From c596e17cf3d69cf5e10d28ee2a8ee35162786da1 Mon Sep 17 00:00:00 2001
From: Sverker Eriksson <sverker@erlang.org>
Date: Wed, 23 Jan 2013 16:04:38 +0100
Subject: erl_interface: Changed erlang_char_encoding interface

to allow bitwise-or'd combinations.
---
 lib/erl_interface/doc/src/ei.xml           | 13 +++++++------
 lib/erl_interface/include/ei.h             |  7 ++++++-
 lib/erl_interface/src/decode/decode_atom.c |  2 +-
 lib/erl_interface/src/encode/encode_atom.c |  3 +--
 lib/erl_interface/src/legacy/erl_marshal.c |  6 +++---
 5 files changed, 18 insertions(+), 13 deletions(-)

(limited to 'lib/erl_interface')

diff --git a/lib/erl_interface/doc/src/ei.xml b/lib/erl_interface/doc/src/ei.xml
index e9c7c644b5..117c787da6 100644
--- a/lib/erl_interface/doc/src/ei.xml
+++ b/lib/erl_interface/doc/src/ei.xml
@@ -91,14 +91,13 @@
 	 <p/>
 	 <code type="none">
 enum erlang_char_encoding {
-    ERLANG_ASCII, ERLANG_LATIN1, ERLANG_UTF8, ERLANG_WHATEVER
+    ERLANG_ASCII, ERLANG_LATIN1, ERLANG_UTF8
 };
 </code>
          <p>The character encoding used for atoms. <c>ERLANG_ASCII</c> represents 7-bit ASCII.
 	 Latin1 and UTF8 are different extensions of 7-bit ASCII. All 7-bit ASCII characters
 	 are valid Latin1 and UTF8 characters. ASCII and Latin1 both represent each character
-	 by one byte. A UTF8 character can consist of one to four bytes. <c>ERLANG_WHATEVER</c>
-	 is not an encoding but rather used as a wildcard.</p>
+	 by one byte. A UTF8 character can consist of one to four bytes.</p>
       </item>
     </taglist>
   </section>
@@ -545,11 +544,13 @@ ei_x_encode_empty_list(&amp;x);
 	want</seealso></c>. The original encoding used in the
 	binary format (latin1 or utf8) can be obtained from <c>*was</c>. The actual encoding of the resulting string
 	(7-bit ascii, latin1 or utf8) can be obtained from <c>*result</c>. Both <c>was</c> and <c>result</c> can be <c>NULL</c>.
-	<c>*result</c> may differ from <c>want</c> if <c>want</c> is <c>ERLANG_WHATEVER</c> or if
-	<c>*result</c> turn out to be pure 7-bit ascii (compatible with both latin1 and utf8).</p>
+	
+	<c>*result</c> may differ from <c>want</c> if <c>want</c> is a bitwise-or'd combination like
+	<c>ERLANG_LATIN1|ERLANG_UTF8</c> or if <c>*result</c> turn out to be pure 7-bit ascii
+	(compatible with both latin1 and utf8).</p>
 	<p>This function fails if the atom is too long for the buffer
 	or if it can not be represented with encoding <c>want</c>.</p>
-	<p>This functions was introduced in R16 release of Erlang/OTP as part of a first step
+	<p>This function was introduced in R16 release of Erlang/OTP as part of a first step
 	to support UTF8 atoms.</p>
       </desc>
     </func>
diff --git a/lib/erl_interface/include/ei.h b/lib/erl_interface/include/ei.h
index 20e575f64d..2278a28adb 100644
--- a/lib/erl_interface/include/ei.h
+++ b/lib/erl_interface/include/ei.h
@@ -190,7 +190,12 @@ extern volatile int __erl_errno;
 #define MAXATOMLEN_UTF8 (255*4 + 1)
 #define MAXNODELEN EI_MAXALIVELEN+1+EI_MAXHOSTNAMELEN
 
-enum erlang_char_encoding { ERLANG_ASCII, ERLANG_LATIN1, ERLANG_UTF8, ERLANG_WHATEVER };
+enum erlang_char_encoding { 
+    ERLANG_ASCII = 1,
+    ERLANG_LATIN1 = 2,
+    ERLANG_UTF8 = 4,
+    ERLANG_ANY = ERLANG_ASCII|ERLANG_LATIN1|ERLANG_UTF8
+};
 
 /* a pid */
 typedef struct {
diff --git a/lib/erl_interface/src/decode/decode_atom.c b/lib/erl_interface/src/decode/decode_atom.c
index 2ada418243..556c400cb3 100644
--- a/lib/erl_interface/src/decode/decode_atom.c
+++ b/lib/erl_interface/src/decode/decode_atom.c
@@ -58,7 +58,7 @@ int ei_decode_atom_as(const char *buf, int *index, char* p, int destlen,
 	return -1;
     }
     
-    if (want_enc == got_enc || want_enc == ERLANG_WHATEVER || want_enc == ERLANG_ASCII) {
+    if ((want_enc & got_enc) || want_enc == ERLANG_ASCII) {
 	int i, found_non_ascii = 0;
 	if (len >= destlen)
 	    return -1;
diff --git a/lib/erl_interface/src/encode/encode_atom.c b/lib/erl_interface/src/encode/encode_atom.c
index 8bbe962396..044f17cb60 100644
--- a/lib/erl_interface/src/encode/encode_atom.c
+++ b/lib/erl_interface/src/encode/encode_atom.c
@@ -59,8 +59,7 @@ int ei_encode_atom_len_as(char *buf, int *index, const char *p, int len,
   char *s0 = s;
   int offs;
 
-  if (len >= MAXATOMLEN && (from_enc == ERLANG_LATIN1 ||
-			    from_enc == ERLANG_ASCII)) {
+  if (len >= MAXATOMLEN && (from_enc & (ERLANG_LATIN1|ERLANG_ASCII))) {
       return -1;
   }
 
diff --git a/lib/erl_interface/src/legacy/erl_marshal.c b/lib/erl_interface/src/legacy/erl_marshal.c
index 884e9d421b..4c45cebb02 100644
--- a/lib/erl_interface/src/legacy/erl_marshal.c
+++ b/lib/erl_interface/src/legacy/erl_marshal.c
@@ -662,7 +662,7 @@ static int read_atom(unsigned char** ext, Erl_Atom_data* a)
     int offs = 0;
     enum erlang_char_encoding enc;
     int ret = ei_decode_atom_as((char*)*ext, &offs, buf, MAXATOMLEN_UTF8,
-				ERLANG_WHATEVER, NULL, &enc);
+				ERLANG_LATIN1|ERLANG_UTF8, NULL, &enc);
     *ext += offs;
 
     if (ret == 0) {
@@ -674,11 +674,11 @@ static int read_atom(unsigned char** ext, Erl_Atom_data* a)
 	a->lenL = 0;
 	a->utf8 = NULL;
 	a->lenU = 0;
-	if (enc == ERLANG_LATIN1 || enc == ERLANG_ASCII) {
+	if (enc & (ERLANG_LATIN1 | ERLANG_ASCII)) {
 	    a->latin1 = clone; 	    
 	    a->lenL = i;
 	}
-	if (enc == ERLANG_UTF8 || enc == ERLANG_ASCII) {
+	if (enc & (ERLANG_UTF8 | ERLANG_ASCII)) {
 	    a->utf8 = clone;
 	    a->lenU = i;
 	}
-- 
cgit v1.2.3