ash: change CRLF handling

As noted in commit 2d848eba5 (ash: fix CRLF handling) all CRs are
removed when scripts are read.  Allow an alternative approach
(configurable at build-time, enabled by default):

- Do not strip CRs from input.

- Treat CR as similar to space or tab in base syntax mode.

- Adjust pgetc_eatbnl() to handle backslash-CRLF in the same way
  as backslash-LF.

With these changes scripts containing CRLF line endings are more
likely to work.

Adds 48-56 bytes.

(GitHub issue #285)

3 files changed, 55 insertions, 7 deletions

diff --git a/configs/mingw32_defconfig b/configs/mingw32_defconfig
index f4019afe6..78035dd25 100644
--- a/configs/mingw32_defconfig
+++ b/configs/mingw32_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Busybox version: 1.36.0.git
-# Wed Oct 12 10:45:30 2022
+# Busybox version: 1.37.0.git
+# Tue Feb  7 09:34:52 2023
 #
 CONFIG_HAVE_DOT_CONFIG=y
 # CONFIG_PLATFORM_POSIX is not set
@@ -145,6 +145,9 @@ CONFIG_LAST_SUPPORTED_WCHAR=0
 # CONFIG_UNICODE_BIDI_SUPPORT is not set
 # CONFIG_UNICODE_NEUTRAL_TABLE is not set
 # CONFIG_UNICODE_PRESERVE_BROKEN is not set
+# CONFIG_LOOP_CONFIGURE is not set
+# CONFIG_NO_LOOP_CONFIGURE is not set
+CONFIG_TRY_LOOP_CONFIGURE=y
 
 #
 # Applets
@@ -1027,6 +1030,7 @@ CONFIG_DHCPD_LEASES_FILE=""
 # CONFIG_FEATURE_UDHCPC_ARPING is not set
 # CONFIG_FEATURE_UDHCPC_SANITIZEOPT is not set
 CONFIG_UDHCPC_DEFAULT_SCRIPT=""
+CONFIG_UDHCPC6_DEFAULT_SCRIPT=""
 # CONFIG_UDHCPC6 is not set
 # CONFIG_FEATURE_UDHCPC6_RFC3646 is not set
 # CONFIG_FEATURE_UDHCPC6_RFC4704 is not set
@@ -1163,6 +1167,7 @@ CONFIG_ASH_GETOPTS=y
 CONFIG_ASH_CMDCMD=y
 CONFIG_ASH_NOCONSOLE=y
 CONFIG_ASH_NOCASEGLOB=y
+CONFIG_ASH_IGNORE_CR=y
 # CONFIG_CTTYHACK is not set
 # CONFIG_HUSH is not set
 # CONFIG_SHELL_HUSH is not set
diff --git a/configs/mingw64_defconfig b/configs/mingw64_defconfig
index bf8cc5452..1ef6024c1 100644
--- a/configs/mingw64_defconfig
+++ b/configs/mingw64_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Busybox version: 1.36.0.git
-# Wed Oct 12 10:45:30 2022
+# Busybox version: 1.37.0.git
+# Tue Feb  7 09:34:52 2023
 #
 CONFIG_HAVE_DOT_CONFIG=y
 # CONFIG_PLATFORM_POSIX is not set
@@ -145,6 +145,9 @@ CONFIG_LAST_SUPPORTED_WCHAR=0
 # CONFIG_UNICODE_BIDI_SUPPORT is not set
 # CONFIG_UNICODE_NEUTRAL_TABLE is not set
 # CONFIG_UNICODE_PRESERVE_BROKEN is not set
+# CONFIG_LOOP_CONFIGURE is not set
+# CONFIG_NO_LOOP_CONFIGURE is not set
+CONFIG_TRY_LOOP_CONFIGURE=y
 
 #
 # Applets
@@ -1027,6 +1030,7 @@ CONFIG_DHCPD_LEASES_FILE=""
 # CONFIG_FEATURE_UDHCPC_ARPING is not set
 # CONFIG_FEATURE_UDHCPC_SANITIZEOPT is not set
 CONFIG_UDHCPC_DEFAULT_SCRIPT=""
+CONFIG_UDHCPC6_DEFAULT_SCRIPT=""
 # CONFIG_UDHCPC6 is not set
 # CONFIG_FEATURE_UDHCPC6_RFC3646 is not set
 # CONFIG_FEATURE_UDHCPC6_RFC4704 is not set
@@ -1163,6 +1167,7 @@ CONFIG_ASH_GETOPTS=y
 CONFIG_ASH_CMDCMD=y
 CONFIG_ASH_NOCONSOLE=y
 CONFIG_ASH_NOCASEGLOB=y
+CONFIG_ASH_IGNORE_CR=y
 # CONFIG_CTTYHACK is not set
 # CONFIG_HUSH is not set
 # CONFIG_SHELL_HUSH is not set
diff --git a/shell/ash.c b/shell/ash.c
index a24ed0ac8..63eb4947c 100644
--- a/shell/ash.c
+++ b/shell/ash.c
@@ -192,6 +192,14 @@
 //config:       Enable support for the 'nocaseglob' option, which allows
 //config:       case-insensitive filename globbing.
 //config:
+//config:config ASH_IGNORE_CR
+//config:       bool "Ignore CR in scripts"
+//config:       default y
+//config:       depends on (ASH || SH_IS_ASH || BASH_IS_ASH) && PLATFORM_MINGW32
+//config:       help
+//config:       Allow CRs to be ignored when shell scripts are parsed.  This
+//config:       makes it possible to run scripts with CRLF line endings.
+//config:
 //config:endif # ash options
 
 //applet:IF_ASH(APPLET(ash, BB_DIR_BIN, BB_SUID_DROP))
@@ -762,7 +770,11 @@ struct strpush {
         struct strpush *spfree;
 
         /* Remember last two characters for pungetc. */
+#if ENABLE_ASH_IGNORE_CR
+        int lastc[3];   /* Ignoring CRs needs more pungetc. */
+#else
         int lastc[2];
+#endif
 
         /* Number of outstanding calls to pungetc. */
         int unget;
@@ -787,7 +799,11 @@ struct parsefile {
         struct strpush *spfree;
 
         /* Remember last two characters for pungetc. */
+#if ENABLE_ASH_IGNORE_CR
+        int lastc[3];   /* Ignoring CRs needs more pungetc. */
+#else
         int lastc[2];
+#endif
 
         /* Number of outstanding calls to pungetc. */
         int unget;
@@ -3596,7 +3612,11 @@ static const uint8_t syntax_index_table[] ALIGN1 = {
         /*  10 "\n" */ CNL_CNL_CNL_CNL,
         /*  11      */ CWORD_CWORD_CWORD_CWORD,
         /*  12      */ CWORD_CWORD_CWORD_CWORD,
+#if ENABLE_ASH_IGNORE_CR
+        /*  13      */ CSPCL_CWORD_CWORD_CWORD,
+#else
         /*  13      */ CWORD_CWORD_CWORD_CWORD,
+#endif
         /*  14      */ CWORD_CWORD_CWORD_CWORD,
         /*  15      */ CWORD_CWORD_CWORD_CWORD,
         /*  16      */ CWORD_CWORD_CWORD_CWORD,
@@ -11846,7 +11866,9 @@ preadbuffer(void)
                 more--;
 
                 c = *q;
-                if (c == '\0' || (ENABLE_PLATFORM_MINGW32 && c == '\r')) {
+                /* Remove CR from input buffer as an alternative to ASH_IGNORE_CR. */
+                if (c == '\0' || (c == '\r' &&
+                                        ENABLE_PLATFORM_MINGW32 && !ENABLE_ASH_IGNORE_CR)) {
                         memmove(q, q + 1, more);
                 } else {
                         q++;
@@ -11934,6 +11956,9 @@ static int __pgetc(void)
         else
                 c = preadbuffer();
 
+#if ENABLE_ASH_IGNORE_CR
+        g_parsefile->lastc[2] = g_parsefile->lastc[1];
+#endif
         g_parsefile->lastc[1] = g_parsefile->lastc[0];
         g_parsefile->lastc[0] = c;
 
@@ -11971,11 +11996,24 @@ pgetc_eatbnl(void)
         int c;
 
         while ((c = pgetc()) == '\\') {
+#if !ENABLE_ASH_IGNORE_CR
                 if (pgetc() != '\n') {
                         pungetc();
                         break;
                 }
-
+#else
+                int c2 = pgetc();
+                if (c2 == '\r') {
+                        if (pgetc() == '\n')
+                                goto eatbnl;
+                        pungetc();
+                }
+                if (c2 != '\n') {
+                        pungetc();
+                        break;
+                }
+ eatbnl:
+#endif
                 nlprompt();
         }
 
@@ -14114,7 +14152,7 @@ xxreadtoken(void)
         setprompt_if(needprompt, 2);
         for (;;) {                      /* until token or start of word found */
                 c = pgetc_eatbnl();
-                if (c == ' ' || c == '\t')
+                if (c == ' ' || c == '\t' || (ENABLE_ASH_IGNORE_CR && c == '\r'))
                         continue;
 
                 if (c == '#') {