From 7a537eb71874cd6c8d0e9ab1f906e6466e7f793c Mon Sep 17 00:00:00 2001
From: Ron Yorston <rmy@pobox.com>
Date: Mon, 30 Jan 2023 10:33:59 +0000
Subject: awk: CRLF handling

Previous efforts at handling DOS-style line endings in awk have
included commits ee7e00dc5 and 1a3717342.

The use of remove_cr() is unwise:

- It's overzealous, removing all CRs, not just those in CRLF pairs.

- Even if that were fixed awk reads input in chunks. There's a
  remote chance a CRLF might appear at a chunk boundary and be
  missed.

remove_cr() will be fixed separately.  In awk treat all data input
as being in text mode.

Skipping CRs in skip_spaces() is also flawed.  Instead read scripts
in text mode.

Add a couple of test cases.  One of these (awk backslash+CRLF eaten
with no trace) fails without this patch.
---
 editors/awk.c       | 37 ++++++++++++++++++-------------------
 testsuite/awk.tests | 14 ++++++++++++++
 2 files changed, 32 insertions(+), 19 deletions(-)

diff --git a/editors/awk.c b/editors/awk.c
index cf9269c6f..2c1272554 100644
--- a/editors/awk.c
+++ b/editors/awk.c
@@ -826,11 +826,7 @@ static char *skip_spaces(char *p)
 		if (*p == '\\' && p[1] == '\n') {
 			p++;
 			t_lineno++;
-#if !ENABLE_PLATFORM_MINGW32
 		} else if (*p != ' ' && *p != '\t') {
-#else
-		} else if (*p != ' ' && *p != '\t' && *p != '\r') {
-#endif
 			break;
 		}
 		p++;
@@ -2231,21 +2227,6 @@ static int ptest(node *pattern)
 	return istrue(evaluate(pattern, &G.ptest__tmpvar));
 }
 
-#if ENABLE_PLATFORM_MINGW32
-static ssize_t FAST_FUNC safe_read_strip_cr(int fd, void *buf, size_t count)
-{
-	ssize_t n;
-
-	do {
-		n = safe_read(fd, buf, count);
-	} while (n > 0 && (n=remove_cr((char *)buf, n)) == 0);
-
-	return n;
-}
-
-#define safe_read safe_read_strip_cr
-#endif
-
 /* read next record from stream rsm into a variable v */
 static int awk_getline(rstream *rsm, var *v)
 {
@@ -2834,6 +2815,15 @@ static int is_assignment(const char *expr)
 	return TRUE;
 }
 
+
+#if ENABLE_PLATFORM_MINGW32
+static void set_text_mode(FILE *f)
+{
+	if (f)
+		_setmode(fileno(f), _O_TEXT);
+}
+#endif
+
 /* switch to next input file */
 static rstream *next_input_file(void)
 {
@@ -2862,6 +2852,9 @@ static rstream *next_input_file(void)
 			break;
 		}
 	}
+#if ENABLE_PLATFORM_MINGW32
+	set_text_mode(rsm.F);
+#endif
 
 	files_happen = TRUE;
 	setvar_s(intvar[FILENAME], fname);
@@ -3242,6 +3235,9 @@ static var *evaluate(node *op, var *res)
 					} else {
 						rsm->F = fopen_for_read(L.s);  /* not xfopen! */
 					}
+#if ENABLE_PLATFORM_MINGW32
+					set_text_mode(rsm->F);
+#endif
 				}
 			} else {
 				if (!iF)
@@ -3695,6 +3691,9 @@ int awk_main(int argc UNUSED_PARAM, char **argv)
 
 		g_progname = llist_pop(&list_f);
 		fd = xopen_stdin(g_progname);
+#if ENABLE_PLATFORM_MINGW32
+		_setmode(fd, _O_TEXT);
+#endif
 		s = xmalloc_read(fd, NULL); /* it's NUL-terminated */
 		close(fd);
 		parse_program(s);
diff --git a/testsuite/awk.tests b/testsuite/awk.tests
index bbf0fbff1..11beb1b10 100755
--- a/testsuite/awk.tests
+++ b/testsuite/awk.tests
@@ -485,4 +485,18 @@ testing 'awk assign while test' \
 	"" \
 	"foo"
 
+optional PLATFORM_MINGW32
+testing 'awk match line ending' \
+	"awk '/world$/'" \
+	"world\n" \
+	"" \
+	"hello\r\nworld\r\n"
+
+testing 'awk backslash+CRLF eaten with no trace' \
+	"awk -f -" \
+	"Hello world\n" \
+	'' \
+	'BEGIN { printf "Hello\\\r\n world\\n" }\n'
+SKIP=
+
 exit $FAILCOUNT
-- 
cgit v1.2.3-55-g6feb