Allow input records longer than INT_MAX.

4 files changed, 46 insertions, 28 deletions

diff --git a/ChangeLog b/ChangeLog
index c1d22b94..732d8af5 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,18 @@
+2023-03-18         Miguel Pineiro Jr.    <mpj@pineiro.cc>
+
+	Allow records longer than INT_MAX. For test cases,
+	see https://lists.gnu.org/archive/html/bug-gawk/2021-05/msg00003.html.
+
+	* awk.h (set_record): `cnt' is now size_t.
+	* field.c (set_record): `cnt' is now size_t. Adjust databuf_size
+	and MAX_SIZE macro.
+	* io.c (get_a_record): Add new second parameter, size_t *len,
+	which holds the size. Adjust return value meaning. Deal with
+	return from an extension's get_a_record() function.
+	(do_getline_redir): Adjust calling get_a_record().
+	(do_getline): Ditto.
+	(inrec): Ditto.
+
 2023-03-09         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* gawkapi.h: Update copyright year. Small edit in leading comment.
diff --git a/awk.h b/awk.h
index 661eb637..afe90f0a 100644
--- a/awk.h
+++ b/awk.h
@@ -1569,7 +1569,7 @@ extern NODE *get_actual_argument(NODE *, int, bool);
 #endif
 /* field.c */
 extern void init_fields(void);
-extern void set_record(const char *buf, int cnt, const awk_fieldwidth_info_t *);
+extern void set_record(const char *buf, size_t cnt, const awk_fieldwidth_info_t *);
 extern void reset_record(void);
 extern void rebuild_record(void);
 extern void set_NF(void);
diff --git a/field.c b/field.c
index 343a3100..9ff37582 100644
--- a/field.c
+++ b/field.c
@@ -261,13 +261,13 @@ rebuild_record()
  * but better correct than fast.
  */
 void
-set_record(const char *buf, int cnt, const awk_fieldwidth_info_t *fw)
+set_record(const char *buf, size_t cnt, const awk_fieldwidth_info_t *fw)
 {
 	NODE *n;
 	static char *databuf;
-	static unsigned long databuf_size;
+	static size_t databuf_size;
 #define INITIAL_SIZE	512
-#define MAX_SIZE	((unsigned long) ~0)	/* maximally portable ... */
+#define MAX_SIZE	((size_t) ~0)	/* maximally portable ... */
 
 	purge_record();
 
diff --git a/io.c b/io.c
index dcbec287..c1057f96 100644
--- a/io.c
+++ b/io.c
@@ -267,7 +267,7 @@ static RECVALUE rsrescan(IOBUF *iop, struct recmatch *recm, SCANSTATE *state);
 
 static RECVALUE (*matchrec)(IOBUF *iop, struct recmatch *recm, SCANSTATE *state) = rs1scan;
 
-static int get_a_record(char **out, IOBUF *iop, int *errcode, const awk_fieldwidth_info_t **field_width);
+static int get_a_record(char **out, size_t *len, IOBUF *iop, int *errcode, const awk_fieldwidth_info_t **field_width);
 
 static void free_rp(struct redirect *rp);
 
@@ -565,21 +565,19 @@ bool
 inrec(IOBUF *iop, int *errcode)
 {
 	char *begin;
-	int cnt;
-	bool retval = true;
+	size_t cnt;
+	bool retval;
 	const awk_fieldwidth_info_t *field_width = NULL;
 
 	if (at_eof(iop) && no_data_left(iop))
-		cnt = EOF;
+		retval = false;
 	else if ((iop->flag & IOP_CLOSED) != 0)
-		cnt = EOF;
+		retval = false;
 	else
-		cnt = get_a_record(& begin, iop, errcode, & field_width);
+		/* Note that get_a_record may return -2 when I/O would block */
+		retval = (get_a_record(& begin, & cnt, iop, errcode, & field_width) == 0);
 
-	/* Note that get_a_record may return -2 when I/O would block */
-	if (cnt < 0) {
-		retval = false;
-	} else {
+	if (retval) {
 		INCREMENT_REC(NR);
 		INCREMENT_REC(FNR);
 		set_record(begin, cnt, field_width);
@@ -2808,7 +2806,8 @@ do_getline_redir(int into_variable, enum redirval redirtype)
 {
 	struct redirect *rp = NULL;
 	IOBUF *iop;
-	int cnt = EOF;
+	size_t cnt;
+	int retval = EOF;
 	char *s = NULL;
 	int errcode;
 	NODE *redir_exp = NULL;
@@ -2843,14 +2842,14 @@ do_getline_redir(int into_variable, enum redirval redirtype)
 		return make_number((AWKNUM) 0.0);
 
 	errcode = 0;
-	cnt = get_a_record(& s, iop, & errcode, (lhs ? NULL : & field_width));
+	retval = get_a_record(& s, & cnt, iop, & errcode, (lhs ? NULL : & field_width));
 	if (errcode != 0) {
 		if (! do_traditional && (errcode != -1))
 			update_ERRNO_int(errcode);
-		return make_number((AWKNUM) cnt);
+		return make_number((AWKNUM) retval);
 	}
 
-	if (cnt == EOF) {
+	if (retval == EOF) {
 		/*
 		 * Don't do iop_close() here if we are
 		 * reading from a pipe; otherwise
@@ -2882,7 +2881,8 @@ do_getline_redir(int into_variable, enum redirval redirtype)
 NODE *
 do_getline(int into_variable, IOBUF *iop)
 {
-	int cnt = EOF;
+	size_t cnt;
+	int retval = EOF;
 	char *s = NULL;
 	int errcode;
 	const awk_fieldwidth_info_t *field_width = NULL;
@@ -2894,16 +2894,16 @@ do_getline(int into_variable, IOBUF *iop)
 	}
 
 	errcode = 0;
-	cnt = get_a_record(& s, iop, & errcode, (into_variable ? NULL : & field_width));
+	retval = get_a_record(& s, & cnt, iop, & errcode, (into_variable ? NULL : & field_width));
 	if (errcode != 0) {
 		if (! do_traditional && (errcode != -1))
 			update_ERRNO_int(errcode);
 		if (into_variable)
 			(void) POP_ADDRESS();
-		return make_number((AWKNUM) cnt);
+		return make_number((AWKNUM) retval);
 	}
 
-	if (cnt == EOF)
+	if (retval == EOF)
 		return NULL;	/* try next file */
 	INCREMENT_REC(NR);
 	INCREMENT_REC(FNR);
@@ -3851,13 +3851,14 @@ errno_io_retry(void)
 
 /*
  * get_a_record --- read a record from IOP into out,
- * return length or EOF, set RT.
+ * its length into len, and set RT.
+ * return 0 on success, EOF when out of data, and -2 if I/O would block.
  * Note that errcode is never NULL, and the caller initializes *errcode to 0.
- * If I/O would block, return -2.
  */
 
 static int
 get_a_record(char **out,        /* pointer to pointer to data */
+        size_t *len,            /* pointer to record length */
         IOBUF *iop,             /* input IOP */
         int *errcode,           /* pointer to error variable */
         const awk_fieldwidth_info_t **field_width)
@@ -3866,7 +3867,6 @@ get_a_record(char **out,        /* pointer to pointer to data */
 	struct recmatch recm;
 	SCANSTATE state;
 	RECVALUE ret;
-	int retval;
 	NODE *rtval = NULL;
 	static RECVALUE (*lastmatchrec)(IOBUF *iop, struct recmatch *recm, SCANSTATE *state) = NULL;
 
@@ -3885,6 +3885,9 @@ get_a_record(char **out,        /* pointer to pointer to data */
 		if (rc == EOF)
 			iop->flag |= IOP_AT_EOF;
 		else {
+			assert(rc >= 0);
+			*len = rc;
+			rc = 0;
 			if (rt_len != 0)
 				set_RT(rt_start, rt_len);
 			else
@@ -4044,11 +4047,11 @@ get_a_record(char **out,        /* pointer to pointer to data */
 
 	if (recm.len == 0) {
 		*out = NULL;
-		retval = 0;
+		*len = 0;
 	} else {
 		assert(recm.start != NULL);
 		*out = recm.start;
-		retval = recm.len;
+		*len = recm.len;
 	}
 
 	iop->off += recm.len + recm.rt_len;
@@ -4056,7 +4059,7 @@ get_a_record(char **out,        /* pointer to pointer to data */
 	if (recm.len == 0 && recm.rt_len == 0 && at_eof(iop))
 		return EOF;
 	else
-		return retval;
+		return 0;
 }
 
 /* set_RS --- update things as appropriate when RS is set */