Mobai
/
magic-Dependency
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #492 from eyougo/master 

fix a bug of RegexSelector when regex has zero-width assertions.
master
Yihua Huang 2017-03-17 06:55:07 +08:00 committed by GitHub
parent c175ea88c0 791520e6a0
commit 676758349f
2 changed files with 41 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
webmagic-core/src/main/java/us/codecraft/webmagic/selector/RegexSelector.java
View File

@ -28,8 +28,7 @@ public class RegexSelector implements Selector {
}
// Check bracket for regex group. Add default group 1 if there is no group.
// Only check if there exists the valid left parenthesis, leave regexp validation for Pattern.
if (StringUtils.countMatches(regexStr, "(") - StringUtils.countMatches(regexStr, "\\(") ==
StringUtils.countMatches(regexStr, "(?:") - StringUtils.countMatches(regexStr, "\\(?:")) {
if ( ! hasGroup(regexStr) ){
regexStr = "(" + regexStr + ")";
}
this.regexStr = regexStr;
@ -45,6 +44,30 @@ public class RegexSelector implements Selector {
this(regexStr, 1);
}
private boolean hasGroup(String regexStr) {
if (StringUtils.countMatches(regexStr, "(") - StringUtils.countMatches(regexStr, "\\(") ==
StringUtils.countMatches(regexStr, "(?:") - StringUtils.countMatches(regexStr, "\\(?:")){
return false;
}
if (StringUtils.countMatches(regexStr, "(") - StringUtils.countMatches(regexStr, "\\(") ==
StringUtils.countMatches(regexStr, "(?=") - StringUtils.countMatches(regexStr, "\\(?=") ) {
return false;
}
if (StringUtils.countMatches(regexStr, "(") - StringUtils.countMatches(regexStr, "\\(") ==
StringUtils.countMatches(regexStr, "(?<") - StringUtils.countMatches(regexStr, "\\(?<") ) {
return false;
}
if (StringUtils.countMatches(regexStr, "(") - StringUtils.countMatches(regexStr, "\\(") ==
StringUtils.countMatches(regexStr, "(?!") - StringUtils.countMatches(regexStr, "\\(?!") ) {
return false;
}
if (StringUtils.countMatches(regexStr, "(") - StringUtils.countMatches(regexStr, "\\(") ==
StringUtils.countMatches(regexStr, "(?#") - StringUtils.countMatches(regexStr, "\\(?#") ) {
return false;
}
return true;
}
@Override
public String select(String text) {
return selectGroup(text).get(group);

16
webmagic-core/src/test/java/us/codecraft/webmagic/selector/RegexSelectorTest.java
View File

@ -22,4 +22,20 @@ public class RegexSelectorTest {
String select = regexSelector.select(source);
Assertions.assertThat(select).isEqualTo(source);
}
@Test
public void testRegexWithZeroWidthAssertions() {
String regex = "^.*(?=\\?)";
String source = "hello world?xxxx";
RegexSelector regexSelector = new RegexSelector(regex);
String select = regexSelector.select(source);
Assertions.assertThat(select).isEqualTo("hello world");
regex = "\\d{3}(?!\\d)";
source = "123456asdf";
regexSelector = new RegexSelector(regex);
select = regexSelector.select(source);
Assertions.assertThat(select).isEqualTo("456");
}
}