revi-archive/phorge-arcanist
1
0
Fork 0
mirror of https://we.phorge.it/source/arcanist.git synced 2024-11-22 06:42:41 +01:00
Code Issues Releases Wiki Activity
phorge-arcanist/support/xhpast/scanner.l
epriestley 2e73916fa2 Initial commit.
2011-01-09 15:22:25 -08:00

1085 lines
28 KiB
Text
Executable file
Raw Blame History

/*
* Copyright 2011 Facebook, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
%{
#include "ast.hpp"
#define push_state(s) xhp_new_push_state(s, yyg)
#define pop_state() xhp_new_pop_state(yyg)
#define set_state(s) xhp_set_state(s, yyg)
#define last_token() yyextra->last_token
#define YY_USER_ACTION \
if (!yyg->yy_more_len) \
yyextra->first_lineno = yyextra->lineno;
#define pttok(t, txt) \
yyextra->token_list.push_back(new xhpast::Token(t, txt, yyextra->list_size++)); \
*yylval = new xhpast::Node(0, yyextra->list_size - 1);
#define ptok(t) \
pttok(t, yytext);
#define tok(t) \
ptok(t); \
return yy_token(t, yyg)
#define tokt(t) pttok(T_XHP_ENTITY, yytext); push_state(XHP_AFTER_ENT); return yyextra->last_token = T_XHP_ENTITY;
#define YY_USER_INIT \
if (yyextra->insert_token) { \
yyg->yy_init = 0; \
int ft = yyextra->insert_token; \
yyextra->insert_token = 0; \
return yy_token(ft, yyg); \
}
using namespace std;
const char* yytokname(int tok);
static int yy_token(int tok, struct yyguts_t* yyg);
static void yy_scan_newlines(const char* text, struct yyguts_t* yyg);
static bool utf8ize(uint32_t v, char* buf /* [5] */) {
if (v <= 0x7f) { // 0xxxxxxx
buf[0] = v;
buf[1] = 0;
} else if (v <= 0x7ff) { // 110yyyxx 10xxxxxx
buf[0] = 0xc0 | (v >> 6);
buf[1] = 0x80 | (v & 0x3f);
buf[2] = 0;
} else if (v <= 0xffff) { // 1110yyyy 10yyyyxx 10xxxxxx
buf[0] = 0xe0 | (v >> 12);
buf[1] = 0x80 | ((v >> 6) & 0x3f);
buf[2] = 0x80 | (v & 0x3f);
buf[3] = 0;
} else if (v <= 0x1fffff) { // 11110zzz 10zzyyyy 10yyyyxx 10xxxxxx
buf[0] = 0xf0 | (v >> 18);
buf[1] = 0x80 | ((v >> 12) & 0x3f);
buf[2] = 0x80 | ((v >> 6) & 0x3f);
buf[3] = 0x80 | (v & 0x3f);
buf[4] = 0;
} else {
return false;
}
return true;
}
%}
%option prefix="xhpast"
%option reentrant
/* PHP allows IF or if */
%option case-insensitive
%option noyywrap nodefault
%option stack
%option bison-bridge
%option 8bit
/* I think an interactive scanner is required because of the bison state
* pushing we do. I'm putting an explicit interactive declaration here in case
* someone tries adding -CF or whatever to the make flags. */
%option interactive
/* The different lexing states. Note that the transitions are done either
* in the lex actions, or in a generic manner in yy_token(). */
%s PHP
%s PHP_COMMENT
%s PHP_EOL_COMMENT
%s PHP_DOC_COMMENT
%s PHP_HEREDOC_START
%s PHP_HEREDOC_NSTART
%s PHP_HEREDOC_NEWLINE
%s PHP_HEREDOC_DATA
%s PHP_NO_RESERVED_WORDS
%s PHP_NO_RESERVED_WORDS_PERSIST
%s PHP_
%s XHP_LABEL
%s XHP_LABEL_WHITESPACE
%s XHP_ATTRS
%s XHP_ATTR_VAL
%s XHP_AFTER_ENT
%s XHP_CHILD
%s XHP_CHILD_START
%s XHP_INVALID_ENTITY
%s XHP_ATTR_TYPE_DECL
%s XHP_CHILDREN_DECL
LNUM [0-9]+
DNUM ([0-9]*"."[0-9]+)|([0-9]+"."[0-9]*)
EXPONENT_DNUM (({LNUM}|{DNUM})[eE][+-]?{LNUM})
HNUM "0x"[0-9a-fA-F]+
LABEL [a-zA-Z_\x7f-\xff][a-zA-Z0-9_\x7f-\xff]*
BYTE (.|\n)
WHITESPACE [ \n\r\t]+
TABS_AND_SPACES [ \t]*
NEWLINE ("\r\n"|"\n"|"\r")
%%
<XHP_ATTR_TYPE_DECL>{
"bool" tok(T_XHP_BOOLEAN);
"int" tok(T_XHP_NUMBER);
"float" tok(T_XHP_FLOAT);
"var" tok(T_VAR);
"array" tok(T_XHP_ARRAY);
"string" tok(T_XHP_STRING);
"enum" tok(T_XHP_ENUM);
@required tok(T_XHP_REQUIRED);
"(" tok('(');
":" tok(T_XHP_COLON);
}
/* Open / close PHP + inline HTML */
<INITIAL>{
"<?php"([ \t]|{NEWLINE}) {
yy_scan_newlines(yytext + 5, yyg);
// the state transition will be done in yy_token()
tok(T_OPEN_TAG);
}
"<?" {
if (yyextra->short_tags) {
tok(T_OPEN_TAG);
} else {
tok(T_INLINE_HTML);
}
}
"<?=" {
if (yyextra->short_tags) {
tok(T_OPEN_TAG_WITH_ECHO);
} else {
tok(T_INLINE_HTML);
}
}
"<%" {
if (yyextra->asp_tags) {
tok(T_OPEN_TAG);
} else {
tok(T_INLINE_HTML);
}
}
"<%=" {
if (yyextra->asp_tags) {
tok(T_OPEN_TAG_WITH_ECHO);
} else {
tok(T_INLINE_HTML);
}
}
"<"|[^<]* {
yy_scan_newlines(yytext, yyg);
tok(T_INLINE_HTML);
}
}
<PHP,PHP_NO_RESERVED_WORDS,PHP_NO_RESERVED_WORDS_PERSIST>{
("?>"|"</script"{WHITESPACE}*">"){NEWLINE}? {
yy_scan_newlines(yytext + 2, yyg);
tok(T_CLOSE_TAG);
}
"%>" {
if (yyextra->asp_tags) {
tok(T_CLOSE_TAG);
} else {
yyless(1);
tok(yytext[0]);
}
}
}
/* Comments and whitespace */
<PHP,PHP_NO_RESERVED_WORDS,PHP_NO_RESERVED_WORDS_PERSIST,XHP_CHILDREN_DECL,XHP_ATTR_TYPE_DECL>{
"#"|"//" {
push_state(PHP_EOL_COMMENT);
yymore();
}
"/**"{WHITESPACE} {
yy_scan_newlines(yytext + 3, yyg);
push_state(PHP_DOC_COMMENT);
yymore();
}
"/*" {
push_state(PHP_COMMENT);
yymore();
}
{WHITESPACE}+ {
yy_scan_newlines(yytext, yyg);
ptok(T_WHITESPACE);
}
}
<PHP_EOL_COMMENT>{
{NEWLINE} {
++yyextra->lineno;
ptok(T_COMMENT);
pop_state();
}
[^\r\n?]+ yymore();
"?>" {
yyless(yyleng - 2);
ptok(T_COMMENT);
pop_state();
}
. yymore();
}
<PHP_DOC_COMMENT,PHP_COMMENT>{
{NEWLINE} {
++yyextra->lineno;
yymore();
}
[^*\r\n]+|"*" yymore();
}
<PHP_DOC_COMMENT>"*/" {
ptok(T_DOC_COMMENT);
pop_state();
}
<PHP_COMMENT>"*/" {
ptok(T_COMMENT);
pop_state();
}
/* Reserved words */
<PHP>{
include tok(T_INCLUDE);
include_once tok(T_INCLUDE_ONCE);
eval tok(T_EVAL);
require tok(T_REQUIRE);
require_once tok(T_REQUIRE_ONCE);
or tok(T_LOGICAL_OR);
xor tok(T_LOGICAL_XOR);
and tok(T_LOGICAL_AND);
print tok(T_PRINT);
instanceof tok(T_INSTANCEOF);
new tok(T_NEW);
clone tok(T_CLONE);
exit tok(T_EXIT);
if tok(T_IF);
elseif tok(T_ELSEIF);
else tok(T_ELSE);
endif tok(T_ENDIF);
echo tok(T_ECHO);
do tok(T_DO);
while tok(T_WHILE);
endwhile tok(T_ENDWHILE);
for tok(T_FOR);
endfor tok(T_ENDFOR);
foreach tok(T_FOREACH);
endforeach tok(T_ENDFOREACH);
declare tok(T_DECLARE);
enddeclare tok(T_ENDDECLARE);
as tok(T_AS);
switch tok(T_SWITCH);
endswitch tok(T_ENDSWITCH);
case tok(T_CASE);
default tok(T_DEFAULT);
break tok(T_BREAK);
continue tok(T_CONTINUE);
goto tok(T_GOTO);
function tok(T_FUNCTION);
const tok(T_CONST);
return tok(T_RETURN);
try tok(T_TRY);
catch tok(T_CATCH);
throw tok(T_THROW);
use tok(T_USE);
global tok(T_GLOBAL);
static tok(T_STATIC);
abstract tok(T_ABSTRACT);
final tok(T_FINAL);
private tok(T_PRIVATE);
protected tok(T_PROTECTED);
public tok(T_PUBLIC);
var tok(T_VAR);
unset tok(T_UNSET);
isset tok(T_ISSET);
empty tok(T_EMPTY);
__halt_compiler tok(T_HALT_COMPILER);
class tok(T_CLASS);
interface tok(T_INTERFACE);
extends tok(T_EXTENDS);
implements tok(T_IMPLEMENTS);
list tok(T_LIST);
array tok(T_ARRAY);
__class__ tok(T_CLASS_C);
__method__ tok(T_METHOD_C);
__function__ tok(T_FUNC_C);
__line__ tok(T_LINE);
__file__ tok(T_FILE);
namespace tok(T_NAMESPACE);
__namespace__ tok(T_NS_C);
__dir__ tok(T_DIR);
attribute {
// expecting_xhp_class_statements is set in some actions in the grammar.
// This means the lexer and parser are interdependent.
if ((last_token() == '{' || last_token() == '}' || last_token() == ';') &&
(yyextra->expecting_xhp_class_statements)) {
tok(T_XHP_ATTRIBUTE);
} else {
tok(T_STRING);
}
}
category {
if ((last_token() == '{' || last_token() == '}' || last_token() == ';') &&
(yyextra->expecting_xhp_class_statements)) {
tok(T_XHP_CATEGORY);
} else {
tok(T_STRING);
}
}
children {
if ((last_token() == '{' || last_token() == '}' || last_token() == ';') &&
(yyextra->expecting_xhp_class_statements)) {
tok(T_XHP_CHILDREN);
} else {
tok(T_STRING);
}
}
}
/* Operators */
<PHP,PHP_NO_RESERVED_WORDS,PHP_NO_RESERVED_WORDS_PERSIST,XHP_ATTR_TYPE_DECL>{
"+=" tok(T_PLUS_EQUAL);
"-=" tok(T_MINUS_EQUAL);
"*=" tok(T_MUL_EQUAL);
"/=" tok(T_DIV_EQUAL);
".=" tok(T_CONCAT_EQUAL);
"%=" tok(T_MOD_EQUAL);
"&=" tok(T_AND_EQUAL);
"|=" tok(T_OR_EQUAL);
"^=" tok(T_XOR_EQUAL);
"<<=" tok(T_SL_EQUAL);
">>=" tok(T_SR_EQUAL);
"||" tok(T_BOOLEAN_OR);
"&&" tok(T_BOOLEAN_AND);
"==" tok(T_IS_EQUAL);
"!="|"<>" tok(T_IS_NOT_EQUAL);
"===" tok(T_IS_IDENTICAL);
"!==" tok(T_IS_NOT_IDENTICAL);
"<=" tok(T_IS_SMALLER_OR_EQUAL);
">=" tok(T_IS_GREATER_OR_EQUAL);
"<<" tok(T_SL);
">>" tok(T_SR);
"++" tok(T_INC);
"--" tok(T_DEC);
"->" tok(T_OBJECT_OPERATOR);
"=>" tok(T_DOUBLE_ARROW);
"::" tok(T_PAAMAYIM_NEKUDOTAYIM);
"\\" tok(T_NS_SEPARATOR);
":" {
// A colon can either mean the start (or component) of an XHP class,
// a ternary expression (as in 1?false:null), the colon of a 'case',
// or finally the start of a block in the old PHP syntax. The following
// disambiguate between the XHP case, which requires a special token,
// and the other cases.
switch (yyextra->last_token) {
// In a ternary expression, the colon must follow a full-fledged
// expression so seeing for instance a binary operator means
// it must be an XHP class.
case ',': case '=': case '|': case '^': case '&': case '<': case '>':
case '+': case '-': case '%': case '!': case '~': case '[': case '(':
case '{': case '.':
case T_LOGICAL_OR: case T_LOGICAL_XOR: case T_LOGICAL_AND:
case T_PLUS_EQUAL: case T_MINUS_EQUAL: case T_MUL_EQUAL:
case T_DIV_EQUAL: case T_CONCAT_EQUAL: case T_MOD_EQUAL:
case T_AND_EQUAL: case T_OR_EQUAL: case T_XOR_EQUAL:
case T_SL_EQUAL: case T_SR_EQUAL: case T_BOOLEAN_OR:
case T_BOOLEAN_AND: case T_IS_EQUAL: case T_IS_NOT_EQUAL:
case T_IS_IDENTICAL: case T_IS_NOT_IDENTICAL: case T_IS_SMALLER_OR_EQUAL:
case T_IS_GREATER_OR_EQUAL:
// An XHP class can also occur after certain keywords. Not sure
// we got them all covered though.
case T_ECHO: case T_RETURN:
case T_EXTENDS: case T_INSTANCEOF: case T_DOUBLE_ARROW:
case T_XHP_ATTRIBUTE:
tok(T_XHP_COLON);
break;
default:
tok(':');
break;
}
}
}
/* Casts */
<PHP,PHP_NO_RESERVED_WORDS,PHP_NO_RESERVED_WORDS_PERSIST>{
"("{TABS_AND_SPACES}(int|integer){TABS_AND_SPACES}")" tok(T_INT_CAST);
"("{TABS_AND_SPACES}(real|double|float){TABS_AND_SPACES}")" tok(T_DOUBLE_CAST);
"("{TABS_AND_SPACES}string{TABS_AND_SPACES}")" tok(T_STRING_CAST);
"("{TABS_AND_SPACES}unicode{TABS_AND_SPACES}")" tok(T_UNICODE_CAST);
"("{TABS_AND_SPACES}binary{TABS_AND_SPACES}")" tok(T_BINARY_CAST);
"("{TABS_AND_SPACES}array{TABS_AND_SPACES}")" tok(T_ARRAY_CAST);
"("{TABS_AND_SPACES}object{TABS_AND_SPACES}")" tok(T_OBJECT_CAST);
"("{TABS_AND_SPACES}(bool|boolean){TABS_AND_SPACES}")" tok(T_BOOL_CAST);
"("{TABS_AND_SPACES}unset{TABS_AND_SPACES}")" tok(T_UNSET_CAST);
}
/* Scalars (parsing these doesn't really matter since we just pass them through literally) */
<PHP,PHP_NO_RESERVED_WORDS,PHP_NO_RESERVED_WORDS_PERSIST,XHP_ATTR_TYPE_DECL>{
{LNUM}|{HNUM} tok(T_LNUMBER);
{DNUM}|{EXPONENT_DNUM} tok(T_DNUMBER);
{LABEL} tok(T_STRING);
"$"{LABEL} tok(T_VARIABLE);
b?'(\\.|\\\n|[^\\']+)*'|b?\"(\\.|\\\n|[^\\\"]+)*\" {
yy_scan_newlines(yytext, yyg);
tok(T_CONSTANT_ENCAPSED_STRING);
}
`[^`]*` {
yy_scan_newlines(yytext, yyg);
tok(T_BACKTICKS_EXPR);
}
}
/* (HERE|NOW)DOC's */
<PHP,PHP_NO_RESERVED_WORDS,PHP_NO_RESERVED_WORDS_PERSIST>b?"<<<"{TABS_AND_SPACES} {
push_state(PHP_HEREDOC_START);
yyextra->heredoc_yyleng = yyleng;
yymore();
}
<PHP_HEREDOC_START>{
"'"{LABEL}"'"|\"{LABEL}\" {
// Create a new string for the heredoc label. Since we're using yymore above
// yytext will actually start at the "<<<" and not the label. Use of
// heredoc_yyleng jumps past that. Then we add 1 to get past the " or '. The
// match is similar to calculate length.
yyextra->heredoc_label = string(yytext + yyextra->heredoc_yyleng + 1, yyleng - yyextra->heredoc_yyleng - 2);
set_state(PHP_HEREDOC_NSTART);
yyextra->heredoc_yyleng = yyleng;
yymore();
}
{LABEL} {
yyextra->heredoc_label = string(yytext + yyextra->heredoc_yyleng);
set_state(PHP_HEREDOC_NSTART);
yyextra->heredoc_yyleng = yyleng;
yymore();
}
}
<PHP_HEREDOC_NSTART>{NEWLINE} {
++yyextra->lineno;
yyextra->heredoc_data = yytext + yyleng;
set_state(PHP_HEREDOC_DATA);
yymore();
}
<PHP_HEREDOC_DATA>{
[^\r\n]*{NEWLINE} {
++yyextra->lineno;
set_state(PHP_HEREDOC_NEWLINE);
yyextra->heredoc_yyleng = yyleng;
yymore();
}
}
<PHP_HEREDOC_NEWLINE>{
{LABEL};?{NEWLINE} {
if (strncmp(yyextra->heredoc_label.c_str(), yytext + yyextra->heredoc_yyleng, yyextra->heredoc_label.size()) == 0) {
switch (yytext[yyextra->heredoc_yyleng + yyextra->heredoc_label.size()]) {
case ';': case '\n': case '\r':
yyless(yyleng - (yyleng - yyextra->heredoc_yyleng - yyextra->heredoc_label.size()));
pop_state();
tok(T_HEREDOC);
}
}
++yyextra->lineno;
yyextra->heredoc_yyleng = yyleng;
yymore();
}
[^\r\n]+ {
set_state(PHP_HEREDOC_DATA);
yyextra->heredoc_yyleng = yyleng;
yymore();
}
{NEWLINE} {
++yyextra->lineno;
yyextra->heredoc_yyleng = yyleng;
yymore();
}
}
/* XHP */
<XHP_LABEL_WHITESPACE>{
{WHITESPACE}+ {
yy_scan_newlines(yytext, yyg);
ptok(T_WHITESPACE);
}
}
<XHP_LABEL,XHP_LABEL_WHITESPACE>{
":" tok(T_XHP_COLON);
"-" tok(T_XHP_HYPHEN);
"::" {
pop_state();
yyextra->colon_hack = true;
tok(T_PAAMAYIM_NEKUDOTAYIM);
}
"--" {
pop_state();
tok(T_DEC);
}
{WHITESPACE} {
yy_scan_newlines(yytext, yyg);
pop_state();
tok(T_XHP_WHITESPACE);
}
{LABEL} tok(T_STRING);
. {
pop_state();
tok(yytext[0]);
}
}
<XHP_ATTRS>{
"="|"/"|">" tok(yytext[0]);
{WHITESPACE}+ {
yy_scan_newlines(yytext, yyg);
ptok(T_WHITESPACE);
}
{LABEL} tok(T_STRING);
}
<XHP_ATTR_VAL>{
[^&'\\"]+ tok(T_XHP_TEXT);
\" {
pop_state();
tok('"');
}
}
<XHP_CHILD_START>{
{WHITESPACE}+ {
/* ignore whitespace at the start */
yy_scan_newlines(yytext, yyg);
ptok(T_WHITESPACE);
set_state(XHP_CHILD);
}
. {
yyless(0);
set_state(XHP_CHILD);
}
}
/* Below we use tokt() (and not tok) which internally transits to
* the XHP_AFTER_ENT state. */
<XHP_CHILD,XHP_AFTER_ENT,XHP_ATTR_VAL>{
/* xml entities */
(?-i:&quot;) tokt("\"");
(?-i:&amp;) tokt("&");
(?-i:&apos;) tokt("\\'");
(?-i:&lt;) tokt("<")
(?-i:&gt;) tokt(">");
/* html entities */
(?-i:&nbsp;) tokt("\u00A0");
(?-i:&iexcl;) tokt("\u00A1");
(?-i:&cent;) tokt("\u00A2");
(?-i:&pound;) tokt("\u00A3");
(?-i:&curren;) tokt("\u00A4");
(?-i:&yen;) tokt("\u00A5");
(?-i:&brvbar;) tokt("\u00A6");
(?-i:&sect;) tokt("\u00A7");
(?-i:&uml;) tokt("\u00A8");
(?-i:&copy;) tokt("\u00A9");
(?-i:&ordf;) tokt("\u00AA");
(?-i:&laquo;) tokt("\u00AB");
(?-i:&not;) tokt("\u00AC");
(?-i:&shy;) tokt("\u00AD");
(?-i:&reg;) tokt("\u00AE");
(?-i:&macr;) tokt("\u00AF");
(?-i:&deg;) tokt("\u00B0");
(?-i:&plusmn;) tokt("\u00B1");
(?-i:&sup2;) tokt("\u00B2");
(?-i:&sup3;) tokt("\u00B3");
(?-i:&acute;) tokt("\u00B4");
(?-i:&micro;) tokt("\u00B5");
(?-i:&para;) tokt("\u00B6");
(?-i:&middot;) tokt("\u00B7");
(?-i:&cedil;) tokt("\u00B8");
(?-i:&sup1;) tokt("\u00B9");
(?-i:&ordm;) tokt("\u00BA");
(?-i:&raquo;) tokt("\u00BB");
(?-i:&frac14;) tokt("\u00BC");
(?-i:&frac12;) tokt("\u00BD");
(?-i:&frac34;) tokt("\u00BE");
(?-i:&iquest;) tokt("\u00BF");
(?-i:&Agrave;) tokt("\u00C0");
(?-i:&Aacute;) tokt("\u00C1");
(?-i:&Acirc;) tokt("\u00C2");
(?-i:&Atilde;) tokt("\u00C3");
(?-i:&Auml;) tokt("\u00C4");
(?-i:&Aring;) tokt("\u00C5");
(?-i:&AElig;) tokt("\u00C6");
(?-i:&Ccedil;) tokt("\u00C7");
(?-i:&Egrave;) tokt("\u00C8");
(?-i:&Eacute;) tokt("\u00C9");
(?-i:&Ecirc;) tokt("\u00CA");
(?-i:&Euml;) tokt("\u00CB");
(?-i:&Igrave;) tokt("\u00CC");
(?-i:&Iacute;) tokt("\u00CD");
(?-i:&Icirc;) tokt("\u00CE");
(?-i:&Iuml;) tokt("\u00CF");
(?-i:&ETH;) tokt("\u00D0");
(?-i:&Ntilde;) tokt("\u00D1");
(?-i:&Ograve;) tokt("\u00D2");
(?-i:&Oacute;) tokt("\u00D3");
(?-i:&Ocirc;) tokt("\u00D4");
(?-i:&Otilde;) tokt("\u00D5");
(?-i:&Ouml;) tokt("\u00D6");
(?-i:&times;) tokt("\u00D7");
(?-i:&Oslash;) tokt("\u00D8");
(?-i:&Ugrave;) tokt("\u00D9");
(?-i:&Uacute;) tokt("\u00DA");
(?-i:&Ucirc;) tokt("\u00DB");
(?-i:&Uuml;) tokt("\u00DC");
(?-i:&Yacute;) tokt("\u00DD");
(?-i:&THORN;) tokt("\u00DE");
(?-i:&szlig;) tokt("\u00DF");
(?-i:&agrave;) tokt("\u00E0");
(?-i:&aacute;) tokt("\u00E1");
(?-i:&acirc;) tokt("\u00E2");
(?-i:&atilde;) tokt("\u00E3");
(?-i:&auml;) tokt("\u00E4");
(?-i:&aring;) tokt("\u00E5");
(?-i:&aelig;) tokt("\u00E6");
(?-i:&ccedil;) tokt("\u00E7");
(?-i:&egrave;) tokt("\u00E8");
(?-i:&eacute;) tokt("\u00E9");
(?-i:&ecirc;) tokt("\u00EA");
(?-i:&euml;) tokt("\u00EB");
(?-i:&igrave;) tokt("\u00EC");
(?-i:&iacute;) tokt("\u00ED");
(?-i:&icirc;) tokt("\u00EE");
(?-i:&iuml;) tokt("\u00EF");
(?-i:&eth;) tokt("\u00F0");
(?-i:&ntilde;) tokt("\u00F1");
(?-i:&ograve;) tokt("\u00F2");
(?-i:&oacute;) tokt("\u00F3");
(?-i:&ocirc;) tokt("\u00F4");
(?-i:&otilde;) tokt("\u00F5");
(?-i:&ouml;) tokt("\u00F6");
(?-i:&divide;) tokt("\u00F7");
(?-i:&oslash;) tokt("\u00F8");
(?-i:&ugrave;) tokt("\u00F9");
(?-i:&uacute;) tokt("\u00FA");
(?-i:&ucirc;) tokt("\u00FB");
(?-i:&uuml;) tokt("\u00FC");
(?-i:&yacute;) tokt("\u00FD");
(?-i:&thorn;) tokt("\u00FE");
(?-i:&yuml;) tokt("\u00FF");
(?-i:&OElig;) tokt("\u0152");
(?-i:&oelig;) tokt("\u0153");
(?-i:&Scaron;) tokt("\u0160");
(?-i:&scaron;) tokt("\u0161");
(?-i:&Yuml;) tokt("\u0178");
(?-i:&fnof;) tokt("\u0192");
(?-i:&circ;) tokt("\u02C6");
(?-i:&tilde;) tokt("\u02DC");
(?-i:&Alpha;) tokt("\u0391");
(?-i:&Beta;) tokt("\u0392");
(?-i:&Gamma;) tokt("\u0393");
(?-i:&Delta;) tokt("\u0394");
(?-i:&Epsilon;) tokt("\u0395");
(?-i:&Zeta;) tokt("\u0396");
(?-i:&Eta;) tokt("\u0397");
(?-i:&Theta;) tokt("\u0398");
(?-i:&Iota;) tokt("\u0399");
(?-i:&Kappa;) tokt("\u039A");
(?-i:&Lambda;) tokt("\u039B");
(?-i:&Mu;) tokt("\u039C");
(?-i:&Nu;) tokt("\u039D");
(?-i:&Xi;) tokt("\u039E");
(?-i:&Omicron;) tokt("\u039F");
(?-i:&Pi;) tokt("\u03A0");
(?-i:&Rho;) tokt("\u03A1");
(?-i:&Sigma;) tokt("\u03A3");
(?-i:&Tau;) tokt("\u03A4");
(?-i:&Upsilon;) tokt("\u03A5");
(?-i:&Phi;) tokt("\u03A6");
(?-i:&Chi;) tokt("\u03A7");
(?-i:&Psi;) tokt("\u03A8");
(?-i:&Omega;) tokt("\u03A9");
(?-i:&alpha;) tokt("\u03B1");
(?-i:&beta;) tokt("\u03B2");
(?-i:&gamma;) tokt("\u03B3");
(?-i:&delta;) tokt("\u03B4");
(?-i:&epsilon;) tokt("\u03B5");
(?-i:&zeta;) tokt("\u03B6");
(?-i:&eta;) tokt("\u03B7");
(?-i:&theta;) tokt("\u03B8");
(?-i:&iota;) tokt("\u03B9");
(?-i:&kappa;) tokt("\u03BA");
(?-i:&lambda;) tokt("\u03BB");
(?-i:&mu;) tokt("\u03BC");
(?-i:&nu;) tokt("\u03BD");
(?-i:&xi;) tokt("\u03BE");
(?-i:&omicron;) tokt("\u03BF");
(?-i:&pi;) tokt("\u03C0");
(?-i:&rho;) tokt("\u03C1");
(?-i:&sigmaf;) tokt("\u03C2");
(?-i:&sigma;) tokt("\u03C3");
(?-i:&tau;) tokt("\u03C4");
(?-i:&upsilon;) tokt("\u03C5");
(?-i:&phi;) tokt("\u03C6");
(?-i:&chi;) tokt("\u03C7");
(?-i:&psi;) tokt("\u03C8");
(?-i:&omega;) tokt("\u03C9");
(?-i:&thetasym;) tokt("\u03D1");
(?-i:&upsih;) tokt("\u03D2");
(?-i:&piv;) tokt("\u03D6");
(?-i:&ensp;) tokt("\u2002");
(?-i:&emsp;) tokt("\u2003");
(?-i:&thinsp;) tokt("\u2009");
(?-i:&zwnj;) tokt("\u200C");
(?-i:&zwj;) tokt("\u200D");
(?-i:&lrm;) tokt("\u200E");
(?-i:&rlm;) tokt("\u200F");
(?-i:&ndash;) tokt("\u2013");
(?-i:&mdash;) tokt("\u2014");
(?-i:&lsquo;) tokt("\u2018");
(?-i:&rsquo;) tokt("\u2019");
(?-i:&sbquo;) tokt("\u201A");
(?-i:&ldquo;) tokt("\u201C");
(?-i:&rdquo;) tokt("\u201D");
(?-i:&bdquo;) tokt("\u201E");
(?-i:&dagger;) tokt("\u2020");
(?-i:&Dagger;) tokt("\u2021");
(?-i:&bull;) tokt("\u2022");
(?-i:&hellip;) tokt("\u2026");
(?-i:&permil;) tokt("\u2030");
(?-i:&prime;) tokt("\u2032");
(?-i:&Prime;) tokt("\u2033");
(?-i:&lsaquo;) tokt("\u2039");
(?-i:&rsaquo;) tokt("\u203A");
(?-i:&oline;) tokt("\u203E");
(?-i:&frasl;) tokt("\u2044");
(?-i:&euro;) tokt("\u20AC");
(?-i:&image;) tokt("\u2111");
(?-i:&weierp;) tokt("\u2118");
(?-i:&real;) tokt("\u211C");
(?-i:&trade;) tokt("\u2122");
(?-i:&alefsym;) tokt("\u2135");
(?-i:&larr;) tokt("\u2190");
(?-i:&uarr;) tokt("\u2191");
(?-i:&rarr;) tokt("\u2192");
(?-i:&darr;) tokt("\u2193");
(?-i:&harr;) tokt("\u2194");
(?-i:&crarr;) tokt("\u21B5");
(?-i:&lArr;) tokt("\u21D0");
(?-i:&uArr;) tokt("\u21D1");
(?-i:&rArr;) tokt("\u21D2");
(?-i:&dArr;) tokt("\u21D3");
(?-i:&hArr;) tokt("\u21D4");
(?-i:&forall;) tokt("\u2200");
(?-i:&part;) tokt("\u2202");
(?-i:&exist;) tokt("\u2203");
(?-i:&empty;) tokt("\u2205");
(?-i:&nabla;) tokt("\u2207");
(?-i:&isin;) tokt("\u2208");
(?-i:&notin;) tokt("\u2209");
(?-i:&ni;) tokt("\u220B");
(?-i:&prod;) tokt("\u220F");
(?-i:&sum;) tokt("\u2211");
(?-i:&minus;) tokt("\u2212");
(?-i:&lowast;) tokt("\u2217");
(?-i:&radic;) tokt("\u221A");
(?-i:&prop;) tokt("\u221D");
(?-i:&infin;) tokt("\u221E");
(?-i:&ang;) tokt("\u2220");
(?-i:&and;) tokt("\u2227");
(?-i:&or;) tokt("\u2228");
(?-i:&cap;) tokt("\u2229");
(?-i:&cup;) tokt("\u222A");
(?-i:&int;) tokt("\u222B");
(?-i:&there4;) tokt("\u2234");
(?-i:&sim;) tokt("\u223C");
(?-i:&cong;) tokt("\u2245");
(?-i:&asymp;) tokt("\u2248");
(?-i:&ne;) tokt("\u2260");
(?-i:&equiv;) tokt("\u2261");
(?-i:&le;) tokt("\u2264");
(?-i:&ge;) tokt("\u2265");
(?-i:&sub;) tokt("\u2282");
(?-i:&sup;) tokt("\u2283");
(?-i:&nsub;) tokt("\u2284");
(?-i:&sube;) tokt("\u2286");
(?-i:&supe;) tokt("\u2287");
(?-i:&oplus;) tokt("\u2295");
(?-i:&otimes;) tokt("\u2297");
(?-i:&perp;) tokt("\u22A5");
(?-i:&sdot;) tokt("\u22C5");
(?-i:&lceil;) tokt("\u2308");
(?-i:&rceil;) tokt("\u2309");
(?-i:&lfloor;) tokt("\u230A");
(?-i:&rfloor;) tokt("\u230B");
(?-i:&lang;) tokt("\u2329");
(?-i:&rang;) tokt("\u232A");
(?-i:&loz;) tokt("\u25CA");
(?-i:&spades;) tokt("\u2660");
(?-i:&clubs;) tokt("\u2663");
(?-i:&hearts;) tokt("\u2665");
(?-i:&diams;) tokt("\u2666");
/* awesome entities */
(?-i:&cloud;) tokt("\u2601");
(?-i:&umbrella;) tokt("\u2602");
(?-i:&snowman;) tokt("\u2603");
(?-i:&snowflake;) tokt("\u2745");
(?-i:&comet;) tokt("\u2604");
(?-i:&thunderstorm;) tokt("\u2608");
/* pseudo entities */
' tokt("\\'");
"\\" tokt("\\\\");
/* meta entities */
(?-i:&#[0-9]+;) {
char buf[5];
utf8ize(atoi(yytext + 2), buf);
tokt(buf);
}
(?-i:&#x)[A-F0-9]+; {
char buf[5];
char *_;
utf8ize(strtol(yytext + 3, &_, 16), buf);
tokt(buf);
}
/* not entities */
& {
yymore();
BEGIN(XHP_INVALID_ENTITY);
}
}
<XHP_INVALID_ENTITY>{
{BYTE}{1,10} {
for (char* ii = yytext; *ii; ++ii) {
if (*ii == ';') {
ii[1] = 0;
break;
}
}
if (!yyextra->terminated) {
yyextra->error = string("Invalid entity: (") + yytext + ")";
yyextra->terminated = true;
}
}
}
<XHP_AFTER_ENT>{
[ \t\x0b\x0c\xa0\r\n]|\r\n {
if (*yytext == '\r' || *yytext == '\n') {
// Since we rewrite newlines into space we need to increment both line
// counters. The first_lineno increment is quite a hack, and makes it so
// that this ent is on the wrong line but it doesn't mess up the rest of
// the file.
++yyextra->lineno;
++yyextra->first_lineno;
}
pop_state();
tok(T_XHP_TEXT);
}
. {
pop_state();
yyless(0);
}
}
<XHP_CHILD>{
[^&'<>\\{ \t\x0b\x0c\xa0\r\n]+{WHITESPACE}? {
yy_scan_newlines(yytext, yyg);
tok(T_XHP_TEXT);
}
{WHITESPACE}* {
yy_scan_newlines(yytext, yyg);
tok(T_XHP_TEXT);
}
/* TODO: I removed {WHITESPACE}* from all of these but it needs to be restored
if the tree is unflattened. */
"{" {
yy_scan_newlines(yytext, yyg);
tok('{');
}
"<" {
yy_scan_newlines(yytext, yyg);
tok('<');
}
"</" {
yy_scan_newlines(yytext, yyg);
tok(T_XHP_LT_DIV);
}
"</>" {
yy_scan_newlines(yytext, yyg);
tok(T_XHP_LT_DIV_GT);
}
}
<XHP_CHILDREN_DECL>{
any tok(T_XHP_ANY);
pcdata tok(T_XHP_PCDATA);
empty tok(T_XHP_EMPTY);
{LABEL} tok(T_STRING);
";" {
pop_state();
tok(';');
}
":" {
tok(T_XHP_COLON);
}
}
/* Other */
<*>{BYTE} {
tok(yytext[0]);
// fix unused function warnings
yy_top_state(NULL);
yyunput(0, 0, NULL);
}
%%
#ifdef DEBUG
static const char* yy_state_name(int state) {
switch (state) {
case INITIAL:
return "INITIAL";
case PHP:
return "PHP";
case PHP_COMMENT:
return "PHP_COMMENT";
case PHP_EOL_COMMENT:
return "PHP_EOL_COMMENT";
case PHP_DOC_COMMENT:
return "PHP_DOC_COMMENT";
case PHP_HEREDOC_START:
return "PHP_HEREDOC_START";
case PHP_HEREDOC_NSTART:
return "PHP_HEREDOC_NSTART";
case PHP_HEREDOC_NEWLINE:
return "PHP_HEREDOC_NEWLINE";
case PHP_HEREDOC_DATA:
return "PHP_HEREDOC_DATA";
case PHP_NO_RESERVED_WORDS:
return "PHP_NO_RESERVED_WORDS";
case PHP_NO_RESERVED_WORDS_PERSIST:
return "PHP_NO_RESERVED_WORDS_PERSIST";
case XHP_LABEL:
return "XHP_LABEL";
case XHP_LABEL_WHITESPACE:
return "XHP_LABEL_WHITESPACE";
case XHP_ATTRS:
return "XHP_ATTRS";
case XHP_ATTR_VAL:
return "XHP_ATTR_VAL";
case XHP_AFTER_ENT:
return "XHP_AFTER_ENT";
case XHP_CHILD:
return "XHP_CHILD";
case XHP_CHILD_START:
return "XHP_CHILD_START";
case XHP_INVALID_ENTITY:
return "XHP_INVALID_ENTITY";
case XHP_ATTR_TYPE_DECL:
return "XHP_ATTR_TYPE_DECL";
case XHP_CHILDREN_DECL:
return "XHP_CHILDREN_DECL";
default:
return "???";
}
}
static void yy_log_token(int tok) {
const char* tokname = yytokname(tok);
if (tokname) {
fprintf(stderr, "--> %s\n", tokname);
} else {
fprintf(stderr, "--> '%c'\n", tok);
}
}
#endif
static int yy_token(int tok, yyguts_t* yyg) {
if (YY_START == PHP_NO_RESERVED_WORDS) {
pop_state();
}
switch (tok) {
case T_OPEN_TAG:
case T_OPEN_TAG_WITH_ECHO:
case T_OPEN_TAG_FAKE:
push_state(PHP);
break;
case T_CLOSE_TAG:
pop_state();
// We need to return a ';', not a T_CLOSE_TAG, because a construct like
// "<?php echo $x ?>" is valid and there are about a billion parser rules
// which terminate with ';' so making a new rule like
// "semicolon_or_close_tag" would be hard. The token in yylval has the
// correct type and value, we just don't generate a node.
return ';';
// In PHP it's ok to use keywords such as 'if' as field names
// or function names.
case T_OBJECT_OPERATOR:
case T_FUNCTION:
push_state(PHP_NO_RESERVED_WORDS);
break;
case T_PAAMAYIM_NEKUDOTAYIM:
if (yyextra->colon_hack) {
yyextra->colon_hack = false;
} else {
push_state(PHP_NO_RESERVED_WORDS);
}
break;
case '{':
// not used anymore
yyextra->curly_stack.push(tok);
break;
}
#ifdef DEBUG
yy_log_token(tok);
#endif
return yyextra->last_token = tok;
}
static inline void yy_scan_newlines(const char* text, struct yyguts_t* yyg) {
for (; *text; ++text) {
if (*text == '\r') {
if (text[1] == '\n') {
++text;
}
++yyextra->lineno;
} else if (*text == '\n') {
++yyextra->lineno;
}
}
}
void xhp_new_push_state(int s, struct yyguts_t* yyg) {
#ifdef DEBUG
fprintf(stderr, "--> PUSH(%s -> %s)\n", yy_state_name(YY_START), yy_state_name(s));
#endif
yy_push_state(s, yyg);
}
void xhp_new_pop_state(struct yyguts_t* yyg) {
#ifdef DEBUG
int s = YY_START;
#endif
yy_pop_state(yyg);
#ifdef DEBUG
fprintf(stderr, "--> POP(%s -> %s)\n", yy_state_name(s), yy_state_name(YY_START));
#endif
}
void xhp_set_state(int s, struct yyguts_t* yyg) {
#ifdef DEBUG
fprintf(stderr, "--> SET(%s)\n", yy_state_name(s));
#endif
BEGIN(s);
}
View git blame Copy permalink