freemarker.template.utility.StringUtil Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of org.apache.servicemix.bundles.freemarker
Show all versions of org.apache.servicemix.bundles.freemarker
This OSGi bundle wraps ${pkgArtifactId} ${pkgVersion} jar file.
/*
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package freemarker.template.utility;
import static freemarker.template.utility.StringUtil.JsStringEncQuotation.*;
import java.io.IOException;
import java.io.UnsupportedEncodingException;
import java.io.Writer;
import java.util.HashMap;
import java.util.Locale;
import java.util.Map;
import java.util.StringTokenizer;
import java.util.regex.Pattern;
import freemarker.core.Environment;
import freemarker.core.ParseException;
import freemarker.ext.dom._ExtDomApi;
import freemarker.template.Version;
/**
* Some text related utilities.
*/
public class StringUtil {
/**
* Used to look up if the chars with low code needs to be escaped, but note that it gives bad result for '=', as
* there the it matters if it's after '['.
*/
private static final char[] ESCAPES = createEscapes();
private static final char[] LT = new char[] { '&', 'l', 't', ';' };
private static final char[] GT = new char[] { '&', 'g', 't', ';' };
private static final char[] AMP = new char[] { '&', 'a', 'm', 'p', ';' };
private static final char[] QUOT = new char[] { '&', 'q', 'u', 'o', 't', ';' };
private static final char[] HTML_APOS = new char[] { '&', '#', '3', '9', ';' };
private static final char[] XML_APOS = new char[] { '&', 'a', 'p', 'o', 's', ';' };
/*
* For better performance most methods are folded down. Don't you scream... :)
*/
/**
* HTML encoding (does not convert line breaks and apostrophe-quote).
* Replaces all '>' '<' '&' and '"' with entity reference, but not "'" (apostrophe-quote).
* The last is not escaped as back then when this was written some user agents didn't understood
* "'" nor "'".
*
* @deprecated Use {@link #XHTMLEnc(String)} instead, because it escapes apostrophe-quote too.
*/
@Deprecated
public static String HTMLEnc(String s) {
return XMLEncNA(s);
}
/**
* XML Encoding.
* Replaces all '>' '<' '&', "'" and '"' with entity reference
*/
public static String XMLEnc(String s) {
return XMLOrHTMLEnc(s, true, true, XML_APOS);
}
/**
* Like {@link #XMLEnc(String)}, but writes the result into a {@link Writer}.
*
* @since 2.3.24
*/
public static void XMLEnc(String s, Writer out) throws IOException {
XMLOrHTMLEnc(s, XML_APOS, out);
}
/**
* XHTML Encoding.
* Replaces all '>' '<' '&', "'" and '"' with entity reference
* suitable for XHTML decoding in common user agents (including legacy
* user agents, which do not decode "'" to "'", so "'" is used
* instead [see http://www.w3.org/TR/xhtml1/#C_16])
*/
public static String XHTMLEnc(String s) {
return XMLOrHTMLEnc(s, true, true, HTML_APOS);
}
/**
* Like {@link #XHTMLEnc(String)}, but writes the result into a {@link Writer}.
*
* @since 2.3.24
*/
public static void XHTMLEnc(String s, Writer out) throws IOException {
XMLOrHTMLEnc(s, HTML_APOS, out);
}
private static String XMLOrHTMLEnc(String s, boolean escGT, boolean escQuot, char[] apos) {
final int ln = s.length();
// First we find out if we need to escape, and if so, what the length of the output will be:
int firstEscIdx = -1;
int lastEscIdx = 0;
int plusOutLn = 0;
for (int i = 0; i < ln; i++) {
escape: do {
final char c = s.charAt(i);
switch (c) {
case '<':
plusOutLn += LT.length - 1;
break;
case '>':
if (!(escGT || maybeCDataEndGT(s, i))) {
break escape;
}
plusOutLn += GT.length - 1;
break;
case '&':
plusOutLn += AMP.length - 1;
break;
case '"':
if (!escQuot) {
break escape;
}
plusOutLn += QUOT.length - 1;
break;
case '\'': // apos
if (apos == null) {
break escape;
}
plusOutLn += apos.length - 1;
break;
default:
break escape;
}
if (firstEscIdx == -1) {
firstEscIdx = i;
}
lastEscIdx = i;
} while (false);
}
if (firstEscIdx == -1) {
return s; // Nothing to escape
} else {
final char[] esced = new char[ln + plusOutLn];
if (firstEscIdx != 0) {
s.getChars(0, firstEscIdx, esced, 0);
}
int dst = firstEscIdx;
scan: for (int i = firstEscIdx; i <= lastEscIdx; i++) {
final char c = s.charAt(i);
switch (c) {
case '<':
dst = shortArrayCopy(LT, esced, dst);
continue scan;
case '>':
if (!(escGT || maybeCDataEndGT(s, i))) {
break;
}
dst = shortArrayCopy(GT, esced, dst);
continue scan;
case '&':
dst = shortArrayCopy(AMP, esced, dst);
continue scan;
case '"':
if (!escQuot) {
break;
}
dst = shortArrayCopy(QUOT, esced, dst);
continue scan;
case '\'': // apos
if (apos == null) {
break;
}
dst = shortArrayCopy(apos, esced, dst);
continue scan;
}
esced[dst++] = c;
}
if (lastEscIdx != ln - 1) {
s.getChars(lastEscIdx + 1, ln, esced, dst);
}
return String.valueOf(esced);
}
}
private static boolean maybeCDataEndGT(String s, int i) {
if (i == 0) return true;
if (s.charAt(i - 1) != ']') return false;
if (i == 1 || s.charAt(i - 2) == ']') return true;
return false;
}
private static void XMLOrHTMLEnc(String s, char[] apos, Writer out) throws IOException {
int writtenEnd = 0; // exclusive end
int ln = s.length();
for (int i = 0; i < ln; i++) {
char c = s.charAt(i);
if (c == '<' || c == '>' || c == '&' || c == '"' || c == '\'') {
int flushLn = i - writtenEnd;
if (flushLn != 0) {
out.write(s, writtenEnd, flushLn);
}
writtenEnd = i + 1;
switch (c) {
case '<': out.write(LT); break;
case '>': out.write(GT); break;
case '&': out.write(AMP); break;
case '"': out.write(QUOT); break;
default: out.write(apos); break;
}
}
}
if (writtenEnd < ln) {
out.write(s, writtenEnd, ln - writtenEnd);
}
}
/**
* For efficiently copying very short char arrays.
*/
private static int shortArrayCopy(char[] src, char[] dst, int dstOffset) {
int ln = src.length;
for (int i = 0; i < ln; i++) {
dst[dstOffset++] = src[i];
}
return dstOffset;
}
/**
* XML encoding without replacing apostrophes.
* @see #XMLEnc(String)
*/
public static String XMLEncNA(String s) {
return XMLOrHTMLEnc(s, true, true, null);
}
/**
* XML encoding for attribute values quoted with {@code "} (not with {@code '}!).
* Also can be used for HTML attributes that are quoted with {@code "}.
* @see #XMLEnc(String)
*/
public static String XMLEncQAttr(String s) {
return XMLOrHTMLEnc(s, false, true, null);
}
/**
* XML encoding without replacing apostrophes and quotation marks and
* greater-thans (except in {@code ]]>}).
* @see #XMLEnc(String)
*/
public static String XMLEncNQG(String s) {
return XMLOrHTMLEnc(s, false, false, null);
}
/**
* Rich Text Format encoding (does not replace line breaks).
* Escapes all '\' '{' '}'.
*/
public static String RTFEnc(String s) {
int ln = s.length();
// First we find out if we need to escape, and if so, what the length of the output will be:
int firstEscIdx = -1;
int lastEscIdx = 0;
int plusOutLn = 0;
for (int i = 0; i < ln; i++) {
char c = s.charAt(i);
if (c == '{' || c == '}' || c == '\\') {
if (firstEscIdx == -1) {
firstEscIdx = i;
}
lastEscIdx = i;
plusOutLn++;
}
}
if (firstEscIdx == -1) {
return s; // Nothing to escape
} else {
char[] esced = new char[ln + plusOutLn];
if (firstEscIdx != 0) {
s.getChars(0, firstEscIdx, esced, 0);
}
int dst = firstEscIdx;
for (int i = firstEscIdx; i <= lastEscIdx; i++) {
char c = s.charAt(i);
if (c == '{' || c == '}' || c == '\\') {
esced[dst++] = '\\';
}
esced[dst++] = c;
}
if (lastEscIdx != ln - 1) {
s.getChars(lastEscIdx + 1, ln, esced, dst);
}
return String.valueOf(esced);
}
}
/**
* Like {@link #RTFEnc(String)}, but writes the result into a {@link Writer}.
*
* @since 2.3.24
*/
public static void RTFEnc(String s, Writer out) throws IOException {
int writtenEnd = 0; // exclusive end
int ln = s.length();
for (int i = 0; i < ln; i++) {
char c = s.charAt(i);
if (c == '{' || c == '}' || c == '\\') {
int flushLn = i - writtenEnd;
if (flushLn != 0) {
out.write(s, writtenEnd, flushLn);
}
out.write('\\');
writtenEnd = i; // Not i + 1, so c will be written out later
}
}
if (writtenEnd < ln) {
out.write(s, writtenEnd, ln - writtenEnd);
}
}
/**
* URL encoding (like%20this) for query parameter values, path segments, fragments; this encodes all
* characters that are reserved anywhere.
*/
public static String URLEnc(String s, String charset) throws UnsupportedEncodingException {
return URLEnc(s, charset, false);
}
/**
* Like {@link #URLEnc(String, String)} but doesn't escape the slash character ({@code /}).
* This can be used to encode a path only if you know that no folder or file name will contain {@code /}
* character (not in the path, but in the name itself), which usually stands, as the commonly used OS-es don't
* allow that.
*
* @since 2.3.21
*/
public static String URLPathEnc(String s, String charset) throws UnsupportedEncodingException {
return URLEnc(s, charset, true);
}
private static String URLEnc(String s, String charset, boolean keepSlash)
throws UnsupportedEncodingException {
int ln = s.length();
int i;
for (i = 0; i < ln; i++) {
char c = s.charAt(i);
if (!safeInURL(c, keepSlash)) {
break;
}
}
if (i == ln) {
// Nothing to escape
return s;
}
StringBuilder b = new StringBuilder(ln + ln / 3 + 2);
b.append(s.substring(0, i));
int encStart = i;
for (i++; i < ln; i++) {
char c = s.charAt(i);
if (safeInURL(c, keepSlash)) {
if (encStart != -1) {
byte[] o = s.substring(encStart, i).getBytes(charset);
for (int j = 0; j < o.length; j++) {
b.append('%');
byte bc = o[j];
int c1 = bc & 0x0F;
int c2 = (bc >> 4) & 0x0F;
b.append((char) (c2 < 10 ? c2 + '0' : c2 - 10 + 'A'));
b.append((char) (c1 < 10 ? c1 + '0' : c1 - 10 + 'A'));
}
encStart = -1;
}
b.append(c);
} else {
if (encStart == -1) {
encStart = i;
}
}
}
if (encStart != -1) {
byte[] o = s.substring(encStart, i).getBytes(charset);
for (int j = 0; j < o.length; j++) {
b.append('%');
byte bc = o[j];
int c1 = bc & 0x0F;
int c2 = (bc >> 4) & 0x0F;
b.append((char) (c2 < 10 ? c2 + '0' : c2 - 10 + 'A'));
b.append((char) (c1 < 10 ? c1 + '0' : c1 - 10 + 'A'));
}
}
return b.toString();
}
private static boolean safeInURL(char c, boolean keepSlash) {
return c >= 'a' && c <= 'z' || c >= 'A' && c <= 'Z'
|| c >= '0' && c <= '9'
|| c == '_' || c == '-' || c == '.' || c == '!' || c == '~'
|| c >= '\'' && c <= '*'
|| keepSlash && c == '/';
}
private static char[] createEscapes() {
char[] escapes = new char['\\' + 1];
for (int i = 0; i < 32; ++i) {
escapes[i] = 1;
}
escapes['\\'] = '\\';
escapes['\''] = '\'';
escapes['"'] = '"';
escapes['<'] = 'l';
// As '=' is only escaped if it's after '[', we can't handle it here
escapes['>'] = 'g';
escapes['&'] = 'a';
escapes['\b'] = 'b';
escapes['\t'] = 't';
escapes['\n'] = 'n';
escapes['\f'] = 'f';
escapes['\r'] = 'r';
return escapes;
}
/**
* Escapes a string according the FTL string literal escaping rules, assuming the literal is quoted with
* {@code quotation}; it doesn't add the quotation marks itself.
*
* @param quotation
* Either {@code '"'} or {@code '\''}. It's assumed that the string literal whose part we calculate is
* enclosed within this kind of quotation mark. Thus, the other kind of quotation character will not be
* escaped in the result.
*
* @since 2.3.22
*/
public static String FTLStringLiteralEnc(String s, char quotation) {
return FTLStringLiteralEnc(s, quotation, false);
}
/**
* Escapes a string according the FTL string literal escaping rules; it doesn't add the quotation marks. As this
* method doesn't know if the string literal is quoted with reuglar quotation marks or apostrophe quute, it will
* escape both.
*
* @see #FTLStringLiteralEnc(String, char)
*/
public static String FTLStringLiteralEnc(String s) {
return FTLStringLiteralEnc(s, (char) 0, false);
}
private static String FTLStringLiteralEnc(String s, char quotation, boolean addQuotation) {
final int ln = s.length();
final char otherQuotation;
if (quotation == 0) {
otherQuotation = 0;
} else if (quotation == '"') {
otherQuotation = '\'';
} else if (quotation == '\'') {
otherQuotation = '"';
} else {
throw new IllegalArgumentException("Unsupported quotation character: " + quotation);
}
final int escLn = ESCAPES.length;
StringBuilder buf = null;
for (int i = 0; i < ln; i++) {
char c = s.charAt(i);
char escape;
if (c == '=') {
escape = i > 0 && s.charAt(i - 1) == '[' ? '=' : 0;
} else if (c < escLn) {
escape = ESCAPES[c]; //
} else if (c == '{' && i > 0 && isInterpolationStart(s.charAt(i - 1))) {
escape = '{';
} else {
escape = 0;
}
if (escape == 0 || escape == otherQuotation) {
if (buf != null) {
buf.append(c);
}
} else {
if (buf == null) {
buf = new StringBuilder(s.length() + 4 + (addQuotation ? 2 : 0));
if (addQuotation) {
buf.append(quotation);
}
buf.append(s.substring(0, i));
}
if (escape == 1) {
// hex encoding for characters below 0x20
// that have no other escape representation
buf.append("\\x00");
int c2 = (c >> 4) & 0x0F;
c = (char) (c & 0x0F);
buf.append((char) (c2 < 10 ? c2 + '0' : c2 - 10 + 'A'));
buf.append((char) (c < 10 ? c + '0' : c - 10 + 'A'));
} else {
buf.append('\\');
buf.append(escape);
}
}
}
if (buf == null) {
return addQuotation ? quotation + s + quotation : s;
} else {
if (addQuotation) {
buf.append(quotation);
}
return buf.toString();
}
}
private static boolean isInterpolationStart(char c) {
return c == '$' || c == '#';
}
/**
* FTL string literal decoding.
*
* \\, \", \', \n, \t, \r, \b and \f will be replaced according to
* Java rules. In additional, it knows \g, \l, \a and \{ which are
* replaced with <, >, & and { respectively.
* \x works as hexadecimal character code escape. The character
* codes are interpreted according to UCS basic plane (Unicode).
* "f\x006Fo", "f\x06Fo" and "f\x6Fo" will be "foo".
* "f\x006F123" will be "foo123" as the maximum number of digits is 4.
*
* All other \X (where X is any character not mentioned above or End-of-string)
* will cause a ParseException.
*
* @param s String literal without the surrounding quotation marks
* @return String with all escape sequences resolved
* @throws ParseException if there string contains illegal escapes
*/
public static String FTLStringLiteralDec(String s) throws ParseException {
int idx = s.indexOf('\\');
if (idx == -1) {
return s;
}
int lidx = s.length() - 1;
int bidx = 0;
StringBuilder buf = new StringBuilder(lidx);
do {
buf.append(s.substring(bidx, idx));
if (idx >= lidx) {
throw new ParseException("The last character of string literal is backslash", 0,0);
}
char c = s.charAt(idx + 1);
switch (c) {
case '"':
buf.append('"');
bidx = idx + 2;
break;
case '\'':
buf.append('\'');
bidx = idx + 2;
break;
case '\\':
buf.append('\\');
bidx = idx + 2;
break;
case 'n':
buf.append('\n');
bidx = idx + 2;
break;
case 'r':
buf.append('\r');
bidx = idx + 2;
break;
case 't':
buf.append('\t');
bidx = idx + 2;
break;
case 'f':
buf.append('\f');
bidx = idx + 2;
break;
case 'b':
buf.append('\b');
bidx = idx + 2;
break;
case 'g':
buf.append('>');
bidx = idx + 2;
break;
case 'l':
buf.append('<');
bidx = idx + 2;
break;
case 'a':
buf.append('&');
bidx = idx + 2;
break;
case '{':
case '=':
buf.append(c);
bidx = idx + 2;
break;
case 'x': {
idx += 2;
int x = idx;
int y = 0;
int z = lidx > idx + 3 ? idx + 3 : lidx;
while (idx <= z) {
char b = s.charAt(idx);
if (b >= '0' && b <= '9') {
y <<= 4;
y += b - '0';
} else if (b >= 'a' && b <= 'f') {
y <<= 4;
y += b - 'a' + 10;
} else if (b >= 'A' && b <= 'F') {
y <<= 4;
y += b - 'A' + 10;
} else {
break;
}
idx++;
}
if (x < idx) {
buf.append((char) y);
} else {
throw new ParseException("Invalid \\x escape in a string literal",0,0);
}
bidx = idx;
break;
}
default:
throw new ParseException("Invalid escape sequence (\\" + c + ") in a string literal",0,0);
}
idx = s.indexOf('\\', bidx);
} while (idx != -1);
buf.append(s.substring(bidx));
return buf.toString();
}
public static Locale deduceLocale(String input) {
if (input == null) return null;
Locale locale = Locale.getDefault();
if (input.length() > 0 && input.charAt(0) == '"') input = input.substring(1, input.length() - 1);
StringTokenizer st = new StringTokenizer(input, ",_ ");
String lang = "", country = "";
if (st.hasMoreTokens()) {
lang = st.nextToken();
}
if (st.hasMoreTokens()) {
country = st.nextToken();
}
if (!st.hasMoreTokens()) {
locale = new Locale(lang, country);
} else {
locale = new Locale(lang, country, st.nextToken());
}
return locale;
}
public static String capitalize(String s) {
StringTokenizer st = new StringTokenizer(s, " \t\r\n", true);
StringBuilder buf = new StringBuilder(s.length());
while (st.hasMoreTokens()) {
String tok = st.nextToken();
buf.append(tok.substring(0, 1).toUpperCase());
buf.append(tok.substring(1).toLowerCase());
}
return buf.toString();
}
public static boolean getYesNo(String s) {
if (s.startsWith("\"")) {
s = s.substring(1, s.length() - 1);
}
if (s.equalsIgnoreCase("n")
|| s.equalsIgnoreCase("no")
|| s.equalsIgnoreCase("f")
|| s.equalsIgnoreCase("false")) {
return false;
} else if (s.equalsIgnoreCase("y")
|| s.equalsIgnoreCase("yes")
|| s.equalsIgnoreCase("t")
|| s.equalsIgnoreCase("true")) {
return true;
}
throw new IllegalArgumentException("Illegal boolean value: " + s);
}
/**
* Splits a string at the specified character.
*/
public static String[] split(String s, char c) {
int i, b, e;
int cnt;
String res[];
int ln = s.length();
i = 0;
cnt = 1;
while ((i = s.indexOf(c, i)) != -1) {
cnt++;
i++;
}
res = new String[cnt];
i = 0;
b = 0;
while (b <= ln) {
e = s.indexOf(c, b);
if (e == -1) e = ln;
res[i++] = s.substring(b, e);
b = e + 1;
}
return res;
}
/**
* Splits a string at the specified string.
*
* @param sep
* The string that separates the items of the resulting array. Since 2.3.28, if this is 0 length, then
* each character will be a separate item in the array.
*/
public static String[] split(String s, String sep, boolean caseInsensitive) {
int sepLn = sep.length();
String convertedS = caseInsensitive ? s.toLowerCase() : s;
int sLn = s.length();
if (sepLn == 0) {
String[] res = new String[sLn];
for (int i = 0; i < sLn; i++) {
res[i] = String.valueOf(s.charAt(i));
}
return res;
}
String splitString = caseInsensitive ? sep.toLowerCase() : sep;
String res[];
{
int next = 0;
int count = 1;
while ((next = convertedS.indexOf(splitString, next)) != -1) {
count++;
next += sepLn;
}
res = new String[count];
}
int dst = 0;
int next = 0;
while (next <= sLn) {
int end = convertedS.indexOf(splitString, next);
if (end == -1) end = sLn;
res[dst++] = s.substring(next, end);
next = end + sepLn;
}
return res;
}
/**
* Same as {@link #replace(String, String, String, boolean, boolean)} with two {@code false} parameters.
* @since 2.3.20
*/
public static String replace(String text, String oldSub, String newSub) {
return replace(text, oldSub, newSub, false, false);
}
/**
* Replaces all occurrences of a sub-string in a string.
* @param text The string where it will replace oldsub with
* newsub.
* @return String The string after the replacements.
*/
public static String replace(String text,
String oldsub,
String newsub,
boolean caseInsensitive,
boolean firstOnly) {
StringBuilder buf;
int tln;
int oln = oldsub.length();
if (oln == 0) {
int nln = newsub.length();
if (nln == 0) {
return text;
} else {
if (firstOnly) {
return newsub + text;
} else {
tln = text.length();
buf = new StringBuilder(tln + (tln + 1) * nln);
buf.append(newsub);
for (int i = 0; i < tln; i++) {
buf.append(text.charAt(i));
buf.append(newsub);
}
return buf.toString();
}
}
} else {
oldsub = caseInsensitive ? oldsub.toLowerCase() : oldsub;
String input = caseInsensitive ? text.toLowerCase() : text;
int e = input.indexOf(oldsub);
if (e == -1) {
return text;
}
int b = 0;
tln = text.length();
buf = new StringBuilder(
tln + Math.max(newsub.length() - oln, 0) * 3);
do {
buf.append(text.substring(b, e));
buf.append(newsub);
b = e + oln;
e = input.indexOf(oldsub, b);
} while (e != -1 && !firstOnly);
buf.append(text.substring(b));
return buf.toString();
}
}
/**
* Removes a line-break from the end of the string (if there's any).
*/
public static String chomp(String s) {
if (s.endsWith("\r\n")) return s.substring(0, s.length() - 2);
if (s.endsWith("\r") || s.endsWith("\n"))
return s.substring(0, s.length() - 1);
return s;
}
/**
* Converts a 0-length string to null, leaves the string as is otherwise.
* @param s maybe {@code null}.
*/
public static String emptyToNull(String s) {
if (s == null) return null;
return s.length() == 0 ? null : s;
}
/**
* Converts the parameter with toString (if it's not null) and passes it to
* {@link #jQuote(String)}.
*/
public static String jQuote(Object obj) {
return jQuote(obj != null ? obj.toString() : null);
}
/**
* Quotes string as Java Language string literal.
* Returns string "null" if s
* is null.
*/
public static String jQuote(String s) {
if (s == null) {
return "null";
}
return javaStringEnc(s, true);
}
/**
* Converts the parameter with toString (if not
* null)and passes it to {@link #jQuoteNoXSS(String)}.
*/
public static String jQuoteNoXSS(Object obj) {
return jQuoteNoXSS(obj != null ? obj.toString() : null);
}
/**
* Same as {@link #jQuoteNoXSS(String)} but also escapes '<'
* as \u003C. This is used for log messages to prevent XSS
* on poorly written Web-based log viewers.
*/
public static String jQuoteNoXSS(String s) {
if (s == null) {
return "null";
}
int ln = s.length();
StringBuilder b = new StringBuilder(ln + 6);
b.append('"');
for (int i = 0; i < ln; i++) {
char c = s.charAt(i);
if (c == '"') {
b.append("\\\"");
} else if (c == '\\') {
b.append("\\\\");
} else if (c == '<') {
b.append("\\u003C");
} else if (c < 0x20) {
if (c == '\n') {
b.append("\\n");
} else if (c == '\r') {
b.append("\\r");
} else if (c == '\f') {
b.append("\\f");
} else if (c == '\b') {
b.append("\\b");
} else if (c == '\t') {
b.append("\\t");
} else {
b.append("\\u00");
b.append(toHexDigitLowerCase(c / 0x10));
b.append(toHexDigitLowerCase(c & 0xF));
}
} else {
b.append(c);
}
} // for each character
b.append('"');
return b.toString();
}
/**
* Creates a quoted FTL string literal from a string, using escaping where necessary. The result either
* uses regular quotation marks (UCS 0x22) or apostrophe-quotes (UCS 0x27), depending on the string content.
* (Currently, apostrophe-quotes will be chosen exactly when the string contains regular quotation character and
* doesn't contain apostrophe-quote character.)
*
* @param s
* The value that should be converted to an FTL string literal whose evaluated value equals to {@code s}
*
* @since 2.3.22
*/
public static String ftlQuote(String s) {
char quotation;
if (s.indexOf('"') != -1 && s.indexOf('\'') == -1) {
quotation = '\'';
} else {
quotation = '\"';
}
return FTLStringLiteralEnc(s, quotation, true);
}
/**
* Tells if a character can occur on the beginning of an FTL identifier expression (without escaping).
*
* @since 2.3.22
*/
public static boolean isFTLIdentifierStart(final char c) {
// This code was generated on JDK 1.8.0_20 Win64 with src/main/misc/identifierChars/IdentifierCharGenerator.java
if (c < 0xAA) { // This branch was edited for speed.
if (c >= 'a' && c <= 'z' || c >= '@' && c <= 'Z') {
return true;
} else {
return c == '$' || c == '_';
}
} else { // c >= 0xAA
if (c < 0xA7F8) {
if (c < 0x2D6F) {
if (c < 0x2128) {
if (c < 0x2090) {
if (c < 0xD8) {
if (c < 0xBA) {
return c == 0xAA || c == 0xB5;
} else { // c >= 0xBA
return c == 0xBA || c >= 0xC0 && c <= 0xD6;
}
} else { // c >= 0xD8
if (c < 0x2071) {
return c >= 0xD8 && c <= 0xF6 || c >= 0xF8 && c <= 0x1FFF;
} else { // c >= 0x2071
return c == 0x2071 || c == 0x207F;
}
}
} else { // c >= 0x2090
if (c < 0x2115) {
if (c < 0x2107) {
return c >= 0x2090 && c <= 0x209C || c == 0x2102;
} else { // c >= 0x2107
return c == 0x2107 || c >= 0x210A && c <= 0x2113;
}
} else { // c >= 0x2115
if (c < 0x2124) {
return c == 0x2115 || c >= 0x2119 && c <= 0x211D;
} else { // c >= 0x2124
return c == 0x2124 || c == 0x2126;
}
}
}
} else { // c >= 0x2128
if (c < 0x2C30) {
if (c < 0x2145) {
if (c < 0x212F) {
return c == 0x2128 || c >= 0x212A && c <= 0x212D;
} else { // c >= 0x212F
return c >= 0x212F && c <= 0x2139 || c >= 0x213C && c <= 0x213F;
}
} else { // c >= 0x2145
if (c < 0x2183) {
return c >= 0x2145 && c <= 0x2149 || c == 0x214E;
} else { // c >= 0x2183
return c >= 0x2183 && c <= 0x2184 || c >= 0x2C00 && c <= 0x2C2E;
}
}
} else { // c >= 0x2C30
if (c < 0x2D00) {
if (c < 0x2CEB) {
return c >= 0x2C30 && c <= 0x2C5E || c >= 0x2C60 && c <= 0x2CE4;
} else { // c >= 0x2CEB
return c >= 0x2CEB && c <= 0x2CEE || c >= 0x2CF2 && c <= 0x2CF3;
}
} else { // c >= 0x2D00
if (c < 0x2D2D) {
return c >= 0x2D00 && c <= 0x2D25 || c == 0x2D27;
} else { // c >= 0x2D2D
return c == 0x2D2D || c >= 0x2D30 && c <= 0x2D67;
}
}
}
}
} else { // c >= 0x2D6F
if (c < 0x31F0) {
if (c < 0x2DD0) {
if (c < 0x2DB0) {
if (c < 0x2DA0) {
return c == 0x2D6F || c >= 0x2D80 && c <= 0x2D96;
} else { // c >= 0x2DA0
return c >= 0x2DA0 && c <= 0x2DA6 || c >= 0x2DA8 && c <= 0x2DAE;
}
} else { // c >= 0x2DB0
if (c < 0x2DC0) {
return c >= 0x2DB0 && c <= 0x2DB6 || c >= 0x2DB8 && c <= 0x2DBE;
} else { // c >= 0x2DC0
return c >= 0x2DC0 && c <= 0x2DC6 || c >= 0x2DC8 && c <= 0x2DCE;
}
}
} else { // c >= 0x2DD0
if (c < 0x3031) {
if (c < 0x2E2F) {
return c >= 0x2DD0 && c <= 0x2DD6 || c >= 0x2DD8 && c <= 0x2DDE;
} else { // c >= 0x2E2F
return c == 0x2E2F || c >= 0x3005 && c <= 0x3006;
}
} else { // c >= 0x3031
if (c < 0x3040) {
return c >= 0x3031 && c <= 0x3035 || c >= 0x303B && c <= 0x303C;
} else { // c >= 0x3040
return c >= 0x3040 && c <= 0x318F || c >= 0x31A0 && c <= 0x31BA;
}
}
}
} else { // c >= 0x31F0
if (c < 0xA67F) {
if (c < 0xA4D0) {
if (c < 0x3400) {
return c >= 0x31F0 && c <= 0x31FF || c >= 0x3300 && c <= 0x337F;
} else { // c >= 0x3400
return c >= 0x3400 && c <= 0x4DB5 || c >= 0x4E00 && c <= 0xA48C;
}
} else { // c >= 0xA4D0
if (c < 0xA610) {
return c >= 0xA4D0 && c <= 0xA4FD || c >= 0xA500 && c <= 0xA60C;
} else { // c >= 0xA610
return c >= 0xA610 && c <= 0xA62B || c >= 0xA640 && c <= 0xA66E;
}
}
} else { // c >= 0xA67F
if (c < 0xA78B) {
if (c < 0xA717) {
return c >= 0xA67F && c <= 0xA697 || c >= 0xA6A0 && c <= 0xA6E5;
} else { // c >= 0xA717
return c >= 0xA717 && c <= 0xA71F || c >= 0xA722 && c <= 0xA788;
}
} else { // c >= 0xA78B
if (c < 0xA7A0) {
return c >= 0xA78B && c <= 0xA78E || c >= 0xA790 && c <= 0xA793;
} else { // c >= 0xA7A0
return c >= 0xA7A0 && c <= 0xA7AA;
}
}
}
}
}
} else { // c >= 0xA7F8
if (c < 0xAB20) {
if (c < 0xAA44) {
if (c < 0xA8FB) {
if (c < 0xA840) {
if (c < 0xA807) {
return c >= 0xA7F8 && c <= 0xA801 || c >= 0xA803 && c <= 0xA805;
} else { // c >= 0xA807
return c >= 0xA807 && c <= 0xA80A || c >= 0xA80C && c <= 0xA822;
}
} else { // c >= 0xA840
if (c < 0xA8D0) {
return c >= 0xA840 && c <= 0xA873 || c >= 0xA882 && c <= 0xA8B3;
} else { // c >= 0xA8D0
return c >= 0xA8D0 && c <= 0xA8D9 || c >= 0xA8F2 && c <= 0xA8F7;
}
}
} else { // c >= 0xA8FB
if (c < 0xA984) {
if (c < 0xA930) {
return c == 0xA8FB || c >= 0xA900 && c <= 0xA925;
} else { // c >= 0xA930
return c >= 0xA930 && c <= 0xA946 || c >= 0xA960 && c <= 0xA97C;
}
} else { // c >= 0xA984
if (c < 0xAA00) {
return c >= 0xA984 && c <= 0xA9B2 || c >= 0xA9CF && c <= 0xA9D9;
} else { // c >= 0xAA00
return c >= 0xAA00 && c <= 0xAA28 || c >= 0xAA40 && c <= 0xAA42;
}
}
}
} else { // c >= 0xAA44
if (c < 0xAAC0) {
if (c < 0xAA80) {
if (c < 0xAA60) {
return c >= 0xAA44 && c <= 0xAA4B || c >= 0xAA50 && c <= 0xAA59;
} else { // c >= 0xAA60
return c >= 0xAA60 && c <= 0xAA76 || c == 0xAA7A;
}
} else { // c >= 0xAA80
if (c < 0xAAB5) {
return c >= 0xAA80 && c <= 0xAAAF || c == 0xAAB1;
} else { // c >= 0xAAB5
return c >= 0xAAB5 && c <= 0xAAB6 || c >= 0xAAB9 && c <= 0xAABD;
}
}
} else { // c >= 0xAAC0
if (c < 0xAAF2) {
if (c < 0xAADB) {
return c == 0xAAC0 || c == 0xAAC2;
} else { // c >= 0xAADB
return c >= 0xAADB && c <= 0xAADD || c >= 0xAAE0 && c <= 0xAAEA;
}
} else { // c >= 0xAAF2
if (c < 0xAB09) {
return c >= 0xAAF2 && c <= 0xAAF4 || c >= 0xAB01 && c <= 0xAB06;
} else { // c >= 0xAB09
return c >= 0xAB09 && c <= 0xAB0E || c >= 0xAB11 && c <= 0xAB16;
}
}
}
}
} else { // c >= 0xAB20
if (c < 0xFB46) {
if (c < 0xFB13) {
if (c < 0xAC00) {
if (c < 0xABC0) {
return c >= 0xAB20 && c <= 0xAB26 || c >= 0xAB28 && c <= 0xAB2E;
} else { // c >= 0xABC0
return c >= 0xABC0 && c <= 0xABE2 || c >= 0xABF0 && c <= 0xABF9;
}
} else { // c >= 0xAC00
if (c < 0xD7CB) {
return c >= 0xAC00 && c <= 0xD7A3 || c >= 0xD7B0 && c <= 0xD7C6;
} else { // c >= 0xD7CB
return c >= 0xD7CB && c <= 0xD7FB || c >= 0xF900 && c <= 0xFB06;
}
}
} else { // c >= 0xFB13
if (c < 0xFB38) {
if (c < 0xFB1F) {
return c >= 0xFB13 && c <= 0xFB17 || c == 0xFB1D;
} else { // c >= 0xFB1F
return c >= 0xFB1F && c <= 0xFB28 || c >= 0xFB2A && c <= 0xFB36;
}
} else { // c >= 0xFB38
if (c < 0xFB40) {
return c >= 0xFB38 && c <= 0xFB3C || c == 0xFB3E;
} else { // c >= 0xFB40
return c >= 0xFB40 && c <= 0xFB41 || c >= 0xFB43 && c <= 0xFB44;
}
}
}
} else { // c >= 0xFB46
if (c < 0xFF21) {
if (c < 0xFDF0) {
if (c < 0xFD50) {
return c >= 0xFB46 && c <= 0xFBB1 || c >= 0xFBD3 && c <= 0xFD3D;
} else { // c >= 0xFD50
return c >= 0xFD50 && c <= 0xFD8F || c >= 0xFD92 && c <= 0xFDC7;
}
} else { // c >= 0xFDF0
if (c < 0xFE76) {
return c >= 0xFDF0 && c <= 0xFDFB || c >= 0xFE70 && c <= 0xFE74;
} else { // c >= 0xFE76
return c >= 0xFE76 && c <= 0xFEFC || c >= 0xFF10 && c <= 0xFF19;
}
}
} else { // c >= 0xFF21
if (c < 0xFFCA) {
if (c < 0xFF66) {
return c >= 0xFF21 && c <= 0xFF3A || c >= 0xFF41 && c <= 0xFF5A;
} else { // c >= 0xFF66
return c >= 0xFF66 && c <= 0xFFBE || c >= 0xFFC2 && c <= 0xFFC7;
}
} else { // c >= 0xFFCA
if (c < 0xFFDA) {
return c >= 0xFFCA && c <= 0xFFCF || c >= 0xFFD2 && c <= 0xFFD7;
} else { // c >= 0xFFDA
return c >= 0xFFDA && c <= 0xFFDC;
}
}
}
}
}
}
}
}
/**
* Tells if a character can occur in an FTL identifier expression (without escaping) as other than the first
* character.
*
* @since 2.3.22
*/
public static boolean isFTLIdentifierPart(final char c) {
return isFTLIdentifierStart(c) || (c >= '0' && c <= '9');
}
/**
* Tells if a character can occur in an FTL identifier if it's preceded with a backslash. For example, {@code "-"}
* is a such character (as you can have an identifier like {@code foo\-bar} in FTL), but {@code "f"} is not, as
* it needn't be, and can't be escaped.
*
* @since 2.3.31
*/
public static boolean isBackslashEscapedFTLIdentifierCharacter(final char c) {
return c == '-' || c == '.' || c == ':' || c == '#';
}
/**
* Escapes the {@code String} with the escaping rules of Java language
* string literals, so it's safe to insert the value into a string literal.
* The resulting string will not be quoted.
*
* See more details at {@link #javaStringEnc(String, boolean)}, as this just calls that with {@code false} as the
* 2nd argument.
*/
public static String javaStringEnc(String s) {
return javaStringEnc(s, false);
}
/**
* Escapes the {@code String} with the escaping rules of Java language string literals, and then if {@code quote} is
* true, it also adds quotation marks before and after it.
*
* All characters under UCS code point 0x20 will be escaped.
* Where they have no dedicated escape sequence in Java, they will
* be replaced with hexadecimal escape ({@code \}uXXXX).
*
* @see #jQuote(String)
*/
public static String javaStringEnc(String s, boolean quote) {
int ln = s.length();
for (int i = 0; i < ln; i++) {
char c = s.charAt(i);
if (c == '"' || c == '\\' || c < 0x20) {
StringBuilder b = new StringBuilder(ln + (quote ? 6 : 4));
if (quote) {
b.append("\"");
}
b.append(s, 0, i);
while (true) {
if (c == '"') {
b.append("\\\"");
} else if (c == '\\') {
b.append("\\\\");
} else if (c < 0x20) {
if (c == '\n') {
b.append("\\n");
} else if (c == '\r') {
b.append("\\r");
} else if (c == '\f') {
b.append("\\f");
} else if (c == '\b') {
b.append("\\b");
} else if (c == '\t') {
b.append("\\t");
} else {
b.append("\\u00");
b.append(toHexDigitLowerCase(c / 0x10));
b.append(toHexDigitLowerCase(c & 0xF));
}
} else {
b.append(c);
}
i++;
if (i >= ln) {
if (quote) {
b.append("\"");
}
return b.toString();
}
c = s.charAt(i);
}
} // if has to be escaped
} // for each character
return quote ? '"' + s + '"' : s;
}
/**
* Escapes a {@link String} to be safely insertable into a JavaScript string literal; for more see
* {@link #jsStringEnc(String, JsStringEncCompatibility, JsStringEncQuotation)
* jsStringEnc(s, JsStringEncCompatibility.JAVA_SCRIPT, null)}.
*/
public static String javaScriptStringEnc(String s) {
return jsStringEnc(s, JsStringEncCompatibility.JAVA_SCRIPT);
}
/**
* Escapes a {@link String} to be safely insertable into a JSON string literal; for more see
* {@link #jsStringEnc(String, JsStringEncCompatibility, JsStringEncQuotation)
* jsStringEnc(s, JsStringEncCompatibility.JSON, null)}.
*/
public static String jsonStringEnc(String s) {
return jsStringEnc(s, JsStringEncCompatibility.JSON);
}
private static final int NO_ESC = 0;
private static final int ESC_HEXA = 1;
private static final int ESC_BACKSLASH = 3;
/**
* Escapes a {@link String} to be safely insertable into a JSON or JavaScript string literal; for more see
* {@link #jsStringEnc(String, JsStringEncCompatibility, JsStringEncQuotation)
* jsStringEnc(s, json ? JsStringEncCompatibility.JSON : JsStringEncCompatibility.JAVA_SCRIPT, null)}.
*
* @since 2.3.20
* @deprecated Use {@link #jsStringEnc(String, JsStringEncCompatibility)} instead.
*/
@Deprecated
public static String jsStringEnc(String s, boolean json) {
return jsStringEnc(s, json ? JsStringEncCompatibility.JSON : JsStringEncCompatibility.JAVA_SCRIPT, null);
}
/**
* Escapes a {@link String} to be safely insertable into a JSON or JavaScript string literal; for more see
* {@link #jsStringEnc(String, JsStringEncCompatibility, JsStringEncQuotation) jsStringEnc(s, compatibility, null)}.
*
* @since 2.3.32
*/
public static String jsStringEnc(String s, JsStringEncCompatibility compatibility) {
return jsStringEnc(s, compatibility, null);
}
/**
* Escapes a {@link String} to be safely insertable into a JavaScript or a JSON string literal, and if the 3rd
* argument is {@code true}, also adds quotation marks around it.
* If instead the caller ensures that the quotation marks are there, then in JSON mode (2nd argument), the quotation
* marks must be {@code "}, not {@code '}, because for JSON we won't escape {@code '}.
*
*
The escaping rules guarantee that if the inside of the JavaScript/JSON string literal is from one or more
* touching pieces that were escaped with this, no character sequence can occur that closes the
* JavaScript/JSON string literal, or has a meaning in HTML/XML that causes the HTML script section to be closed.
* (If, however, the escaped section is preceded by or followed by strings from other sources, this can't be
* guaranteed in some rare cases. Like x = "</${a?js_string}" might closes the "script"
* element if {@code a} is {@code "script>"}.)
*
* The escaped characters are:
*
*
*
*
* Input
* Output
*
* {@code "}
* {@code \"}
*
* {@code '} if not in JSON-mode, nor is the {@code quited} argument {@code true}
* {@code \'}
*
* {@code \}
* {@code \\}
*
* {@code /} if the method can't know that it won't be directly after <
* {@code \/}
*
* > if the method can't know that it won't be directly after {@code ]]} or {@code --}
* JavaScript: \>; JSON: {@code \}{@code u003E}
*
* < if the method can't know that it won't be directly followed by {@code !} or {@code ?}
* \{@code u}003C
*
*
* u0000-u001f (UNICODE control characters - disallowed by JSON)
* u007f-u009f (UNICODE control characters - disallowed by JSON)
* {@code \n}, {@code \r} and such, or if there's no such dedicated escape:
* JavaScript: \xXX, JSON: \{@code u}XXXX
*
*
* u2028 (Line separator - source code line-break in ECMAScript)
* u2029 (Paragraph separator - source code line-break in ECMAScript)
* \{@code u}XXXX
*
*
* @param s The string to escape
* @param compatibility If escaping should restrict itself to rules that are valid in JSON, in JavaScript, or in both.
* @param quotation In not {@code null}, quotation marks of this type are added around the value.
*
* @since 2.3.32
*/
public static String jsStringEnc(String s, JsStringEncCompatibility compatibility, JsStringEncQuotation quotation) {
NullArgumentException.check("s", s);
int ln = s.length();
StringBuilder sb;
if (quotation == null) {
sb = null;
} else {
if (quotation == APOSTROPHE && compatibility.jsonCompatible) {
throw new IllegalArgumentException("JSON compatible mode doesn't allow quotationMode=" + quotation);
}
sb = new StringBuilder(ln + 8);
sb.append(quotation.getSymbol());
}
for (int i = 0; i < ln; i++) {
final char c = s.charAt(i);
final int escapeType; //
if (!(c > '>' && c < 0x7F && c != '\\') && c != ' ' && !(c >= 0xA0 && c < 0x2028)) { // skip common chars
if (c <= 0x1F) { // control chars range 1
if (c == '\n') {
escapeType = 'n';
} else if (c == '\r') {
escapeType = 'r';
} else if (c == '\f') {
escapeType = 'f';
} else if (c == '\b') {
escapeType = 'b';
} else if (c == '\t') {
escapeType = 't';
} else {
escapeType = ESC_HEXA;
}
} else if (c == '"') {
escapeType = quotation == APOSTROPHE ? NO_ESC : ESC_BACKSLASH;
} else if (c == '\'') {
escapeType = !compatibility.javaScriptCompatible || quotation == QUOTATION_MARK ? NO_ESC
: (compatibility.jsonCompatible ? ESC_HEXA : ESC_BACKSLASH);
} else if (c == '\\') {
escapeType = ESC_BACKSLASH;
} else if (c == '/'
&& (i == 0 && quotation == null || i != 0 && s.charAt(i - 1) == '<')) {
// against closing elements with "') {
// against "]]> and "-->"
final boolean dangerous;
if (quotation != null && i < 2) {
dangerous = false;
} else if (i == 0) {
dangerous = true;
} else {
final char prevC = s.charAt(i - 1);
if (prevC == ']' || prevC == '-') {
if (i == 1) {
dangerous = true;
} else {
final char prevPrevC = s.charAt(i - 2);
dangerous = prevPrevC == prevC;
}
} else {
dangerous = false;
}
}
escapeType = dangerous ? (compatibility.jsonCompatible ? ESC_HEXA : ESC_BACKSLASH) : NO_ESC;
} else if (c == '<') {
// against "= 0x7F && c <= 0x9F) // control chars range 2
|| (c == 0x2028 || c == 0x2029) // UNICODE line terminators
) {
escapeType = ESC_HEXA;
} else {
escapeType = NO_ESC;
}
if (escapeType != NO_ESC) { // If needs escaping
if (sb == null) {
sb = new StringBuilder(ln + 6);
sb.append(s, 0, i);
}
sb.append('\\');
if (escapeType > 0x20) {
sb.append((char) escapeType);
} else if (escapeType == ESC_HEXA) {
if (!compatibility.jsonCompatible && c < 0x100) {
sb.append('x');
sb.append(toHexDigitUpperCase(c >> 4));
sb.append(toHexDigitUpperCase(c & 0xF));
} else {
sb.append('u');
int cp = c;
sb.append(toHexDigitUpperCase((cp >> 12) & 0xF));
sb.append(toHexDigitUpperCase((cp >> 8) & 0xF));
sb.append(toHexDigitUpperCase((cp >> 4) & 0xF));
sb.append(toHexDigitUpperCase(cp & 0xF));
}
} else { // escapeType == ESC_BACKSLASH
sb.append(c);
}
continue;
}
// Falls through when escapeType == NO_ESC
}
// Needs no escaping
if (sb != null) sb.append(c);
} // for each character
if (quotation != null) {
sb.append(quotation.getSymbol());
}
return sb == null ? s : sb.toString();
}
private static char toHexDigitLowerCase(int d) {
return (char) (d < 0xA ? d + '0' : d - 0xA + 'a');
}
private static char toHexDigitUpperCase(int d) {
return (char) (d < 0xA ? d + '0' : d - 0xA + 'A');
}
/**
* Parses a name-value pair list, where the pairs are separated with comma,
* and the name and value is separated with colon.
* The keys and values can contain only letters, digits and {@code _}. They
* can't be quoted. White-space around the keys and values are ignored. The
* value can be omitted if defaultValue is not null. When a
* value is omitted, then the colon after the key must be omitted as well.
* The same key can't be used for multiple times.
*
* @param s the string to parse.
* For example: "strong:100, soft:900".
* @param defaultValue the value used when the value is omitted in a
* key-value pair.
*
* @return the map that contains the name-value pairs.
*
* @throws java.text.ParseException if the string is not a valid name-value
* pair list.
*/
public static Map parseNameValuePairList(String s, String defaultValue)
throws java.text.ParseException {
Map map = new HashMap();
char c = ' ';
int ln = s.length();
int p = 0;
int keyStart;
int valueStart;
String key;
String value;
fetchLoop: while (true) {
// skip ws
while (p < ln) {
c = s.charAt(p);
if (!Character.isWhitespace(c)) {
break;
}
p++;
}
if (p == ln) {
break fetchLoop;
}
keyStart = p;
// seek key end
while (p < ln) {
c = s.charAt(p);
if (!(Character.isLetterOrDigit(c) || c == '_')) {
break;
}
p++;
}
if (keyStart == p) {
throw new java.text.ParseException(
"Expecting letter, digit or \"_\" "
+ "here, (the first character of the key) but found "
+ jQuote(String.valueOf(c))
+ " at position " + p + ".",
p);
}
key = s.substring(keyStart, p);
// skip ws
while (p < ln) {
c = s.charAt(p);
if (!Character.isWhitespace(c)) {
break;
}
p++;
}
if (p == ln) {
if (defaultValue == null) {
throw new java.text.ParseException(
"Expecting \":\", but reached "
+ "the end of the string "
+ " at position " + p + ".",
p);
}
value = defaultValue;
} else if (c != ':') {
if (defaultValue == null || c != ',') {
throw new java.text.ParseException(
"Expecting \":\" here, but found "
+ jQuote(String.valueOf(c))
+ " at position " + p + ".",
p);
}
// skip ","
p++;
value = defaultValue;
} else {
// skip ":"
p++;
// skip ws
while (p < ln) {
c = s.charAt(p);
if (!Character.isWhitespace(c)) {
break;
}
p++;
}
if (p == ln) {
throw new java.text.ParseException(
"Expecting the value of the key "
+ "here, but reached the end of the string "
+ " at position " + p + ".",
p);
}
valueStart = p;
// seek value end
while (p < ln) {
c = s.charAt(p);
if (!(Character.isLetterOrDigit(c) || c == '_')) {
break;
}
p++;
}
if (valueStart == p) {
throw new java.text.ParseException(
"Expecting letter, digit or \"_\" "
+ "here, (the first character of the value) "
+ "but found "
+ jQuote(String.valueOf(c))
+ " at position " + p + ".",
p);
}
value = s.substring(valueStart, p);
// skip ws
while (p < ln) {
c = s.charAt(p);
if (!Character.isWhitespace(c)) {
break;
}
p++;
}
// skip ","
if (p < ln) {
if (c != ',') {
throw new java.text.ParseException(
"Excpecting \",\" or the end "
+ "of the string here, but found "
+ jQuote(String.valueOf(c))
+ " at position " + p + ".",
p);
} else {
p++;
}
}
}
// store the key-value pair
if (map.put(key, value) != null) {
throw new java.text.ParseException(
"Dublicated key: "
+ jQuote(key), keyStart);
}
}
return map;
}
/**
* Used internally by the XML DOM wrapper to check if the subvariable name is just an element name, or a more
* complex XPath expression.
*
* @return whether the name is a valid XML element name. (This routine might only be 99% accurate. REVISIT)
*
* @deprecated Don't use this outside FreeMarker; it's name if misleading, and it doesn't follow the XML specs.
*/
@Deprecated
static public boolean isXMLID(String name) {
return _ExtDomApi.isXMLNameLike(name);
}
/**
* @return whether the qname matches the combination of nodeName, nsURI, and environment prefix settings.
*/
static public boolean matchesName(String qname, String nodeName, String nsURI, Environment env) {
return _ExtDomApi.matchesName(qname, nodeName, nsURI, env);
}
/**
* Pads the string at the left with spaces until it reaches the desired
* length. If the string is longer than this length, then it returns the
* unchanged string.
*
* @param s the string that will be padded.
* @param minLength the length to reach.
*/
public static String leftPad(String s, int minLength) {
return leftPad(s, minLength, ' ');
}
/**
* Pads the string at the left with the specified character until it reaches
* the desired length. If the string is longer than this length, then it
* returns the unchanged string.
*
* @param s the string that will be padded.
* @param minLength the length to reach.
* @param filling the filling pattern.
*/
public static String leftPad(String s, int minLength, char filling) {
int ln = s.length();
if (minLength <= ln) {
return s;
}
StringBuilder res = new StringBuilder(minLength);
int dif = minLength - ln;
for (int i = 0; i < dif; i++) {
res.append(filling);
}
res.append(s);
return res.toString();
}
/**
* Pads the string at the left with a filling pattern until it reaches the
* desired length. If the string is longer than this length, then it returns
* the unchanged string. For example: leftPad('ABC', 9, '1234')
* returns "123412ABC".
*
* @param s the string that will be padded.
* @param minLength the length to reach.
* @param filling the filling pattern. Must be at least 1 characters long.
* Can't be null.
*/
public static String leftPad(String s, int minLength, String filling) {
int ln = s.length();
if (minLength <= ln) {
return s;
}
StringBuilder res = new StringBuilder(minLength);
int dif = minLength - ln;
int fln = filling.length();
if (fln == 0) {
throw new IllegalArgumentException(
"The \"filling\" argument can't be 0 length string.");
}
int cnt = dif / fln;
for (int i = 0; i < cnt; i++) {
res.append(filling);
}
cnt = dif % fln;
for (int i = 0; i < cnt; i++) {
res.append(filling.charAt(i));
}
res.append(s);
return res.toString();
}
/**
* Pads the string at the right with spaces until it reaches the desired
* length. If the string is longer than this length, then it returns the
* unchanged string.
*
* @param s the string that will be padded.
* @param minLength the length to reach.
*/
public static String rightPad(String s, int minLength) {
return rightPad(s, minLength, ' ');
}
/**
* Pads the string at the right with the specified character until it
* reaches the desired length. If the string is longer than this length,
* then it returns the unchanged string.
*
* @param s the string that will be padded.
* @param minLength the length to reach.
* @param filling the filling pattern.
*/
public static String rightPad(String s, int minLength, char filling) {
int ln = s.length();
if (minLength <= ln) {
return s;
}
StringBuilder res = new StringBuilder(minLength);
res.append(s);
int dif = minLength - ln;
for (int i = 0; i < dif; i++) {
res.append(filling);
}
return res.toString();
}
/**
* Pads the string at the right with a filling pattern until it reaches the
* desired length. If the string is longer than this length, then it returns
* the unchanged string. For example: rightPad('ABC', 9, '1234')
* returns "ABC412341". Note that the filling pattern is
* started as if you overlay "123412341" with the left-aligned
* "ABC", so it starts with "4".
*
* @param s the string that will be padded.
* @param minLength the length to reach.
* @param filling the filling pattern. Must be at least 1 characters long.
* Can't be null.
*/
public static String rightPad(String s, int minLength, String filling) {
int ln = s.length();
if (minLength <= ln) {
return s;
}
StringBuilder res = new StringBuilder(minLength);
res.append(s);
int dif = minLength - ln;
int fln = filling.length();
if (fln == 0) {
throw new IllegalArgumentException(
"The \"filling\" argument can't be 0 length string.");
}
int start = ln % fln;
int end = fln - start <= dif
? fln
: start + dif;
for (int i = start; i < end; i++) {
res.append(filling.charAt(i));
}
dif -= end - start;
int cnt = dif / fln;
for (int i = 0; i < cnt; i++) {
res.append(filling);
}
cnt = dif % fln;
for (int i = 0; i < cnt; i++) {
res.append(filling.charAt(i));
}
return res.toString();
}
/**
* Converts a version number string to an integer for easy comparison.
* The version number must start with numbers separated with
* dots. There can be any number of such dot-separated numbers, but only
* the first three will be considered. After the numbers arbitrary text can
* follow, and will be ignored.
*
* The string will be trimmed before interpretation.
*
* @return major * 1000000 + minor * 1000 + micro
*/
public static int versionStringToInt(String version) {
return new Version(version).intValue();
}
/**
* Tries to run {@code toString()}, but if that fails, returns a
* {@code "[com.example.SomeClass.toString() failed: " + e + "]"} instead. Also, it returns {@code null} for
* {@code null} parameter.
*
* @since 2.3.20
*/
public static String tryToString(Object object) {
if (object == null) return null;
try {
return object.toString();
} catch (Throwable e) {
return failedToStringSubstitute(object, e);
}
}
private static String failedToStringSubstitute(Object object, Throwable e) {
String eStr;
try {
eStr = e.toString();
} catch (Throwable e2) {
eStr = ClassUtil.getShortClassNameOfObject(e);
}
return "[" + ClassUtil.getShortClassNameOfObject(object) + ".toString() failed: " + eStr + "]";
}
/**
* Converts {@code 1}, {@code 2}, {@code 3} and so forth to {@code "A"}, {@code "B"}, {@code "C"} and so fort. When
* reaching {@code "Z"}, it continues like {@code "AA"}, {@code "AB"}, etc. The lowest supported number is 1, but
* there's no upper limit.
*
* @throws IllegalArgumentException
* If the argument is 0 or less.
*
* @since 2.3.22
*/
public static String toUpperABC(int n) {
return toABC(n, 'A');
}
/**
* Same as {@link #toUpperABC(int)}, but produces lower case result, like {@code "ab"}.
*
* @since 2.3.22
*/
public static String toLowerABC(int n) {
return toABC(n, 'a');
}
/**
* @param oneDigit
* The character that stands for the value 1.
*/
private static String toABC(final int n, char oneDigit) {
if (n < 1) {
throw new IllegalArgumentException("Can't convert 0 or negative "
+ "numbers to latin-number: " + n);
}
// First find out how many "digits" will we need. We start from A, then
// try AA, then AAA, etc. (Note that the smallest digit is "A", which is
// 1, not 0. Hence this isn't like a usual 26-based number-system):
int reached = 1;
int weight = 1;
while (true) {
int nextWeight = weight * 26;
int nextReached = reached + nextWeight;
if (nextReached <= n) {
// So we will have one more digit
weight = nextWeight;
reached = nextReached;
} else {
// No more digits
break;
}
}
// Increase the digits of the place values until we get as close
// to n as possible (but don't step over it).
StringBuilder sb = new StringBuilder();
while (weight != 0) {
// digitIncrease: how many we increase the digit which is already 1
final int digitIncrease = (n - reached) / weight;
sb.append((char) (oneDigit + digitIncrease));
reached += digitIncrease * weight;
weight /= 26;
}
return sb.toString();
}
/**
* Behaves exactly like {@link String#trim()}, but works on arrays. If the resulting array would have the same
* content after trimming, it returns the original array instance. Otherwise it returns a new array instance (or
* {@link CollectionUtils#EMPTY_CHAR_ARRAY}).
*
* @since 2.3.22
*/
public static char[] trim(final char[] cs) {
if (cs.length == 0) {
return cs;
}
int start = 0;
int end = cs.length;
while (start < end && cs[start] <= ' ') {
start++;
}
while (start < end && cs[end - 1] <= ' ') {
end--;
}
if (start == 0 && end == cs.length) {
return cs;
}
if (start == end) {
return CollectionUtils.EMPTY_CHAR_ARRAY;
}
char[] newCs = new char[end - start];
System.arraycopy(cs, start, newCs, 0, end - start);
return newCs;
}
/**
* Tells if {@link String#trim()} will return a 0-length string for the {@link String} equivalent of the argument.
*
* @since 2.3.22
*/
public static boolean isTrimmableToEmpty(char[] text) {
return isTrimmableToEmpty(text, 0, text.length);
}
/**
* Like {@link #isTrimmableToEmpty(char[])}, but acts on a sub-array that starts at {@code start} (inclusive index).
*
* @since 2.3.23
*/
public static boolean isTrimmableToEmpty(char[] text, int start) {
return isTrimmableToEmpty(text, start, text.length);
}
/**
* Like {@link #isTrimmableToEmpty(char[])}, but acts on a sub-array that starts at {@code start} (inclusive index)
* and ends at {@code end} (exclusive index).
*
* @since 2.3.23
*/
public static boolean isTrimmableToEmpty(char[] text, int start, int end) {
for (int i = start; i < end; i++) {
// We follow Java's String.trim() here, which simply states that c <= ' ' is whitespace.
if (text[i] > ' ') {
return false;
}
}
return true;
}
/**
* Same as {@link #globToRegularExpression(String, boolean)} with {@code caseInsensitive} argument {@code false}.
*
* @since 2.3.24
*/
public static Pattern globToRegularExpression(String glob) {
return globToRegularExpression(glob, false);
}
/**
* Creates a regular expression from a glob. The glob must use {@code /} for as file separator, not {@code \}
* (backslash), and is always case sensitive.
*
* This glob implementation recognizes these special characters:
*
* - {@code ?}: Wildcard that matches exactly one character, other than {@code /}
*
- {@code *}: Wildcard that matches zero, one or multiple characters, other than {@code /}
*
- {@code **}: Wildcard that matches zero, one or multiple directories. For example, {@code **}{@code /head.ftl}
* matches {@code foo/bar/head.ftl}, {@code foo/head.ftl} and {@code head.ftl} too. {@code **} must be either
* preceded by {@code /} or be at the beginning of the glob. {@code **} must be either followed by {@code /} or be
* at the end of the glob. When {@code **} is at the end of the glob, it also matches file names, like
* {@code a/**} matches {@code a/b/c.ftl}. If the glob only consist of a {@code **}, it will be a match for
* everything.
*
- {@code \} (backslash): Makes the next character non-special (a literal). For example {@code How\?.ftl} will
* match {@code How?.ftl}, but not {@code HowX.ftl}. Naturally, two backslashes produce one literal backslash.
*
- {@code [}: Reserved for future purposes; can't be used
*
{: Reserved for future purposes; can't be used
*
*
* @since 2.3.24
*/
public static Pattern globToRegularExpression(String glob, boolean caseInsensitive) {
StringBuilder regex = new StringBuilder();
int nextStart = 0;
boolean escaped = false;
int ln = glob.length();
for (int idx = 0; idx < ln; idx++) {
char c = glob.charAt(idx);
if (!escaped) {
if (c == '?') {
appendLiteralGlobSection(regex, glob, nextStart, idx);
regex.append("[^/]");
nextStart = idx + 1;
} else if (c == '*') {
appendLiteralGlobSection(regex, glob, nextStart, idx);
if (idx + 1 < ln && glob.charAt(idx + 1) == '*') {
if (!(idx == 0 || glob.charAt(idx - 1) == '/')) {
throw new IllegalArgumentException(
"The \"**\" wildcard must be directly after a \"/\" or it must be at the "
+ "beginning, in this glob: " + glob);
}
if (idx + 2 == ln) { // trailing "**"
regex.append(".*");
idx++;
} else { // "**/"
if (!(idx + 2 < ln && glob.charAt(idx + 2) == '/')) {
throw new IllegalArgumentException(
"The \"**\" wildcard must be followed by \"/\", or must be at tehe end, "
+ "in this glob: " + glob);
}
regex.append("(.*?/)*");
idx += 2; // "*/".length()
}
} else {
regex.append("[^/]*");
}
nextStart = idx + 1;
} else if (c == '\\') {
escaped = true;
} else if (c == '[' || c == '{') {
throw new IllegalArgumentException(
"The \"" + c + "\" glob operator is currently unsupported "
+ "(precede it with \\ for literal matching), "
+ "in this glob: " + glob);
}
} else {
escaped = false;
}
}
appendLiteralGlobSection(regex, glob, nextStart, glob.length());
return Pattern.compile(regex.toString(), caseInsensitive ? Pattern.CASE_INSENSITIVE | Pattern.UNICODE_CASE : 0);
}
private static void appendLiteralGlobSection(StringBuilder regex, String glob, int start, int end) {
if (start == end) return;
String part = unescapeLiteralGlobSection(glob.substring(start, end));
regex.append(Pattern.quote(part));
}
private static String unescapeLiteralGlobSection(String s) {
int backslashIdx = s.indexOf('\\');
if (backslashIdx == -1) {
return s;
}
int ln = s.length();
StringBuilder sb = new StringBuilder(ln - 1);
int nextStart = 0;
do {
sb.append(s, nextStart, backslashIdx);
nextStart = backslashIdx + 1;
} while ((backslashIdx = s.indexOf('\\', nextStart + 1)) != -1);
if (nextStart < ln) {
sb.append(s, nextStart, ln);
}
return sb.toString();
}
/**
* Used as the argument of {@link StringUtil#jsStringEnc(String, JsStringEncCompatibility, JsStringEncQuotation)}.
*
* @since 2.3.32
*/
public enum JsStringEncCompatibility {
/**
* Output is expected to be used in JavaScript, not in JSON.
*/
JAVA_SCRIPT(true, false),
/**
* Output is expected to be used in JSON, not in JavaScript. While JSON is compatible with JavaScript, in this
* mode we don't care about escaping apostrophe, as it's not special in JSON.
*/
JSON(false, true),
/**
* Output is expected to be used both in JSON and JavaScript.
*/
JAVA_SCRIPT_OR_JSON(true, true);
JsStringEncCompatibility(boolean javaScriptCompatible, boolean jsonCompatible) {
this.javaScriptCompatible = javaScriptCompatible;
this.jsonCompatible = jsonCompatible;
}
private final boolean javaScriptCompatible;
private final boolean jsonCompatible;
boolean isJSONCompatible() {
return jsonCompatible;
}
}
/**
* Used as the argument of {@link StringUtil#jsStringEnc(String, JsStringEncCompatibility, JsStringEncQuotation)}.
*
* @since 2.3.32
*/
public enum JsStringEncQuotation {
QUOTATION_MARK('"'),
APOSTROPHE('\'');
private final char symbol;
JsStringEncQuotation(char symbol) {
this.symbol = symbol;
}
public char getSymbol() {
return symbol;
}
}
}