1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/dom/xslt/xpath/txExprLexer.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,226 @@
1.4 +/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
1.5 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.6 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.7 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.8 +
1.9 +
1.10 +#ifndef MITREXSL_EXPRLEXER_H
1.11 +#define MITREXSL_EXPRLEXER_H
1.12 +
1.13 +#include "txCore.h"
1.14 +#include "nsString.h"
1.15 +
1.16 +/**
1.17 + * A Token class for the ExprLexer.
1.18 + *
1.19 + * This class was ported from XSL:P, an open source Java based
1.20 + * XSLT processor, written by yours truly.
1.21 + */
1.22 +class Token
1.23 +{
1.24 +public:
1.25 +
1.26 + /**
1.27 + * Token types
1.28 + */
1.29 + enum Type {
1.30 + //-- Trivial Tokens
1.31 + NULL_TOKEN = 1,
1.32 + LITERAL,
1.33 + NUMBER,
1.34 + CNAME,
1.35 + VAR_REFERENCE,
1.36 + PARENT_NODE,
1.37 + SELF_NODE,
1.38 + R_PAREN,
1.39 + R_BRACKET, // 9
1.40 + /**
1.41 + * start of tokens for 3.7, bullet 1
1.42 + * ExprLexer::nextIsOperatorToken bails if the tokens aren't
1.43 + * consecutive.
1.44 + */
1.45 + COMMA,
1.46 + AT_SIGN,
1.47 + L_PAREN,
1.48 + L_BRACKET,
1.49 + AXIS_IDENTIFIER,
1.50 +
1.51 + // These tokens include their following left parenthesis
1.52 + FUNCTION_NAME_AND_PAREN, // 15
1.53 + COMMENT_AND_PAREN,
1.54 + NODE_AND_PAREN,
1.55 + PROC_INST_AND_PAREN,
1.56 + TEXT_AND_PAREN,
1.57 +
1.58 + /**
1.59 + * operators
1.60 + */
1.61 + //-- boolean ops
1.62 + AND_OP, // 20
1.63 + OR_OP,
1.64 +
1.65 + //-- relational
1.66 + EQUAL_OP, // 22
1.67 + NOT_EQUAL_OP,
1.68 + LESS_THAN_OP,
1.69 + GREATER_THAN_OP,
1.70 + LESS_OR_EQUAL_OP,
1.71 + GREATER_OR_EQUAL_OP,
1.72 + //-- additive operators
1.73 + ADDITION_OP, // 28
1.74 + SUBTRACTION_OP,
1.75 + //-- multiplicative
1.76 + DIVIDE_OP, // 30
1.77 + MULTIPLY_OP,
1.78 + MODULUS_OP,
1.79 + //-- path operators
1.80 + PARENT_OP, // 33
1.81 + ANCESTOR_OP,
1.82 + UNION_OP,
1.83 + /**
1.84 + * end of tokens for 3.7, bullet 1 -/
1.85 + */
1.86 + //-- Special endtoken
1.87 + END // 36
1.88 + };
1.89 +
1.90 +
1.91 + /**
1.92 + * Constructors
1.93 + */
1.94 + typedef nsASingleFragmentString::const_char_iterator iterator;
1.95 +
1.96 + Token(iterator aStart, iterator aEnd, Type aType)
1.97 + : mStart(aStart),
1.98 + mEnd(aEnd),
1.99 + mType(aType),
1.100 + mNext(nullptr)
1.101 + {
1.102 + }
1.103 + Token(iterator aChar, Type aType)
1.104 + : mStart(aChar),
1.105 + mEnd(aChar + 1),
1.106 + mType(aType),
1.107 + mNext(nullptr)
1.108 + {
1.109 + }
1.110 +
1.111 + const nsDependentSubstring Value()
1.112 + {
1.113 + return Substring(mStart, mEnd);
1.114 + }
1.115 +
1.116 + iterator mStart, mEnd;
1.117 + Type mType;
1.118 + Token* mNext;
1.119 +};
1.120 +
1.121 +/**
1.122 + * A class for splitting an "Expr" String into tokens and
1.123 + * performing basic Lexical Analysis.
1.124 + *
1.125 + * This class was ported from XSL:P, an open source Java based XSL processor
1.126 + */
1.127 +
1.128 +class txExprLexer
1.129 +{
1.130 +public:
1.131 +
1.132 + txExprLexer();
1.133 + ~txExprLexer();
1.134 +
1.135 + /**
1.136 + * Parse the given string.
1.137 + * returns an error result if lexing failed.
1.138 + * The given string must outlive the use of the lexer, as the
1.139 + * generated Tokens point to Substrings of it.
1.140 + * mPosition points to the offending location in case of an error.
1.141 + */
1.142 + nsresult parse(const nsASingleFragmentString& aPattern);
1.143 +
1.144 + typedef nsASingleFragmentString::const_char_iterator iterator;
1.145 + iterator mPosition;
1.146 +
1.147 + /**
1.148 + * Functions for iterating over the TokenList
1.149 + */
1.150 +
1.151 + Token* nextToken();
1.152 + Token* peek()
1.153 + {
1.154 + NS_ASSERTION(mCurrentItem, "peek called uninitialized lexer");
1.155 + return mCurrentItem;
1.156 + }
1.157 + Token* peekAhead()
1.158 + {
1.159 + NS_ASSERTION(mCurrentItem, "peekAhead called on uninitialized lexer");
1.160 + // Don't peek past the end node
1.161 + return (mCurrentItem && mCurrentItem->mNext) ? mCurrentItem->mNext : mCurrentItem;
1.162 + }
1.163 + bool hasMoreTokens()
1.164 + {
1.165 + NS_ASSERTION(mCurrentItem, "HasMoreTokens called on uninitialized lexer");
1.166 + return (mCurrentItem && mCurrentItem->mType != Token::END);
1.167 + }
1.168 +
1.169 + /**
1.170 + * Trivial Tokens
1.171 + */
1.172 + //-- LF, changed to enum
1.173 + enum _TrivialTokens {
1.174 + D_QUOTE = '\"',
1.175 + S_QUOTE = '\'',
1.176 + L_PAREN = '(',
1.177 + R_PAREN = ')',
1.178 + L_BRACKET = '[',
1.179 + R_BRACKET = ']',
1.180 + L_ANGLE = '<',
1.181 + R_ANGLE = '>',
1.182 + COMMA = ',',
1.183 + PERIOD = '.',
1.184 + ASTERIX = '*',
1.185 + FORWARD_SLASH = '/',
1.186 + EQUAL = '=',
1.187 + BANG = '!',
1.188 + VERT_BAR = '|',
1.189 + AT_SIGN = '@',
1.190 + DOLLAR_SIGN = '$',
1.191 + PLUS = '+',
1.192 + HYPHEN = '-',
1.193 + COLON = ':',
1.194 + //-- whitespace tokens
1.195 + SPACE = ' ',
1.196 + TX_TAB = '\t',
1.197 + TX_CR = '\n',
1.198 + TX_LF = '\r'
1.199 + };
1.200 +
1.201 +private:
1.202 +
1.203 + Token* mCurrentItem;
1.204 + Token* mFirstItem;
1.205 + Token* mLastItem;
1.206 +
1.207 + int mTokenCount;
1.208 +
1.209 + void addToken(Token* aToken);
1.210 +
1.211 + /**
1.212 + * Returns true if the following Token should be an operator.
1.213 + * This is a helper for the first bullet of [XPath 3.7]
1.214 + * Lexical Structure
1.215 + */
1.216 + bool nextIsOperatorToken(Token* aToken);
1.217 +
1.218 + /**
1.219 + * Returns true if the given character represents a numeric letter (digit)
1.220 + * Implemented in ExprLexerChars.cpp
1.221 + */
1.222 + static bool isXPathDigit(char16_t ch)
1.223 + {
1.224 + return (ch >= '0' && ch <= '9');
1.225 + }
1.226 +};
1.227 +
1.228 +#endif
1.229 +
