wordsplit: fix ws trimming and return_delim option

* src/wordsplit.c (_WSNF_DELIM): New node flag
(wsnode_flagstr): Print 'd' for that flag.
(expvar): Don't set WRDSF_WS when expanding variable
value.  This fixes, e.g., expansion of a$x if x begins with
a whitespace.
(wsnode_nullelim): A _WSNF_DELIM node breaks join chain.
(wordsplit_trimws): Don't affect nodes within join chains.
(scan_word): Set _WSNF_DELIM for returned delimiters.
* tests/wordsplit.at (TESTWSP): 8th argument - prologue
shell code.  Use it to unset environment variables in
env/getenv tests.
Test ws elimination with return_delims set.

2 files changed, 47 insertions, 13 deletions

diff --git a/src/wordsplit.c b/src/wordsplit.c
index 29547f6..4715547 100644
--- a/src/wordsplit.c
+++ b/src/wordsplit.c
@@ -309,2 +309,3 @@ alloc_space (struct wordsplit *wsp, size_t count)
 #define _WSNF_SEXP     0x20	/* is a sed expression */
+#define _WSNF_DELIM    0x40     /* node is a delimiter */
 
@@ -333,3 +334,3 @@ wsnode_flagstr (int flags)
 {
-  static char retbuf[6];
+  static char retbuf[7];
   char *p = retbuf;
@@ -358,2 +359,6 @@ wsnode_flagstr (int flags)
     *p++ = '-';
+  if (flags & _WSNF_DELIM)
+    *p++ = 'd';
+  else
+    *p++ = '-';
   *p = 0;
@@ -1150,3 +1155,3 @@ expvar (struct wordsplit *wsp, const char *str, size_t len,
 				WRDSF_NOVAR | WRDSF_NOCMD |
-				WRDSF_WS | WRDSF_QUOTE);
+				WRDSF_QUOTE);
 	  free (value);
@@ -1270,2 +1275,4 @@ wsnode_nullelim (struct wordsplit *wsp)
       struct wordsplit_node *next = p->next;
+      if (p->flags & _WSNF_DELIM && p->prev)
+	p->prev->flags &= ~_WSNF_JOIN;
       if (p->flags & _WSNF_NULL)
@@ -1448,5 +1455,4 @@ wordsplit_trimws (struct wordsplit *wsp)
 
-      if (p->flags & _WSNF_QUOTE)
-	continue;
-
+      if (!(p->flags & _WSNF_QUOTE))
+	{
 	  /* Skip leading whitespace: */
@@ -1456,2 +1462,10 @@ wordsplit_trimws (struct wordsplit *wsp)
 	  p->v.segm.beg = n;
+	}
+      
+      while (p->next && (p->flags & _WSNF_JOIN))
+	p = p->next;
+      
+      if (p->flags & _WSNF_QUOTE)
+	continue;
+      
       /* Trim trailing whitespace */
@@ -1837,2 +1851,3 @@ scan_word (struct wordsplit *wsp, size_t start)
       i++;
+      flags |= _WSNF_DELIM;
     }
diff --git a/tests/wordsplit.at b/tests/wordsplit.at
index 49bba19..d41b60c 100644
--- a/tests/wordsplit.at
+++ b/tests/wordsplit.at
@@ -37,3 +37,3 @@ dnl ------------------------------------------------------------
 dnl TESTWSP([NAME], [KW = `'], [OPTS], [INPUT], [STDOUT = `'],
-dnl         [STDERR = `'], [ENV])
+dnl         [STDERR = `'], [ENV], [PROLOGUE])
 dnl
@@ -42,3 +42,4 @@ AT_SETUP([$1])
 AT_KEYWORDS([wordsplit wsp ]genkw[ $2])
-AT_CHECK([$7 wsp $3 <<'EOT'
+AT_CHECK([$8
+$7 wsp $3 <<'EOT'
 [$4]
@@ -374,8 +375,11 @@ TESTWSP([getvar],[wsp-getvar],
 3: end
-])
+],
+[],
+[],
+[unset foo; unset x])
 
 TESTWSP([getvar and env],[wsp-getvar],
-[foo=bar x=quux],
-[begin $foo $TVAR $x end],
-[NF: 5
+[foo=bar x=quux y=xur],
+[begin $foo $TVAR $x $y end],
+[NF: 6
 0: begin
@@ -384,6 +388,8 @@ TESTWSP([getvar and env],[wsp-getvar],
 3: quux
-4: end
+4: zwar
+5: end
 ],
 [],
-[TVAR=12])
+[TVAR=12 y=zwar],
+[unset foo; unset x])
 
@@ -501,2 +507,15 @@ TESTWSP([ws elimination],[wsp-ws-elim],[delim ' ()' ws return_delims],
 
+TESTWSP([ws elimination + return delim],[wsp-ws-elim-ret],
+[-default novar nocmd delim ":," return_delims ws dquote],
+["foo" : "bar", "quux" : "baaz" ],
+[NF: 7
+0: foo
+1: :
+2: bar
+3: ,
+4: quux
+5: :
+6: baaz
+])
+
 TESTWSP([empty quotes],[wsp-empty-quotes],[delim : ws return_delims],