问题
I generated a Lua Parser using the Lua.g grammar with antlr in java.
My lua code which i wish to parse basically looks like this
function uniqueid_some_event (e)
if (e:HasString("string1")) then
-- do something
end
if (e:HasString("string2")) then
-- do something
end
end
I have hundred of these events in different files for specific actor bindings.
Now i want to parse these files and gather for each what conditions are checked - in the above case - i want to extract "string1" and "string2" as event triggers. (To be more precise, i want to create a report that will display just the triggers for each file)
I gather i need to somehow modify the Lua.g to add my own logic in there but im lost because i dont find any documentation on this - i looked into LuaEclise which basically does some things but it does not work for me either.
So - is it possible to add to the generated LuaParser some kind of w3c DOM return value? Or something like getFunctions() which returns all functions found, and within each function getHasStringStatements() which would return the conditions?
回答1:
I don't recommend using the Lua grammar from the ANTLR wiki. AFAIK, there are quite a few things wrong with it (no proper long-strings and long-comments, invalid number/hex tokens, global backtracking to name just a few).
Here's a (IMO) better grammar for Lua 5.2:
/*
Copyright (c) 2011-2012 by Bart Kiers
Permission is hereby granted, free of charge, to any person
obtaining a copy of this software and associated documentation
files (the "Software"), to deal in the Software without
restriction, including without limitation the rights to use,
copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following
conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.
*/
grammar Lua52;
options {
output=AST;
ASTLabelType=CommonTree;
}
tokens {
// literals
And = 'and';
Break = 'break';
Do = 'do';
Else = 'else';
Elseif = 'elseif';
End = 'end';
False = 'false';
For = 'for';
Function = 'function';
Goto = 'goto';
If = 'if';
In = 'in';
Local = 'local';
Nil = 'nil';
Not = 'not';
Or = 'or';
Repeat = 'repeat';
Return = 'return';
Then = 'then';
True = 'true';
Until = 'until';
While = 'while';
Add = '+';
Minus = '-';
Mult = '*';
Div = '/';
Mod = '%';
Pow = '^';
Length = '#';
Eq = '==';
NEq = '~=';
LTEq = '<=';
GTEq = '>=';
LT = '<';
GT = '>';
Assign = '=';
OPar = '(';
CPar = ')';
OBrace = '{';
CBrace = '}';
OBrack = '[';
CBrack = ']';
ColCol = '::';
SCol = ';';
Col = ':';
Comma = ',';
DotDotDot = '...';
DotDot = '..';
Dot = '.';
// imaginary tokens
ASSIGNMENT;
LOCAL_ASSIGNMENT;
CONDITION;
UNARY_MINUS;
CALL;
COL_CALL;
INDEX;
EXPR_LIST;
VAR_LIST;
CHUNK;
NAME_LIST;
LABEL;
TABLE;
FIELD_LIST;
FIELD;
FOR_IN;
PARAM_LIST;
FUNCTION;
ASSIGNMENT_VAR;
VAR;
}
@parser::header {
package luja.parser;
}
@lexer::header {
package luja.parser;
import java.math.*;
}
@parser::members {
private boolean addSelf = false;
private CommonTree createPowAST(List tokens) {
int n = tokens.size();
CommonTree ast = new CommonTree(new CommonToken(Pow, "^"));
ast.addChild((CommonTree)tokens.get(n - 2));
ast.addChild((CommonTree)tokens.get(n - 1));
for(int i = n - 3; i >= 0; i--) {
CommonTree temp = new CommonTree(new CommonToken(Pow, "^"));
temp.addChild((CommonTree)tokens.get(i));
temp.addChild(ast);
ast = temp;
}
return ast;
}
private CommonTree namesToVar(List<String> names, String name) {
names.add(name);
return namesToVar(names);
}
private CommonTree namesToVar(List<String> names) {
if(names.size() == 1) {
return new CommonTree(new CommonToken(Name, names.get(0)));
}
CommonTree ast = new CommonTree(new CommonToken(VAR, "VAR"));
ast.addChild(new CommonTree(new CommonToken(Name, names.get(0))));
for(int i = 1; i < names.size(); i++) {
CommonTree indexNode = new CommonTree(new CommonToken(INDEX, "INDEX"));
indexNode.addChild(new CommonTree(new CommonToken(Name, names.get(i))));
ast.addChild(indexNode);
}
return ast;
}
@Override
public void reportError(RecognitionException e) {
throw new RuntimeException(e);
}
}
@lexer::members {
private boolean ahead(CharSequence chars) {
for(int i = 0; i < chars.length(); i++) {
if(input.LA(i + 1) != chars.charAt(i)) {
return false;
}
}
return true;
}
@Override
public void reportError(RecognitionException e) {
throw new RuntimeException(e);
}
private String unescape(String text) {
StringBuilder b = new StringBuilder();
String regex = "\\\\([\\\\abfnrtv\"']|\r?\n|\r|\\d{1,3}|x[0-9a-fA-F]{2}|z\\s*)|(?s).";
java.util.regex.Matcher m = java.util.regex.Pattern.compile(regex).matcher(text);
while(m.find()) {
if(m.group(1) != null) {
// an escaped char
String matched = m.group(1);
if(matched.equals("\\")) b.append("\\");
else if(matched.equals("a")) b.append("\u0007");
else if(matched.equals("b")) b.append("\u0008");
else if(matched.equals("f")) b.append("\u000C");
else if(matched.equals("n")) b.append("\n");
else if(matched.equals("r")) b.append("\r");
else if(matched.equals("t")) b.append("\t");
else if(matched.equals("v")) b.append("\u000B");
else if(matched.equals("\"")) b.append("\"");
else if(matched.equals("'")) b.append("'");
else if(matched.matches("\r?\n|\r")) b.append(matched);
else if(matched.matches("\\d{1,3}")) b.append((char)Integer.parseInt(matched));
else if(matched.matches("x[0-9a-fA-F]{2}")) b.append((char)Integer.parseInt(matched.substring(1), 16));
else if(matched.equals("z\\s*")) { /* do nothing, remove from string */ }
}
else {
// a normal char, append "as is"
b.append(m.group());
}
}
return b.toString();
}
}
//////////////////////////////// parser rules ////////////////////////////////
parse
: chunk EOF -> chunk
;
chunk
: stat* ret_stat? -> ^(CHUNK stat* ret_stat?)
;
stat
: (assignment)=> assignment
| var[false] // must be a function call, not an index: check and throw exception
| do_block
| while_stat
| repeat_stat
| local
| goto_stat
| if_stat
| for_stat
| function
| label
| Break
| ';' -> /* remove from AST (empty rewrite rule) */
;
do_block
: Do chunk End -> ^(Do chunk)
;
while_stat
: While expr do_block -> ^(While expr do_block)
;
repeat_stat
: Repeat chunk Until expr -> ^(Repeat chunk expr)
;
assignment
: var_list '=' expr_list // in every 'var' in 'var_list', the last must be an 'index', not a 'call'
-> ^(ASSIGNMENT ^(VAR_LIST var_list) ^(EXPR_LIST expr_list))
;
local
: Local ( name_list '=' expr_list -> ^(LOCAL_ASSIGNMENT ^(NAME_LIST name_list) ^(EXPR_LIST expr_list))
| Function Name func_body -> ^(LOCAL_ASSIGNMENT ^(NAME_LIST Name) ^(EXPR_LIST func_body))
)
;
goto_stat
: Goto Name -> ^(Goto Name)
;
if_stat
: If expr Then chunk elseif_stat* else_stat? End -> ^(If ^(CONDITION expr chunk) elseif_stat* else_stat?)
;
elseif_stat
: Elseif expr Then chunk -> ^(CONDITION expr chunk)
;
else_stat
: Else chunk -> ^(CONDITION True chunk)
;
for_stat
: For ( Name '=' a=expr ',' b=expr (',' c=expr)? do_block -> ^(For Name $a $b $c? do_block)
| name_list In expr_list do_block -> ^(FOR_IN ^(NAME_LIST name_list) ^(EXPR_LIST expr_list) do_block)
)
;
function
: Function names ( Col Name {addSelf=true;} func_body {addSelf=false;}
-> ^(ASSIGNMENT ^(VAR_LIST {namesToVar($names.list, $Name.text)}) ^(EXPR_LIST func_body))
| func_body
-> ^(ASSIGNMENT ^(VAR_LIST {namesToVar($names.list)}) ^(EXPR_LIST func_body))
)
;
names returns [List<String> list]
@init{$list = new ArrayList<String>();}
: a=Name {$list.add($a.text);} ('.' b=Name {$list.add($b.text);})*
;
function_literal
: Function func_body -> func_body
;
func_body
: '(' param_list ')' chunk End -> ^(FUNCTION param_list chunk)
;
param_list
: name_list (',' DotDotDot)? -> ^(PARAM_LIST name_list DotDotDot?)
| DotDotDot? -> ^(PARAM_LIST DotDotDot?)
;
ret_stat
: Return expr_list? ';'? -> ^(Return expr_list?)
;
expr
: or_expr
;
or_expr
: and_expr (Or^ and_expr)*
;
and_expr
: rel_expr (And^ rel_expr)*
;
rel_expr
: concat_expr ((LT | GT | LTEq | GTEq | NEq | Eq)^ concat_expr)?
;
concat_expr
: add_expr (DotDot^ add_expr)*
;
add_expr
: mult_expr ((Add | Minus)^ mult_expr)*
;
mult_expr
: unary_expr ((Mult | Div | Mod)^ unary_expr)*
;
unary_expr
: Minus unary_expr -> ^(UNARY_MINUS unary_expr)
| Length pow_expr -> ^(Length pow_expr)
| Not unary_expr -> ^(Not unary_expr)
| pow_expr
;
// right associative
pow_expr
// : (a=atom -> $a) ((Pow atom)+ -> ^(Pow atom+))?
: (a+=atom -> $a) ((Pow a+=atom)+ -> {createPowAST($a)})?
;
atom
: var[false]
| function_literal
| table_constructor
| DotDotDot
| Number
| String
| Nil
| True
| False
;
var[boolean assign]
: (callee[assign] -> callee) ( (tail)=> (((tail)=> t=tail)+ -> {assign}? ^(ASSIGNMENT_VAR callee tail+)
-> ^(VAR callee tail+))
)?
;
callee[boolean assign]
: '(' expr ')' -> expr
| Name
;
tail
: '.' Name -> ^(INDEX String[$Name.text])
| '[' expr ']' -> ^(INDEX expr)
| ':' Name '(' expr_list? ')' -> ^(INDEX {new CommonTree(new CommonToken(String, $Name.text))}) ^(COL_CALL expr_list?)
| ':' Name table_constructor -> ^(INDEX {new CommonTree(new CommonToken(String, $Name.text))}) ^(COL_CALL table_constructor)
| ':' Name String -> ^(INDEX {new CommonTree(new CommonToken(String, $Name.text))}) ^(COL_CALL String)
| '(' expr_list? ')' -> ^(CALL expr_list?)
| table_constructor -> ^(CALL table_constructor)
| String -> ^(CALL String)
;
table_constructor
: '{' field_list? '}' -> ^(TABLE field_list?)
;
field_list
: field (field_sep field)* field_sep? -> field+
;
field
: '[' expr ']' '=' expr -> ^(FIELD expr expr)
| Name '=' expr -> ^(FIELD {new CommonTree(new CommonToken(String, $Name.text))} expr)
| expr -> ^(FIELD expr)
;
field_sep
: ','
| ';'
;
label
: '::' Name '::' -> ^(LABEL Name)
;
var_list
: var[true] (',' var[true])* -> var+
;
expr_list
: expr (',' expr)* -> expr+
;
name_list
: Name (',' Name)* -> {addSelf}? {new CommonTree(new CommonToken(Name, "self"))} Name+
-> Name+
;
//////////////////////////////// lexer rules ////////////////////////////////
Name
: (Letter | '_') (Letter | '_' | Digit)*
;
Number
: (Digit+ ('.' Digit*)? Exponent? | '.' Digit+ Exponent?) {setText(new java.math.BigDecimal($text).toPlainString().replaceAll("\\.0*$", ""));}
| '0' ('x' | 'X') a=HexDigits ('.' b=HexDigits?)? c=BinaryExponent?
{
double num = Long.parseLong($a.text, 16);
if($b != null) {
double fraction = Long.parseLong($b.text, 16) / Math.pow(16, $b.text.length());
num += fraction;
}
if($c != null) {
int binExp = Integer.valueOf($c.text.contains("+") ? $c.text.substring(2) : $c.text.substring(1));
for(int i = 0; i < Math.abs(binExp); i++) {
num = binExp < 0 ? num/2 : num*2;
}
}
setText(new BigDecimal(Double.toString(num)).toPlainString().replaceAll("\\.0*$", ""));
}
;
String
: '"' (EscapeSequence | ~('\\' | '"' | '\r' | '\n'))* '"' {setText(unescape($text.substring(1, $text.length()-1)));}
| '\'' (EscapeSequence | ~('\\' | '\'' | '\r' | '\n'))* '\'' {setText(unescape($text.substring(1, $text.length()-1)));}
| LongBracket {setText($text.replaceAll("^\\[=*\\[|]=*]$", ""));}
;
//////////////////////////////// lexer rules to skip ////////////////////////////////
Comment
: '--' ( LongBracket
| '[' '='* ~('=' | '[') ~('\r' | '\n')* // matches '--[=====...' as a single line comment
| (~'[' ~('\r' | '\n')*)?
)
{skip();}
;
Space
: (' ' | '\t' | '\r' | '\n' | '\u000C')+ {skip();}
;
//////////////////////////////// fragment lexer rules ////////////////////////////////
fragment Letter
: 'a'..'z'
| 'A'..'Z'
;
fragment Digit
: '0'..'9'
;
fragment HexDigit
: Digit
| 'a'..'f'
| 'A'..'F'
;
fragment HexDigits
: HexDigit+
;
fragment Exponent
: ('e' | 'E') ('-' | '+')? Digit+
;
fragment BinaryExponent
: ('p' | 'P') ('-' | '+')? Digit+
;
fragment EscapeSequence
: '\\' ( ('a' | 'b' | 'f' | 'n' | 'r' | 't' | 'v' | '\\' | '"' | '\'' | 'z' | LineBreak)
| Digit (Digit Digit?)?
| 'x' HexDigit HexDigit
)
;
fragment LineBreak
: '\r'? '\n'
| '\r'
;
fragment LongBracket
@init{StringBuilder b = new StringBuilder("]");}
:
// match opening bracket and build equal sized closing bracket
'[' ('=' {b.append("=");})* '[' {b.append("]");}
// keep matching chars until the closing bracket is ahead
({!ahead(b)}?=> (~'\\' | EscapeSequence) )*
{
if(input.LA(1) == EOF) {
throw new RuntimeException("unfinished long comment or string near '<eof>'");
}
// let the lexer match the closing bracket
match(b.toString());
}
;
//////////////////////////////// a fall through rule throwing an exception ////////////////////////////////
Any
@after {throw new RuntimeException("unexpected symbol near: '" + $text + "'");}
: .
;
If you now parse the following input:
--[===[
function uniqueid_some_event (e)
if (e:HasString("ignore string1")) then
-- do something
end
if (e:HasString("ignore string2")) then
-- do something
end
end
some invalid closing comment tags: ]==] ]====]
]===]
function uniqueid_some_event (e)
if (e:HasString("string1")) then
-- do something
end
if (e:HasString("string2")) then
-- do something
end
end
if (e:HasString("outside function...")) then end
you will get the following AST returned from the generated parser:
Now all you need to do is walk through the AST, and when you stumble upon a ASSIGNMENT node, check if the right child is an expression list with a FUNCTION in it. If this happens, walk through this node to look for if-statements that have string-expressions in them.
Here's a start:
public class LUjATest {
private static void findFunctions(CommonTree tree) {
if (tree == null) return;
if(tree.getType() == Lua52Parser.ASSIGNMENT) {
String name = tree.getChild(0).getChild(0).getText();
CommonTree expressions = (CommonTree) tree.getChild(1);
if(expressions.getChildCount() > 0 &&
expressions.getChild(0).getType() == Lua52Parser.FUNCTION) {
System.out.println("walk the tree:\n " + expressions.toStringTree() +
"\nto find all strings for event: '" + name + "'");
}
}
else {
for (int i = 0; i < tree.getChildCount(); i++) {
findFunctions((CommonTree) tree.getChild(i));
}
}
}
public static void main(String[] args) throws Exception {
Lua52Lexer lexer = new Lua52Lexer(new ANTLRFileStream("src/lua/test.lua"));
Lua52Parser parser = new Lua52Parser(new CommonTokenStream(lexer));
CommonTree tree = (CommonTree) parser.parse().getTree();
findFunctions(tree);
}
}
When running the class above, you will see the following output:
walk the tree: (EXPR_LIST (FUNCTION (PARAM_LIST e) (CHUNK (if (CONDITION (VAR e (INDEX HasString) (COL_CALL string1)) CHUNK)) (if (CONDITION (VAR e (INDEX HasString) (COL_CALL string2)) CHUNK))))) to find all strings for event: 'uniqueid_some_event'
来源：https://stackoverflow.com/questions/11955707/using-antlr-to-parse-lua-if-statements-in-specific-functions