dorkbox/SystemTray
2
1
Fork 1
Code Issues Pull Requests Releases 5 Wiki Activity

Fixed issues with unbalanced () during parsing CSS attribute values.

Browse Source
This commit is contained in:
nathan 2017-07-02 02:34:23 +02:00
parent e7e9cddda9
commit 0caf22de16
1 changed files with 50 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

66
src/dorkbox/systemTray/util/CssParser.java
View File

@ -3,11 +3,11 @@ package dorkbox.systemTray.util;
import java.io.File;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Iterator;
import java.util.List;
import dorkbox.util.FileUtil;
import dorkbox.util.OS;
/**
* A simple, basic CSS parser
@ -42,7 +42,16 @@ class CssParser {
@Override
public
String toString() {
return label + "\n\t" + Arrays.toString(attributes.toArray());
StringBuilder builder = new StringBuilder();
for (CssAttribute attribute : attributes) {
builder.append("\t")
.append(attribute.key)
.append(" : ")
.append(attribute.value)
.append(OS.LINE_SEPARATOR);
}
return label + "\n" + builder.toString();
}
}
@ -129,26 +138,51 @@ class CssParser {
// split the section into an arrayList, one per item. Split by attribute element
String nodeSection = css.substring(endOfNodeLabels, endOfSection);
int start = nodeSection.indexOf('{') + 1;
while (start != -1) {
int end = nodeSection.indexOf(';', start);
int sectionStart = nodeSection.indexOf('{') + 1;
int sectionEnd = nodeSection.indexOf('}');
while (sectionStart != -1) {
int end = nodeSection.indexOf(';', sectionStart);
if (end != -1) {
int separator = nodeSection.indexOf(':', start);
int separator = nodeSection.indexOf(':', sectionStart);
// String.charAt() is a bit slow because of all the extra range checking it performes
if (separator < end) {
String key = nodeSection.substring(start, separator);
String value = nodeSection.substring(separator + 1, end);
// the parenthesis must be balanced for the value
short parenCount = 0;
if (value.contains("url(") || value.contains("url (")) {
// this is really a URL , so we have to goto to the closing ');'
end = nodeSection.indexOf(')', end);
end = nodeSection.indexOf(';', end);
value = nodeSection.substring(separator + 1, end);
int j = separator;
while (j < end) {
j++;
if (nodeSection.charAt(j) == '(') {
parenCount++;
}
else if (nodeSection.charAt(j) == ')') {
parenCount--;
}
}
j--;
if (parenCount > 0) {
do {
// find the extended balancing paren
if (nodeSection.charAt(j) == ')') {
parenCount--;
}
j++;
} while (parenCount > 0 && j < sectionEnd);
end = j;
} else {
end = j+1;
}
String key = nodeSection.substring(sectionStart, separator);
String value = nodeSection.substring(separator + 1, end);
attributes.add(new CssAttribute(key, value));
}
start = end + 1;
sectionStart = end + 1;
}
else {
break;
@ -382,7 +416,7 @@ class CssParser {
int maxValue = -1;
CssNode maxNode = null; // guaranteed to be non-null
CssNode maxNode = null;
// not the CLEANEST way to do this, but it works by only choosing css nodes that are important
for (CssNode cssNode : sectionsWithAttribute) {