bin/134856: [PATCH] [sed] Implementation of "addr1,+N" ranges

Jeremie Le Hen jeremie at le-hen.org
Fri May 22 11:50:03 UTC 2009


>Number:         134856
>Category:       bin
>Synopsis:       [PATCH] [sed] Implementation of "addr1,+N" ranges
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-bugs
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          change-request
>Submitter-Id:   current-users
>Arrival-Date:   Fri May 22 11:50:01 UTC 2009
>Closed-Date:
>Last-Modified:
>Originator:     Jeremie Le Hen
>Release:        FreeBSD 7.2-STABLE i386
>Organization:
N/A
>Environment:
System: FreeBSD 7.2-STABLE

>Description:
	This patch implements the "addr1,+N" range in sed(1).
	It is stated in the manpage that this is a non-standard extension
	(although both GNU sed and vim supports this).


--- sed_range.diff begins here ---
Index: compile.c
===================================================================
RCS file: /mnt/space/cvsroot/src/usr.bin/sed/compile.c,v
retrieving revision 1.33
diff -u -p -r1.33 compile.c
--- compile.c	11 Nov 2008 17:15:57 -0000	1.33
+++ compile.c	22 May 2009 11:26:38 -0000
@@ -181,7 +181,7 @@ semicolon:	EATSPACE();
 		if ((*link = cmd = malloc(sizeof(struct s_command))) == NULL)
 			err(1, "malloc");
 		link = &cmd->next;
-		cmd->nonsel = cmd->inrange = 0;
+		cmd->startline = cmd->nonsel = 0;
 		/* First parse the addresses */
 		naddr = 0;
 
@@ -775,6 +775,7 @@ compile_addr(char *p, struct s_addr *a)
 
 	icase = 0;
 
+	a->type = 0;
 	switch (*p) {
 	case '\\':				/* Context address */
 		++p;
@@ -798,10 +799,16 @@ compile_addr(char *p, struct s_addr *a)
 	case '$':				/* Last line */
 		a->type = AT_LAST;
 		return (p + 1);
+
+	case '+':				/* Relative line number */
+		a->type = AT_RELLINE;
+		p++;
+		/* FALLTHROUGH */
 						/* Line number */
 	case '0': case '1': case '2': case '3': case '4':
 	case '5': case '6': case '7': case '8': case '9':
-		a->type = AT_LINE;
+		if (a->type == 0)
+			a->type = AT_LINE;
 		a->u.l = strtol(p, &end, 10);
 		return (end);
 	default:
Index: defs.h
===================================================================
RCS file: /mnt/space/cvsroot/src/usr.bin/sed/defs.h,v
retrieving revision 1.7
diff -u -p -r1.7 defs.h
--- defs.h	9 Feb 2008 09:12:02 -0000	1.7
+++ defs.h	22 May 2009 11:26:19 -0000
@@ -38,8 +38,9 @@
  * Types of address specifications
  */
 enum e_atype {
-	AT_RE,					/* Line that match RE */
+	AT_RE	    = 1,			/* Line that match RE */
 	AT_LINE,				/* Specific line */
+	AT_RELLINE,				/* Relative line */
 	AT_LAST,				/* Last line */
 };
 
@@ -91,6 +92,7 @@ struct s_tr {
 struct s_command {
 	struct s_command *next;			/* Pointer to next command */
 	struct s_addr *a1, *a2;			/* Start and end address */
+	u_long startline;			/* Start line number or zero */
 	char *t;				/* Text for : a c i r w */
 	union {
 		struct s_command *c;		/* Command(s) for b t { */
@@ -100,7 +102,6 @@ struct s_command {
 	} u;
 	char code;				/* Command code */
 	u_int nonsel:1;				/* True if ! */
-	u_int inrange:1;			/* True if in range */
 };
 
 /*
Index: process.c
===================================================================
RCS file: /mnt/space/cvsroot/src/usr.bin/sed/process.c,v
retrieving revision 1.49
diff -u -p -r1.49 process.c
--- process.c	9 Feb 2008 09:12:02 -0000	1.49
+++ process.c	22 May 2009 11:29:47 -0000
@@ -275,8 +275,8 @@ new:		if (!nflag && !pd)
 	    (a)->type == AT_LINE ? linenum == (a)->u.l : lastline())
 
 /*
- * Return TRUE if the command applies to the current line.  Sets the inrange
- * flag to process ranges.  Interprets the non-select (``!'') flag.
+ * Return TRUE if the command applies to the current line.  Sets the start
+ * line for process ranges.  Interprets the non-select (``!'') flag.
  */
 static __inline int
 applies(struct s_command *cp)
@@ -287,18 +287,22 @@ applies(struct s_command *cp)
 	if (cp->a1 == NULL && cp->a2 == NULL)
 		r = 1;
 	else if (cp->a2)
-		if (cp->inrange) {
+		if (cp->startline > 0) {
 			if (MATCH(cp->a2)) {
-				cp->inrange = 0;
+				cp->startline = 0;
 				lastaddr = 1;
 				r = 1;
-			} else if (cp->a2->type == AT_LINE &&
-				   linenum > cp->a2->u.l) {
+			} else if (linenum - cp->startline <= cp->a2->u.l)
+				r = 1;
+			else if ((cp->a2->type == AT_LINE &&
+				   linenum > cp->a2->u.l) ||
+				   (cp->a2->type == AT_RELLINE &&
+				   linenum - cp->startline > cp->a2->u.l)) {
 				/*
 				 * We missed the 2nd address due to a branch,
 				 * so just close the range and return false.
 				 */
-				cp->inrange = 0;
+				cp->startline = 0;
 				r = 0;
 			} else
 				r = 1;
@@ -308,12 +312,15 @@ applies(struct s_command *cp)
 			 * equal to the line number first selected, only
 			 * one line shall be selected.
 			 *	-- POSIX 1003.2
+			 * Likewise if the relative second line address is zero.
 			 */
-			if (cp->a2->type == AT_LINE &&
-			    linenum >= cp->a2->u.l)
+			if ((cp->a2->type == AT_LINE &&
+			    linenum >= cp->a2->u.l) ||
+			    (cp->a2->type == AT_RELLINE && cp->a2->u.l == 0))
 				lastaddr = 1;
-			else
-				cp->inrange = 1;
+			else {
+				cp->startline = linenum;
+			}
 			r = 1;
 		} else
 			r = 0;
@@ -331,11 +338,11 @@ resetstate(void)
 	struct s_command *cp;
 
 	/*
-	 * Reset all inrange markers.
+	 * Reset all in-range markers.
 	 */
 	for (cp = prog; cp; cp = cp->code == '{' ? cp->u.c : cp->next)
 		if (cp->a2)
-			cp->inrange = 0;
+			cp->startline = 0;
 
 	/*
 	 * Clear out the hold space.
Index: sed.1
===================================================================
RCS file: /mnt/space/cvsroot/src/usr.bin/sed/sed.1,v
retrieving revision 1.48
diff -u -p -r1.48 sed.1
--- sed.1	1 Sep 2008 17:48:40 -0000	1.48
+++ sed.1	22 May 2009 11:03:04 -0000
@@ -211,6 +211,9 @@ that matches the second address.
 If the second address is a number
 less than or equal to the line number first selected, only that
 line is selected.
+The number in the second address may be prefixed with a
+.Pq Dq \&+
+to specify the number of lines to match after the first pattern.
 In the case when the second address is a context
 address,
 .Nm
@@ -594,7 +597,10 @@ The
 .Fl E , I , a
 and
 .Fl i
-options, as well as the
+options, the prefixing
+.Dq \&+
+in the second member of an address range,
+as well as the
 .Dq I
 flag to the address regular expression and substitution command are
 non-standard
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:


More information about the freebsd-bugs mailing list