X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FIdentifyFile.java;h=cab3e60c2163931c436bf55b39091788e3bb6da7;hb=3c7d4cf5ef1331e2fad9c617c228486badf6f6a0;hp=7f3db40720fda78e4b677235514f8715ed01c61e;hpb=e8b5b1e551794bf54de00c610363ffd140c62e55;p=jalview.git

diff --git a/src/jalview/io/IdentifyFile.java b/src/jalview/io/IdentifyFile.java
index 7f3db40..cab3e60 100755
--- a/src/jalview/io/IdentifyFile.java
+++ b/src/jalview/io/IdentifyFile.java
@@ -1,25 +1,29 @@
 /*
- * Jalview - A Sequence Alignment Editor and Viewer
- * Copyright (C) 2007 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
  *
- * This program is free software; you can redistribute it and/or
+ * This file is part of Jalview.
+ *
+ * Jalview is free software: you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version 2
+ * as published by the Free Software Foundation, either version 3
  * of the License, or (at your option) any later version.
  *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
+ * Jalview is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ * PURPOSE.  See the GNU General Public License for more details.
  *
  * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA
+ * along with Jalview.  If not, see <http://www.gnu.org/licenses/>.
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
  */
 package jalview.io;
 
-import java.io.*;
-import java.net.*;
+import java.util.Locale;
+
+import java.io.File;
+import java.io.IOException;
 
 /**
  * DOCUMENT ME!
@@ -29,154 +33,484 @@ import java.net.*;
  */
 public class IdentifyFile
 {
+
+  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. 
-   * @note Do not use this method
-   * for stream sources - create a FileParse object instead. 
+   * Identify a datasource's file content.
+   *
+   * @note Do not use this method for stream sources - create a FileParse object
+   *       instead.
    *
-   * @param file DOCUMENT ME!
-   * @param protocol DOCUMENT ME! 
-   * @return ID String
+   * @param file
+   * @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);
-      if (parser.isValid()) {
-        return Identify(parser);
+    try
+    {
+      parser = new FileParse(file, sourceType);
+      if (parser.isValid())
+      {
+        return identify(parser);
       }
-    } catch (Exception e) {
-      System.err.println("Error whilst identifying");
+    } 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);
     }
-    if (parser!=null)
-      return parser.errormessage;
-    return "UNIDENTIFIED FILE PARSING ERROR";
+    throw new FileFormatException(emessage);
   }
-  public String Identify(FileParse source) {
-    return Identify(source, true); // preserves original behaviour prior to version 2.3
+
+  public FileFormatI identify(FileParse source) throws FileFormatException
+  {
+    return identify(source, true);
+    // preserves original behaviour prior to version 2.3
+  }
+
+  public FileFormatI identify(AlignmentFileReaderI file,
+          boolean closeSource) throws IOException
+  {
+    FileParse fp = new FileParse(file.getInFile(),
+            file.getDataSourceType());
+    return identify(fp, closeSource);
   }
+
   /**
-   * Identify contents of source, closing it or resetting source to start afterwards.
+   * Identify contents of source, closing it or resetting source to start
+   * afterwards.
+   *
    * @param source
    * @param closeSource
-   * @return filetype string
+   * @return (best guess at) file format
+   * @throws FileFormatException
    */
-  public String Identify(FileParse source, boolean closeSource) {
-    String reply = "PFAM";
+  public FileFormatI identify(FileParse source, boolean closeSource)
+          throws FileFormatException
+  {
+    FileFormatI reply = FileFormat.Pfam;
     String data;
-    try {
-      while ( (data = source.nextLine()) != null)
+    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)
       {
-        data = data.toUpperCase();
+        source.mark();
+      }
+      boolean aaIndexHeaderRead = false;
 
-        if ( (data.indexOf("# STOCKHOLM") > -1))
+      while ((data = source.nextLine()) != null)
+      {
+        bytesRead += data.length();
+        trimmedLength += data.trim().length();
+        if (!lineswereskipped)
         {
-          reply = "STH";
+          for (int i = 0; !isBinary && i < data.length(); i++)
+          {
+            char c = data.charAt(i);
+            isBinary = (c < 32 && c != '\t' && c != '\n' && c != '\r'
+                    && c != 5 && c != 27); // nominal binary character filter
+            // excluding CR, LF, tab,DEL and ^E
+            // for certain blast ids
+          }
+        }
+        if (isBinary)
+        {
+          // jar files are special - since they contain all sorts of random
+          // characters.
+          if (source.inFile != null)
+          {
+            String fileStr = source.inFile.getName();
+            if (fileStr.contains(".jar") || fileStr.contains(".zip")
+                    || fileStr.contains(".jvp"))
+            {
+              // possibly a Jalview archive (but check further)
+              reply = FileFormat.Jalview;
+            }
+          }
+          if (!lineswereskipped && data.startsWith("PK"))
+          {
+            reply = FileFormat.Jalview; // archive
+            break;
+          }
+        }
+        data = data.toUpperCase(Locale.ROOT);
 
+        if (data.startsWith(ScoreMatrixFile.SCOREMATRIX))
+        {
+          reply = FileFormat.ScoreMatrix;
           break;
         }
-
-        if ( (data.indexOf("#") == 0) || (data.length() < 1))
+        if (data.startsWith("LOCUS"))
         {
-          continue;
+          reply = FileFormat.GenBank;
+          break;
         }
-
-        if (data.indexOf("PILEUP") > -1)
+        if (data.startsWith("ID "))
         {
-          reply = "PileUp";
-
+          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.indexOf("//") == 0) ||
-            ( (data.indexOf("!!") > -1) &&
-             (data.indexOf("!!") < data.indexOf(
-                 "_MULTIPLE_ALIGNMENT "))))
+        if (data.startsWith("##GFF-VERSION"))
         {
-          reply = "MSF";
-
+          // GFF - possibly embedded in a Jalview features file!
+          reply = FileFormat.Features;
           break;
         }
-        else if (data.indexOf("CLUSTAL") > -1)
+        if (looksLikeFeatureData(data))
         {
-          reply = "CLUSTAL";
-
+          reply = FileFormat.Features;
           break;
         }
-        else if ( (data.indexOf(">P1;") > -1) ||
-                 (data.indexOf(">DL;") > -1))
+        if (data.indexOf("# STOCKHOLM") > -1)
         {
-          reply = "PIR";
-
+          reply = FileFormat.Stockholm;
           break;
         }
-        else if (data.indexOf(">") > -1)
+        if (data.indexOf("_ENTRY.ID") > -1
+                || data.indexOf("_AUDIT_AUTHOR.NAME") > -1
+                || data.indexOf("_ATOM_SITE.") > -1)
         {
-          // could be BLC file, read next line to confirm
+          reply = FileFormat.MMCif;
+          break;
+        }
+        // if (data.indexOf(">") > -1)
+        if (data.startsWith(">"))
+        {
+          // FASTA, PIR file or BLC file
+          boolean checkPIR = false, starterm = false;
+          if ((data.indexOf(">P1;") > -1) || (data.indexOf(">DL;") > -1))
+          {
+            // watch for PIR file attributes
+            checkPIR = true;
+            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
           {
-            //Is this a single line BLC file?
-            source.nextLine();
+            // Is this a single line BLC file?
+            String data1 = source.nextLine();
             String data2 = source.nextLine();
-            if (data2 != null
-                && data.indexOf("*") > -1
-                && data.indexOf("*") == data2.indexOf("*"))
+            int c1;
+            if (checkPIR)
+            {
+              starterm = (data1 != null && data1.indexOf("*") > -1)
+                      || (data2 != null && data2.indexOf("*") > -1);
+            }
+            if (data2 != null && (c1 = data.indexOf("*")) > -1)
+            {
+              if (c1 == 0 && c1 == data2.indexOf("*"))
+              {
+                reply = FileFormat.BLC;
+              }
+              else
+              {
+                reply = FileFormat.Fasta; // possibly a bad choice - may be
+                                          // recognised as
+                // PIR
+              }
+              // otherwise can still possibly be a PIR file
+            }
+            else
+            {
+              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.
+              if (!checkPIR)
+              {
+                break;
+              }
+            }
+          }
+          // final check for PIR content. require
+          // >P1;title\n<blah>\nterminated sequence to occur at least once.
+
+          // TODO the PIR/fasta ambiguity may be the use case that is needed to
+          // have
+          // a 'Parse as type XXX' parameter for the applet/application.
+          if (checkPIR)
+          {
+            String dta = null;
+            if (!starterm)
             {
-              reply = "BLC";
+              do
+              {
+                try
+                {
+                  dta = source.nextLine();
+                } catch (IOException ex)
+                {
+                }
+                if (dta != null && dta.indexOf("*") > -1)
+                {
+                  starterm = true;
+                }
+              } while (dta != null && !starterm);
+            }
+            if (starterm)
+            {
+              reply = FileFormat.PIR;
+              break;
             }
             else
             {
-              reply = "FASTA";
+              reply = FileFormat.Fasta; // probably a bad choice!
             }
           }
+          // read as a FASTA (probably)
+          break;
+        }
+        if (data.indexOf("{\"") > -1)
+        {
+          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"))
+          {
+            reply = FileFormat.Html;
+            break;
+          }
+          if (upper.substring(lessThan).startsWith("<RNAML"))
+          {
+            reply = FileFormat.Rnaml;
+            break;
+          }
+        }
+
+        if ((data.length() < 1) || (data.indexOf("#") == 0))
+        {
+          lineswereskipped = true;
+          continue;
+        }
+
+        if (data.indexOf("PILEUP") > -1)
+        {
+          reply = FileFormat.Pileup;
 
           break;
         }
-        else if (data.indexOf("HEADER") == 0 ||
-                 data.indexOf("ATOM") == 0)
+
+        if ((data.indexOf("//") == 0) || ((data.indexOf("!!") > -1) && (data
+                .indexOf("!!") < data.indexOf("_MULTIPLE_ALIGNMENT "))))
         {
-          reply = "PDB";
+          reply = FileFormat.MSF;
+
           break;
         }
-        else if (data.indexOf(":") < data.indexOf(",")) //  && data.indexOf(",")<data.indexOf(",", data.indexOf(",")))
+        else if (data.indexOf("CLUSTAL") > -1)
         {
-          // file looks like a concise JNet file
-          reply = "JnetFile";
+          reply = FileFormat.Clustal;
+
           break;
         }
-        else  if (source.inFile!=null) 
+
+        else if (data.indexOf("HEADER") == 0 || data.indexOf("ATOM") == 0)
         {
-            String fileStr=source.inFile.getName();
-            // possibly a Jalview archive. 
-            if (fileStr.lastIndexOf(".jar")>-1 || fileStr.lastIndexOf(".zip")>-1) 
-            {
-              reply = "Jalview";
-            }
-        } else if (data.startsWith("PK")) {
-          reply="Jalview"; // archive.
+          reply = FileFormat.PDB;
+          break;
+        }
+        else if (data.matches("\\s*\\d+\\s+\\d+\\s*"))
+        {
+          reply = FileFormat.Phylip;
           break;
         }
+        else if (!lineswereskipped && looksLikeJnetData(data))
+        {
+            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
       }
-      if (closeSource) {
+      if (closeSource)
+      {
         source.close();
-      } else {
-        source.reset(); // so the file can be parsed from the beginning again.
       }
-    }
-    catch (Exception ex)
+      else
+      {
+        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 (trimmedLength == 0)
+    {
+      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;
   }
+
+  /**
+   * 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();
+      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)
+    {
+      System.err.println("Usage: <Filename> [<Filename> ...]");
+    }
+  }
+
 }