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.
24 import java.io.IOException;
25 import java.util.Locale;
27 import jalview.bin.Console;
35 public class IdentifyFile
38 public FileFormatI identify(Object file, DataSourceType protocol)
39 throws FileFormatException
42 return (file instanceof File ? identify((File) file, protocol)
43 : identify((String) file, protocol));
47 public FileFormatI identify(File file, DataSourceType sourceType)
48 throws FileFormatException
51 String emessage = "UNIDENTIFIED FILE PARSING ERROR";
52 FileParse parser = null;
55 parser = new FileParse(file, sourceType);
58 return identify(parser);
62 Console.error("Error whilst identifying " + file, e);
63 emessage = e.getMessage();
67 throw new FileFormatException(parser.errormessage);
69 throw new FileFormatException(emessage);
73 * Identify a datasource's file content.
75 * @note Do not use this method for stream sources - create a FileParse object
81 * @throws FileFormatException
83 public FileFormatI identify(String file, DataSourceType sourceType)
84 throws FileFormatException
86 String emessage = "UNIDENTIFIED FILE PARSING ERROR";
87 FileParse parser = null;
90 parser = new FileParse(file, sourceType);
93 return identify(parser);
97 Console.error("Error whilst identifying " + file, e);
98 emessage = e.getMessage();
102 throw new FileFormatException(parser.errormessage);
104 throw new FileFormatException(emessage);
107 public FileFormatI identify(FileParse source) throws FileFormatException
109 return identify(source, true);
110 // preserves original behaviour prior to version 2.3
113 public FileFormatI identify(AlignmentFileReaderI file,
114 boolean closeSource) throws IOException
116 FileParse fp = new FileParse(file.getInFile(),
117 file.getDataSourceType());
118 return identify(fp, closeSource);
122 * Identify contents of source, closing it or resetting source to start
127 * @return (best guess at) file format
128 * @throws FileFormatException
130 public FileFormatI identify(FileParse source, boolean closeSource)
131 throws FileFormatException
133 FileFormatI reply = FileFormat.Pfam;
136 int trimmedLength = 0;
137 boolean lineswereskipped = false;
138 boolean isBinary = false; // true if length is non-zero and non-printable
139 // characters are encountered
147 boolean aaIndexHeaderRead = false;
149 while ((data = source.nextLine()) != null)
151 bytesRead += data.length();
152 trimmedLength += data.trim().length();
153 if (!lineswereskipped)
155 for (int i = 0; !isBinary && i < data.length(); i++)
157 char c = data.charAt(i);
158 isBinary = (c < 32 && c != '\t' && c != '\n' && c != '\r'
159 && c != 5 && c != 27); // nominal binary character filter
160 // excluding CR, LF, tab,DEL and ^E
161 // for certain blast ids
166 // jar files are special - since they contain all sorts of random
168 if (source.inFile != null)
170 String fileStr = source.inFile.getName();
171 if (fileStr.contains(".jar") || fileStr.contains(".zip")
172 || fileStr.contains(".jvp"))
174 // possibly a Jalview archive (but check further)
175 reply = FileFormat.Jalview;
178 if (!lineswereskipped && data.startsWith("PK"))
180 reply = FileFormat.Jalview; // archive
184 data = data.toUpperCase(Locale.ROOT);
186 if (data.startsWith(ScoreMatrixFile.SCOREMATRIX))
188 reply = FileFormat.ScoreMatrix;
191 if (data.startsWith("LOCUS"))
193 reply = FileFormat.GenBank;
196 if (data.startsWith("ID "))
198 if (data.substring(2).trim().split(";").length == 7)
200 reply = FileFormat.Embl;
204 if (data.startsWith("H ") && !aaIndexHeaderRead)
206 aaIndexHeaderRead = true;
208 if (data.startsWith("D ") && aaIndexHeaderRead)
210 reply = FileFormat.ScoreMatrix;
213 if (data.startsWith("##GFF-VERSION"))
215 // GFF - possibly embedded in a Jalview features file!
216 reply = FileFormat.Features;
219 if (looksLikeFeatureData(data))
221 reply = FileFormat.Features;
224 if (data.indexOf("# STOCKHOLM") > -1)
226 reply = FileFormat.Stockholm;
229 if (data.indexOf("_ENTRY.ID") > -1
230 || data.indexOf("_AUDIT_AUTHOR.NAME") > -1
231 || data.indexOf("_ATOM_SITE.") > -1)
233 reply = FileFormat.MMCif;
236 // if (data.indexOf(">") > -1)
237 if (data.startsWith(">"))
239 // FASTA, PIR file or BLC file
240 boolean checkPIR = false, starterm = false;
241 if ((data.indexOf(">P1;") > -1) || (data.indexOf(">DL;") > -1))
243 // watch for PIR file attributes
245 reply = FileFormat.PIR;
247 // could also be BLC file, read next line to confirm
248 data = source.nextLine();
250 if (data.indexOf(">") > -1)
252 reply = FileFormat.BLC;
256 // Is this a single line BLC file?
257 String data1 = source.nextLine();
258 String data2 = source.nextLine();
262 starterm = (data1 != null && data1.indexOf("*") > -1)
263 || (data2 != null && data2.indexOf("*") > -1);
265 if (data2 != null && (c1 = data.indexOf("*")) > -1)
267 if (c1 == 0 && c1 == data2.indexOf("*"))
269 reply = FileFormat.BLC;
273 reply = FileFormat.Fasta; // possibly a bad choice - may be
277 // otherwise can still possibly be a PIR file
281 reply = FileFormat.Fasta;
282 // TODO : AMSA File is indicated if there is annotation in the
283 // FASTA file - but FASTA will automatically generate this at the
291 // final check for PIR content. require
292 // >P1;title\n<blah>\nterminated sequence to occur at least once.
294 // TODO the PIR/fasta ambiguity may be the use case that is needed to
296 // a 'Parse as type XXX' parameter for the applet/application.
306 dta = source.nextLine();
307 } catch (IOException ex)
310 if (dta != null && dta.indexOf("*") > -1)
314 } while (dta != null && !starterm);
318 reply = FileFormat.PIR;
323 reply = FileFormat.Fasta; // probably a bad choice!
326 // read as a FASTA (probably)
329 if (data.indexOf("{\"") > -1)
331 reply = FileFormat.Json;
334 int lessThan = data.indexOf("<");
335 if ((lessThan > -1)) // possible Markup Language data i.e HTML,
338 String upper = data.toUpperCase(Locale.ROOT);
339 if (upper.substring(lessThan).startsWith("<HTML"))
341 reply = FileFormat.Html;
344 if (upper.substring(lessThan).startsWith("<RNAML"))
346 reply = FileFormat.Rnaml;
351 if ((data.length() < 1) || (data.indexOf("#") == 0))
353 lineswereskipped = true;
357 if (data.indexOf("PILEUP") > -1)
359 reply = FileFormat.Pileup;
364 if ((data.indexOf("//") == 0) || ((data.indexOf("!!") > -1) && (data
365 .indexOf("!!") < data.indexOf("_MULTIPLE_ALIGNMENT "))))
367 reply = FileFormat.MSF;
371 else if (data.indexOf("CLUSTAL") > -1)
373 reply = FileFormat.Clustal;
378 else if (data.indexOf("HEADER") == 0 || data.indexOf("ATOM") == 0)
380 reply = FileFormat.PDB;
383 else if (data.matches("\\s*\\d+\\s+\\d+\\s*"))
385 reply = FileFormat.Phylip;
390 if (!lineswereskipped && looksLikeJnetData(data))
392 reply = FileFormat.Jnet;
397 lineswereskipped = true; // this means there was some junk before any
398 // key file signature
406 source.reset(bytesRead); // so the file can be parsed from the mark
408 } catch (Exception ex)
410 Console.error("File Identification failed!\n" + ex);
411 throw new FileFormatException(source.errormessage);
413 if (trimmedLength == 0)
415 Console.error("File Identification failed! - Empty file was read.");
416 throw new FileFormatException("EMPTY DATA FILE");
418 Console.debug("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 Console.debug("Type of " + args[i] + " is " + type);
492 if (args == null || args.length == 0)
494 Console.error("Usage: <Filename> [<Filename> ...]");