From afec2544f1793d11c47356c32ef74ce8c4035263 Mon Sep 17 00:00:00 2001
From: Stephen Dolan <mu@netsoc.tcd.ie>
Date: Sat, 20 Oct 2012 00:09:20 +0100
Subject: [PATCH] A string interpolation syntax that I don't hate. Also tests.

You can interpolate values into strings using \(this syntax), e.g.

    "best \("str" + "ing") ever"
---
 lexer.l       | 24 +++++-------------------
 opcode_list.h |  1 +
 parser.y      | 18 ++++++++++--------
 testdata      | 14 +++++++++++++-
 4 files changed, 29 insertions(+), 28 deletions(-)

diff --git a/lexer.l b/lexer.l
index a353d89bf2..1921f380fb 100644
--- a/lexer.l
+++ b/lexer.l
@@ -61,45 +61,31 @@
 }
 
 
-\"(\\.|[^\\\"])*\" |
 -?[0-9.]+([eE][+-]?[0-9]+)? { 
    yylval->literal = jv_parse_sized(yytext, yyleng); return LITERAL; 
 }
 
-\"(\\.|[^\\\"])* { 
-  yylval->literal = jv_invalid_with_msg(jv_string("Unterminated string"));
-  return LITERAL;
-}
-
-"@(" {
+"\"" {
   yy_push_state(IN_QQSTRING, yyscanner);
   return QQSTRING_START;
 }
 
 <IN_QQSTRING>{
-  "%(" {
+  "\\(" {
     return enter(QQSTRING_INTERP_START, YY_START, yyscanner);
   }
-  "%"[^(] {
-    return INVALID_CHARACTER;
-  }
-  ")" {
+  "\"" {
     yy_pop_state(yyscanner);
     return QQSTRING_END;
   }
-  "\\"[)%] {
-    char text[2] = {yytext[1], 0};
-    yylval->literal = jv_string(text);
-    return QQSTRING_TEXT;
-  }
-  (\\[^u)%]|\\u[a-zA-Z0-9]{0,4})+ {
+  (\\[^u(]|\\u[a-zA-Z0-9]{0,4})+ {
     /* pass escapes to the json parser */
     jv escapes = jv_string_fmt("\"%.*s\"", yyleng, yytext);
     yylval->literal = jv_parse_sized(jv_string_value(escapes), jv_string_length(jv_copy(escapes)));
     jv_free(escapes);
     return QQSTRING_TEXT;
   }
-  [^\\)%]+ {
+  [^\\\"]+ {
     yylval->literal = jv_string_sized(yytext, yyleng);
     return QQSTRING_TEXT;
   }
diff --git a/opcode_list.h b/opcode_list.h
index 30e7203d6f..306373fbf0 100644
--- a/opcode_list.h
+++ b/opcode_list.h
@@ -18,6 +18,7 @@ OP(INSERT, NONE,    4, 2)
 OP(ASSIGN, VARIABLE, 3, 0)
 
 OP(CALL_BUILTIN_1_1, CFUNC, 1, 1)
+OP(CALL_BUILTIN_2_1, CFUNC, 2, 1)
 OP(CALL_BUILTIN_3_1, CFUNC, 3, 1)
 
 OP(CALL_1_1, UFUNC, 1, 1)
diff --git a/parser.y b/parser.y
index 8c4b45663a..0effe07cdf 100644
--- a/parser.y
+++ b/parser.y
@@ -81,7 +81,7 @@
 %left '*' '/'
 
 
-%type <blk> Exp Term MkDict MkDictPair ExpD ElseBody QQString FuncDef FuncDefs
+%type <blk> Exp Term MkDict MkDictPair ExpD ElseBody QQString FuncDef FuncDefs String
 %{
 #include "lexer.gen.h"
 #define FAIL(loc, msg)                                   \
@@ -295,14 +295,19 @@ Exp ">=" Exp {
   $$ = gen_binop($1, $3, GREATEREQ);
 } |
 
-QQSTRING_START QQString QQSTRING_END {
-  $$ = $2;
+String {
+  $$ = $1;
 } |
 
 Term { 
   $$ = $1; 
 }
 
+String:
+QQSTRING_START QQString QQSTRING_END {
+  $$ = $2;
+}
+
 FuncDef:
 "def" IDENT ':' Exp ';' {
   block body = block_join($4, gen_op_simple(RET));
@@ -416,11 +421,8 @@ MkDictPair
 : IDENT ':' ExpD { 
   $$ = gen_dictpair(gen_op_const(LOADK, $1), $3);
  }
-| LITERAL ':' ExpD {
-  if (jv_get_kind($1) != JV_KIND_STRING) {
-    FAIL(@1, "Object keys must be strings");
-  }
-  $$ = gen_dictpair(gen_op_const(LOADK, $1), $3);
+| String ':' ExpD {
+  $$ = gen_dictpair($1, $3);
   }
 | IDENT {
   $$ = gen_dictpair(gen_op_const(LOADK, jv_copy($1)),
diff --git a/testdata b/testdata
index c37dee9304..ea0b1cc6f1 100644
--- a/testdata
+++ b/testdata
@@ -31,7 +31,19 @@ null
 null
 []
 
-# FIXME: string literals
+# We test escapes by matching them against Unicode codepoints
+# FIXME: more tests needed for weird unicode stuff (e.g. utf16 pairs)
+"Aa\r\n\t\b\f\u03bc"
+null
+"Aa\u000d\u000a\u0009\u0008\u000c\u03bc"
+
+.
+"Aa\r\n\t\b\f\u03bc"
+"Aa\u000d\u000a\u0009\u0008\u000c\u03bc"
+
+"inter\("pol" + "ation")"
+null
+"interpolation"
 
 #
 # Dictionary construction syntax