support negative LEN in ${VAR:OFFSET:LEN}

[Mikael Magnusson <mikachu@xxxxxxxxx>]

bash 4.2 adds support for this.

http://cgit.mika.l3ib.org/cgit/zsh-cvs/patch/?id=3950051089ce1c87679abf0335edc16fd9289878

From: Mikael Magnusson <mikachu@xxxxxxxxx>
Date: Thu, 10 Mar 2011 10:56:19 +0100
Subject: [PATCH] support LEN < 0 in ${VAR:OFFSET:LEN}, new in bash 4.2

---
 Doc/Zsh/expn.yo        |   12 ++++++++----
 Src/subst.c            |   36 +++++++++++++++++++++++++++---------
 Test/D04parameter.ztst |   14 ++++++++++++++
 3 files changed, 49 insertions(+), 13 deletions(-)

diff --git a/Doc/Zsh/expn.yo b/Doc/Zsh/expn.yo
index 9c44913..b48ddd9 100644
--- a/Doc/Zsh/expn.yo
+++ b/Doc/Zsh/expn.yo
@@ -588,7 +588,7 @@ remove the non-matched elements).
 xitem(tt(${)var(name)tt(:)var(offset)tt(}))
 item(tt(${)var(name)tt(:)var(offset)tt(:)var(length)tt(}))(
 This syntax gives effects similar to parameter subscripting
-in the form tt($)var(name)tt({)var(start)tt(,)var(end)tt(}), but is
+in the form tt($)var(name)tt([)var(start)tt(,)var(end)tt(]), but is
 compatible with other shells; note that both var(offset) and var(length)
 are interpreted differently from the components of a subscript.

@@ -608,8 +608,12 @@ the option tt(KSH_ARRAYS).
 A negative offset counts backwards from the end of the scalar or array,
 so that -1 corresponds to the last character or element, and so on.

-var(length) is always treated directly as a length and hence may not be
-negative.  The option tt(MULTIBYTE) is obeyed, i.e. the offset and length
+var(length) is treated directly as a length when it is positive.
+When it is negative, it works as an offset just like var(offset). If
+this results in a negative length, a diagnostic will be printed and
+nothing will be substituted.
+
+The option tt(MULTIBYTE) is obeyed, i.e. the offset and length
 count multibyte characters where appropriate.

 var(offset) and var(length) undergo the same set of shell substitutions
@@ -635,7 +639,7 @@ tt(${)var(name)tt(:-)var(word)tt(}) form of
substitution.  Instead, a space
 may be inserted before the tt(-).  Furthermore, neither var(offset) nor
 var(length) may begin with an alphabetic character or tt(&) as these are
 used to indicate history-style modifiers.  To substitute a value from a
-variable, the recommended approach is to proceed it with a tt($) as this
+variable, the recommended approach is to precede it with a tt($) as this
 signifies the intention (parameter substitution can easily be rendered
 unreadable); however, as arithmetic substitution is performed, the
 expression tt(${var: offs}) does work, retrieving the offset from
diff --git a/Src/subst.c b/Src/subst.c
index 377aba8..141a353 100644
--- a/Src/subst.c
+++ b/Src/subst.c
@@ -2799,7 +2799,8 @@ paramsubst(LinkList l, LinkNode n, char **str,
int qt, int ssub)
 	    char *check_offset = check_colon_subscript(s, &check_offset2);
 	    if (check_offset) {
 		zlong offset = mathevali(check_offset);
-		zlong length = (zlong)-1;
+		zlong length;
+		int length_set = 0;
 		int offset_hack_argzero = 0;
 		if (errflag)
 		    return NULL;
@@ -2814,14 +2815,11 @@ paramsubst(LinkList l, LinkNode n, char **str,
int qt, int ssub)
 			zerr("invalid length: %s", check_offset);
 			return NULL;
 		    }
-                    if (check_offset) {
+		    if (check_offset) {
 			length = mathevali(check_offset);
+			length_set = 1;
 			if (errflag)
 			    return NULL;
-			if (length < (zlong)0) {
-			    zerr("invalid length: %s", check_offset);
-			    return NULL;
-			}
 		    }
 		}
 		if (horrible_offset_hack) {
@@ -2849,8 +2847,15 @@ paramsubst(LinkList l, LinkNode n, char **str,
int qt, int ssub)
 		    }
 		    if (offset_hack_argzero)
 			alen++;
-		    if (length < 0)
-		      length = alen;
+		    if (length_set) {
+			if (length < 0)
+			    length += alen - offset;
+			if (length < 0) {
+			    zerr("substring expression < 0: %d", length);
+			    return NULL;
+			}
+		    } else
+			length = alen;
 		    if (offset > alen)
 			offset = alen;
 		    if (offset + length > alen)
@@ -2879,11 +2884,24 @@ paramsubst(LinkList l, LinkNode n, char **str,
int qt, int ssub)
 			    offset = 0;
 		    }
 		    MB_METACHARINIT();
+		    if (length_set && length < 0)
+			length -= offset;
 		    for (sptr = val; *sptr && offset; ) {
 			sptr += MB_METACHARLEN(sptr);
 			offset--;
 		    }
-		    if (length >= 0) {
+		    if (length_set) {
+			if (length < 0) {
+			    MB_METACHARINIT();
+			    for (eptr = val; *eptr; ) {
+				eptr += MB_METACHARLEN(eptr);
+				length++;
+			    }
+			    if (length < 0) {
+				zerr("substring expression < 0: %d", length);
+				return NULL;
+			    }
+			}
 			for (eptr = sptr; *eptr && length; ) {
 			    eptr += MB_METACHARLEN(eptr);
 			    length--;
diff --git a/Test/D04parameter.ztst b/Test/D04parameter.ztst
index c0ad1d2..272fb87 100644
--- a/Test/D04parameter.ztst
+++ b/Test/D04parameter.ztst
@@ -1338,6 +1338,7 @@
    print ${foo:$(echo 3 + 3):`echo 4 - 3`}
    print ${foo: -1}
    print ${foo: -10}
+   print ${foo:5:-2}
 0:Bash-style offsets, scalar
 >456789
 >56789
@@ -1349,6 +1350,7 @@
 >7
 >9
 >123456789
+>67

    foo=(1 2 3 4 5 6 7 8 9)
    print ${foo:3}
@@ -1361,6 +1363,7 @@
    print ${foo:$(echo 3 + 3):`echo 4 - 3`}
    print ${foo: -1}
    print ${foo: -10}
+   print ${foo:5:-2}
 0:Bash-style offsets, array
 >4 5 6 7 8 9
 >5 6 7 8 9
@@ -1372,6 +1375,7 @@
 >7
 >9
 >1 2 3 4 5 6 7 8 9
+>6 7

    testfn() {
      emulate -L sh
@@ -1410,3 +1414,13 @@
    print ${str:0:}
 1:Regression test for missing length after offset
 ?(eval):2: unrecognized modifier
+
+   foo="123456789"
+   print ${foo:5:-6}
+1:Regression test for total length < 0 in string
+?(eval):2: substring expression < 0: -2
+
+   foo=(1 2 3 4 5 6 7 8 9)
+   print ${foo:5:-6}
+1:Regression test for total length < 0 in array
+?(eval):2: substring expression < 0: -2
-- 
1.7.4-rc1

-- 
Mikael Magnusson