Merge branch 'gawk-4.1-stable'

13 files changed, 112 insertions, 13 deletions

diff --git a/ChangeLog b/ChangeLog
index bb4a4b44..3a32a238 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,19 @@
+2014-05-09         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* debug.c (do_eval): Don't free `f' which points into the context
+	that was previously freed. Bug reported by Jan Chaloupka
+	<jchaloup@redhat.com>.  Apparently introduced with move to
+	SYMTAB and FUNCTAB, but only showed up on Fedora 20 and Ubuntu 14.04,
+	which have a newer glibc.
+	(do_eval): Fix a memory leak seen by valgrind on Fedora 20 and
+	Ubuntu 14.04: the new SRCFILE that is added wasn't released.
+
+	Unrelated:
+
+	* io.c (get_a_record): Handle return of TERMNEAREND when the
+	entire file has been read into the buffer and we're using a
+	regex for RS. Bug report by Grail Dane <grail69@hotmail.com>.
+
 2014-05-04         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* debug.c (debug_prog): Change check for GAWK_RESTART so that it
diff --git a/awk.h b/awk.h
index aefdd072..9255b45b 100644
--- a/awk.h
+++ b/awk.h
@@ -1373,6 +1373,7 @@ extern NODE *stopme(int nargs);
 extern void shadow_funcs(void);
 extern int check_special(const char *name);
 extern SRCFILE *add_srcfile(enum srctype stype, char *src, SRCFILE *curr, bool *already_included, int *errcode);
+extern void free_srcfile(SRCFILE *thisfile);
 extern void register_deferred_variable(const char *name, NODE *(*load_func)(void));
 extern int files_are_same(char *path, SRCFILE *src);
 extern void valinfo(NODE *n, Func_print print_func, FILE *fp);
diff --git a/awkgram.c b/awkgram.c
index 80d6ea05..7d444fb8 100644
--- a/awkgram.c
+++ b/awkgram.c
@@ -4672,6 +4672,15 @@ parse_program(INSTRUCTION **pcode)
 	return (ret || errcount);
 }
 
+/* free_srcfile --- free a SRCFILE struct */
+
+void
+free_srcfile(SRCFILE *thisfile)
+{
+	efree(thisfile->src);
+	efree(thisfile);
+}
+
 /* do_add_srcfile --- add one item to srcfiles */
 
 static SRCFILE *
diff --git a/awkgram.y b/awkgram.y
index 7d530b7f..07131d22 100644
--- a/awkgram.y
+++ b/awkgram.y
@@ -2333,6 +2333,15 @@ parse_program(INSTRUCTION **pcode)
 	return (ret || errcount);
 }
 
+/* free_srcfile --- free a SRCFILE struct */
+
+void
+free_srcfile(SRCFILE *thisfile)
+{
+	efree(thisfile->src);
+	efree(thisfile);
+}
+
 /* do_add_srcfile --- add one item to srcfiles */
 
 static SRCFILE *
diff --git a/debug.c b/debug.c
index 84e69d47..32b308af 100644
--- a/debug.c
+++ b/debug.c
@@ -5448,6 +5448,7 @@ do_eval(CMDARG *arg, int cmd ATTRIBUTE_UNUSED)
 	int ecount = 0, pcount = 0;
 	int ret;
 	int save_flags = do_flags;
+	SRCFILE *the_source;
 	
 	if (prog_running) {
 		this_frame = find_frame(0);
@@ -5458,7 +5459,7 @@ do_eval(CMDARG *arg, int cmd ATTRIBUTE_UNUSED)
 	ctxt = new_context();
 	ctxt->install_func = append_symbol;	/* keep track of newly installed globals */
 	push_context(ctxt);
-	(void) add_srcfile(SRC_CMDLINE, arg->a_string, srcfiles, NULL, NULL);
+	the_source = add_srcfile(SRC_CMDLINE, arg->a_string, srcfiles, NULL, NULL);
 	do_flags = false;
 	ret = parse_program(&code);
 	do_flags = save_flags;
@@ -5565,8 +5566,18 @@ do_eval(CMDARG *arg, int cmd ATTRIBUTE_UNUSED)
 
 	pop_context();	/* switch to prev context */
 	free_context(ctxt, (ret_val != NULL));   /* free all instructions and optionally symbols */
-	if (ret_val != NULL)
-		destroy_symbol(f);	/* destroy "@eval" */
+
+	/*
+	 * May 2014:
+	 * Don't do this. f points into the context we just released.
+	 * Only showed up on Fedora 20 / Ubuntu 14.04.
+	 *
+	 * if (ret_val != NULL)
+	 *	destroy_symbol(f);	// destroy "@eval"
+	 */
+
+	free_srcfile(the_source);
+
 	return false;
 }
 
diff --git a/io.c b/io.c
index 2bb8f28a..b1c9fa18 100644
--- a/io.c
+++ b/io.c
@@ -206,6 +206,7 @@ typedef enum { CLOSE_ALL, CLOSE_TO, CLOSE_FROM } two_way_close_type;
 #define at_eof(iop)     (((iop)->flag & IOP_AT_EOF) != 0)
 #define has_no_data(iop)        ((iop)->dataend == NULL)
 #define no_data_left(iop)	((iop)->off >= (iop)->dataend)
+#define buffer_has_all_data(iop) ((iop)->dataend - (iop)->off == (iop)->public.sbuf.st_size)
 
 /*
  * The key point to the design is to split out the code that searches through
@@ -3474,6 +3475,8 @@ get_a_record(char **out,        /* pointer to pointer to data */
 		iop->flag &= ~IOP_AT_START;
 		if (ret == REC_OK)
 			break;
+		if (ret == TERMNEAREND && buffer_has_all_data(iop))
+			break;
 
 		/* need to add more data to buffer */
 		/* shift data down in buffer */
diff --git a/test/ChangeLog b/test/ChangeLog
index 6d4cd1c2..878766c7 100644
--- a/test/ChangeLog
+++ b/test/ChangeLog
@@ -1,3 +1,10 @@
+2014-05-09         Andrew J. Schorr     <aschorr@telemetry-investments.com>
+
+	* Makefile.am (rebuf): Force buffer size to 4096 via AWKBUFSIZE
+	environment variable.
+	(rsgetline): New test.
+	* rsgetline.awk, rsgetline.in, rsgetline.ok: New files.
+
 2014-04-11         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* Makefile.am (charset-msg-start): Add a warning message that tests
diff --git a/test/Makefile.am b/test/Makefile.am
index 238f2ed7..ba6bbeec 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -759,6 +759,9 @@ EXTRA_DIST = \
 	rs.awk \
 	rs.in \
 	rs.ok \
+	rsgetline.awk \
+	rsgetline.in \
+	rsgetline.ok \
 	rsnul1nl.awk \
 	rsnul1nl.in \
 	rsnul1nl.ok \
@@ -1000,7 +1003,7 @@ GAWK_EXT_TESTS = \
 	nastyparm next nondec nondec2 \
 	patsplit posix printfbad1 printfbad2 printfbad3 procinfs \
 	profile1 profile2 profile3 profile4 profile5 pty1 \
-	rebuf regx8bit reginttrad reint reint2 rsstart1 \
+	rebuf regx8bit reginttrad reint reint2 rsgetline rsstart1 \
 	rsstart2 rsstart3 rstest6 shadow sortfor sortu split_after_fpat \
 	splitarg4 strftime \
 	strtonum switch2 symtab1 symtab2 symtab3 symtab4 symtab5 symtab6 \
@@ -1329,6 +1332,11 @@ fmtspcl: fmtspcl.ok
 	$(CMP) "$(srcdir)"/$@-mpfr.ok _$@ && rm -f _$@ ; \
 	fi
 
+rebuf::
+	@echo $@
+	@AWKBUFSIZE=4096 AWKPATH="$(srcdir)" $(AWK) -f $@.awk  < "$(srcdir)"/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
+
 reint::
 	@echo $@
 	@$(AWK) --re-interval -f "$(srcdir)"/reint.awk "$(srcdir)"/reint.in >_$@
diff --git a/test/Makefile.in b/test/Makefile.in
index 0831c597..b192b561 100644
--- a/test/Makefile.in
+++ b/test/Makefile.in
@@ -1005,6 +1005,9 @@ EXTRA_DIST = \
 	rs.awk \
 	rs.in \
 	rs.ok \
+	rsgetline.awk \
+	rsgetline.in \
+	rsgetline.ok \
 	rsnul1nl.awk \
 	rsnul1nl.in \
 	rsnul1nl.ok \
@@ -1245,7 +1248,7 @@ GAWK_EXT_TESTS = \
 	nastyparm next nondec nondec2 \
 	patsplit posix printfbad1 printfbad2 printfbad3 procinfs \
 	profile1 profile2 profile3 profile4 profile5 pty1 \
-	rebuf regx8bit reginttrad reint reint2 rsstart1 \
+	rebuf regx8bit reginttrad reint reint2 rsgetline rsstart1 \
 	rsstart2 rsstart3 rstest6 shadow sortfor sortu split_after_fpat \
 	splitarg4 strftime \
 	strtonum switch2 symtab1 symtab2 symtab3 symtab4 symtab5 symtab6 \
@@ -1754,6 +1757,11 @@ fmtspcl: fmtspcl.ok
 	$(CMP) "$(srcdir)"/$@-mpfr.ok _$@ && rm -f _$@ ; \
 	fi
 
+rebuf::
+	@echo $@
+	@AWKBUFSIZE=4096 AWKPATH="$(srcdir)" $(AWK) -f $@.awk  < "$(srcdir)"/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
+
 reint::
 	@echo $@
 	@$(AWK) --re-interval -f "$(srcdir)"/reint.awk "$(srcdir)"/reint.in >_$@
@@ -3498,14 +3506,14 @@ pty1:
 	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
 
-rebuf:
+regx8bit:
 	@echo $@
-	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  < "$(srcdir)"/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
 
-regx8bit:
+rsgetline:
 	@echo $@
-	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  < "$(srcdir)"/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
 
 rstest6:
diff --git a/test/Maketests b/test/Maketests
index b9b713c4..e847d765 100644
--- a/test/Maketests
+++ b/test/Maketests
@@ -1152,14 +1152,14 @@ pty1:
 	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
 
-rebuf:
+regx8bit:
 	@echo $@
-	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  < "$(srcdir)"/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
 
-regx8bit:
+rsgetline:
 	@echo $@
-	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@AWKPATH="$(srcdir)" $(AWK) -f $@.awk  < "$(srcdir)"/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/$@.ok _$@ && rm -f _$@
 
 rstest6:
diff --git a/test/rsgetline.awk b/test/rsgetline.awk
new file mode 100644
index 00000000..fa327fcf
--- /dev/null
+++ b/test/rsgetline.awk
@@ -0,0 +1,23 @@
+# Date: Sun, 4 May 2014 18:09:01 +0200
+# From: Davide Brini <dave_br@gmx.com>
+# To: bug-gawk@gnu.org
+# Subject: Re: [bug-gawk] Computed regex and getline bug / issue
+# 
+# I have been able to reduce the behavior to these simple test cases, which
+# (unless I'm missing something obvious) should behave identically but don't:
+# 
+# $ printf '1,2,' | gawk 'BEGIN{RS="[,]+"}{print; a = getline; print "-"a"-"; print}'
+# 1
+# -0-
+# 1
+
+BEGIN {
+	RS = "[,]+"
+}
+
+{
+	printf "[%s] [%s]\n", $0, RT
+	a = getline
+	print "-"a"-"
+	printf "[%s] [%s]\n", $0, RT
+}
diff --git a/test/rsgetline.in b/test/rsgetline.in
new file mode 100644
index 00000000..f1782346
--- /dev/null
+++ b/test/rsgetline.in
@@ -0,0 +1 @@
+1,2,
\ No newline at end of file
diff --git a/test/rsgetline.ok b/test/rsgetline.ok
new file mode 100644
index 00000000..1388369a
--- /dev/null
+++ b/test/rsgetline.ok
@@ -0,0 +1,3 @@
+[1] [,]
+-1-
+[2] [,]