Add a 6th argument to the API get_record function instead of having a separate field_width array pointer in the input buf.

3 files changed, 41 insertions, 19 deletions

diff --git a/ChangeLog b/ChangeLog
index bc1fe393..223d6fba 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,17 @@
+2017-03-06         Andrew J. Schorr     <aschorr@telemetry-investments.com>
+
+	* gawkapi.h (awk_input_buf_t): Remove field_width array and instead
+	add it as a 6th argument to the get_record function. This should
+	not break existing code, since it's fine to ignore the additional
+	argument. Document the behavior of the field_width argument.
+	* io.c (inrec): Pass pointer to field_width array to get_a_record,
+	and then hand it off to set_record.
+	(do_getline_redir): If not reading into a variable, pass pointer to
+	field_width array to get_a_record and then hand it off to set_record.
+	(do_getline): Ditto.
+	(get_a_record): Add a 4th field_width argument to pass through to
+	the API get_record method.
+
 2017-03-05         Andrew J. Schorr     <aschorr@telemetry-investments.com>
 
 	* awk.h (set_record): Add a new argument containing a field-width
diff --git a/gawkapi.h b/gawkapi.h
index 6d552b8f..6fa022ca 100644
--- a/gawkapi.h
+++ b/gawkapi.h
@@ -146,21 +146,24 @@ typedef struct awk_input {
 	 * than zero, gawk will automatically update the ERRNO variable based
 	 * on the value of *errcode (e.g., setting *errcode = errno should do
 	 * the right thing).
-	 */
-	int (*get_record)(char **out, struct awk_input *iobuf, int *errcode,
-			char **rt_start, size_t *rt_len);
-
-	/*
-	 * If this pointer is non-NULL, then this record should be parsed
-	 * using the supplied field widths instead of the default gawk
-	 * field parsing mechanism. The field_width array should have
+	 *
+	 * If field_width is non-NULL, then its value will be initialized
+	 * to NULL, and the function may set it to point to an array of
+	 * integers supplying field width information to override the default
+	 * gawk field parsing mechanism.  The field_width array should have
 	 * at least 2*NF+1 elements, and the value of field_width[2*NF]
 	 * must be negative. The first entry field_width[0] should contain
 	 * the number of bytes to skip before $1; field_width[1] contains
 	 * the number of bytes in $1. Note that these values are specified
-	 * in bytes, not (potentially multi-byte) characters!
+	 * in bytes, not (potentially multi-byte) characters! And note that this
+	 * array will not be copied by gawk; it must persist at least until the
+	 * next call to get_record or close_func. Note that field_width will
+	 * be NULL when getline is assigning the results to a variable, thus
+	 * field parsing is not needed.
 	 */
-	const int *field_width;
+	int (*get_record)(char **out, struct awk_input *iobuf, int *errcode,
+			char **rt_start, size_t *rt_len,
+			const int **field_width);
 
 	/*
 	 * No argument prototype on read_func to allow for older systems
diff --git a/io.c b/io.c
index 040b485c..8aa9da82 100644
--- a/io.c
+++ b/io.c
@@ -287,7 +287,7 @@ 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);
+static int get_a_record(char **out, IOBUF *iop, int *errcode, const int **field_width);
 
 static void free_rp(struct redirect *rp);
 
@@ -590,13 +590,14 @@ inrec(IOBUF *iop, int *errcode)
 	char *begin;
 	int cnt;
 	bool retval = true;
+	const int *field_width = NULL;
 
 	if (at_eof(iop) && no_data_left(iop))
 		cnt = EOF;
 	else if ((iop->flag & IOP_CLOSED) != 0)
 		cnt = EOF;
 	else
-		cnt = get_a_record(& begin, iop, errcode);
+		cnt = get_a_record(& begin, iop, errcode, & field_width);
 
 	/* Note that get_a_record may return -2 when I/O would block */
 	if (cnt < 0) {
@@ -604,7 +605,7 @@ inrec(IOBUF *iop, int *errcode)
 	} else {
 		INCREMENT_REC(NR);
 		INCREMENT_REC(FNR);
-		set_record(begin, cnt, iop->public.field_width);
+		set_record(begin, cnt, field_width);
 		if (*errcode > 0)
 			retval = false;
 	}
@@ -2618,6 +2619,7 @@ do_getline_redir(int into_variable, enum redirval redirtype)
 	NODE *redir_exp = NULL;
 	NODE **lhs = NULL;
 	int redir_error = 0;
+	const int *field_width = NULL;
 
 	if (into_variable)
 		lhs = POP_ADDRESS();
@@ -2646,7 +2648,7 @@ do_getline_redir(int into_variable, enum redirval redirtype)
 		return make_number((AWKNUM) 0.0);
 
 	errcode = 0;
-	cnt = get_a_record(& s, iop, & errcode);
+	cnt = get_a_record(& s, iop, & errcode, (lhs ? NULL : & field_width));
 	if (errcode != 0) {
 		if (! do_traditional && (errcode != -1))
 			update_ERRNO_int(errcode);
@@ -2668,7 +2670,7 @@ do_getline_redir(int into_variable, enum redirval redirtype)
 	}
 
 	if (lhs == NULL)	/* no optional var. */
-		set_record(s, cnt, iop->public.field_width);
+		set_record(s, cnt, field_width);
 	else {			/* assignment to variable */
 		unref(*lhs);
 		*lhs = make_string(s, cnt);
@@ -2686,6 +2688,7 @@ do_getline(int into_variable, IOBUF *iop)
 	int cnt = EOF;
 	char *s = NULL;
 	int errcode;
+	const int *field_width = NULL;
 
 	if (iop == NULL) {	/* end of input */
 		if (into_variable)
@@ -2694,7 +2697,7 @@ do_getline(int into_variable, IOBUF *iop)
 	}
 
 	errcode = 0;
-	cnt = get_a_record(& s, iop, & errcode);
+	cnt = get_a_record(& s, iop, & errcode, (into_variable ? NULL : & field_width));
 	if (errcode != 0) {
 		if (! do_traditional && (errcode != -1))
 			update_ERRNO_int(errcode);
@@ -2709,7 +2712,7 @@ do_getline(int into_variable, IOBUF *iop)
 	INCREMENT_REC(FNR);
 
 	if (! into_variable)	/* no optional var. */
-		set_record(s, cnt, iop->public.field_width);
+		set_record(s, cnt, field_width);
 	else {			/* assignment to variable */
 		NODE **lhs;
 		lhs = POP_ADDRESS();
@@ -3653,7 +3656,8 @@ errno_io_retry(void)
 static int
 get_a_record(char **out,        /* pointer to pointer to data */
         IOBUF *iop,             /* input IOP */
-        int *errcode)           /* pointer to error variable */
+        int *errcode,           /* pointer to error variable */
+        const int **field_width)/* pointer to pointer to field_width array */
 {
 	struct recmatch recm;
 	SCANSTATE state;
@@ -3672,7 +3676,8 @@ get_a_record(char **out,        /* pointer to pointer to data */
 		char *rt_start;
 		size_t rt_len;
 		int rc = iop->public.get_record(out, &iop->public, errcode,
-						&rt_start, &rt_len);
+						&rt_start, &rt_len,
+						field_width);
 		if (rc == EOF)
 			iop->flag |= IOP_AT_EOF;
 		else {