X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FIdentifyFile.java;h=cab3e60c2163931c436bf55b39091788e3bb6da7;hb=3c7d4cf5ef1331e2fad9c617c228486badf6f6a0;hp=ff0f0ccb0b0bda684f60e637b8b576317bb2db8c;hpb=4d7f98a6dd54d9863ba449ec79dcd95d25ed863d;p=jalview.git

diff --git a/src/jalview/io/IdentifyFile.java b/src/jalview/io/IdentifyFile.java
index ff0f0cc..cab3e60 100755
--- a/src/jalview/io/IdentifyFile.java
+++ b/src/jalview/io/IdentifyFile.java
@@ -20,6 +20,9 @@
  */
 package jalview.io;
 
+import java.util.Locale;
+
+import java.io.File;
 import java.io.IOException;
 
 /**
@@ -30,7 +33,41 @@ import java.io.IOException;
  */
 public class IdentifyFile
 {
-  public static final String GFF3File = "GFF v2 or v3";
+
+  public FileFormatI identify(Object file, DataSourceType protocol)
+          throws FileFormatException
+  {
+    // BH 2018
+    return (file instanceof File ? identify((File) file, protocol)
+            : identify((String) file, protocol));
+
+  }
+
+  public FileFormatI identify(File file, DataSourceType sourceType)
+          throws FileFormatException
+  {
+    // BH 2018
+    String emessage = "UNIDENTIFIED FILE PARSING ERROR";
+    FileParse parser = null;
+    try
+    {
+      parser = new FileParse(file, sourceType);
+      if (parser.isValid())
+      {
+        return identify(parser);
+      }
+    } catch (Exception e)
+    {
+      System.err.println("Error whilst identifying " + file);
+      e.printStackTrace(System.err);
+      emessage = e.getMessage();
+    }
+    if (parser != null)
+    {
+      throw new FileFormatException(parser.errormessage);
+    }
+    throw new FileFormatException(emessage);
+  }
 
   /**
    * Identify a datasource's file content.
@@ -39,39 +76,47 @@ public class IdentifyFile
    *       instead.
    *
    * @param file
-   *          DOCUMENT ME!
-   * @param protocol
-   *          DOCUMENT ME!
-   * @return ID String
+   * @param sourceType
+   * @return
+   * @throws FileFormatException
    */
-  public String Identify(String file, String protocol)
+  public FileFormatI identify(String file, DataSourceType sourceType)
+          throws FileFormatException
   {
     String emessage = "UNIDENTIFIED FILE PARSING ERROR";
     FileParse parser = null;
     try
     {
-      parser = new FileParse(file, protocol);
+      parser = new FileParse(file, sourceType);
       if (parser.isValid())
       {
-        return Identify(parser);
+        return identify(parser);
       }
     } catch (Exception e)
     {
-      System.err.println("Error whilst identifying");
+      System.err.println("Error whilst identifying " + file);
       e.printStackTrace(System.err);
       emessage = e.getMessage();
     }
     if (parser != null)
     {
-      return parser.errormessage;
+      throw new FileFormatException(parser.errormessage);
     }
-    return emessage;
+    throw new FileFormatException(emessage);
+  }
+
+  public FileFormatI identify(FileParse source) throws FileFormatException
+  {
+    return identify(source, true);
+    // preserves original behaviour prior to version 2.3
   }
 
-  public String Identify(FileParse source)
+  public FileFormatI identify(AlignmentFileReaderI file,
+          boolean closeSource) throws IOException
   {
-    return Identify(source, true); // preserves original behaviour prior to
-    // version 2.3
+    FileParse fp = new FileParse(file.getInFile(),
+            file.getDataSourceType());
+    return identify(fp, closeSource);
   }
 
   /**
@@ -80,25 +125,32 @@ public class IdentifyFile
    *
    * @param source
    * @param closeSource
-   * @return filetype string
+   * @return (best guess at) file format
+   * @throws FileFormatException
    */
-  public String Identify(FileParse source, boolean closeSource)
+  public FileFormatI identify(FileParse source, boolean closeSource)
+          throws FileFormatException
   {
-    String reply = "PFAM";
+    FileFormatI reply = FileFormat.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
+
     try
     {
       if (!closeSource)
       {
         source.mark();
       }
+      boolean aaIndexHeaderRead = false;
+
       while ((data = source.nextLine()) != null)
       {
-        length += data.trim().length();
+        bytesRead += data.length();
+        trimmedLength += data.trim().length();
         if (!lineswereskipped)
         {
           for (int i = 0; !isBinary && i < data.length(); i++)
@@ -117,29 +169,69 @@ public class IdentifyFile
           if (source.inFile != null)
           {
             String fileStr = source.inFile.getName();
-            // possibly a Jalview archive.
-            if (fileStr.lastIndexOf(".jar") > -1
-                    || fileStr.lastIndexOf(".zip") > -1)
+            if (fileStr.contains(".jar") || fileStr.contains(".zip")
+                    || fileStr.contains(".jvp"))
             {
-              reply = "Jalview";
+              // possibly a Jalview archive (but check further)
+              reply = FileFormat.Jalview;
             }
           }
           if (!lineswereskipped && data.startsWith("PK"))
           {
-            reply = "Jalview"; // archive.
+            reply = FileFormat.Jalview; // archive
             break;
           }
         }
-        data = data.toUpperCase();
+        data = data.toUpperCase(Locale.ROOT);
 
+        if (data.startsWith(ScoreMatrixFile.SCOREMATRIX))
+        {
+          reply = FileFormat.ScoreMatrix;
+          break;
+        }
+        if (data.startsWith("LOCUS"))
+        {
+          reply = FileFormat.GenBank;
+          break;
+        }
+        if (data.startsWith("ID "))
+        {
+          if (data.substring(2).trim().split(";").length == 7)
+          {
+            reply = FileFormat.Embl;
+            break;
+          }
+        }
+        if (data.startsWith("H ") && !aaIndexHeaderRead)
+        {
+          aaIndexHeaderRead = true;
+        }
+        if (data.startsWith("D ") && aaIndexHeaderRead)
+        {
+          reply = FileFormat.ScoreMatrix;
+          break;
+        }
         if (data.startsWith("##GFF-VERSION"))
         {
-          reply = GFF3File;
+          // GFF - possibly embedded in a Jalview features file!
+          reply = FileFormat.Features;
+          break;
+        }
+        if (looksLikeFeatureData(data))
+        {
+          reply = FileFormat.Features;
           break;
         }
         if (data.indexOf("# STOCKHOLM") > -1)
         {
-          reply = "STH";
+          reply = FileFormat.Stockholm;
+          break;
+        }
+        if (data.indexOf("_ENTRY.ID") > -1
+                || data.indexOf("_AUDIT_AUTHOR.NAME") > -1
+                || data.indexOf("_ATOM_SITE.") > -1)
+        {
+          reply = FileFormat.MMCif;
           break;
         }
         // if (data.indexOf(">") > -1)
@@ -151,14 +243,14 @@ public class IdentifyFile
           {
             // watch for PIR file attributes
             checkPIR = true;
-            reply = "PIR";
+            reply = FileFormat.PIR;
           }
           // could also be BLC file, read next line to confirm
           data = source.nextLine();
 
           if (data.indexOf(">") > -1)
           {
-            reply = "BLC";
+            reply = FileFormat.BLC;
           }
           else
           {
@@ -175,18 +267,19 @@ public class IdentifyFile
             {
               if (c1 == 0 && c1 == data2.indexOf("*"))
               {
-                reply = "BLC";
+                reply = FileFormat.BLC;
               }
               else
               {
-                reply = "FASTA"; // possibly a bad choice - may be recognised as
+                reply = FileFormat.Fasta; // possibly a bad choice - may be
+                                          // recognised as
                 // PIR
               }
               // otherwise can still possibly be a PIR file
             }
             else
             {
-              reply = "FASTA";
+              reply = FileFormat.Fasta;
               // TODO : AMSA File is indicated if there is annotation in the
               // FASTA file - but FASTA will automatically generate this at the
               // mo.
@@ -215,7 +308,6 @@ public class IdentifyFile
                 } catch (IOException ex)
                 {
                 }
-                ;
                 if (dta != null && dta.indexOf("*") > -1)
                 {
                   starterm = true;
@@ -224,49 +316,39 @@ public class IdentifyFile
             }
             if (starterm)
             {
-              reply = "PIR";
+              reply = FileFormat.PIR;
               break;
             }
             else
             {
-              reply = "FASTA"; // probably a bad choice!
+              reply = FileFormat.Fasta; // probably a bad choice!
             }
           }
           // read as a FASTA (probably)
           break;
         }
-        if ((data.indexOf("<") > -1)) // possible Markup Language data i.e HTML,
-                                      // RNAML, XML
+        if (data.indexOf("{\"") > -1)
         {
-          boolean identified = false;
-          do
+          reply = FileFormat.Json;
+          break;
+        }
+        int lessThan = data.indexOf("<");
+        if ((lessThan > -1)) // possible Markup Language data i.e HTML,
+                             // RNAML, XML
+        {
+          String upper = data.toUpperCase(Locale.ROOT);
+          if (upper.substring(lessThan).startsWith("<HTML"))
           {
-            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)
+            reply = FileFormat.Html;
+            break;
+          }
+          if (upper.substring(lessThan).startsWith("<RNAML"))
           {
+            reply = FileFormat.Rnaml;
             break;
           }
         }
 
-        if (data.indexOf("{\"") > -1)
-        {
-          reply = JSONFile.FILE_DESC;
-          break;
-        }
         if ((data.length() < 1) || (data.indexOf("#") == 0))
         {
           lineswereskipped = true;
@@ -275,55 +357,60 @@ public class IdentifyFile
 
         if (data.indexOf("PILEUP") > -1)
         {
-          reply = "PileUp";
+          reply = FileFormat.Pileup;
 
           break;
         }
 
-        if ((data.indexOf("//") == 0)
-                || ((data.indexOf("!!") > -1) && (data.indexOf("!!") < data
-                        .indexOf("_MULTIPLE_ALIGNMENT "))))
+        if ((data.indexOf("//") == 0) || ((data.indexOf("!!") > -1) && (data
+                .indexOf("!!") < data.indexOf("_MULTIPLE_ALIGNMENT "))))
         {
-          reply = "MSF";
+          reply = FileFormat.MSF;
 
           break;
         }
         else if (data.indexOf("CLUSTAL") > -1)
         {
-          reply = "CLUSTAL";
+          reply = FileFormat.Clustal;
 
           break;
         }
 
         else if (data.indexOf("HEADER") == 0 || data.indexOf("ATOM") == 0)
         {
-          reply = "PDB";
+          reply = FileFormat.PDB;
           break;
         }
         else if (data.matches("\\s*\\d+\\s+\\d+\\s*"))
         {
-          reply = PhylipFile.FILE_DESC;
+          reply = FileFormat.Phylip;
           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 if (!lineswereskipped && looksLikeJnetData(data))
         {
-          // file looks like a concise JNet file
-          reply = "JnetFile";
-          break;
+            reply = FileFormat.Jnet;
+            break;
+
         }
+        else // phylogenetic file
+        {
+          // may also have a <xml header before hand, probably need to check it 
+          if (data.toLowerCase(Locale.ROOT).startsWith("<phyloxml"))
+          {
+            reply = FileFormat.PhyloXML;
+            break;
+          }
+          // commented out, nexus parser is troublesome
 
+          // else if (((identifier.startsWith("nexus"))
+          // || (identifier.startsWith("#nexus"))
+          // || (identifier.startsWith("# nexus"))
+          // || (identifier.startsWith("begin"))))
+          // {
+          // reply = FileFormat.Nexus;
+          // break;
+          // }
+        }
         lineswereskipped = true; // this means there was some junk before any
         // key file signature
       }
@@ -333,29 +420,91 @@ 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;
+      throw new FileFormatException(source.errormessage);
     }
-    if (length == 0)
+    if (trimmedLength == 0)
     {
-      System.err
-              .println("File Identification failed! - Empty file was read.");
-      return "EMPTY DATA FILE";
+      System.err.println(
+              "File Identification failed! - Empty file was read.");
+      throw new FileFormatException("EMPTY DATA FILE");
     }
+    System.out.println("File format identified as " + reply.toString());
     return reply;
   }
 
-  public static void main(String[] args)
+  /**
+   * 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;
+  }
+
+  /**
+   * 
+   * @param args
+   * @j2sIgnore
+   */
+  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);
+      FileFormatI type = null;
+      try
+      {
+        type = ider.identify(args[i], DataSourceType.FILE);
+      } catch (FileFormatException e)
+      {
+        System.err.println(
+                String.format("Error '%s' identifying file type for %s",
+                        args[i], e.getMessage()));
+      }
       System.out.println("Type of " + args[i] + " is " + type);
     }
     if (args == null || args.length == 0)
@@ -363,4 +512,5 @@ public class IdentifyFile
       System.err.println("Usage: <Filename> [<Filename> ...]");
     }
   }
+
 }