/*
- * 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.
*
* 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)
{
}
if (parser != null)
{
- return parser.errormessage;
+ throw new FileFormatException(parser.errormessage);
}
- return emessage;
+ throw new FileFormatException(emessage);
}
- public String Identify(FileParse source)
+ public FileFormatI identify(FileParse source) throws FileFormatException
{
- return Identify(source, true); // preserves original behaviour prior to
- // version 2.3
+ 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);
}
/**
*
* @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.length();
+ bytesRead += data.length();
+ trimmedLength += data.trim().length();
if (!lineswereskipped)
{
for (int i = 0; !isBinary && i < data.length(); i++)
if (fileStr.lastIndexOf(".jar") > -1
|| fileStr.lastIndexOf(".zip") > -1)
{
- reply = "Jalview";
+ reply = FileFormat.Jalview;
}
}
if (!lineswereskipped && data.startsWith("PK"))
{
- reply = "Jalview"; // archive.
+ reply = FileFormat.Jalview; // archive.
break;
}
}
data = data.toUpperCase();
- if ((data.indexOf("# STOCKHOLM") > -1))
+ if (data.startsWith(ScoreMatrixFile.SCOREMATRIX))
{
- reply = "STH";
-
+ reply = FileFormat.ScoreMatrix;
break;
}
- // if (data.matches("<(\"[^\"]*\"|'[^']*'|[^'\">])*>"))
- if (data.matches("<(?i)html(\"[^\"]*\"|'[^']*'|[^'\">])*>"))
+ if (data.startsWith("HMMER3"))
{
- reply = "HTML";
+ reply = FileFormat.HMMER3;
break;
}
-
- if (data.matches("<(?i)rnaml (\"[^\"]*\"|'[^']*'|[^'\">])*>"))
+ if (data.startsWith("H ") && !aaIndexHeaderRead)
{
- reply = "RNAML";
-
+ aaIndexHeaderRead = true;
+ }
+ if (data.startsWith("D ") && aaIndexHeaderRead)
+ {
+ reply = FileFormat.ScoreMatrix;
break;
}
-
- if ((data.length() < 1) || (data.indexOf("#") == 0))
+ if (data.startsWith("##GFF-VERSION"))
{
- lineswereskipped = true;
- continue;
+ // GFF - possibly embedded in a Jalview features file!
+ reply = FileFormat.Features;
+ break;
}
-
- if (data.indexOf("PILEUP") > -1)
+ if (looksLikeFeatureData(data))
{
- reply = "PileUp";
-
+ reply = FileFormat.Features;
break;
}
-
- if ((data.indexOf("//") == 0)
- || ((data.indexOf("!!") > -1) && (data.indexOf("!!") < data
- .indexOf("_MULTIPLE_ALIGNMENT "))))
+ if (data.indexOf("# STOCKHOLM") > -1)
{
- reply = "MSF";
-
+ reply = FileFormat.Stockholm;
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 = FileFormat.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;
{
// 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
{
{
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.
} catch (IOException ex)
{
}
- ;
if (dta != null && dta.indexOf("*") > -1)
{
starterm = true;
}
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;
}
- else if (data.indexOf("HEADER") == 0 || data.indexOf("ATOM") == 0)
+ if (data.indexOf("{\"") > -1)
{
- reply = "PDB";
+ reply = FileFormat.Json;
break;
}
- else if (data.matches("\\s*\\d+\\s+\\d+\\s*"))
+ int lessThan = data.indexOf("<");
+ if ((lessThan > -1)) // possible Markup Language data i.e HTML,
+ // RNAML, XML
+ {
+ String upper = data.toUpperCase();
+ 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;
+ }
+
+ if ((data.indexOf("//") == 0) || ((data.indexOf("!!") > -1) && (data
+ .indexOf("!!") < data.indexOf("_MULTIPLE_ALIGNMENT "))))
{
- reply = PhylipFile.FILE_DESC;
+ reply = FileFormat.MSF;
+
break;
}
+ else if (data.indexOf("CLUSTAL") > -1)
+ {
+ reply = FileFormat.Clustal;
+ 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 (data.indexOf("HEADER") == 0 || data.indexOf("ATOM") == 0)
{
- // file looks like a concise JNet file
- reply = "JnetFile";
+ 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;
+ }
+ }
lineswereskipped = true; // this means there was some junk before any
// key file signature
}
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;
+ }
+
+ 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)