Bytestring: implement bytestring literals and constants

Implement byte string literals on the format b"xxx" and b64"xxx" which can be used as literals
and in constants.

The format b"xxx" supports character data and octal and
hexadecimal data using C escapes (\n, \nn, \nnn, \xn and \xnn).

The format b64"xxx" supports base64 encoded strings (RFC1341).

2 files changed, 68 insertions, 2 deletions

diff --git a/conf/cf-lex.l b/conf/cf-lex.l
index 704a1750..41520116 100644
--- a/conf/cf-lex.l
+++ b/conf/cf-lex.l
@@ -50,6 +50,7 @@
 #include "conf/cf-parse.tab.h"
 #include "lib/string.h"
 #include "lib/hash.h"
+#include "lib/base64.h"
 
 struct keyword {
   byte *name;
@@ -124,10 +125,11 @@ static enum yytokentype cf_lex_symbol(const char *data);
 %option nounput
 %option noreject
 
-%x COMMENT CCOMM CLI QUOTED APOSTROPHED INCLUDE
+%x COMMENT CCOMM CLI QUOTED BQUOTED B64QUOTED APOSTROPHED INCLUDE
 
 ALPHA [a-zA-Z_]
 DIGIT [0-9]
+OIGIT [0-7]
 XIGIT [0-9a-fA-F]
 ALNUM [a-zA-Z_0-9]
 WHITE [ \t]
@@ -367,6 +369,61 @@ else: {
 
 <QUOTED>.	BUFFER_PUSH(quoted_buffer) = yytext[0];
 
+[b]["] {
+  BEGIN(BQUOTED);
+  quoted_buffer_init();
+}
+
+<BQUOTED>\n	cf_error("Unterminated byte string");
+<BQUOTED><<EOF>> cf_error("Unterminated byte string");
+<BQUOTED>["]	{
+  BEGIN(INITIAL);
+  struct bytestring *bytes;
+
+  bytes = cfg_allocz(sizeof(*bytes) + quoted_buffer.used);
+  memcpy(bytes->data, quoted_buffer.data, quoted_buffer.used);
+  bytes->length = quoted_buffer.used;
+
+  cf_lval.bs = bytes;
+  return BYTESTRING;
+}
+
+<BQUOTED>.	BUFFER_PUSH(quoted_buffer) = yytext[0];
+
+<BQUOTED>[\\]({OIGIT}|{OIGIT}{OIGIT}|{OIGIT}{OIGIT}{OIGIT})	{
+  char buf[4];
+  char *end = NULL;
+  memcpy(buf, yytext + 1, yyleng);
+  buf[yyleng] = 0;
+  BUFFER_PUSH(quoted_buffer) = bstrtoul10(buf, &end);
+}
+
+<BQUOTED>[\\][x]({XIGIT}|{XIGIT}{XIGIT})	{
+  char buf[3];
+  char *end = NULL;
+  memcpy(buf, yytext + 2, yyleng);
+  buf[yyleng] = 0;
+  BUFFER_PUSH(quoted_buffer) = bstrtoul16(buf, &end);
+}
+
+[b][6][4]["] {
+  BEGIN(B64QUOTED);
+  quoted_buffer_init();
+}
+
+<B64QUOTED>\n	cf_error("Unterminated base64 string");
+<B64QUOTED><<EOF>> cf_error("Unterminated base64 string");
+<B64QUOTED>["]	{
+  BEGIN(INITIAL);
+  cf_lval.bs = base64_decode_bs(cfg_mem, quoted_buffer.data, quoted_buffer.used, NULL);
+  if (!cf_lval.bs)
+    cf_error("Invalid base64 string");
+
+  return BYTESTRING;
+}
+
+<B64QUOTED>.	BUFFER_PUSH(quoted_buffer) = yytext[0];
+
 <INITIAL,COMMENT><<EOF>>	{ if (check_eof()) return END; }
 
 {WHITE}+
diff --git a/conf/confbase.Y b/conf/confbase.Y
index 6985783b..99349c4e 100644
--- a/conf/confbase.Y
+++ b/conf/confbase.Y
@@ -92,7 +92,7 @@ CF_DECLS
   struct channel_limit cl;
   struct timeformat *tf;
   mpls_label_stack *mls;
-  struct bytestring *bs;
+  const struct bytestring *bs;
 }
 
 %token END CLI_MARKER INVALID_TOKEN ELSECOL DDOT
@@ -115,6 +115,7 @@ CF_DECLS
 %type <mls> label_stack_start label_stack
 
 %type <t> text opttext
+%type <bs> bytestring
 %type <s> symbol
 
 %nonassoc PREFIX_DUMMY
@@ -383,6 +384,14 @@ opttext:
  | /* empty */ { $$ = NULL; }
  ;
 
+bytestring:
+   BYTESTRING
+ | CF_SYM_KNOWN {
+     if ($1->class != (SYM_CONSTANT | T_BYTESTRING)) cf_error("Bytestring constant expected");
+     $$ = SYM_VAL($1).bs;
+   }
+ ;
+
 
 CF_CODE