// 过滤整数字符，把所有非0～9的字符全部删除
private void filterInt(StringBuilder builder) {
for (int i = builder.length() - 1; i >= 0; i--) {
int cp = builder.codePointAt(i);
if (cp > '9' || cp < '0') {
builder.deleteCharAt(i);
}
}
}
public void insertString(FilterBypass fb, int offset, String string, AttributeSet attr)
throws BadLocationException {
StringBuilder builder = new StringBuilder(string);
for (int i = builder.length() - 1; i >= 0; i--) {
int cp = builder.codePointAt(i);
if (!Character.isDigit(cp) && cp != '-') {
builder.deleteCharAt(i);
if (Character.isSupplementaryCodePoint(cp)) {
i--;
builder.deleteCharAt(i);
}
}
}
super.insertString(fb, offset, builder.toString(), attr);
}
private StringBuilder sanitize(final StringBuilder aContent) {
int i = 0;
int lastBreak = 0;
while (i < aContent.length()) {
// Check valid unicode char
if (!isValidXMLChar(aContent.codePointAt(i))) {
aContent.setCharAt(i, ' ');
i++;
continue;
}
// Set up how many characters we want to skip
int seek = i + 1;
// Do we maybe have an entity?
if (aContent.charAt(i) == '&') {
// REC 2006-10-21 Some PDFs seem to have entities and others
// don't
// so we may encounter &'s that do not introduce an entity and
// just ignore them.
final int end = aContent.indexOf(";", i);
if (end != -1) {
final String cand = aContent.substring(i, end + 1);
String r = null;
try {
if (cand.startsWith("&#x")) {
final int cp = Integer.parseInt(cand.substring(2, cand.length() - 1), 16);
r = isValidXMLChar(cp) ? String.valueOf(Character.toChars(cp)) : " ";
} else if (cand.startsWith("&#")) {
final int cp = Integer.parseInt(cand.substring(2, cand.length() - 1));
r = isValidXMLChar(cp) ? String.valueOf(Character.toChars(cp)) : " ";
} else {
// RE 2006-10-22 The chance that there is a & and a
// ;
// together in a string is quite big. Let's be
// tolerant.
}
} catch (final NumberFormatException e) {
log.warn("Invalid numeric entity in fragment [" + cand + "] - Dropping it.");
}
// Expand the entity and set proper skip (if found)
if (r != null) {
aContent.replace(i, i + cand.length(), r);
seek = i + r.length();
}
}
}
// Match against the Trie after numeric entity expansion is over
if (substitutionTable != null) {
final Trie<String>.Node match = substitutionTable.getNode(aContent, i);
if (match != null) {
aContent.replace(i, i + match.level, match.value);
seek = i + match.value.length();
}
}
// Check line breaks
while (i < seek) {
if (aContent.charAt(i) == '\n') {
lastBreak = i;
} else if (Character.isWhitespace(aContent.codePointAt(i)) && (i > (lastBreak + 79))) {
lastBreak = i;
aContent.replace(i, i + 1, "\n");
}
i++;
}
}
return aContent;
}