021
+============================
+
+Design Notes for a Boolean Type in Gawk
+
+1. A new function bool(val) converts val to bool, returning Boolean TRUE
+or FALSE. This is the generator for boolean values and is enough to have
+instead of predefining new variables TRUE and FALSE.
+
+2. Assigning from a boolean value copies the bool type.
+
+3. Boolean variables have numeric values 1 and 0 respectively, and string
+values "TRUE" and "FALSE". Thus they differ from other variables where a
+"false" value must be zero and null.
+
+Given:
+
+ true = bool(1)
+ false = bool(0)
+
+this implies all of the following:
+
+ print(true) --> "TRUE"
+ print(false) --> "FALSE"
+ Same for %s in printf
+ Same for bool_var ""
+ printf %d gives 0/1
+
+4. typeof() returns "bool".
+
+5. Numeric operators treat booleans as numbers. asort() sorts booleans before
+numbers, and false before true.
+
+6. These string function generate a runtime fatal error
+if given an argument / target of boolean type:
+
+ gsub sub
+
+These functions merely treat the value as a string
+but issue a lint warning.
+
+ substr match index gensub
+ length split patsplit
+ tolower toupper
+
+7. Updates to API needed for an additional type, and the table
+for requested vs. returns.
+
+8. The following extensions need revising:
+
+ - JSON extension
+ - dump / read array extensions
+ - what else?
diff --git a/builtin.c b/builtin.c
index 1b612f4a..1158fad2 100644
--- a/builtin.c
+++ b/builtin.c
@@ -381,6 +381,8 @@ do_index(int nargs)
s1 = force_string(s1);
s2 = force_string(s2);
+ warn_bool("index", 1, s1);
+ warn_bool("index", 2, s2);
p1 = s1->stptr;
p2 = s2->stptr;
@@ -552,6 +554,7 @@ do_length(int nargs)
if (do_lint && (fixtype(tmp)->flags & STRING) == 0)
lintwarn(_("%s: received non-string argument"), "length");
tmp = force_string(tmp);
+ warn_bool("length", 1, tmp);
if (gawk_mb_cur_max > 1) {
tmp = force_wstring(tmp);
@@ -1779,6 +1782,16 @@ do_sqrt(int nargs)
return make_number((AWKNUM) sqrt(arg));
}
+/* warn_bool --- warn that bool parameter is used as a string */
+
+void
+warn_bool(const char *func, int argnum, NODE *n)
+{
+ if (do_lint && (n->flags & BOOL) != 0)
+ lintwarn(_("%s: argument %d of type bool used as a string"),
+ func, argnum);
+}
+
/* do_substr --- do the substr function */
NODE *
@@ -1802,6 +1815,7 @@ do_substr(int nargs)
DEREF(t1);
t1 = POP_STRING();
+ warn_bool("substr", 1, t1);
if (nargs == 3) {
if (! (d_length >= 1)) {
@@ -2407,6 +2421,7 @@ do_tolower(int nargs)
NODE *t1, *t2;
t1 = POP_SCALAR();
+ warn_bool("tolower", 1, t1);
if (do_lint && (fixtype(t1)->flags & STRING) == 0)
lintwarn(_("%s: received non-string argument"), "tolower");
t1 = force_string(t1);
@@ -2438,6 +2453,7 @@ do_toupper(int nargs)
NODE *t1, *t2;
t1 = POP_SCALAR();
+ warn_bool("toupper", 1, t1);
if (do_lint && (fixtype(t1)->flags & STRING) == 0)
lintwarn(_("%s: received non-string argument"), "toupper");
t1 = force_string(t1);
@@ -2662,6 +2678,7 @@ do_match(int nargs)
tre = POP();
rp = re_update(tre);
t1 = POP_STRING();
+ warn_bool("mastch", 1, t1);
rstart = research(rp, t1->stptr, 0, t1->stlen, RE_NEED_START);
if (rstart >= 0) { /* match succeded */
@@ -2882,6 +2899,7 @@ do_sub(int nargs, unsigned int flags)
rp = re_update(tmp);
target = POP_STRING(); /* original string */
+ warn_bool("gensub", 3, target);
glob_flag = POP_SCALAR(); /* value of global flag */
if ( (glob_flag->flags & STRING) != 0
@@ -2924,6 +2942,10 @@ do_sub(int nargs, unsigned int flags)
}
}
+ if ((target->flags & BOOL) != 0)
+ fatal(_("%s: target cannot be of type bool"),
+ (flags & GSUB) != 0 ? "gsub" : "sub");
+
global = (how_many == -1);
rep_node = POP_STRING(); /* replacement text */
@@ -4112,7 +4134,10 @@ do_typeof(int nargs)
}
break;
case Node_val:
- switch (fixtype(arg)->flags & (STRING|NUMBER|USER_INPUT|REGEX)) {
+ switch (fixtype(arg)->flags & (STRING|NUMBER|USER_INPUT|REGEX|BOOL)) {
+ case BOOL:
+ res = "bool";
+ break;
case NUMBER:
res = "number";
break;
@@ -4321,3 +4346,18 @@ check_symtab_functab(NODE *dest, const char *fname, const char *msg)
else if (dest == func_table)
fatal(msg, fname, "FUNCTAB");
}
+
+/* do_bool --- create boolean values */
+
+NODE *
+do_bool(int nargs)
+{
+ NODE *tmp;
+ bool result;
+
+ tmp = POP_SCALAR();
+ result = boolval(tmp);
+ DEREF(tmp);
+
+ return make_bool_node(result);
+}
diff --git a/eval.c b/eval.c
index 640f939f..75c39327 100644
--- a/eval.c
+++ b/eval.c
@@ -455,6 +455,7 @@ flags2str(int flagval)
{ XARRAY, "XARRAY" },
{ NUMCONSTSTR, "NUMCONSTSTR" },
{ REGEX, "REGEX" },
+ { BOOL, "BOOL" },
{ 0, NULL },
};
diff --git a/field.c b/field.c
index c21046b9..2bd58863 100644
--- a/field.c
+++ b/field.c
@@ -1019,6 +1019,7 @@ do_split(int nargs)
assoc_clear(arr);
src = TOP_STRING();
+ warn_bool("split", 1, src);
if (src->stlen == 0) {
/*
* Skip the work if first arg is the null string.
@@ -1096,6 +1097,7 @@ do_patsplit(int nargs)
_("%s: cannot use %s as second argument"));
src = TOP_STRING();
+ warn_bool("patsplit", 1, src);
if ((sep->flags & REGEX) != 0)
sep = sep->typed_re;
diff --git a/gawkapi.h b/gawkapi.h
index 54130b15..3fd02df7 100644
--- a/gawkapi.h
+++ b/gawkapi.h
@@ -366,7 +366,8 @@ typedef enum {
AWK_STRNUM,
AWK_ARRAY,
AWK_SCALAR, /* opaque access to a variable */
- AWK_VALUE_COOKIE /* for updating a previously created value */
+ AWK_VALUE_COOKIE, /* for updating a previously created value */
+ AWK_BOOL
} awk_valtype_t;
/*
@@ -381,6 +382,7 @@ typedef struct awk_value {
awk_array_t a;
awk_scalar_t scl;
awk_value_cookie_t vc;
+ awk_bool_t b;
} u;
#define str_value u.s
#define strnum_value str_value
@@ -391,6 +393,7 @@ typedef struct awk_value {
#define array_cookie u.a
#define scalar_cookie u.scl
#define value_cookie u.vc
+#define bool_value u.b
} awk_value_t;
/*
@@ -567,26 +570,28 @@ typedef struct gawk_api {
+-------------------------------------------------------+
| Type of Actual Value: |
- +--------+--------+--------+--------+-------+-----------+
- | String | Strnum | Number | Regex | Array | Undefined |
- +-----------+-----------+--------+--------+--------+--------+-------+-----------+
- | | String | String | String | String | String | false | false |
- | +-----------+--------+--------+--------+--------+-------+-----------+
- | | Strnum | false | Strnum | Strnum | false | false | false |
- | +-----------+--------+--------+--------+--------+-------+-----------+
- | | Number | Number | Number | Number | false | false | false |
- | +-----------+--------+--------+--------+--------+-------+-----------+
- | | Regex | false | false | false | Regex | false | false |
- | +-----------+--------+--------+--------+--------+-------+-----------+
- | Type | Array | false | false | false | false | Array | false |
- | Requested +-----------+--------+--------+--------+--------+-------+-----------+
- | | Scalar | Scalar | Scalar | Scalar | Scalar | false | false |
- | +-----------+--------+--------+--------+--------+-------+-----------+
- | | Undefined | String | Strnum | Number | Regex | Array | Undefined |
- | +-----------+--------+--------+--------+--------+-------+-----------+
- | | Value | false | false | false | false | false | false |
- | | Cookie | | | | | | |
- +-----------+-----------+--------+--------+--------+--------+-------+-----------+
+ +--------+--------+--------+--------+--------+-------+-----------+
+ | String | Strnum | Number | Regex | Bool | Array | Undefined |
+ +-----------+-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | String | String | String | String | String | String | false | false |
+ | +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | Strnum | false | Strnum | Strnum | false | false | false | false |
+ | +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | Number | Number | Number | Number | false | Number | false | false |
+ | +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | Regex | false | false | false | Regex | false | false | false |
+ | +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | Bool | false | false | false | false | Bool | false | false |
+ | +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | Type | Array | false | false | false | false | false | Array | false |
+ | Requested +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | Scalar | Scalar | Scalar | Scalar | Scalar | Scalar | false | false |
+ | +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | Undefined | String | Strnum | Number | Regex | Bool | Array | Undefined |
+ | +-----------+--------+--------+--------+--------+--------+-------+-----------+
+ | | Value | false | false | false | false | false | false | false |
+ | | Cookie | | | | | | | |
+ +-----------+-----------+--------+--------+--------+--------+--------+-------+-----------+
*/
/* Functions to handle parameters passed to the extension. */
diff --git a/main.c b/main.c
index 5b153bc3..76f28289 100644
--- a/main.c
+++ b/main.c
@@ -980,7 +980,6 @@ load_procinfo_argv()
// hook it into PROCINFO
sub = make_string("argv", 4);
assoc_set(PROCINFO_node, sub, argv_array);
-
}
/* load_procinfo --- populate the PROCINFO array */
diff --git a/node.c b/node.c
index c22c06ab..dda90f0f 100644
--- a/node.c
+++ b/node.c
@@ -1082,3 +1082,25 @@ more_blocks(int id)
}
#endif
+
+/* make_bool_node --- make a boolean-valued node */
+
+extern NODE *
+make_bool_node(bool value)
+{
+ NODE *val;
+ const char *sval;
+ AWKNUM nval;
+
+ sval = (value ? "TRUE" : "FALSE");
+ nval = (value ? 1.0 : 0.0);
+
+ val = make_number(nval);
+ val->stptr = estrdup(sval, strlen(sval));
+ val->stlen = strlen(sval);
+ val->flags &= ~NUMBER;
+ val->flags |= NUMCUR|STRCUR|BOOL;
+ val->stfmt = STFMT_UNUSED;
+
+ return val;
+}
diff --git a/test/ChangeLog b/test/ChangeLog
index 53fe5627..2604c270 100644
--- a/test/ChangeLog
+++ b/test/ChangeLog
@@ -1,3 +1,8 @@
+2021-03-08 Arnold D. Robbins <arnold@skeeve.com>
+
+ * dumpvars.ok, functab5.ok, id.ok, intest.awk, symtab11.ok,
+ symtab8.ok: Updated after code changes.
+
2021-02-13 Arnold D. Robbins <arnold@skeeve.com>
* Makefile.am (EXTRA_DIST): argcasfile, new test.
diff --git a/test/dumpvars.ok b/test/dumpvars.ok
index 85d1c859..7caecd35 100644
--- a/test/dumpvars.ok
+++ b/test/dumpvars.ok
@@ -9,7 +9,7 @@ FILENAME: "-"
FNR: 3
FPAT: "[^[:space:]]+"
FS: " "
-FUNCTAB: array, 41 elements
+FUNCTAB: array, 42 elements
IGNORECASE: 0
LINT: 0
NF: 1
diff --git a/test/functab5.ok b/test/functab5.ok
index 9ac4295d..ef110989 100644
--- a/test/functab5.ok
+++ b/test/functab5.ok
@@ -3,6 +3,7 @@ asort'
asorti'
atan2'
bindtextdomain'
+bool'
chdir'
close'
compl'
diff --git a/test/id.ok b/test/id.ok
index ab4df74a..f990546f 100644
--- a/test/id.ok
+++ b/test/id.ok
@@ -34,6 +34,7 @@ asort -> builtin
asorti -> builtin
atan2 -> builtin
bindtextdomain -> builtin
+bool -> builtin
close -> builtin
compl -> builtin
cos -> builtin
diff --git a/test/intest.awk b/test/intest.awk
index f030d07a..18e0cc4d 100644
--- a/test/intest.awk
+++ b/test/intest.awk
@@ -1,4 +1,4 @@
BEGIN {
- bool = ((b = 1) in c);
- print bool, b # gawk-3.0.1 prints "0 "; should print "0 1"
+ bool_result = ((b = 1) in c);
+ print bool_result, b # gawk-3.0.1 prints "0 "; should print "0 1"
}
diff --git a/test/symtab11.ok b/test/symtab11.ok
index 7d4be46c..da2cfcba 100644
--- a/test/symtab11.ok
+++ b/test/symtab11.ok
@@ -37,6 +37,7 @@ BEGIN -- Functab is next
[asorti] = asorti
[atan2] = atan2
[bindtextdomain] = bindtextdomain
+[bool] = bool
[close] = close
[compl] = compl
[cos] = cos
diff --git a/test/symtab8.ok b/test/symtab8.ok
index da29b585..0cf40fe9 100644
--- a/test/symtab8.ok
+++ b/test/symtab8.ok
@@ -9,7 +9,7 @@ FIELDWIDTHS: ""
FNR: 1
FPAT: "[^[:space:]]+"
FS: " "
-FUNCTAB: array, 41 elements
+FUNCTAB: array, 42 elements
IGNORECASE: 0
LINT: 0
NF: 1
generated by cgit v1.2.3 (git 2.25.1) at 2025-05-07 13:56:10 +0000