com.itextpdf.styledxmlparser.jsoup.select.QueryParser Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of styled-xml-parser Show documentation
Show all versions of styled-xml-parser Show documentation
Styled XML parser is used by iText modules to parse HTML and XML
/*
This file is part of the iText (R) project.
Copyright (c) 1998-2021 iText Group NV
Authors: iText Software.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License version 3
as published by the Free Software Foundation with the addition of the
following permission added to Section 15 as permitted in Section 7(a):
FOR ANY PART OF THE COVERED WORK IN WHICH THE COPYRIGHT IS OWNED BY
ITEXT GROUP. ITEXT GROUP DISCLAIMS THE WARRANTY OF NON INFRINGEMENT
OF THIRD PARTY RIGHTS
This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
or FITNESS FOR A PARTICULAR PURPOSE.
See the GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program; if not, see http://www.gnu.org/licenses or write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
Boston, MA, 02110-1301 USA, or download the license from the following URL:
http://itextpdf.com/terms-of-use/
The interactive user interfaces in modified source and object code versions
of this program must display Appropriate Legal Notices, as required under
Section 5 of the GNU Affero General Public License.
In accordance with Section 7(b) of the GNU Affero General Public License,
a covered work must retain the producer line in every PDF that is created
or manipulated using iText.
You can be released from the requirements of the license by purchasing
a commercial license. Buying such a license is mandatory as soon as you
develop commercial activities involving the iText software without
disclosing the source code of your own applications.
These activities include: offering paid services to customers as an ASP,
serving PDFs on the fly in a web application, shipping iText with a closed
source product.
For more information, please contact iText Software Corp. at this
address: [email protected]
*/
package com.itextpdf.styledxmlparser.jsoup.select;
import com.itextpdf.styledxmlparser.jsoup.PortUtil;
import com.itextpdf.styledxmlparser.jsoup.helper.StringUtil;
import com.itextpdf.styledxmlparser.jsoup.helper.Validate;
import com.itextpdf.styledxmlparser.jsoup.parser.TokenQueue;
import java.util.ArrayList;
import java.util.List;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
/**
* Parses a CSS selector into an Evaluator tree.
*/
public class QueryParser {
private final static String[] combinators = {",", ">", "+", "~", " "};
private static final String[] AttributeEvals = new String[]{"=", "!=", "^=", "$=", "*=", "~="};
private TokenQueue tq;
private String query;
private List evals = new ArrayList();
/**
* Create a new QueryParser.
* @param query CSS query
*/
private QueryParser(String query) {
this.query = query;
this.tq = new TokenQueue(query);
}
/**
* Parse a CSS query into an Evaluator.
* @param query CSS query
* @return Evaluator
*/
public static Evaluator parse(String query) {
QueryParser p = new QueryParser(query);
return p.parse();
}
/**
* Parse the query
* @return Evaluator
*/
Evaluator parse() {
tq.consumeWhitespace();
if (tq.matchesAny(combinators)) { // if starts with a combinator, use root as elements
evals.add(new StructuralEvaluator.Root());
combinator(tq.consume());
} else {
findElements();
}
while (!tq.isEmpty()) {
// hierarchy and extras
boolean seenWhite = tq.consumeWhitespace();
if (tq.matchesAny(combinators)) {
combinator(tq.consume());
} else if (seenWhite) {
combinator(' ');
} else { // E.class, E#id, E[attr] etc. AND
findElements(); // take next el, #. etc off queue
}
}
if (evals.size() == 1)
return evals.get(0);
return new CombiningEvaluator.And(evals);
}
private void combinator(char combinator) {
tq.consumeWhitespace();
String subQuery = consumeSubQuery(); // support multi > childs
Evaluator rootEval; // the new topmost evaluator
Evaluator currentEval; // the evaluator the new eval will be combined to. could be root, or rightmost or.
Evaluator newEval = parse(subQuery); // the evaluator to add into target evaluator
boolean replaceRightMost = false;
if (evals.size() == 1) {
rootEval = currentEval = evals.get(0);
// make sure OR (,) has precedence:
if (rootEval instanceof CombiningEvaluator.Or && combinator != ',') {
currentEval = ((CombiningEvaluator.Or) currentEval).rightMostEvaluator();
replaceRightMost = true;
}
}
else {
rootEval = currentEval = new CombiningEvaluator.And(evals);
}
evals.clear();
// for most combinators: change the current eval into an AND of the current eval and the new eval
if (combinator == '>')
currentEval = new CombiningEvaluator.And(newEval, new StructuralEvaluator.ImmediateParent(currentEval));
else if (combinator == ' ')
currentEval = new CombiningEvaluator.And(newEval, new StructuralEvaluator.Parent(currentEval));
else if (combinator == '+')
currentEval = new CombiningEvaluator.And(newEval, new StructuralEvaluator.ImmediatePreviousSibling(currentEval));
else if (combinator == '~')
currentEval = new CombiningEvaluator.And(newEval, new StructuralEvaluator.PreviousSibling(currentEval));
else if (combinator == ',') { // group or.
CombiningEvaluator.Or or;
if (currentEval instanceof CombiningEvaluator.Or) {
or = (CombiningEvaluator.Or) currentEval;
or.add(newEval);
} else {
or = new CombiningEvaluator.Or();
or.add(currentEval);
or.add(newEval);
}
currentEval = or;
}
else
throw new Selector.SelectorParseException("Unknown combinator: " + combinator);
if (replaceRightMost)
((CombiningEvaluator.Or) rootEval).replaceRightMostEvaluator(currentEval);
else rootEval = currentEval;
evals.add(rootEval);
}
private String consumeSubQuery() {
StringBuilder sq = new StringBuilder();
while (!tq.isEmpty()) {
if (tq.matches("("))
sq.append("(").append(tq.chompBalanced('(', ')')).append(")");
else if (tq.matches("["))
sq.append("[").append(tq.chompBalanced('[', ']')).append("]");
else if (tq.matchesAny(combinators))
break;
else
sq.append(tq.consume());
}
return sq.toString();
}
private void findElements() {
if (tq.matchChomp("#"))
byId();
else if (tq.matchChomp("."))
byClass();
else if (tq.matchesWord())
byTag();
else if (tq.matches("["))
byAttribute();
else if (tq.matchChomp("*"))
allElements();
else if (tq.matchChomp(":lt("))
indexLessThan();
else if (tq.matchChomp(":gt("))
indexGreaterThan();
else if (tq.matchChomp(":eq("))
indexEquals();
else if (tq.matches(":has("))
has();
else if (tq.matches(":contains("))
contains(false);
else if (tq.matches(":containsOwn("))
contains(true);
else if (tq.matches(":matches("))
matches(false);
else if (tq.matches(":matchesOwn("))
matches(true);
else if (tq.matches(":not("))
not();
else if (tq.matchChomp(":nth-child("))
cssNthChild(false, false);
else if (tq.matchChomp(":nth-last-child("))
cssNthChild(true, false);
else if (tq.matchChomp(":nth-of-type("))
cssNthChild(false, true);
else if (tq.matchChomp(":nth-last-of-type("))
cssNthChild(true, true);
else if (tq.matchChomp(":first-child"))
evals.add(new Evaluator.IsFirstChild());
else if (tq.matchChomp(":last-child"))
evals.add(new Evaluator.IsLastChild());
else if (tq.matchChomp(":first-of-type"))
evals.add(new Evaluator.IsFirstOfType());
else if (tq.matchChomp(":last-of-type"))
evals.add(new Evaluator.IsLastOfType());
else if (tq.matchChomp(":only-child"))
evals.add(new Evaluator.IsOnlyChild());
else if (tq.matchChomp(":only-of-type"))
evals.add(new Evaluator.IsOnlyOfType());
else if (tq.matchChomp(":empty"))
evals.add(new Evaluator.IsEmpty());
else if (tq.matchChomp(":root"))
evals.add(new Evaluator.IsRoot());
else // unhandled
throw new Selector.SelectorParseException("Could not parse query " + PortUtil.escapedSingleBracket + "{0}" + PortUtil.escapedSingleBracket + ": unexpected token at " + PortUtil.escapedSingleBracket + "{1}" + PortUtil.escapedSingleBracket, query, tq.remainder());
}
private void byId() {
String id = tq.consumeCssIdentifier();
Validate.notEmpty(id);
evals.add(new Evaluator.Id(id));
}
private void byClass() {
String className = tq.consumeCssIdentifier();
Validate.notEmpty(className);
evals.add(new Evaluator.Class(className.trim().toLowerCase()));
}
private void byTag() {
String tagName = tq.consumeElementSelector();
Validate.notEmpty(tagName);
// namespaces: if element name is "abc:def", selector must be "abc|def", so flip:
if (tagName.contains("|"))
tagName = tagName.replace("|", ":");
evals.add(new Evaluator.Tag(tagName.trim().toLowerCase()));
}
private void byAttribute() {
TokenQueue cq = new TokenQueue(tq.chompBalanced('[', ']')); // content queue
String key = cq.consumeToAny(AttributeEvals); // eq, not, start, end, contain, match, (no val)
Validate.notEmpty(key);
cq.consumeWhitespace();
if (cq.isEmpty()) {
if (key.startsWith("^"))
evals.add(new Evaluator.AttributeStarting(key.substring(1)));
else
evals.add(new Evaluator.Attribute(key));
} else {
if (cq.matchChomp("="))
evals.add(new Evaluator.AttributeWithValue(key, cq.remainder()));
else if (cq.matchChomp("!="))
evals.add(new Evaluator.AttributeWithValueNot(key, cq.remainder()));
else if (cq.matchChomp("^="))
evals.add(new Evaluator.AttributeWithValueStarting(key, cq.remainder()));
else if (cq.matchChomp("$="))
evals.add(new Evaluator.AttributeWithValueEnding(key, cq.remainder()));
else if (cq.matchChomp("*="))
evals.add(new Evaluator.AttributeWithValueContaining(key, cq.remainder()));
else if (cq.matchChomp("~="))
evals.add(new Evaluator.AttributeWithValueMatching(key, Pattern.compile(cq.remainder())));
else
throw new Selector.SelectorParseException("Could not parse attribute query " + PortUtil.escapedSingleBracket + "{0}" + PortUtil.escapedSingleBracket + ": unexpected token at " + PortUtil.escapedSingleBracket + "{1}" + PortUtil.escapedSingleBracket, query, cq.remainder());
}
}
private void allElements() {
evals.add(new Evaluator.AllElements());
}
// pseudo selectors :lt, :gt, :eq
private void indexLessThan() {
evals.add(new Evaluator.IndexLessThan(consumeIndex()));
}
private void indexGreaterThan() {
evals.add(new Evaluator.IndexGreaterThan(consumeIndex()));
}
private void indexEquals() {
evals.add(new Evaluator.IndexEquals(consumeIndex()));
}
//pseudo selectors :first-child, :last-child, :nth-child, ...
private static final Pattern NTH_AB = Pattern.compile("((\\+|-)?(\\d+)?)n(\\s*(\\+|-)?\\s*\\d+)?", Pattern.CASE_INSENSITIVE);
private static final Pattern NTH_B = Pattern.compile("(\\+|-)?(\\d+)");
private void cssNthChild(boolean backwards, boolean ofType) {
String argS = tq.chompTo(")").trim().toLowerCase();
Matcher mAB = NTH_AB.matcher(argS);
Matcher mB = NTH_B.matcher(argS);
final int a, b;
if ("odd".equals(argS)) {
a = 2;
b = 1;
} else if ("even".equals(argS)) {
a = 2;
b = 0;
} else if (mAB.matches()) {
a = mAB.group(3) != null ? Integer.parseInt(mAB.group(1).replaceFirst("^\\+", "")) : 1;
b = mAB.group(4) != null ? Integer.parseInt(mAB.group(4).replaceFirst("^\\+", "")) : 0;
} else if (mB.matches()) {
a = 0;
b = Integer.parseInt(mB.group().replaceFirst("^\\+", ""));
} else {
throw new Selector.SelectorParseException("Could not parse nth-index " + PortUtil.escapedSingleBracket + "{0}" + PortUtil.escapedSingleBracket + ": unexpected format", argS);
}
if (ofType)
if (backwards)
evals.add(new Evaluator.IsNthLastOfType(a, b));
else
evals.add(new Evaluator.IsNthOfType(a, b));
else {
if (backwards)
evals.add(new Evaluator.IsNthLastChild(a, b));
else
evals.add(new Evaluator.IsNthChild(a, b));
}
}
private int consumeIndex() {
String indexS = tq.chompTo(")").trim();
Validate.isTrue(StringUtil.isNumeric(indexS), "Index must be numeric");
return Integer.parseInt(indexS);
}
// pseudo selector :has(el)
private void has() {
tq.consume(":has");
String subQuery = tq.chompBalanced('(', ')');
Validate.notEmpty(subQuery, ":has(el) subselect must not be empty");
evals.add(new StructuralEvaluator.Has(parse(subQuery)));
}
// pseudo selector :contains(text), containsOwn(text)
private void contains(boolean own) {
tq.consume(own ? ":containsOwn" : ":contains");
String searchText = TokenQueue.unescape(tq.chompBalanced('(', ')'));
Validate.notEmpty(searchText, ":contains(text) query must not be empty");
if (own)
evals.add(new Evaluator.ContainsOwnText(searchText));
else
evals.add(new Evaluator.ContainsText(searchText));
}
// :matches(regex), matchesOwn(regex)
private void matches(boolean own) {
tq.consume(own ? ":matchesOwn" : ":matches");
String regex = tq.chompBalanced('(', ')'); // don't unescape, as regex bits will be escaped
Validate.notEmpty(regex, ":matches(regex) query must not be empty");
if (own)
evals.add(new Evaluator.MatchesOwn(Pattern.compile(regex)));
else
evals.add(new Evaluator.Matches(Pattern.compile(regex)));
}
// :not(selector)
private void not() {
tq.consume(":not");
String subQuery = tq.chompBalanced('(', ')');
Validate.notEmpty(subQuery, ":not(selector) subselect must not be empty");
evals.add(new StructuralEvaluator.Not(parse(subQuery)));
}
}