2020-04-27 14:52:31 +02:00
|
|
|
|
2021-02-06 20:12:00 +01:00
|
|
|
package eu.dnetlib.dhp.collection.worker;
|
2019-04-03 16:05:16 +02:00
|
|
|
|
|
|
|
import java.util.HashMap;
|
|
|
|
import java.util.HashSet;
|
|
|
|
import java.util.Map;
|
|
|
|
import java.util.Set;
|
|
|
|
import java.util.regex.Pattern;
|
|
|
|
|
2020-04-18 12:42:58 +02:00
|
|
|
/** @author jochen, Andreas Czerniak */
|
2019-04-03 16:05:16 +02:00
|
|
|
public class XmlCleaner {
|
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
/** Pattern for numeric entities. */
|
2020-04-29 19:09:07 +02:00
|
|
|
private static final Pattern validCharacterEntityPattern = Pattern.compile("^&#x?\\d{2,4};"); // $NON-NLS-1$
|
2020-04-27 14:52:31 +02:00
|
|
|
// private static Pattern validCharacterEntityPattern = Pattern.compile("^&#?\\d{2,4};");
|
|
|
|
// //$NON-NLS-1$
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
// see https://www.w3.org/TR/REC-xml/#charsets , not only limited to 
|
2020-04-29 19:09:07 +02:00
|
|
|
private static final Pattern invalidControlCharPattern = Pattern.compile("&#x?1[0-9a-fA-F];");
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
/**
|
|
|
|
* Pattern that negates the allowable XML 4 byte unicode characters. Valid are: #x9 | #xA | #xD | [#x20-#xD7FF] |
|
|
|
|
* [#xE000-#xFFFD] | [#x10000-#x10FFFF]
|
|
|
|
*/
|
2020-04-29 19:09:07 +02:00
|
|
|
private static final Pattern invalidCharacterPattern = Pattern.compile("[^\t\r\n\u0020-\uD7FF\uE000-\uFFFD]"); // $NON-NLS-1$
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
// Map entities to their unicode equivalent
|
2020-04-29 19:09:07 +02:00
|
|
|
private static final Set<String> goodEntities = new HashSet<>();
|
|
|
|
private static final Map<String, String> badEntities = new HashMap<>();
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
static {
|
|
|
|
// pre-defined XML entities
|
|
|
|
goodEntities.add("""); // $NON-NLS-1$ // quotation mark
|
|
|
|
goodEntities.add("&"); // $NON-NLS-1$ // ampersand
|
|
|
|
goodEntities.add("<"); // $NON-NLS-1$ // less-than sign
|
|
|
|
goodEntities.add(">"); // $NON-NLS-1$ // greater-than sign
|
|
|
|
// control entities
|
|
|
|
// badEntities.put("", "");
|
|
|
|
badEntities.put("", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("€", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("‚", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("ƒ", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("„", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("…", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("†", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("‡", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("ˆ", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("‰", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("Š", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("‹", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("Œ", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("Ž", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("‘", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("’", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("“", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("”", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("•", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("–", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("—", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("˜", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("™", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("š", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("›", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("œ", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("ž", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
badEntities.put("Ÿ", " "); // $NON-NLS-1$ //$NON-NLS-2$ // illegal HTML character
|
|
|
|
// misc entities
|
|
|
|
badEntities.put("€", "\u20AC"); // $NON-NLS-1$ //$NON-NLS-2$ // euro
|
|
|
|
badEntities.put("‘", "\u2018"); // $NON-NLS-1$ //$NON-NLS-2$ // left single quotation
|
|
|
|
// mark
|
|
|
|
badEntities.put("’", "\u2019"); // $NON-NLS-1$ //$NON-NLS-2$ // right single quotation
|
|
|
|
// mark
|
|
|
|
// Latin 1 entities
|
|
|
|
badEntities.put(" ", "\u00A0"); // $NON-NLS-1$ //$NON-NLS-2$ // no-break space
|
|
|
|
badEntities.put("¡", "\u00A1"); // $NON-NLS-1$ //$NON-NLS-2$ // inverted exclamation
|
|
|
|
// mark
|
|
|
|
badEntities.put("¢", "\u00A2"); // $NON-NLS-1$ //$NON-NLS-2$ // cent sign
|
|
|
|
badEntities.put("£", "\u00A3"); // $NON-NLS-1$ //$NON-NLS-2$ // pound sign
|
|
|
|
badEntities.put("¤", "\u00A4"); // $NON-NLS-1$ //$NON-NLS-2$ // currency sign
|
|
|
|
badEntities.put("¥", "\u00A5"); // $NON-NLS-1$ //$NON-NLS-2$ // yen sign
|
|
|
|
badEntities.put("¦", "\u00A6"); // $NON-NLS-1$ //$NON-NLS-2$ // broken vertical bar
|
|
|
|
badEntities.put("§", "\u00A7"); // $NON-NLS-1$ //$NON-NLS-2$ // section sign
|
|
|
|
badEntities.put("¨", "\u00A8"); // $NON-NLS-1$ //$NON-NLS-2$ // diaeresis
|
|
|
|
badEntities.put("©", "\u00A9"); // $NON-NLS-1$ //$NON-NLS-2$ // copyright sign
|
|
|
|
badEntities.put("ª", "\u00AA"); // $NON-NLS-1$ //$NON-NLS-2$ // feminine ordinal
|
|
|
|
// indicator
|
|
|
|
badEntities.put("«", "\u00AB"); // $NON-NLS-2$ // left-pointing double angle
|
|
|
|
// quotation mark
|
|
|
|
badEntities.put("¬", "\u00AC"); // $NON-NLS-1$ //$NON-NLS-2$ // not sign
|
|
|
|
badEntities.put("­", "\u00AD"); // $NON-NLS-1$ //$NON-NLS-2$ // soft hyphen
|
|
|
|
badEntities.put("®", "\u00AE"); // $NON-NLS-1$ //$NON-NLS-2$ // registered sign
|
|
|
|
badEntities.put("¯", "\u00AF"); // $NON-NLS-1$ //$NON-NLS-2$ // macron
|
|
|
|
badEntities.put("°", "\u00B0"); // $NON-NLS-1$ //$NON-NLS-2$ // degree sign
|
|
|
|
badEntities.put("±", "\u00B1"); // $NON-NLS-1$ //$NON-NLS-2$ // plus-minus sign
|
|
|
|
badEntities.put("²", "\u00B2"); // $NON-NLS-1$ //$NON-NLS-2$ // superscript two
|
|
|
|
badEntities.put("³", "\u00B3"); // $NON-NLS-1$ //$NON-NLS-2$ // superscript three
|
|
|
|
badEntities.put("´", "\u00B4"); // $NON-NLS-1$ //$NON-NLS-2$ // acute accent
|
|
|
|
badEntities.put("µ", "\u00B5"); // $NON-NLS-1$ //$NON-NLS-2$ // micro sign
|
|
|
|
badEntities.put("¶", "\u00B6"); // $NON-NLS-1$ //$NON-NLS-2$ // pilcrow sign
|
|
|
|
badEntities.put("·", "\u00B7"); // $NON-NLS-1$ //$NON-NLS-2$ // middle dot
|
|
|
|
badEntities.put("¸", "\u00B8"); // $NON-NLS-1$ //$NON-NLS-2$ // cedilla
|
|
|
|
badEntities.put("¹", "\u00B9"); // $NON-NLS-1$ //$NON-NLS-2$ // superscript one
|
|
|
|
badEntities.put("º", "\u00BA"); // $NON-NLS-1$ //$NON-NLS-2$ // masculine ordinal
|
|
|
|
// indicator
|
|
|
|
badEntities.put("»", "\u00BB"); // $NON-NLS-2$ // right-pointing double
|
|
|
|
// angle quotation
|
|
|
|
// mark
|
|
|
|
badEntities.put("¼", "\u00BC"); // $NON-NLS-1$ //$NON-NLS-2$ // vulgar fraction one
|
|
|
|
// quarter
|
|
|
|
badEntities.put("½", "\u00BD"); // $NON-NLS-1$ //$NON-NLS-2$ // vulgar fraction one
|
|
|
|
// half
|
|
|
|
badEntities.put("¾", "\u00BE"); // $NON-NLS-1$ //$NON-NLS-2$ // vulgar fraction three
|
|
|
|
// quarters
|
|
|
|
badEntities.put("¿", "\u00BF"); // $NON-NLS-1$ //$NON-NLS-2$ // inverted question
|
|
|
|
// mark
|
|
|
|
badEntities.put("À", "\u00C0"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// A
|
|
|
|
// with grave
|
|
|
|
badEntities.put("Á", "\u00C1"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// A
|
|
|
|
// with acute
|
|
|
|
badEntities.put("Â", "\u00C2"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter A
|
|
|
|
// with circumflex
|
|
|
|
badEntities.put("Ã", "\u00C3"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// A
|
|
|
|
// with tilde
|
|
|
|
badEntities.put("Ä", "\u00C4"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter A
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("Å", "\u00C5"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter A
|
|
|
|
// with ring above
|
|
|
|
badEntities.put("Æ", "\u00C6"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// AE
|
|
|
|
badEntities.put("Ç", "\u00C7"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// C
|
|
|
|
// with cedilla
|
|
|
|
badEntities.put("È", "\u00C8"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// E
|
|
|
|
// with grave
|
|
|
|
badEntities.put("É", "\u00C9"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// E
|
|
|
|
// with acute
|
|
|
|
badEntities.put("Ê", "\u00CA"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter E
|
|
|
|
// with circumflex
|
|
|
|
badEntities.put("Ë", "\u00CB"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter E
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("Ì", "\u00CC"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// I
|
|
|
|
// with grave
|
|
|
|
badEntities.put("Í", "\u00CD"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// I
|
|
|
|
// with acute
|
|
|
|
badEntities.put("Î", "\u00CE"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter I
|
|
|
|
// with circumflex
|
|
|
|
badEntities.put("Ï", "\u00CF"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter I
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("Ð", "\u00D0"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter ETH
|
|
|
|
badEntities.put("Ñ", "\u00D1"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// N
|
|
|
|
// with tilde
|
|
|
|
badEntities.put("Ò", "\u00D2"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// O
|
|
|
|
// with grave
|
|
|
|
badEntities.put("Ó", "\u00D3"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// O
|
|
|
|
// with acute
|
|
|
|
badEntities.put("Ô", "\u00D4"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter O
|
|
|
|
// with circumflex
|
|
|
|
badEntities.put("Õ", "\u00D5"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// O
|
|
|
|
// with tilde
|
|
|
|
badEntities.put("Ö", "\u00D6"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter O
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("×", "\u00D7"); // $NON-NLS-1$ //$NON-NLS-2$ // multiplication sign
|
|
|
|
badEntities.put("Ø", "\u00D8"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// O
|
|
|
|
// with stroke
|
|
|
|
badEntities.put("Ù", "\u00D9"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// U
|
|
|
|
// with grave
|
|
|
|
badEntities.put("Ú", "\u00DA"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// U
|
|
|
|
// with acute
|
|
|
|
badEntities.put("Û", "\u00DB"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter U
|
|
|
|
// with circumflex
|
|
|
|
badEntities.put("Ü", "\u00DC"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter U
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("Ý", "\u00DD"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// Y
|
|
|
|
// with acute
|
|
|
|
badEntities.put("Þ", "\u00DE"); // $NON-NLS-1$ //$NON-NLS-2$ // latin capital letter
|
|
|
|
// THORN
|
|
|
|
badEntities.put("ß", "\u00DF"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter
|
|
|
|
// sharp s
|
|
|
|
badEntities.put("à", "\u00E0"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter a
|
|
|
|
// with
|
|
|
|
// grave
|
|
|
|
badEntities.put("á", "\u00E1"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter a
|
|
|
|
// with
|
|
|
|
// acute
|
|
|
|
badEntities.put("â", "\u00E2"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter a
|
|
|
|
// with
|
|
|
|
// circumflex
|
|
|
|
badEntities.put("ã", "\u00E3"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter a
|
|
|
|
// with
|
|
|
|
// tilde
|
|
|
|
badEntities.put("ä", "\u00E4"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter a
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("å", "\u00E5"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter a
|
|
|
|
// with
|
|
|
|
// ring above
|
|
|
|
badEntities.put("æ", "\u00E6"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter ae
|
|
|
|
badEntities.put("ç", "\u00E7"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter c
|
|
|
|
// with
|
|
|
|
// cedilla
|
|
|
|
badEntities.put("è", "\u00E8"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter e
|
|
|
|
// with
|
|
|
|
// grave
|
|
|
|
badEntities.put("é", "\u00E9"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter e
|
|
|
|
// with
|
|
|
|
// acute
|
|
|
|
badEntities.put("ê", "\u00EA"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter e
|
|
|
|
// with
|
|
|
|
// circumflex
|
|
|
|
badEntities.put("ë", "\u00EB"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter e
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("ì", "\u00EC"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter i
|
|
|
|
// with
|
|
|
|
// grave
|
|
|
|
badEntities.put("í", "\u00ED"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter i
|
|
|
|
// with
|
|
|
|
// acute
|
|
|
|
badEntities.put("î", "\u00EE"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter i
|
|
|
|
// with
|
|
|
|
// circumflex
|
|
|
|
badEntities.put("ï", "\u00EF"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter i
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("ð", "\u00F0"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter eth
|
|
|
|
badEntities.put("ñ", "\u00F1"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter n
|
|
|
|
// with
|
|
|
|
// tilde
|
|
|
|
badEntities.put("ò", "\u00F2"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter o
|
|
|
|
// with
|
|
|
|
// grave
|
|
|
|
badEntities.put("ó", "\u00F3"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter o
|
|
|
|
// with
|
|
|
|
// acute
|
|
|
|
badEntities.put("ô", "\u00F4"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter o
|
|
|
|
// with
|
|
|
|
// circumflex
|
|
|
|
badEntities.put("õ", "\u00F5"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter o
|
|
|
|
// with
|
|
|
|
// tilde
|
|
|
|
badEntities.put("ö", "\u00F6"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter o
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("÷", "\u00F7"); // $NON-NLS-1$ //$NON-NLS-2$ // division sign
|
|
|
|
badEntities.put("ø", "\u00F8"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter o
|
|
|
|
// with
|
|
|
|
// stroke
|
|
|
|
badEntities.put("ù", "\u00F9"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter u
|
|
|
|
// with
|
|
|
|
// grave
|
|
|
|
badEntities.put("ú", "\u00FA"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter u
|
|
|
|
// with
|
|
|
|
// acute
|
|
|
|
badEntities.put("û", "\u00FB"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter u
|
|
|
|
// with
|
|
|
|
// circumflex
|
|
|
|
badEntities.put("ü", "\u00FC"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter u
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
badEntities.put("ý", "\u00FD"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter y
|
|
|
|
// with
|
|
|
|
// acute
|
|
|
|
badEntities.put("þ", "\u00FE"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter
|
|
|
|
// thorn
|
|
|
|
badEntities.put("ÿ", "\u00FF"); // $NON-NLS-1$ //$NON-NLS-2$ // latin small letter y
|
|
|
|
// with
|
|
|
|
// diaeresis
|
|
|
|
}
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
/**
|
|
|
|
* For each entity in the input that is not allowed in XML, replace the entity with its unicode equivalent or remove
|
|
|
|
* it. For each instance of a bare {@literal &}, replace it with {@literal
|
|
|
|
* &<br/>
|
|
|
|
} XML only allows 4 entities: {@literal &amp;}, {@literal &quot;}, {@literal
|
|
|
|
* &lt;} and {@literal &gt;}.
|
|
|
|
*
|
|
|
|
* @param broken the string to handle entities
|
|
|
|
* @return the string with entities appropriately fixed up
|
|
|
|
*/
|
|
|
|
public static String cleanAllEntities(final String broken) {
|
|
|
|
if (broken == null) {
|
|
|
|
return null;
|
|
|
|
}
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
String working = invalidControlCharPattern.matcher(broken).replaceAll("");
|
|
|
|
working = invalidCharacterPattern.matcher(working).replaceAll("");
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
int cleanfrom = 0;
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
while (true) {
|
|
|
|
int amp = working.indexOf('&', cleanfrom);
|
|
|
|
// If there are no more amps then we are done
|
|
|
|
if (amp == -1) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
// Skip references of the kind &#ddd;
|
|
|
|
if (validCharacterEntityPattern.matcher(working.substring(amp)).find()) {
|
|
|
|
cleanfrom = working.indexOf(';', amp) + 1;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
int i = amp + 1;
|
|
|
|
while (true) {
|
|
|
|
// if we are at the end of the string then just escape the '&';
|
|
|
|
if (i >= working.length()) {
|
|
|
|
return working.substring(0, amp) + "&" + working.substring(amp + 1); // $NON-NLS-1$
|
|
|
|
}
|
|
|
|
// if we have come to a ; then we have an entity
|
|
|
|
// If it is something that xml can't handle then replace it.
|
|
|
|
final char c = working.charAt(i);
|
|
|
|
if (c == ';') {
|
|
|
|
final String entity = working.substring(amp, i + 1);
|
|
|
|
final String replace = handleEntity(entity);
|
|
|
|
working = working.substring(0, amp) + replace + working.substring(i + 1);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
// Did we end an entity without finding a closing ;
|
|
|
|
// Then treat it as an '&' that needs to be replaced with &
|
|
|
|
if (!Character.isLetterOrDigit(c)) {
|
|
|
|
working = working.substring(0, amp) + "&" + working.substring(amp + 1); // $NON-NLS-1$
|
|
|
|
amp = i + 4; // account for the 4 extra characters
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
i++;
|
|
|
|
}
|
|
|
|
cleanfrom = amp + 1;
|
|
|
|
}
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
if (Pattern.compile("<<").matcher(working).find()) {
|
|
|
|
working = working.replaceAll("<<", "<<");
|
|
|
|
}
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
if (Pattern.compile(">>").matcher(working).find()) {
|
|
|
|
working = working.replaceAll(">>", ">>");
|
|
|
|
}
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
return working;
|
|
|
|
}
|
2019-04-03 16:05:16 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
/**
|
|
|
|
* Replace entity with its unicode equivalent, if it is not a valid XML entity. Otherwise strip it out. XML only
|
|
|
|
* allows 4 entities: &amp;, &quot;, &lt; and &gt;.
|
|
|
|
*
|
|
|
|
* @param entity the entity to be replaced
|
|
|
|
* @return the substitution for the entity, either itself, the unicode equivalent or an empty string.
|
|
|
|
*/
|
|
|
|
private static String handleEntity(final String entity) {
|
|
|
|
if (goodEntities.contains(entity)) {
|
|
|
|
return entity;
|
|
|
|
}
|
2020-04-27 14:45:40 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
final String replace = badEntities.get(entity);
|
|
|
|
if (replace != null) {
|
|
|
|
return replace;
|
|
|
|
}
|
2020-04-27 14:45:40 +02:00
|
|
|
|
2020-04-27 14:52:31 +02:00
|
|
|
return replace != null ? replace : "";
|
|
|
|
}
|
2019-04-03 16:05:16 +02:00
|
|
|
}
|