X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FIdentifyFile.java;h=71f423769bb0d3031d7f126162c953f598f024f8;hb=94b82b9357592d44785de48290b2c9f2af46b2ca;hp=4fb2516873b74e99d53dfa747ac56565da6591aa;hpb=227ee1c1c69c228b67d408a0a83ae51bf5811272;p=jalview.git

diff --git a/src/jalview/io/IdentifyFile.java b/src/jalview/io/IdentifyFile.java
index 4fb2516..71f4237 100755
--- a/src/jalview/io/IdentifyFile.java
+++ b/src/jalview/io/IdentifyFile.java
@@ -1,6 +1,6 @@
 /*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8.2)
- * Copyright (C) 2014 The Jalview Authors
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
  *
  * This file is part of Jalview.
  *
@@ -30,6 +30,8 @@ import java.io.IOException;
  */
 public class IdentifyFile
 {
+  public static final String FeaturesFile = "GFF or Jalview features";
+
   /**
    * Identify a datasource's file content.
    *
@@ -42,7 +44,7 @@ public class IdentifyFile
    *          DOCUMENT ME!
    * @return ID String
    */
-  public String Identify(String file, String protocol)
+  public String identify(String file, String protocol)
   {
     String emessage = "UNIDENTIFIED FILE PARSING ERROR";
     FileParse parser = null;
@@ -51,7 +53,7 @@ public class IdentifyFile
       parser = new FileParse(file, protocol);
       if (parser.isValid())
       {
-        return Identify(parser);
+        return identify(parser);
       }
     } catch (Exception e)
     {
@@ -66,9 +68,9 @@ public class IdentifyFile
     return emessage;
   }
 
-  public String Identify(FileParse source)
+  public String identify(FileParse source)
   {
-    return Identify(source, true); // preserves original behaviour prior to
+    return identify(source, true); // preserves original behaviour prior to
     // version 2.3
   }
 
@@ -80,11 +82,12 @@ public class IdentifyFile
    * @param closeSource
    * @return filetype string
    */
-  public String Identify(FileParse source, boolean closeSource)
+  public String identify(FileParse source, boolean closeSource)
   {
     String reply = "PFAM";
     String data;
-    int length = 0;
+    int bytesRead = 0;
+    int trimmedLength = 0;
     boolean lineswereskipped = false;
     boolean isBinary = false; // true if length is non-zero and non-printable
     // characters are encountered
@@ -96,7 +99,8 @@ public class IdentifyFile
       }
       while ((data = source.nextLine()) != null)
       {
-        length += data.length();
+        bytesRead += data.length();
+        trimmedLength += data.trim().length();
         if (!lineswereskipped)
         {
           for (int i = 0; !isBinary && i < data.length(); i++)
@@ -130,55 +134,31 @@ public class IdentifyFile
         }
         data = data.toUpperCase();
 
-        if ((data.indexOf("# STOCKHOLM") > -1))
+        if (data.startsWith("##GFF-VERSION"))
         {
-          reply = "STH";
-
+          // GFF - possibly embedded in a Jalview features file!
+          reply = FeaturesFile;
           break;
         }
-        // if (data.matches("<(\"[^\"]*\"|'[^']*'|[^'\">])*>"))
-        if (data.matches("<(?i)html(\"[^\"]*\"|'[^']*'|[^'\">])*>"))
-        {
-          reply = "HTML";
-          break;
-        }
-
-        if (data.matches("<(?i)rnaml (\"[^\"]*\"|'[^']*'|[^'\">])*>"))
-        {
-          reply = "RNAML";
-
-          break;
-        }
-
-        if ((data.length() < 1) || (data.indexOf("#") == 0))
-        {
-          lineswereskipped = true;
-          continue;
-        }
-
-        if (data.indexOf("PILEUP") > -1)
+        if (looksLikeFeatureData(data))
         {
-          reply = "PileUp";
-
+          reply = FeaturesFile;
           break;
         }
-
-        if ((data.indexOf("//") == 0)
-                || ((data.indexOf("!!") > -1) && (data.indexOf("!!") < data
-                        .indexOf("_MULTIPLE_ALIGNMENT "))))
+        if (data.indexOf("# STOCKHOLM") > -1)
         {
-          reply = "MSF";
-
+          reply = "STH";
           break;
         }
-        else if (data.indexOf("CLUSTAL") > -1)
+        if (data.indexOf("_ENTRY.ID") > -1
+                || data.indexOf("_AUDIT_AUTHOR.NAME") > -1
+                || data.indexOf("_ATOM_SITE.") > -1)
         {
-          reply = "CLUSTAL";
-
+          reply = "mmCIF";
           break;
         }
-
-        else if (data.indexOf(">") > -1)
+        // if (data.indexOf(">") > -1)
+        if (data.startsWith(">"))
         {
           // FASTA, PIR file or BLC file
           boolean checkPIR = false, starterm = false;
@@ -270,6 +250,71 @@ public class IdentifyFile
           // read as a FASTA (probably)
           break;
         }
+        if ((data.indexOf("<") > -1)) // possible Markup Language data i.e HTML,
+                                      // RNAML, XML
+        {
+          // FIXME this is nuts - it consumes the rest of the file if no match
+          boolean identified = false;
+          do
+          {
+            if (data.matches("<(?i)html(\"[^\"]*\"|'[^']*'|[^'\">])*>"))
+            {
+              reply = HtmlFile.FILE_DESC;
+              identified = true;
+              break;
+            }
+
+            if (data.matches("<(?i)rnaml (\"[^\"]*\"|'[^']*'|[^'\">])*>"))
+            {
+              reply = "RNAML";
+              identified = true;
+              break;
+            }
+          } while ((data = source.nextLine()) != null);
+
+          if (identified)
+          {
+            break;
+          }
+          if (data == null)
+          {
+            break;
+          }
+        }
+
+        if (data.indexOf("{\"") > -1)
+        {
+          reply = JSONFile.FILE_DESC;
+          break;
+        }
+        if ((data.length() < 1) || (data.indexOf("#") == 0))
+        {
+          lineswereskipped = true;
+          continue;
+        }
+
+        if (data.indexOf("PILEUP") > -1)
+        {
+          reply = "PileUp";
+
+          break;
+        }
+
+        if ((data.indexOf("//") == 0)
+                || ((data.indexOf("!!") > -1) && (data.indexOf("!!") < data
+                        .indexOf("_MULTIPLE_ALIGNMENT "))))
+        {
+          reply = "MSF";
+
+          break;
+        }
+        else if (data.indexOf("CLUSTAL") > -1)
+        {
+          reply = "CLUSTAL";
+
+          break;
+        }
+
         else if (data.indexOf("HEADER") == 0 || data.indexOf("ATOM") == 0)
         {
           reply = "PDB";
@@ -280,24 +325,13 @@ public class IdentifyFile
           reply = PhylipFile.FILE_DESC;
           break;
         }
-
-
-        /*
-         * // TODO comment out SimpleBLAST identification for Jalview 2.4.1 else
-         * if (!lineswereskipped && data.indexOf("BLAST")<4) { reply =
-         * "SimpleBLAST"; break;
-         *
-         * } // end comments for Jalview 2.4.1
-         */
-        else if (!lineswereskipped && data.charAt(0) != '*'
-                && data.charAt(0) != ' '
-                && data.indexOf(":") < data.indexOf(",")) // &&
-          // data.indexOf(",")<data.indexOf(",",
-          // data.indexOf(",")))
+        else
         {
-          // file looks like a concise JNet file
-          reply = "JnetFile";
-          break;
+          if (!lineswereskipped && looksLikeJnetData(data))
+          {
+            reply = "JnetFile";
+            break;
+          }
         }
 
         lineswereskipped = true; // this means there was some junk before any
@@ -309,29 +343,73 @@ public class IdentifyFile
       }
       else
       {
-        source.reset(); // so the file can be parsed from the beginning again.
+        source.reset(bytesRead); // so the file can be parsed from the mark
       }
     } catch (Exception ex)
     {
       System.err.println("File Identification failed!\n" + ex);
       return source.errormessage;
     }
-    if (length == 0)
+    if (trimmedLength == 0)
     {
       System.err
-      .println("File Identification failed! - Empty file was read.");
+              .println("File Identification failed! - Empty file was read.");
       return "EMPTY DATA FILE";
     }
     return reply;
   }
 
+  /**
+   * Returns true if the data appears to be Jnet concise annotation format
+   * 
+   * @param data
+   * @return
+   */
+  protected boolean looksLikeJnetData(String data)
+  {
+    char firstChar = data.charAt(0);
+    int colonPos = data.indexOf(":");
+    int commaPos = data.indexOf(",");
+    boolean isJnet = firstChar != '*' && firstChar != ' ' && colonPos > -1
+            && commaPos > -1 && colonPos < commaPos;
+    // && data.indexOf(",")<data.indexOf(",", data.indexOf(","))) / ??
+    return isJnet;
+  }
+
+  /**
+   * Returns true if the data has at least 6 tab-delimited fields _and_ 
+   * fields 4 and 5 are integer (start/end) 
+   * @param data
+   * @return
+   */
+  protected boolean looksLikeFeatureData(String data)
+  {
+    if (data == null)
+    {
+      return false;
+    }
+    String[] columns = data.split("\t");
+    if (columns.length < 6) {
+      return false;
+    }
+    for (int col = 3; col < 5; col++)
+    {
+      try {
+        Integer.parseInt(columns[col]);
+      } catch (NumberFormatException e) {
+        return false;
+      }
+    }
+    return true;
+  }
+
   public static void main(String[] args)
   {
 
     for (int i = 0; args != null && i < args.length; i++)
     {
       IdentifyFile ider = new IdentifyFile();
-      String type = ider.Identify(args[i], AppletFormatAdapter.FILE);
+      String type = ider.identify(args[i], AppletFormatAdapter.FILE);
       System.out.println("Type of " + args[i] + " is " + type);
     }
     if (args == null || args.length == 0)