*/
package jalview.util;
+import java.io.UnsupportedEncodingException;
+import java.net.URLEncoder;
import java.util.ArrayList;
-import java.util.HashMap;
import java.util.List;
-import java.util.Map;
+import java.util.Locale;
+import java.util.regex.Matcher;
import java.util.regex.Pattern;
public class StringUtils
private static final Pattern DELIMITERS_PATTERN = Pattern
.compile(".*='[^']*(?!')");
+ private static final char PERCENT = '%';
+
private static final boolean DEBUG = false;
+ /*
+ * URL encoded characters, indexed by char value
+ * e.g. urlEncodings['='] = urlEncodings[61] = "%3D"
+ */
+ private static String[] urlEncodings = new String[255];
+
/**
* Returns a new character array, after inserting characters into the given
* character array.
* @param delimiter
* @return elements separated by separator
*/
- public static String[] separatorListToArray(String input, String delimiter)
+ public static String[] separatorListToArray(String input,
+ String delimiter)
{
int seplen = delimiter.length();
if (input == null || input.equals("") || input.equals(delimiter))
{
return null;
}
- List<String> jv = new ArrayList<String>();
+ List<String> jv = new ArrayList<>();
int cp = 0, pos, escape;
boolean wasescaped = false, wasquoted = false;
String lstitem = null;
if (wasescaped || wasquoted)
{
// append to previous pos
- jv.set(jv.size() - 1,
- lstitem = lstitem + delimiter
- + input.substring(cp, pos + escape));
+ jv.set(jv.size() - 1, lstitem = lstitem + delimiter
+ + input.substring(cp, pos + escape));
}
else
{
}
if (DEBUG)
{
- System.err.println("Empty Array from '" + delimiter
- + "' separated List");
+ System.err.println(
+ "Empty Array from '" + delimiter + "' separated List");
}
return null;
}
}
if (DEBUG)
{
- System.err.println("Returning '" + separator
- + "' separated List:\n");
+ System.err
+ .println("Returning '" + separator + "' separated List:\n");
System.err.println(v);
}
return v.toString();
}
if (DEBUG)
{
- System.err.println("Returning empty '" + separator
- + "' separated List\n");
+ System.err.println(
+ "Returning empty '" + separator + "' separated List\n");
}
return "" + separator;
}
-
- /**
- * Parses the input line to a map of name / value(s) pairs. For example the
- * line <br>
- * Notes=Fe-S;Method=manual curation; source = Pfam; Notes = Metal <br>
- * if parsed with delimiter=";" and separators {' ', '='} <br>
- * would return a map with { Notes={Fe=S, Metal}, Method={manual curation},
- * source={Pfam}} <br>
- * Note the name/value strings are trimmed of leading / trailing spaces; the
- * first separator encountered is used
- *
- * @param line
- * @param delimiter
- * the major delimiter between name-value pairs
- * @param separators
- * one or more separators used between name and value
- * @return the name-values map (which may be empty but never null)
- */
- public static Map<String, List<String>> parseNameValuePairs(String line,
- String delimiter, char[] separators)
- {
- Map<String, List<String>> map = new HashMap<String, List<String>>();
- if (line == null || line.trim().length() == 0)
- {
- return map;
- }
-
- for (String pair : line.trim().split(delimiter))
- {
- pair = pair.trim();
- if (pair.length() == 0)
- {
- continue;
- }
-
- int sepPos = -1;
- for (char sep : separators)
- {
- int pos = pair.indexOf(sep);
- if (pos > -1 && (sepPos == -1 || pos < sepPos))
- {
- sepPos = pos;
- }
- }
-
- if (sepPos == -1)
- {
- // no name=value detected
- continue;
- }
-
- String key = pair.substring(0, sepPos).trim();
- String value = pair.substring(sepPos + 1).trim();
- if (value.length() > 0)
- {
- List<String> vals = map.get(key);
- if (vals == null)
- {
- vals = new ArrayList<String>();
- map.put(key, vals);
- }
- vals.add(value);
- }
- }
- return map;
- }
/**
* Converts a list to a string with a delimiter before each term except the
}
return result;
}
+
+ /**
+ * Compares two versions formatted as e.g. "3.4.5" and returns -1, 0 or 1 as
+ * the first version precedes, is equal to, or follows the second
+ *
+ * @param v1
+ * @param v2
+ * @return
+ */
+ public static int compareVersions(String v1, String v2)
+ {
+ return compareVersions(v1, v2, null);
+ }
+
+ /**
+ * Compares two versions formatted as e.g. "3.4.5b1" and returns -1, 0 or 1 as
+ * the first version precedes, is equal to, or follows the second
+ *
+ * @param v1
+ * @param v2
+ * @param pointSeparator
+ * a string used to delimit point increments in sub-tokens of the
+ * version
+ * @return
+ */
+ public static int compareVersions(String v1, String v2,
+ String pointSeparator)
+ {
+ if (v1 == null || v2 == null)
+ {
+ return 0;
+ }
+ String[] toks1 = v1.split("\\.");
+ String[] toks2 = v2.split("\\.");
+ int i = 0;
+ for (; i < toks1.length; i++)
+ {
+ if (i >= toks2.length)
+ {
+ /*
+ * extra tokens in v1
+ */
+ return 1;
+ }
+ String tok1 = toks1[i];
+ String tok2 = toks2[i];
+ if (pointSeparator != null)
+ {
+ /*
+ * convert e.g. 5b2 into decimal 5.2 for comparison purposes
+ */
+ tok1 = tok1.replace(pointSeparator, ".");
+ tok2 = tok2.replace(pointSeparator, ".");
+ }
+ try
+ {
+ float f1 = Float.valueOf(tok1);
+ float f2 = Float.valueOf(tok2);
+ int comp = Float.compare(f1, f2);
+ if (comp != 0)
+ {
+ return comp;
+ }
+ } catch (NumberFormatException e)
+ {
+ System.err
+ .println("Invalid version format found: " + e.getMessage());
+ return 0;
+ }
+ }
+
+ if (i < toks2.length)
+ {
+ /*
+ * extra tokens in v2
+ */
+ return -1;
+ }
+
+ /*
+ * same length, all tokens match
+ */
+ return 0;
+ }
+
+ /**
+ * Converts the string to all lower-case except the first character which is
+ * upper-cased
+ *
+ * @param s
+ * @return
+ */
+ public static String toSentenceCase(String s)
+ {
+ if (s == null)
+ {
+ return s;
+ }
+ if (s.length() <= 1)
+ {
+ return s.toUpperCase(Locale.ROOT);
+ }
+ return s.substring(0, 1).toUpperCase(Locale.ROOT)
+ + s.substring(1).toLowerCase(Locale.ROOT);
+ }
+
+ /**
+ * A helper method that strips off any leading or trailing html and body tags.
+ * If no html tag is found, then also html-encodes angle bracket characters.
+ *
+ * @param text
+ * @return
+ */
+ public static String stripHtmlTags(String text)
+ {
+ if (text == null)
+ {
+ return null;
+ }
+ String tmp2up = text.toUpperCase(Locale.ROOT);
+ int startTag = tmp2up.indexOf("<HTML>");
+ if (startTag > -1)
+ {
+ text = text.substring(startTag + 6);
+ tmp2up = tmp2up.substring(startTag + 6);
+ }
+ // is omission of "<BODY>" intentional here??
+ int endTag = tmp2up.indexOf("</BODY>");
+ if (endTag > -1)
+ {
+ text = text.substring(0, endTag);
+ tmp2up = tmp2up.substring(0, endTag);
+ }
+ endTag = tmp2up.indexOf("</HTML>");
+ if (endTag > -1)
+ {
+ text = text.substring(0, endTag);
+ }
+
+ if (startTag == -1 && (text.contains("<") || text.contains(">")))
+ {
+ text = text.replaceAll("<", "<");
+ text = text.replaceAll(">", ">");
+ }
+ return text;
+ }
+
+ /**
+ * Answers the input string with any occurrences of the 'encodeable'
+ * characters replaced by their URL encoding
+ *
+ * @param s
+ * @param encodable
+ * @return
+ */
+ public static String urlEncode(String s, String encodable)
+ {
+ if (s == null || s.isEmpty())
+ {
+ return s;
+ }
+
+ /*
+ * do % encoding first, as otherwise it may double-encode!
+ */
+ if (encodable.indexOf(PERCENT) != -1)
+ {
+ s = urlEncode(s, PERCENT);
+ }
+
+ for (char c : encodable.toCharArray())
+ {
+ if (c != PERCENT)
+ {
+ s = urlEncode(s, c);
+ }
+ }
+ return s;
+ }
+
+ /**
+ * Answers the input string with any occurrences of {@code c} replaced with
+ * their url encoding. Answers the input string if it is unchanged.
+ *
+ * @param s
+ * @param c
+ * @return
+ */
+ static String urlEncode(String s, char c)
+ {
+ String decoded = String.valueOf(c);
+ if (s.indexOf(decoded) != -1)
+ {
+ String encoded = getUrlEncoding(c);
+ if (!encoded.equals(decoded))
+ {
+ s = s.replace(decoded, encoded);
+ }
+ }
+ return s;
+ }
+
+ /**
+ * Answers the input string with any occurrences of the specified (unencoded)
+ * characters replaced by their URL decoding.
+ * <p>
+ * Example: {@code urlDecode("a%3Db%3Bc", "-;=,")} should answer
+ * {@code "a=b;c"}.
+ *
+ * @param s
+ * @param encodable
+ * @return
+ */
+ public static String urlDecode(String s, String encodable)
+ {
+ if (s == null || s.isEmpty())
+ {
+ return s;
+ }
+
+ for (char c : encodable.toCharArray())
+ {
+ String encoded = getUrlEncoding(c);
+ if (s.indexOf(encoded) != -1)
+ {
+ String decoded = String.valueOf(c);
+ s = s.replace(encoded, decoded);
+ }
+ }
+ return s;
+ }
+
+ /**
+ * Does a lazy lookup of the url encoding of the given character, saving the
+ * value for repeat lookups
+ *
+ * @param c
+ * @return
+ */
+ private static String getUrlEncoding(char c)
+ {
+ if (c < 0 || c >= urlEncodings.length)
+ {
+ return String.valueOf(c);
+ }
+
+ String enc = urlEncodings[c];
+ if (enc == null)
+ {
+ try
+ {
+ enc = urlEncodings[c] = URLEncoder.encode(String.valueOf(c),
+ "UTF-8");
+ } catch (UnsupportedEncodingException e)
+ {
+ enc = urlEncodings[c] = String.valueOf(c);
+ }
+ }
+ return enc;
+ }
+
+ public static int firstCharPosIgnoreCase(String text, String chars)
+ {
+ int min = text.length() + 1;
+ for (char c : chars.toLowerCase(Locale.ROOT).toCharArray())
+ {
+ int i = text.toLowerCase(Locale.ROOT).indexOf(c);
+ if (0 <= i && i < min)
+ {
+ min = i;
+ }
+ }
+ return min < text.length() + 1 ? min : -1;
+ }
+
+ public static int indexOfFirstWhitespace(String text)
+ {
+ int index = -1;
+ Pattern pat = Pattern.compile("\\s");
+ Matcher m = pat.matcher(text);
+ if (m.find())
+ {
+ index = m.start();
+ }
+ return index;
+ }
+
+ /*
+ * implementation of String.replaceLast.
+ * Replaces only the last occurrence of toReplace in string with replacement.
+ */
+ public static String replaceLast(String string, String toReplace,
+ String replacement)
+ {
+ int pos = string.lastIndexOf(toReplace);
+ if (pos > -1)
+ {
+ return new StringBuilder().append(string.substring(0, pos))
+ .append(replacement)
+ .append(string.substring(pos + toReplace.length()))
+ .toString();
+ }
+ else
+ {
+ return string;
+ }
+ }
+
}