// (C) 2001, 2002 Ximian, Inc (http://www.ximian.com)
// (C) 2004 Novell, Inc
//
-
-/*
- * TODO:
- * Make sure we accept the proper Unicode ranges, per the spec.
- * Report error 1032
-*/
+//
using System;
using System.Text;
bool handle_assembly = false;
bool handle_constraints = false;
bool handle_typeof = false;
+ bool query_parsing;
Location current_location;
Location current_comment_location = Location.Null;
ArrayList escapedIdentifiers = new ArrayList ();
+ static bool linq;
+
//
// XML documentation buffer. The save point is used to divide
// comments on types and comments on members.
}
}
+ public static bool LinqEnabled {
+ set {
+ linq = value;
+ }
+ }
+
public bool TypeOfParsing {
get {
return handle_typeof;
}
}
+ public bool QueryParsing {
+ set {
+ query_parsing = value;
+ }
+ }
+
public XmlCommentState doc_state {
get { return xmlDocState; }
set {
// Class variables
//
static CharArrayHashtable[] keywords;
- static Hashtable keywordStrings = new Hashtable ();
+ static Hashtable keywordStrings;
static NumberStyles styles;
static NumberFormatInfo csharp_format_info;
public int putback_char;
public int previous_col;
public Stack ifstack;
-#if GMCS_SOURCES
public int parsing_generic_less_than;
-#endif
+ public int current_token;
+
public Position (Tokenizer t)
{
position = t.reader.Position;
previous_col = t.previous_col;
if (t.ifstack != null && t.ifstack.Count != 0)
ifstack = (Stack)t.ifstack.Clone ();
-#if GMCS_SOURCES
parsing_generic_less_than = t.parsing_generic_less_than;
-#endif
+ current_token = t.current_token;
}
}
putback_char = p.putback_char;
previous_col = p.previous_col;
ifstack = p.ifstack;
+ parsing_generic_less_than = p.parsing_generic_less_than;
+ current_token = p.current_token;
}
// Do not reset the position, ignore it.
position_stack.Pop ();
}
- static void AddKeyword (string kw, int token) {
+ static void AddKeyword (string kw, int token)
+ {
keywordStrings.Add (kw, kw);
if (keywords [kw.Length] == null) {
keywords [kw.Length] = new CharArrayHashtable (kw.Length);
static void InitTokens ()
{
+ keywordStrings = new Hashtable ();
keywords = new CharArrayHashtable [64];
AddKeyword ("__arglist", Token.ARGLIST);
AddKeyword ("partial", Token.PARTIAL);
#if GMCS_SOURCE
AddKeyword ("where", Token.WHERE);
+
+ if (linq) {
+ AddKeyword ("from", Token.FROM);
+ AddKeyword ("join", Token.JOIN);
+ AddKeyword ("on", Token.ON);
+ AddKeyword ("equals", Token.EQUALS);
+ AddKeyword ("select", Token.SELECT);
+ AddKeyword ("group", Token.GROUP);
+ AddKeyword ("by", Token.BY);
+ AddKeyword ("let", Token.LET);
+ AddKeyword ("orderby", Token.ORDERBY);
+ AddKeyword ("ascending", Token.ASCENDING);
+ AddKeyword ("descending", Token.DESCENDING);
+ AddKeyword ("into", Token.INTO);
+ }
#endif
}
// Class initializer
//
static Tokenizer ()
+ {
+ Reset ();
+ }
+
+ public static void Reset ()
{
InitTokens ();
csharp_format_info = NumberFormatInfo.InvariantInfo;
styles = NumberStyles.Float;
-
+
string_builder = new System.Text.StringBuilder ();
}
if (handle_assembly == false && res == Token.ASSEMBLY)
return -1;
#if GMCS_SOURCE
- if (handle_constraints == false && res == Token.WHERE)
+ if (linq) {
+ if (res == Token.FROM &&
+ (current_token == Token.ASSIGN || current_token == Token.OPEN_BRACKET ||
+ current_token == Token.RETURN || current_token == Token.IN)) {
+ query_parsing = true;
+ return res;
+ }
+
+ if (!query_parsing && res > Token.QUERY_FIRST_TOKEN && res < Token.QUERY_LAST_TOKEN)
+ return -1;
+
+ return res;
+ }
+
+ if (!handle_constraints && res == Token.WHERE)
return -1;
#endif
return res;
static bool is_identifier_part_character (char c)
{
- return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || c == '_' || (c >= '0' && c <= '9') || Char.IsLetter (c);
+ return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || c == '_' || (c >= '0' && c <= '9') ||
+ Char.IsLetter (c) || Char.GetUnicodeCategory (c) == UnicodeCategory.ConnectorPunctuation;
}
public static bool IsKeyword (string s)
return true;
}
-#if GMCS_SOURCE
bool parse_generic_dimension (out int dimension)
{
dimension = 1;
return false;
}
- int parsing_generic_less_than = 0;
-
+#if GMCS_SOURCE
public void PutbackNullable ()
{
if (nullable_pos < 0)
nullable_pos = -1;
}
#endif
+
+ public int peek_token ()
+ {
+ int the_token;
+
+ PushPosition ();
+ the_token = token ();
+ PopPosition ();
+
+ return the_token;
+ }
+
+ bool parse_namespace_or_typename (int next)
+ {
+ if (next == -1)
+ next = peek_token ();
+ while (next == Token.IDENTIFIER){
+ token ();
+ again:
+ next = peek_token ();
+ if (next == Token.DOT || next == Token.DOUBLE_COLON){
+ token ();
+ next = peek_token ();
+ continue;
+ }
+ if (next == Token.OP_GENERICS_LT){
+ token ();
+ if (!parse_less_than ())
+ return false;
+ goto again;
+ }
+ return true;
+ }
+
+ return false;
+ }
+
+ bool is_simple_type (int token)
+ {
+ return (token == Token.BOOL ||
+ token == Token.DECIMAL ||
+ token == Token.SBYTE ||
+ token == Token.BYTE ||
+ token == Token.SHORT ||
+ token == Token.USHORT ||
+ token == Token.INT ||
+ token == Token.UINT ||
+ token == Token.LONG ||
+ token == Token.ULONG ||
+ token == Token.CHAR ||
+ token == Token.FLOAT ||
+ token == Token.DOUBLE);
+ }
+
+ bool is_builtin_reference_type (int token)
+ {
+ return (token == Token.OBJECT || token == Token.STRING);
+ }
+
+ bool parse_opt_rank (int next)
+ {
+ while (true){
+ if (next != Token.OPEN_BRACKET)
+ return true;
+
+ token ();
+ while (true){
+ next = token ();
+ if (next == Token.CLOSE_BRACKET){
+ next = peek_token ();
+ break;
+ }
+ if (next == Token.COMMA)
+ continue;
+
+ return false;
+ }
+ }
+ }
+
+ bool parse_type ()
+ {
+ int next = peek_token ();
+
+ if (is_simple_type (next)){
+ token ();
+ next = peek_token ();
+ if (next == Token.INTERR)
+ token ();
+ return parse_opt_rank (peek_token ());
+ }
+ if (parse_namespace_or_typename (next)){
+ next = peek_token ();
+ if (next == Token.INTERR)
+ token ();
+ return parse_opt_rank (peek_token ());
+ } else if (is_builtin_reference_type (next)){
+ token ();
+ return parse_opt_rank (peek_token ());
+ }
+
+ return false;
+ }
+
+ //
+ // Invoked after '(' has been seen and tries to parse:
+ // type identifier [, type identifier]*
+ //
+ // if this is the case, instead of returning an
+ // OPEN_PARENS token we return a special token that
+ // triggers lambda parsing.
+ //
+ // This is needed because we can not introduce the
+ // explicitly_typed_lambda_parameter_list after a '(' in the
+ // grammar without introducing reduce/reduce conflicts.
+ //
+ // We need to parse a type and if it is followed by an
+ // identifier, we know it has to be parsed as a lambda
+ // expression.
+ //
+ // the type expression can be prefixed with `ref' or `out'
+ //
+ public bool parse_lambda_parameters ()
+ {
+ while (true){
+ int next = peek_token ();
+
+ if (next == Token.REF || next == Token.OUT)
+ token ();
+
+ if (parse_type ()){
+ next = peek_token ();
+ if (next == Token.IDENTIFIER){
+ token ();
+ next = peek_token ();
+ if (next == Token.COMMA){
+ token ();
+ continue;
+ }
+ if (next == Token.CLOSE_PARENS)
+ return true;
+ }
+ }
+ return false;
+ }
+ }
+
+ int parsing_generic_less_than = 0;
+
int is_punct (char c, ref bool doread)
{
int d;
case ']':
return Token.CLOSE_BRACKET;
case '(':
- return Token.OPEN_PARENS;
+ if (linq){
+ PushPosition ();
+ bool have_lambda_parameter = parse_lambda_parameters ();
+ PopPosition ();
+
+ if (have_lambda_parameter)
+ return Token.OPEN_PARENS_LAMBDA;
+ else
+ return Token.OPEN_PARENS;
+ } else
+ return Token.OPEN_PARENS;
case ')': {
if (deambiguate_close_parens == 0)
return Token.CLOSE_PARENS;
PopPosition ();
if (is_generic_lt) {
- parsing_generic_less_than++;
return Token.OP_GENERICS_LT;
} else
parsing_generic_less_than = 0;
doread = true;
return Token.OP_EQ;
}
+ if (d == '>'){
+ doread = true;
+ val = Location;
+ return Token.ARROW;
+ }
return Token.ASSIGN;
}
case "pragma":
if (RootContext.Version == LanguageVersion.ISO_1) {
- Report.FeatureIsNotStandardized (Location, "#pragma");
+ Report.FeatureIsNotISO1 (Location, "#pragma");
return true;
}
// Whether we have seen comments on the current line
bool comments_seen = false;
-
val = null;
for (;(c = getChar ()) != -1;) {
- if (c == ' ' || c == '\t' || c == '\f' || c == '\v' || c == 0xa0)
+ if (c == '\t'){
+ col = ((col + 8) / 8) * 8;
+ continue;
+ }
+
+ if (c == ' ' || c == '\f' || c == '\v' || c == 0xa0 || c == 0)
continue;
if (c == '\r') {
}
}
while ((d = getChar ()) != -1 && (d != '\n') && d != '\r')
- if (d == '\n'){
- }
+ if (d == '\n'){
+ }
any_token_seen |= tokens_seen;
tokens_seen = false;
comments_seen = false;