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

Merge pull request !337 from Almark Ming/master

master
黄亿华 2013-12-21 07:54:57 +08:00
parent 0c3ff3d6b1 91ed66ecac
commit 1bdffc1e56
1 changed files with 102 additions and 93 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
webmagic-core/src/main/java/us/codecraft/webmagic/selector/RegexSelector.java
View File

@ -26,12 +26,21 @@ public class RegexSelector implements Selector {
if (StringUtils.isBlank(regexStr)) {
throw new IllegalArgumentException("regex must not be empty");
}
/* Can't detect '\(', '(?:)' so that would result in ArrayIndexOutOfBoundsException
if (!StringUtils.contains(regexStr, "(") && !StringUtils.contains(regexStr, ")")) {
regexStr = "(" + regexStr + ")";
}
if (!StringUtils.contains(regexStr, "(") || !StringUtils.contains(regexStr, ")")) {
throw new IllegalArgumentException("regex must have capture group 1");
}
*/
// Try to fix: 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, "\\\\\\(?:")) {
regexStr = "(" + regexStr + ")";
}
this.regexStr = regexStr;
try {
regex = Pattern.compile(regexStr, Pattern.DOTALL | Pattern.CASE_INSENSITIVE);