1 # Copyright (C) 2007-2010 Martin A. Hansen.
3 # This program is free software; you can redistribute it and/or
4 # modify it under the terms of the GNU General Public License
5 # as published by the Free Software Foundation; either version 2
6 # of the License, or (at your option) any later version.
8 # This program is distributed in the hope that it will be useful,
9 # but WITHOUT ANY WARRANTY; without even the implied warranty of
10 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11 # GNU General Public License for more details.
13 # You should have received a copy of the GNU General Public License
14 # along with this program; if not, write to the Free Software
15 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
17 # http://www.gnu.org/copyleft/gpl.html
19 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
21 # This software is part of the Biopieces framework (www.biopieces.org).
23 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
31 # Biopieces are command line scripts and uses OptionParser to parse command line
32 # options according to a list of casts. Each cast prescribes the long and short
33 # name of the option, the type, if it is mandatory, the default value, and allowed
34 # and disallowed values. An optional list of extra casts can be supplied, and the
35 # integrity of the casts are checked. Following the command line parsing, the
36 # options are checked according to the casts. Methods are also included for handling
37 # the parsing and emitting of Biopiece records, which are ASCII text records consisting
38 # of lines with a key/value pair seperated by a colon and a white space ': '.
39 # Each record is separated by a line with three dashes '---'.
41 # Initialize a Biopiece and write the status to file.
42 # Options are for testing purposes only.
43 def initialize(test=nil, input=STDIN, output=STDOUT)
48 @status.set unless @test
51 # Check the integrity of a list of casts, followed by parsion options from argv
52 # and finally checking the options according to the casts. Returns nil if argv
53 # is empty, otherwise an options hash.
54 def parse(argv, cast_list=[], script_path=$0)
55 casts = Casts.new(cast_list)
56 option_handler = OptionHandler.new(argv, casts, script_path, @test)
57 @options = option_handler.options_parse
60 # Open Biopiece input stream if not open and iterate over all Biopiece
61 # records in the stream.
63 @in = Stream::open(@options, mode="r", @input) unless @in.is_a? IO
67 @in.each_line do |line|
69 when /^([^:]+): (.*)$/
72 yield record unless record.empty?
75 raise "Bad record format: #{line}"
79 yield record unless record.empty?
84 alias :each :each_record
86 # Open Biopiece output stream if not open and puts record to the stream.
88 @out = Stream::open(@options, mode="w", @output) unless @out.is_a? IO
90 record.each do |key,value|
91 @out.print "#{key}: #{value}\n"
97 # Create a temporary directory inside the ENV["BP_TMP"] dir.
102 path = ENV["BP_TMP"] + "/" + [user, time + pid, pid, "bp_tmp"].join("_")
104 @status.set_tmpdir(path)
110 # Error class for all exceptions to do with option casts.
111 class CastError < StandardError; end
114 # Class to handle casts of command line options. Each cast prescribes the long and
115 # short name of the option, the type, if it is mandatory, the default value, and
116 # allowed and disallowed values. An optional list of extra casts can be supplied,
117 # and the integrity of the casts are checked.
119 TYPES = %w[flag string list int uint float file file! files files! dir dir! genome]
120 MANDATORY = %w[long short type mandatory default allowed disallowed]
122 # Initialize cast object with an optional options cast list to which
123 # ubiquitous casts are added after which all casts are checked.
124 def initialize(cast_list=[])
125 @cast_list = cast_list
128 self.push *@cast_list
133 # Add ubiquitous options casts.
135 @cast_list << {:long => 'help', :short => '?', :type => 'flag', :mandatory => false, :default => nil, :allowed => nil, :disallowed => nil}
136 @cast_list << {:long => 'stream_in', :short => 'I', :type => 'files!', :mandatory => false, :default => nil, :allowed => nil, :disallowed => nil}
137 @cast_list << {:long => 'stream_out', :short => 'O', :type => 'file', :mandatory => false, :default => nil, :allowed => nil, :disallowed => nil}
138 @cast_list << {:long => 'verbose', :short => 'v', :type => 'flag', :mandatory => false, :default => nil, :allowed => nil, :disallowed => nil}
141 # Check integrity of the casts.
148 # Check if all mandatory keys are present in casts and raise if not.
150 @cast_list.each do |cast|
151 MANDATORY.each do |mandatory|
152 raise CastError, "Missing symbol in cast: '#{mandatory.to_sym}'" unless cast.has_key? mandatory.to_sym
157 # Check if all values in casts are valid.
159 @cast_list.each do |cast|
161 check_val_short(cast)
163 check_val_mandatory(cast)
164 check_val_default(cast)
165 check_val_allowed(cast)
166 check_val_disallowed(cast)
170 # Check if the values to long are legal and raise if not.
171 def check_val_long(cast)
172 unless cast[:long].is_a? String and cast[:long].length > 1
173 raise CastError, "Illegal cast of long: '#{cast[:long]}'"
177 # Check if the values to short are legal and raise if not.
178 def check_val_short(cast)
179 unless cast[:short].is_a? String and cast[:short].length == 1
180 raise CastError, "Illegal cast of short: '#{cast[:short]}'"
184 # Check if values to type are legal and raise if not.
185 def check_val_type(cast)
188 type_hash[type] = true
191 unless type_hash.has_key? cast[:type]
192 raise CastError, "Illegal cast of type: '#{cast[:type]}'"
196 # Check if values to mandatory are legal and raise if not.
197 def check_val_mandatory(cast)
198 unless cast[:mandatory] == true or cast[:mandatory] == false
199 raise CastError, "Illegal cast of mandatory: '#{cast[:mandatory]}'"
203 # Check if values to default are legal and raise if not.
204 def check_val_default(cast)
205 unless cast[:default].nil? or
206 cast[:default].is_a? String or
207 cast[:default].is_a? Integer or
208 cast[:default].is_a? Float
209 raise CastError, "Illegal cast of default: '#{cast[:default]}'"
213 # Check if values to allowed are legal and raise if not.
214 def check_val_allowed(cast)
215 unless cast[:allowed].is_a? String or cast[:allowed].nil?
216 raise CastError, "Illegal cast of allowed: '#{cast[:allowed]}'"
220 # Check if values to disallowed are legal and raise if not.
221 def check_val_disallowed(cast)
222 unless cast[:disallowed].is_a? String or cast[:disallowed].nil?
223 raise CastError, "Illegal cast of disallowed: '#{cast[:disallowed]}'"
227 # Check cast for duplicate long or short options names.
230 @cast_list.each do |cast|
231 raise CastError, "Duplicate argument: '--#{cast[:long]}'" if check_hash.has_key? cast[:long]
232 raise CastError, "Duplicate argument: '-#{cast[:short]}'" if check_hash.has_key? cast[:short]
233 check_hash[cast[:long]] = true
234 check_hash[cast[:short]] = true
240 # Class for parsing argv using OptionParser according to given casts.
241 # Default options are set, file glob expressions expanded, and options are
242 # checked according to the casts. Usage information is printed and exit called
245 REGEX_LIST = /^(list|files|files!)$/
246 REGEX_INT = /^(int|uint)$/
247 REGEX_STRING = /^(string|file|file!|dir|dir!|genome)$/
249 def initialize(argv, casts, script_path, test=nil)
252 @script_path = script_path
256 # Parse options from argv using OptionParser and casts denoting long and
257 # short option names. Usage information is printed and exit called.
258 # A hash with options is returned.
262 option_parser = OptionParser.new do |option|
263 @casts.each do |cast|
266 option.on("-#{cast[:short]}", "--#{cast[:long]}") do |o|
267 @options[cast[:long]] = o
270 option.on("-#{cast[:short]}", "--#{cast[:long]} F", Float) do |f|
271 @options[cast[:long]] = f
274 option.on( "-#{cast[:short]}", "--#{cast[:long]} A", Array) do |a|
275 @options[cast[:long]] = a
278 option.on("-#{cast[:short]}", "--#{cast[:long]} I", Integer) do |i|
279 @options[cast[:long]] = i
282 option.on("-#{cast[:short]}", "--#{cast[:long]} S", String) do |s|
283 @options[cast[:long]] = s
286 raise ArgumentError, "Unknown option type: '#{cast[:type]}'"
291 option_parser.parse!(@argv)
294 print_usage_and_exit(true)
295 elsif print_usage_short?
306 # Given the script name determine the path of the wiki file with the usage info.
308 path = ENV["BP_DIR"] + "/bp_usage/" + File.basename(@script_path) + ".wiki"
309 raise "No such wiki file: #{path}" unless File.file? path
313 # Check if full "usage info" should be printed.
314 def print_usage_full?
318 # Check if short "usage info" should be printed.
319 def print_usage_short?
322 elsif @options["stream_in"]
324 elsif @options["data_in"]
326 elsif wiki_path =~ /^(list_biopieces|list_genomes|list_mysql_databases|biostat)$/ # TODO get rid of this!
333 # Print usage info by Calling an external script 'print_wiki'
334 # using a system() call and exit. An optional 'full' flag
335 # outputs the full usage info.
336 def print_usage_and_exit(full=nil)
341 system("print_wiki --data_in #{wiki_path} --help")
343 system("print_wiki --data_in #{wiki_path}")
346 raise "Failed printing wiki: #{wiki_path}" unless $?.success?
352 # Set default options value from cast unless a value is set.
354 @casts.each do |cast|
356 @options[cast[:long]] = cast[:default] unless @options.has_key? cast[:long]
361 # Expands glob expressions to a full list of paths.
362 # Examples: "*.fna" or "foo.fna,*.fna" or "foo.fna,/bar/*.fna"
364 @casts.each do |cast|
365 if cast[:type] == 'files' or cast[:type] == 'files!'
366 if @options.has_key? cast[:long]
369 @options[cast[:long]].each do |path|
371 Dir.glob(path).each do |file|
372 files << file if File.file? file
379 @options[cast[:long]] = files
385 # Check all options according to casts.
387 @casts.each do |cast|
388 options_check_mandatory(cast)
389 options_check_int(cast)
390 options_check_uint(cast)
391 options_check_file(cast)
392 options_check_files(cast)
393 options_check_dir(cast)
394 options_check_allowed(cast)
395 options_check_disallowed(cast)
399 # Check if a mandatory option is set and raise if it isn't.
400 def options_check_mandatory(cast)
402 raise ArgumentError, "Mandatory argument: --#{cast[:long]}" unless @options.has_key? cast[:long]
406 # Check int type option and raise if not an integer.
407 def options_check_int(cast)
408 if cast[:type] == 'int' and @options.has_key? cast[:long]
409 unless @options[cast[:long]].is_a? Integer
410 raise ArgumentError, "Argument to --#{cast[:long]} must be an integer, not '#{@options[cast[:long]]}'"
415 # Check uint type option and raise if not an unsinged integer.
416 def options_check_uint(cast)
417 if cast[:type] == 'uint' and @options.has_key? cast[:long]
418 unless @options[cast[:long]].is_a? Integer and @options[cast[:long]] >= 0
419 raise ArgumentError, "Argument to --#{cast[:long]} must be an unsigned integer, not '#{@options[cast[:long]]}'"
424 # Check file! type argument and raise if file don't exists.
425 def options_check_file(cast)
426 if cast[:type] == 'file!' and @options.has_key? cast[:long]
427 raise ArgumentError, "No such file: '#{@options[cast[:long]]}'" unless File.file? @options[cast[:long]]
431 # Check files! type argument and raise if files don't exists.
432 def options_check_files(cast)
433 if cast[:type] == 'files!' and @options.has_key? cast[:long]
434 @options[cast[:long]].each do |path|
435 raise ArgumentError, "No such file: '#{path}'" unless File.file? path
440 # Check dir! type argument and raise if directory don't exist.
441 def options_check_dir(cast)
442 if cast[:type] == 'dir!' and @options.has_key? cast[:long]
443 raise ArgumentError, "No such directory: '#{@options[cast[:long]]}'" unless File.directory? @options[cast[:long]]
447 # Check options and raise unless allowed.
448 def options_check_allowed(cast)
449 if cast[:allowed] and @options.has_key? cast[:long]
451 cast[:allowed].split(',').each { |a| allowed_hash[a] = 1 }
453 raise ArgumentError, "Argument '#{@options[cast[:long]]}' to --#{cast[:long]} not allowed" unless allowed_hash.has_key? @options[cast[:long]]
457 # Check disallowed argument values and raise if disallowed.
458 def options_check_disallowed(cast)
459 if cast[:disallowed] and @options.has_key? cast[:long]
460 cast[:disallowed].split(',').each do |val|
461 raise ArgumentError, "Argument '#{@options[cast[:long]]}' to --#{cast[:long]} is disallowed" if val == @options[cast[:long]]
467 # Class for manipulating the execution status of Biopieces by setting a
468 # status file with a time stamp, process id, and command arguments. The
469 # status file is used for creating log entries and for displaying the
470 # runtime status of Biopieces.
472 # Write the status to a status file.
474 time0 = Time.new.strftime("%Y-%m-%d %X")
476 File.open(path, mode="w") { |file| file.puts [time0, ARGV.join(" ")].join(";") }
479 # Append the a temporary directory path to the status file.
480 def set_tmpdir(tmp_path)
481 status = File.open(path, mode="r").read.chomp
483 File.open(path, mode="w") do |fh|
484 fh << [status, tmp_path].join(";") + "\n"
488 # Extract the temporary directory path from the status file,
489 # and return this or nil if not found.
491 tmp_path = File.open(path, mode="r").read.chomp.split(";").last
493 File.directory?(tmp_path) ? tmp_path : nil
496 # Write the Biopiece status to the log file.
498 time1 = Time.new.strftime("%Y-%m-%d %X")
500 script = File.basename($0)
502 stream = File.open(path)
503 time0, args, tmp_dir = stream.first.split(";")
506 elap = time_diff(time0, time1)
507 command = [script, args].join(" ")
508 log_file = ENV["BP_LOG"] + "/biopieces.log"
510 File.open(log_file, mode="a") { |file| file.puts [time0, time1, elap, user, exit_status, command].join("\t") }
513 # Delete status file.
520 # Path to status file
523 script = File.basename($0)
525 path = ENV["BP_TMP"] + "/" + [user, script, pid, "status"].join(".")
528 # Get the elapsed time from the difference between two time stamps.
529 def time_diff(t0, t1)
530 Time.at((DateTime.parse(t1).to_time - DateTime.parse(t0).to_time).to_i).gmtime.strftime('%X')
536 # Open Biopieces output data stream for reading from stdin or a file
537 # specified in options["stream_in"] OR writing to stdout or a file
538 # specified in options["stream_out"].
539 def self.open(options, mode, stdio)
541 $stdin.tty? ? read(options["stream_in"]) : stdio
543 options["stream_out"] ? self.write(options["stream_out"], options["compress"]) : stdio
545 raise "Bad mode #{mode}"
551 # Opens a reads stream to a list of files.
553 self.zipped?(files) ? self.zread(files) : self.nread(files)
556 # Opens a write stream to a file and returns a _io_ object.
557 def self.write(file, zip=nil)
558 zip ? self.zwrite(file) : self.nwrite(file)
561 # Opens a list of gzipped files for reading and return an _io_ object.
562 def self.zread(files)
563 stdin, stdout, stderr = Open3.popen3("zcat " + files.join(' '));
569 # Opens a file for gzipped writing and return an _io_ object.
570 def self.zwrite(file)
571 stdin, stdout, stderr = Open3.popen3("gzip -f > #{file}")
577 # Opens a list of files for reading and return an _io_ object.
578 def self.nread(files)
579 stdin, stdout, stderr = Open3.popen3("cat " + files.join(' '));
585 # Opens a file for writing and return an _io_ object.
586 def self.nwrite(file)
587 File.open(file, mode="w")
590 # Test if a list of files are gzipped or not.
591 # Raises if files are mixed zipped and unzipped.
592 def self.zipped?(files)
596 type = `file #{file}`
598 if type =~ /gzip compressed/
599 type_hash[:gzip] = true
601 type_hash[:ascii] = true
605 raise "Mixture of zipped and unzipped files" if type_hash.size == 2
615 exit_status = $! ? $!.inspect : "OK"
619 exit_status = "ERROR"
621 exit_status = "INTERRUPTED"
623 exit_status = "TERMINATED"
629 tmpdir = status.get_tmpdir
630 FileUtils.remove_entry_secure(tmpdir) unless tmpdir.nil?
631 status.log(exit_status)