2 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3 * Copyright (C) $$Year-Rel$$ The Jalview Authors
5 * This file is part of Jalview.
7 * Jalview is free software: you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License
9 * as published by the Free Software Foundation, either version 3
10 * of the License, or (at your option) any later version.
12 * Jalview is distributed in the hope that it will be useful, but
13 * WITHOUT ANY WARRANTY; without even the implied warranty
14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
15 * PURPOSE. See the GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
19 * The Jalview Authors are detailed in the 'AUTHORS' file.
23 import java.util.Locale;
26 import java.io.IOException;
34 public class IdentifyFile
37 public FileFormatI identify(Object file, DataSourceType protocol) throws FileFormatException
40 return (file instanceof File ? identify((File) file, protocol) : identify((String) file, protocol));
44 public FileFormatI identify(File file, DataSourceType sourceType)
45 throws FileFormatException
48 String emessage = "UNIDENTIFIED FILE PARSING ERROR";
49 FileParse parser = null;
52 parser = new FileParse(file, sourceType);
55 return identify(parser);
59 System.err.println("Error whilst identifying " + file);
60 e.printStackTrace(System.err);
61 emessage = e.getMessage();
65 throw new FileFormatException(parser.errormessage);
67 throw new FileFormatException(emessage);
71 * Identify a datasource's file content.
73 * @note Do not use this method for stream sources - create a FileParse object
79 * @throws FileFormatException
81 public FileFormatI identify(String file, DataSourceType sourceType)
82 throws FileFormatException
84 String emessage = "UNIDENTIFIED FILE PARSING ERROR";
85 FileParse parser = null;
88 parser = new FileParse(file, sourceType);
91 return identify(parser);
95 System.err.println("Error whilst identifying " + file);
96 e.printStackTrace(System.err);
97 emessage = e.getMessage();
101 throw new FileFormatException(parser.errormessage);
103 throw new FileFormatException(emessage);
106 public FileFormatI identify(FileParse source) throws FileFormatException
108 return identify(source, true);
109 // preserves original behaviour prior to version 2.3
112 public FileFormatI identify(AlignmentFileReaderI file,
113 boolean closeSource) throws IOException
115 FileParse fp = new FileParse(file.getInFile(),
116 file.getDataSourceType());
117 return identify(fp, closeSource);
121 * Identify contents of source, closing it or resetting source to start
126 * @return (best guess at) file format
127 * @throws FileFormatException
129 public FileFormatI identify(FileParse source, boolean closeSource)
130 throws FileFormatException
132 FileFormatI reply = FileFormat.Pfam;
135 int trimmedLength = 0;
136 boolean lineswereskipped = false;
137 boolean isBinary = false; // true if length is non-zero and non-printable
138 // characters are encountered
146 boolean aaIndexHeaderRead = false;
148 while ((data = source.nextLine()) != null)
150 bytesRead += data.length();
151 trimmedLength += data.trim().length();
152 if (!lineswereskipped)
154 for (int i = 0; !isBinary && i < data.length(); i++)
156 char c = data.charAt(i);
157 isBinary = (c < 32 && c != '\t' && c != '\n' && c != '\r'
158 && c != 5 && c != 27); // nominal binary character filter
159 // excluding CR, LF, tab,DEL and ^E
160 // for certain blast ids
165 // jar files are special - since they contain all sorts of random
167 if (source.inFile != null)
169 String fileStr = source.inFile.getName();
170 if (fileStr.contains(".jar")
171 || fileStr.contains(".zip") || fileStr.contains(".jvp"))
173 // possibly a Jalview archive (but check further)
174 reply = FileFormat.Jalview;
177 if (!lineswereskipped && data.startsWith("PK"))
179 reply = FileFormat.Jalview; // archive
183 data = data.toUpperCase(Locale.ROOT);
185 if (data.startsWith(ScoreMatrixFile.SCOREMATRIX))
187 reply = FileFormat.ScoreMatrix;
190 if (data.startsWith("LOCUS"))
192 reply = FileFormat.GenBank;
195 if (data.startsWith("ID "))
197 if (data.substring(2).trim().split(";").length == 7)
199 reply = FileFormat.Embl;
203 if (data.startsWith("H ") && !aaIndexHeaderRead)
205 aaIndexHeaderRead = true;
207 if (data.startsWith("D ") && aaIndexHeaderRead)
209 reply = FileFormat.ScoreMatrix;
212 if (data.startsWith("##GFF-VERSION"))
214 // GFF - possibly embedded in a Jalview features file!
215 reply = FileFormat.Features;
218 if (looksLikeFeatureData(data))
220 reply = FileFormat.Features;
223 if (data.indexOf("# STOCKHOLM") > -1)
225 reply = FileFormat.Stockholm;
228 if (data.indexOf("_ENTRY.ID") > -1
229 || data.indexOf("_AUDIT_AUTHOR.NAME") > -1
230 || data.indexOf("_ATOM_SITE.") > -1)
232 reply = FileFormat.MMCif;
235 // if (data.indexOf(">") > -1)
236 if (data.startsWith(">"))
238 // FASTA, PIR file or BLC file
239 boolean checkPIR = false, starterm = false;
240 if ((data.indexOf(">P1;") > -1) || (data.indexOf(">DL;") > -1))
242 // watch for PIR file attributes
244 reply = FileFormat.PIR;
246 // could also be BLC file, read next line to confirm
247 data = source.nextLine();
249 if (data.indexOf(">") > -1)
251 reply = FileFormat.BLC;
255 // Is this a single line BLC file?
256 String data1 = source.nextLine();
257 String data2 = source.nextLine();
261 starterm = (data1 != null && data1.indexOf("*") > -1)
262 || (data2 != null && data2.indexOf("*") > -1);
264 if (data2 != null && (c1 = data.indexOf("*")) > -1)
266 if (c1 == 0 && c1 == data2.indexOf("*"))
268 reply = FileFormat.BLC;
272 reply = FileFormat.Fasta; // possibly a bad choice - may be
276 // otherwise can still possibly be a PIR file
280 reply = FileFormat.Fasta;
281 // TODO : AMSA File is indicated if there is annotation in the
282 // FASTA file - but FASTA will automatically generate this at the
290 // final check for PIR content. require
291 // >P1;title\n<blah>\nterminated sequence to occur at least once.
293 // TODO the PIR/fasta ambiguity may be the use case that is needed to
295 // a 'Parse as type XXX' parameter for the applet/application.
305 dta = source.nextLine();
306 } catch (IOException ex)
309 if (dta != null && dta.indexOf("*") > -1)
313 } while (dta != null && !starterm);
317 reply = FileFormat.PIR;
322 reply = FileFormat.Fasta; // probably a bad choice!
325 // read as a FASTA (probably)
328 if (data.indexOf("{\"") > -1)
330 reply = FileFormat.Json;
333 int lessThan = data.indexOf("<");
334 if ((lessThan > -1)) // possible Markup Language data i.e HTML,
337 String upper = data.toUpperCase(Locale.ROOT);
338 if (upper.substring(lessThan).startsWith("<HTML"))
340 reply = FileFormat.Html;
343 if (upper.substring(lessThan).startsWith("<RNAML"))
345 reply = FileFormat.Rnaml;
350 if ((data.length() < 1) || (data.indexOf("#") == 0))
352 lineswereskipped = true;
356 if (data.indexOf("PILEUP") > -1)
358 reply = FileFormat.Pileup;
363 if ((data.indexOf("//") == 0) || ((data.indexOf("!!") > -1) && (data
364 .indexOf("!!") < data.indexOf("_MULTIPLE_ALIGNMENT "))))
366 reply = FileFormat.MSF;
370 else if (data.indexOf("CLUSTAL") > -1)
372 reply = FileFormat.Clustal;
377 else if (data.indexOf("HEADER") == 0 || data.indexOf("ATOM") == 0)
379 reply = FileFormat.PDB;
382 else if (data.matches("\\s*\\d+\\s+\\d+\\s*"))
384 reply = FileFormat.Phylip;
389 if (!lineswereskipped && looksLikeJnetData(data))
391 reply = FileFormat.Jnet;
396 lineswereskipped = true; // this means there was some junk before any
397 // key file signature
405 source.reset(bytesRead); // so the file can be parsed from the mark
407 } catch (Exception ex)
409 System.err.println("File Identification failed!\n" + ex);
410 throw new FileFormatException(source.errormessage);
412 if (trimmedLength == 0)
415 "File Identification failed! - Empty file was read.");
416 throw new FileFormatException("EMPTY DATA FILE");
418 System.out.println("File format identified as " + reply.toString());
423 * Returns true if the data appears to be Jnet concise annotation format
428 protected boolean looksLikeJnetData(String data)
430 char firstChar = data.charAt(0);
431 int colonPos = data.indexOf(":");
432 int commaPos = data.indexOf(",");
433 boolean isJnet = firstChar != '*' && firstChar != ' ' && colonPos > -1
434 && commaPos > -1 && colonPos < commaPos;
435 // && data.indexOf(",")<data.indexOf(",", data.indexOf(","))) / ??
440 * Returns true if the data has at least 6 tab-delimited fields _and_ fields 4
441 * and 5 are integer (start/end)
446 protected boolean looksLikeFeatureData(String data)
452 String[] columns = data.split("\t");
453 if (columns.length < 6)
457 for (int col = 3; col < 5; col++)
461 Integer.parseInt(columns[col]);
462 } catch (NumberFormatException e)
475 public static void main(String[] args)
477 for (int i = 0; args != null && i < args.length; i++)
479 IdentifyFile ider = new IdentifyFile();
480 FileFormatI type = null;
483 type = ider.identify(args[i], DataSourceType.FILE);
484 } catch (FileFormatException e)
487 String.format("Error '%s' identifying file type for %s",
488 args[i], e.getMessage()));
490 System.out.println("Type of " + args[i] + " is " + type);
492 if (args == null || args.length == 0)
494 System.err.println("Usage: <Filename> [<Filename> ...]");