X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FStringUtils.java;h=1f33522c7ce5c8324d94bd0e1a1e1b222871be46;hb=3609d4b908fa64cab35f2348401baab3347188fc;hp=6044655b80c327eb3a237d82d72a6e46156284da;hpb=8f920d337154e092f5f9056ffde3cdf2735eca43;p=jalview.git

diff --git a/src/jalview/util/StringUtils.java b/src/jalview/util/StringUtils.java
index 6044655..1f33522 100644
--- a/src/jalview/util/StringUtils.java
+++ b/src/jalview/util/StringUtils.java
@@ -20,6 +20,8 @@
  */
 package jalview.util;
 
+import java.io.UnsupportedEncodingException;
+import java.net.URLEncoder;
 import java.util.ArrayList;
 import java.util.List;
 import java.util.regex.Pattern;
@@ -29,8 +31,16 @@ 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.
@@ -107,29 +117,6 @@ public class StringUtils
   }
 
   /**
-   * Returns the last part of 'input' after the last occurrence of 'token'. For
-   * example to extract only the filename from a full path or URL.
-   * 
-   * @param input
-   * @param token
-   *          a delimiter which must be in regular expression format
-   * @return
-   */
-  public static String getLastToken(String input, String token)
-  {
-    if (input == null)
-    {
-      return null;
-    }
-    if (token == null)
-    {
-      return input;
-    }
-    String[] st = input.split(token);
-    return st[st.length - 1];
-  }
-
-  /**
    * Parses the input string into components separated by the delimiter. Unlike
    * String.split(), this method will ignore occurrences of the delimiter which
    * are nested within single quotes in name-value pair values, e.g. a='b,c'.
@@ -138,14 +125,15 @@ public class StringUtils
    * @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;
@@ -155,9 +143,8 @@ public class StringUtils
       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
       {
@@ -201,8 +188,8 @@ public class StringUtils
     }
     if (DEBUG)
     {
-      System.err.println("Empty Array from '" + delimiter
-              + "' separated List");
+      System.err.println(
+              "Empty Array from '" + delimiter + "' separated List");
     }
     return null;
   }
@@ -235,20 +222,20 @@ public class StringUtils
       }
       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;
   }
-  
+
   /**
    * Converts a list to a string with a delimiter before each term except the
    * first. Returns an empty string given a null or zero-length argument. This
@@ -299,4 +286,263 @@ public class StringUtils
     }
     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();
+    }
+    return s.substring(0, 1).toUpperCase() + s.substring(1).toLowerCase();
+  }
+
+  /**
+   * 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();
+    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("<", "&lt;");
+      text = text.replaceAll(">", "&gt;");
+    }
+    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;
+  }
 }