1 raise "Ruby 1.9 or later required" if RUBY_VERSION < "1.9"
3 # Copyright (C) 2007-2010 Martin A. Hansen.
5 # This program is free software; you can redistribute it and/or
6 # modify it under the terms of the GNU General Public License
7 # as published by the Free Software Foundation; either version 2
8 # of the License, or (at your option) any later version.
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
15 # You should have received a copy of the GNU General Public License
16 # along with this program; if not, write to the Free Software
17 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
19 # http://www.gnu.org/copyleft/gpl.html
21 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
23 # This software is part of the Biopieces framework (www.biopieces.org).
25 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
33 # Biopieces are command line scripts and uses OptionParser to parse command line
34 # options according to a list of casts. Each cast prescribes the long and short
35 # name of the option, the type, if it is mandatory, the default value, and allowed
36 # and disallowed values. An optional list of extra casts can be supplied, and the
37 # integrity of the casts are checked. Following the command line parsing, the
38 # options are checked according to the casts. Methods are also included for handling
39 # the parsing and emitting of Biopiece records, which are ASCII text records consisting
40 # of lines with a key/value pair separated by a colon and a white space ': '.
41 # Each record is separated by a line with three dashes '---'.
45 attr_accessor :out # accessor for out stream _ios_
47 # Initialize a Biopiece and write the status to file.
48 # Options are for testing purposes only.
49 def initialize(test=nil, input=STDIN, output=STDOUT)
55 # Check the integrity of a list of casts, followed by parsion options from argv
56 # and finally checking the options according to the casts. Returns nil if argv
57 # is empty, otherwise an options hash.
58 def parse(argv, cast_list=[], script_path=$0)
59 casts = Casts.new(cast_list)
60 option_handler = OptionHandler.new(argv, casts, script_path, @test)
61 @options = option_handler.options_parse
63 @in = Stream.open(@options, mode="r", @input)
68 # Open Biopiece input stream if not open and iterate over all Biopiece
69 # records in the stream.
75 @in.each_line do |line|
77 when /^([^:]+): (.*)$/
78 record[$1.to_sym] = $2
80 yield record unless record.empty?
83 raise "Bad record format: #{line}"
87 yield record unless record.empty?
92 alias :each :each_record
94 # Open Biopiece output stream if not open and puts record to the stream.
96 @out = Stream.open(@options, mode="w", @output) unless @out.is_a? IO
98 record.each do |key,value|
99 @out.print "#{key.to_s}: #{value}\n"
108 # Create a temporary directory inside the ENV["BP_TMP"] dir.
113 path = ENV["BP_TMP"] + "/" + [user, time + pid, pid, "bp_tmp"].join("_")
115 Status.new.set_tmpdir(path)
121 # Error class for all exceptions to do with option casts.
122 class CastError < StandardError; end
125 # Class to handle casts of command line options. Each cast prescribes the long and
126 # short name of the option, the type, if it is mandatory, the default value, and
127 # allowed and disallowed values. An optional list of extra casts can be supplied,
128 # and the integrity of the casts are checked.
130 TYPES = %w[flag string list int uint float file file! files files! dir dir! genome]
131 MANDATORY = %w[long short type mandatory default allowed disallowed]
133 # Initialize cast object with an optional options cast list to which
134 # ubiquitous casts are added after which all casts are checked.
135 def initialize(cast_list=[])
136 @cast_list = cast_list
140 self.push(*@cast_list)
145 # Add ubiquitous options casts.
147 @cast_list << {:long=>'help', :short=>'?', :type=>'flag', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
148 @cast_list << {:long=>'stream_in', :short=>'I', :type=>'files!', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
149 @cast_list << {:long=>'stream_out', :short=>'O', :type=>'file', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
150 @cast_list << {:long=>'verbose', :short=>'v', :type=>'flag', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
153 # Check integrity of the casts.
160 # Check if all mandatory keys are present in casts and raise if not.
162 @cast_list.each do |cast|
163 MANDATORY.each do |mandatory|
164 raise CastError, "Missing symbol in cast: '#{mandatory.to_sym}'" unless cast.has_key? mandatory.to_sym
169 # Check if all values in casts are valid.
171 @cast_list.each do |cast|
173 check_val_short(cast)
175 check_val_mandatory(cast)
176 check_val_default(cast)
177 check_val_allowed(cast)
178 check_val_disallowed(cast)
182 # Check if the values to long are legal and raise if not.
183 def check_val_long(cast)
184 unless cast[:long].is_a? String and cast[:long].length > 1
185 raise CastError, "Illegal cast of long: '#{cast[:long]}'"
189 # Check if the values to short are legal and raise if not.
190 def check_val_short(cast)
191 unless cast[:short].is_a? String and cast[:short].length == 1
192 raise CastError, "Illegal cast of short: '#{cast[:short]}'"
196 # Check if values to type are legal and raise if not.
197 def check_val_type(cast)
200 type_hash[type] = true
203 unless type_hash.has_key? cast[:type]
204 raise CastError, "Illegal cast of type: '#{cast[:type]}'"
208 # Check if values to mandatory are legal and raise if not.
209 def check_val_mandatory(cast)
210 unless cast[:mandatory] == true or cast[:mandatory] == false
211 raise CastError, "Illegal cast of mandatory: '#{cast[:mandatory]}'"
215 # Check if values to default are legal and raise if not.
216 def check_val_default(cast)
217 unless cast[:default].nil? or
218 cast[:default].is_a? String or
219 cast[:default].is_a? Integer or
220 cast[:default].is_a? Float
221 raise CastError, "Illegal cast of default: '#{cast[:default]}'"
225 # Check if values to allowed are legal and raise if not.
226 def check_val_allowed(cast)
227 unless cast[:allowed].is_a? String or cast[:allowed].nil?
228 raise CastError, "Illegal cast of allowed: '#{cast[:allowed]}'"
232 # Check if values to disallowed are legal and raise if not.
233 def check_val_disallowed(cast)
234 unless cast[:disallowed].is_a? String or cast[:disallowed].nil?
235 raise CastError, "Illegal cast of disallowed: '#{cast[:disallowed]}'"
239 # Check cast for duplicate long or short options names.
242 @cast_list.each do |cast|
243 raise CastError, "Duplicate argument: '--#{cast[:long]}'" if check_hash.has_key? cast[:long]
244 raise CastError, "Duplicate argument: '-#{cast[:short]}'" if check_hash.has_key? cast[:short]
245 check_hash[cast[:long]] = true
246 check_hash[cast[:short]] = true
250 # Convert values to :long keys to symbols for all casts.
252 @cast_list.each do |cast|
253 cast[:long] = cast[:long].to_sym
260 # Class for parsing argv using OptionParser according to given casts.
261 # Default options are set, file glob expressions expanded, and options are
262 # checked according to the casts. Usage information is printed and exit called
265 REGEX_LIST = /^(list|files|files!)$/
266 REGEX_INT = /^(int|uint)$/
267 REGEX_STRING = /^(file|file!|dir|dir!|genome)$/
269 def initialize(argv, casts, script_path, test=nil)
272 @script_path = script_path
276 # Parse options from argv using OptionParser and casts denoting long and
277 # short option names. Usage information is printed and exit called.
278 # A hash with options is returned.
282 option_parser = OptionParser.new do |option|
283 @casts.each do |cast|
286 option.on("-#{cast[:short]}", "--#{cast[:long]}") do |o|
287 @options[cast[:long]] = o
290 option.on("-#{cast[:short]}", "--#{cast[:long]} F", Float) do |f|
291 @options[cast[:long]] = f
294 option.on("-#{cast[:short]}", "--#{cast[:long]} S", String) do |s|
295 @options[cast[:long]] = s.to_sym # TODO: this to_sym - is that needed?
298 option.on( "-#{cast[:short]}", "--#{cast[:long]} A", Array) do |a|
299 @options[cast[:long]] = a
302 option.on("-#{cast[:short]}", "--#{cast[:long]} I", Integer) do |i|
303 @options[cast[:long]] = i
306 option.on("-#{cast[:short]}", "--#{cast[:long]} S", String) do |s|
307 @options[cast[:long]] = s
310 raise ArgumentError, "Unknown option type: '#{cast[:type]}'"
315 option_parser.parse!(@argv)
318 print_usage_and_exit(true)
319 elsif print_usage_short?
330 # Given the script name determine the path of the wiki file with the usage info.
332 path = File.join(ENV["BP_DIR"], "bp_usage", File.basename(@script_path)) + ".wiki"
333 raise "No such wiki file: #{path}" unless File.file? path
337 # Check if full "usage info" should be printed.
338 def print_usage_full?
342 # Check if short "usage info" should be printed.
343 def print_usage_short?
346 elsif @options[:stream_in]
348 elsif @options[:data_in]
350 elsif wiki_path =~ /^(list_biopieces|list_genomes|list_mysql_databases|biostat)$/ # TODO get rid of this!
357 # Print usage info by Calling an external script 'print_wiki'
358 # using a system() call and exit. An optional 'full' flag
359 # outputs the full usage info.
360 def print_usage_and_exit(full=nil)
365 system("print_wiki --data_in #{wiki_path} --help")
367 system("print_wiki --data_in #{wiki_path}")
370 raise "Failed printing wiki: #{wiki_path}" unless $?.success?
376 # Set default options value from cast unless a value is set.
378 @casts.each do |cast|
380 unless @options.has_key? cast[:long]
381 if cast[:type] == 'list'
382 @options[cast[:long]] = cast[:default].split ','
384 @options[cast[:long]] = cast[:default]
391 # Expands glob expressions to a full list of paths.
392 # Examples: "*.fna" or "foo.fna,*.fna" or "foo.fna,/bar/*.fna"
394 @casts.each do |cast|
395 if cast[:type] == 'files' or cast[:type] == 'files!'
396 if @options.has_key? cast[:long]
399 @options[cast[:long]].each do |path|
401 Dir.glob(path).each do |file|
402 files << file if File.file? file
409 @options[cast[:long]] = files
415 # Check all options according to casts.
417 @casts.each do |cast|
418 options_check_mandatory(cast)
419 options_check_int(cast)
420 options_check_uint(cast)
421 options_check_file(cast)
422 options_check_files(cast)
423 options_check_dir(cast)
424 options_check_allowed(cast)
425 options_check_disallowed(cast)
429 # Check if a mandatory option is set and raise if it isn't.
430 def options_check_mandatory(cast)
432 raise ArgumentError, "Mandatory argument: --#{cast[:long]}" unless @options.has_key? cast[:long]
436 # Check int type option and raise if not an integer.
437 def options_check_int(cast)
438 if cast[:type] == 'int' and @options.has_key? cast[:long]
439 unless @options[cast[:long]].is_a? Integer
440 raise ArgumentError, "Argument to --#{cast[:long]} must be an integer, not '#{@options[cast[:long]]}'"
445 # Check uint type option and raise if not an unsinged integer.
446 def options_check_uint(cast)
447 if cast[:type] == 'uint' and @options.has_key? cast[:long]
448 unless @options[cast[:long]].is_a? Integer and @options[cast[:long]] >= 0
449 raise ArgumentError, "Argument to --#{cast[:long]} must be an unsigned integer, not '#{@options[cast[:long]]}'"
454 # Check file! type argument and raise if file don't exists.
455 def options_check_file(cast)
456 if cast[:type] == 'file!' and @options.has_key? cast[:long]
457 raise ArgumentError, "No such file: '#{@options[cast[:long]]}'" unless File.file? @options[cast[:long]]
461 # Check files! type argument and raise if files don't exists.
462 def options_check_files(cast)
463 if cast[:type] == 'files!' and @options.has_key? cast[:long]
464 @options[cast[:long]].each do |path|
466 raise ArgumentError, "File not readable: '#{path}'" unless File.readable? path
471 # Check dir! type argument and raise if directory don't exist.
472 def options_check_dir(cast)
473 if cast[:type] == 'dir!' and @options.has_key? cast[:long]
474 raise ArgumentError, "No such directory: '#{@options[cast[:long]]}'" unless File.directory? @options[cast[:long]]
478 # Check options and raise unless allowed.
479 def options_check_allowed(cast)
480 if cast[:allowed] and @options.has_key? cast[:long]
482 cast[:allowed].split(',').each { |a| allowed_hash[a.to_s] = 1 }
484 raise ArgumentError, "Argument '#{@options[cast[:long]]}' to --#{cast[:long]} not allowed" unless allowed_hash.has_key? @options[cast[:long]].to_s
488 # Check disallowed argument values and raise if disallowed.
489 def options_check_disallowed(cast)
490 if cast[:disallowed] and @options.has_key? cast[:long]
491 cast[:disallowed].split(',').each do |val|
492 raise ArgumentError, "Argument '#{@options[cast[:long]]}' to --#{cast[:long]} is disallowed" if val.to_s == @options[cast[:long]].to_s
498 # Class for manipulating the execution status of Biopieces by setting a
499 # status file with a time stamp, process id, and command arguments. The
500 # status file is used for creating log entries and for displaying the
501 # runtime status of Biopieces.
503 # Write the status to a status file.
505 time0 = Time.new.strftime("%Y-%m-%d %X")
507 File.open(path, mode="w") do |fh|
508 fh.puts [time0, ARGV.join(" ")].join(";")
512 # Append the a temporary directory path to the status file.
513 def set_tmpdir(tmpdir_path)
516 File.open(path, mode="r") do |fh|
517 status = fh.read.chomp
520 status = "#{status};#{tmpdir_path}\n"
522 File.open(path, mode="w") do |fh|
527 # Extract the temporary directory path from the status file,
528 # and return this or nil if not found.
530 File.open(path, mode="r") do |fh|
531 tmpdir_path = fh.read.chomp.split(";").last
532 return tmpdir_path if File.directory?(tmpdir_path)
538 # Write the Biopiece status to the log file.
540 time1 = Time.new.strftime("%Y-%m-%d %X")
542 script = File.basename($0)
544 stream = File.open(path)
545 time0, args, tmp_dir = stream.first.split(";")
548 elap = time_diff(time0, time1)
549 command = [script, args].join(" ")
550 log_file = ENV["BP_LOG"] + "/biopieces.log"
552 File.open(log_file, mode = "a") { |file| file.puts [time0, time1, elap, user, exit_status, command].join("\t") }
555 # Delete status file.
562 # Path to status file
565 script = File.basename($0)
567 path = ENV["BP_TMP"] + "/" + [user, script, pid, "status"].join(".")
570 # Get the elapsed time from the difference between two time stamps.
571 def time_diff(t0, t1)
572 Time.at((DateTime.parse(t1).to_time - DateTime.parse(t0).to_time).to_i).gmtime.strftime('%X')
578 # Open Biopieces output data stream for reading from stdin or a file
579 # specified in options[:stream_in] OR writing to stdout or a file
580 # specified in options[:stream_out] or options[:data_out].
581 def self.open(options, mode, stdio)
583 if options[:data_in] and options[:data_in].first == "-"
586 $stdin.tty? ? read(options[:stream_in]) : stdio
589 options[:stream_out] ? self.write(options[:stream_out], options[:compress]) : stdio
591 raise "Bad mode #{mode}"
597 # Opens a reads stream to a list of files.
599 return if files.nil? #TODO case/when
600 self.zipped?(files) ? self.zread(files) : self.nread(files)
603 # Opens a write stream to a file and returns a _io_ object.
604 def self.write(file, zip=nil)
605 zip ? self.zwrite(file) : self.nwrite(file)
608 # Opens a list of gzipped files for reading and return an _io_ object.
609 def self.zread(files)
610 stdin, stdout, stderr = Open3.popen3("zcat " + files.join(' '));
616 # Opens a file for gzipped writing and return an _io_ object.
617 def self.zwrite(file)
618 stdin, stdout, stderr = Open3.popen3("gzip -f > #{file}")
624 # Opens a list of files for reading and return an _io_ object.
625 def self.nread(files)
626 stdin, stdout, stderr = Open3.popen3("cat " + files.join(' '));
632 # Opens a file for writing and return an _io_ object.
633 def self.nwrite(file)
634 File.open(file, mode="w")
637 # Test if a list of files are gzipped or not.
638 # Raises if files are mixed zipped and unzipped.
639 def self.zipped?(files)
643 type = `file #{file}`
645 if type =~ /gzip compressed/
646 type_hash[:gzip] = true
648 type_hash[:ascii] = true
652 raise "Mixture of zipped and unzipped files" if type_hash.size == 2
658 # Set status when 'biopieces' is required.
661 # Clean up when 'biopieces' exists.
663 exit_status = $! ? $!.inspect : "OK"
667 exit_status = "ERROR"
669 exit_status = "INTERRUPTED"
671 exit_status = "TERMINATED"
677 tmpdir = status.get_tmpdir
678 FileUtils.remove_entry_secure(tmpdir) unless tmpdir.nil?
679 status.log(exit_status)