1 raise "Ruby 1.9 or later required" if RUBY_VERSION < "1.9"
3 # Copyright (C) 2007-2011 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 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
36 # Monkey patch (evil) changing the to_s method of the Hash class to
37 # return a Hash in Biopieces format; keys and value pairs on one line
38 # each seperated with ': ' and terminated by a line of '---'.
43 self.each do |key, value|
44 string << "#{key.to_s}: #{value}\n"
53 # Error class for all exceptions to do with the Biopieces class.
54 class BiopiecesError < StandardError; end
56 # Biopieces are command line scripts and uses OptionParser to parse command line
57 # options according to a list of casts. Each cast prescribes the long and short
58 # name of the option, the type, if it is mandatory, the default value, and allowed
59 # and disallowed values. An optional list of extra casts can be supplied, and the
60 # integrity of the casts are checked. Following the command line parsing, the
61 # options are checked according to the casts. Methods are also included for handling
62 # the parsing and emitting of Biopiece records, which are ASCII text records consisting
63 # of lines with a key/value pair separated by a colon and a white space ': '.
64 # Each record is separated by a line with three dashes '---'.
68 # Class method to check the integrity of a list of casts, followed by parsing
69 # options from argv and finally checking the options according to the casts.
70 def self.options_parse(argv, cast_list=[], script_path=$0)
71 casts = Casts.new(cast_list)
72 option_handler = OptionHandler.new(argv, casts, script_path)
73 options = option_handler.options_parse
78 # Class method for opening data streams for reading and writing Biopiece
79 # records. Records are read from STDIN (default) or file (possibly gzipped)
80 # and written to STDOUT (default) or file.
81 def self.open(input = STDIN, output = STDOUT)
82 io_in = self.open_input(input)
83 io_out = self.open_output(output)
97 # Class method to create a temporary directory inside the ENV["BP_TMP"] directory.
102 path = File.join(ENV["BP_TMP"], [user, time + pid, pid, "bp_tmp"].join("_"))
104 Status.new.set_tmpdir(path)
108 # Initialize a Biopiece object for either reading or writing from _ios_.
109 def initialize(ios, stdio = nil)
114 # Method to write a Biopiece record to _ios_.
119 # Method to close _ios_.
121 @ios.close unless @stdio
124 # Method to parse and yield a Biopiece record from _ios_.
126 while record = get_entry
130 self # conventionally
133 alias :each :each_record
138 @ios.each_line do |line|
140 when /^([^:]+): (.*)$/
141 record[$1.to_sym] = $2
145 raise BiopiecesError, "Bad record format: #{line}"
149 return record unless record.empty?
152 alias :get_record :each_entry
156 # Class method for opening data stream for reading Biopiece records.
157 # Records are read from STDIN (default) or file (possibly gzipped).
158 def self.open_input(input)
161 input = self.new(StringIO.new)
163 input = self.new(STDIN, true)
165 elsif File.exists? input
166 ios = File.open(input, 'r')
169 ios = Zlib::GzipReader.new(ios)
174 input = self.new(ios)
180 # Class method for opening data stream for writing Biopiece records.
181 # Records are written to STDOUT (default) or file.
182 def self.open_output(output)
184 output = self.new(STDOUT, true)
185 elsif not output.is_a? IO
186 output = self.new(File.open(output, 'w'))
194 # Error class for all exceptions to do with option casts.
195 class CastError < StandardError; end
197 # Class to handle casts of command line options. Each cast prescribes the long and
198 # short name of the option, the type, if it is mandatory, the default value, and
199 # allowed and disallowed values. An optional list of extra casts can be supplied,
200 # and the integrity of the casts are checked.
202 TYPES = %w[flag string list int uint float file file! files files! dir dir! genome]
203 MANDATORY = %w[long short type mandatory default allowed disallowed]
205 # Initialize cast object with an optional options cast list to which
206 # ubiquitous casts are added after which all casts are checked.
207 def initialize(cast_list=[])
208 @cast_list = cast_list
212 self.push(*@cast_list)
217 # Add ubiquitous options casts.
219 @cast_list << {:long=>'help', :short=>'?', :type=>'flag', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
220 @cast_list << {:long=>'stream_in', :short=>'I', :type=>'file!', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
221 @cast_list << {:long=>'stream_out', :short=>'O', :type=>'file', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
222 @cast_list << {:long=>'verbose', :short=>'v', :type=>'flag', :mandatory=>false, :default=>nil, :allowed=>nil, :disallowed=>nil}
225 # Check integrity of the casts.
232 # Check if all mandatory keys are present in casts and raise if not.
234 @cast_list.each do |cast|
235 MANDATORY.each do |mandatory|
236 raise CastError, "Missing symbol in cast: '#{mandatory}'" unless cast.has_key? mandatory.to_sym
241 # Check if all values in casts are valid.
243 @cast_list.each do |cast|
245 check_val_short(cast)
247 check_val_mandatory(cast)
248 check_val_default(cast)
249 check_val_allowed(cast)
250 check_val_disallowed(cast)
254 # Check if the values to long are legal and raise if not.
255 def check_val_long(cast)
256 unless cast[:long].is_a? String and cast[:long].length > 1
257 raise CastError, "Illegal cast of long: '#{cast[:long]}'"
261 # Check if the values to short are legal and raise if not.
262 def check_val_short(cast)
263 unless cast[:short].is_a? String and cast[:short].length == 1
264 raise CastError, "Illegal cast of short: '#{cast[:short]}'"
268 # Check if values to type are legal and raise if not.
269 def check_val_type(cast)
272 type_hash[type] = true
275 unless type_hash[cast[:type]]
276 raise CastError, "Illegal cast of type: '#{cast[:type]}'"
280 # Check if values to mandatory are legal and raise if not.
281 def check_val_mandatory(cast)
282 unless cast[:mandatory] == true or cast[:mandatory] == false
283 raise CastError, "Illegal cast of mandatory: '#{cast[:mandatory]}'"
287 # Check if values to default are legal and raise if not.
288 def check_val_default(cast)
289 unless cast[:default].nil? or
290 cast[:default].is_a? String or
291 cast[:default].is_a? Integer or
292 cast[:default].is_a? Float
293 raise CastError, "Illegal cast of default: '#{cast[:default]}'"
297 # Check if values to allowed are legal and raise if not.
298 def check_val_allowed(cast)
299 unless cast[:allowed].is_a? String or cast[:allowed].nil?
300 raise CastError, "Illegal cast of allowed: '#{cast[:allowed]}'"
304 # Check if values to disallowed are legal and raise if not.
305 def check_val_disallowed(cast)
306 unless cast[:disallowed].is_a? String or cast[:disallowed].nil?
307 raise CastError, "Illegal cast of disallowed: '#{cast[:disallowed]}'"
311 # Check cast for duplicate long or short options names.
314 @cast_list.each do |cast|
315 raise CastError, "Duplicate argument: '--#{cast[:long]}'" if check_hash[cast[:long]]
316 raise CastError, "Duplicate argument: '-#{cast[:short]}'" if check_hash[cast[:short]]
317 check_hash[cast[:long]] = true
318 check_hash[cast[:short]] = true
322 # Convert values to :long keys to symbols for all casts.
324 @cast_list.each do |cast|
325 cast[:long] = cast[:long].to_sym
331 # Class for parsing argv using OptionParser according to given casts.
332 # Default options are set, file glob expressions expanded, and options are
333 # checked according to the casts. Usage information is printed and exit called
336 REGEX_LIST = /^(list|files|files!)$/
337 REGEX_INT = /^(int|uint)$/
338 REGEX_STRING = /^(file|file!|dir|dir!|genome)$/
340 def initialize(argv, casts, script_path)
343 @script_path = script_path
347 # Parse options from argv using OptionParser and casts denoting long and
348 # short option names. Usage information is printed and exit called.
349 # A hash with options is returned.
351 option_parser = OptionParser.new do |option|
352 @casts.each do |cast|
355 option.on("-#{cast[:short]}", "--#{cast[:long]}") do |o|
356 @options[cast[:long]] = o
359 option.on("-#{cast[:short]}", "--#{cast[:long]} F", Float) do |f|
360 @options[cast[:long]] = f
363 option.on("-#{cast[:short]}", "--#{cast[:long]} S", String) do |s|
364 @options[cast[:long]] = s
367 option.on( "-#{cast[:short]}", "--#{cast[:long]} A", Array) do |a|
368 @options[cast[:long]] = a
371 option.on("-#{cast[:short]}", "--#{cast[:long]} I", Integer) do |i|
372 @options[cast[:long]] = i
375 option.on("-#{cast[:short]}", "--#{cast[:long]} S", String) do |s|
376 @options[cast[:long]] = s
379 raise ArgumentError, "Unknown option type: '#{cast[:type]}'"
384 option_parser.parse!(@argv)
387 print_usage_and_exit(true)
388 elsif print_usage_short?
399 # Given the script name determine the path of the wiki file with the usage info.
401 path = File.join(ENV["BP_DIR"], "bp_usage", File.basename(@script_path)) + ".wiki"
402 raise "No such wiki file: #{path}" unless File.file? path
406 # Check if full "usage info" should be printed.
407 def print_usage_full?
411 # Check if short "usage info" should be printed.
412 def print_usage_short?
415 elsif @options[:stream_in]
417 elsif @options[:data_in]
419 elsif wiki_path =~ /^(list_biopieces|list_genomes|list_mysql_databases|biostat)$/ # TODO get rid of this!
426 # Print usage info by Calling an external script 'print_wiki'
427 # using a system() call and exit. An optional 'full' flag
428 # outputs the full usage info.
429 def print_usage_and_exit(full=nil)
434 system("print_wiki --data_in #{wiki_path} --help")
436 system("print_wiki --data_in #{wiki_path}")
439 raise "Failed printing wiki: #{wiki_path}" unless $?.success?
445 # Set default options value from cast unless a value is set.
447 @casts.each do |cast|
449 unless @options[cast[:long]]
450 if cast[:type] == 'list'
451 @options[cast[:long]] = cast[:default].split ','
453 @options[cast[:long]] = cast[:default]
460 # Expands glob expressions to a full list of paths.
461 # Examples: "*.fna" or "foo.fna,*.fna" or "foo.fna,/bar/*.fna"
463 @casts.each do |cast|
464 if cast[:type] == 'files' or cast[:type] == 'files!'
465 if @options[cast[:long]]
468 @options[cast[:long]].each do |path|
470 Dir.glob(path).each do |file|
471 files << file if File.file? file
478 @options[cast[:long]] = files
484 # Check all options according to casts.
486 @casts.each do |cast|
487 options_check_mandatory(cast)
488 options_check_int(cast)
489 options_check_uint(cast)
490 options_check_file(cast)
491 options_check_files(cast)
492 options_check_dir(cast)
493 options_check_allowed(cast)
494 options_check_disallowed(cast)
498 # Check if a mandatory option is set and raise if it isn't.
499 def options_check_mandatory(cast)
501 raise ArgumentError, "Mandatory argument: --#{cast[:long]}" unless @options[cast[:long]]
505 # Check int type option and raise if not an integer.
506 def options_check_int(cast)
507 if cast[:type] == 'int' and @options[cast[:long]]
508 unless @options[cast[:long]].is_a? Integer
509 raise ArgumentError, "Argument to --#{cast[:long]} must be an integer, not '#{@options[cast[:long]]}'"
514 # Check uint type option and raise if not an unsinged integer.
515 def options_check_uint(cast)
516 if cast[:type] == 'uint' and @options[cast[:long]]
517 unless @options[cast[:long]].is_a? Integer and @options[cast[:long]] >= 0
518 raise ArgumentError, "Argument to --#{cast[:long]} must be an unsigned integer, not '#{@options[cast[:long]]}'"
523 # Check file! type argument and raise if file don't exists.
524 def options_check_file(cast)
525 if cast[:type] == 'file!' and @options[cast[:long]]
526 raise ArgumentError, "No such file: '#{@options[cast[:long]]}'" unless File.file? @options[cast[:long]]
530 # Check files! type argument and raise if files don't exists.
531 def options_check_files(cast)
532 if cast[:type] == 'files!' and @options[cast[:long]]
533 @options[cast[:long]].each do |path|
535 raise ArgumentError, "File not readable: '#{path}'" unless File.readable? path
540 # Check dir! type argument and raise if directory don't exist.
541 def options_check_dir(cast)
542 if cast[:type] == 'dir!' and @options[cast[:long]]
543 raise ArgumentError, "No such directory: '#{@options[cast[:long]]}'" unless File.directory? @options[cast[:long]]
547 # Check options and raise unless allowed.
548 def options_check_allowed(cast)
549 if cast[:allowed] and @options[cast[:long]]
551 cast[:allowed].split(',').each { |a| allowed_hash[a.to_s] = 1 }
553 raise ArgumentError, "Argument '#{@options[cast[:long]]}' to --#{cast[:long]} not allowed" unless allowed_hash[@options[cast[:long]].to_s]
557 # Check disallowed argument values and raise if disallowed.
558 def options_check_disallowed(cast)
559 if cast[:disallowed] and @options[cast[:long]]
560 cast[:disallowed].split(',').each do |val|
561 raise ArgumentError, "Argument '#{@options[cast[:long]]}' to --#{cast[:long]} is disallowed" if val.to_s == @options[cast[:long]].to_s
567 # Class for manipulating the execution status of Biopieces by setting a
568 # status file with a time stamp, process id, and command arguments. The
569 # status file is used for creating log entries and for displaying the
570 # runtime status of Biopieces.
572 # Write the status to a status file.
574 time0 = Time.new.strftime("%Y-%m-%d %X")
576 File.open(path, "w") do |fh|
577 fh.flock(File::LOCK_EX)
578 fh.puts [time0, ARGV.join(" ")].join(";")
582 # Append the a temporary directory path to the status file.
583 def set_tmpdir(tmpdir_path)
586 File.open(path, "r") do |fh|
587 fh.flock(File::LOCK_SH)
588 status = fh.read.chomp
591 status = "#{status};#{tmpdir_path}\n"
593 File.open(path, "w") do |fh|
594 fh.flock(File::LOCK_EX)
599 # Extract the temporary directory path from the status file,
600 # and return this or nil if not found.
602 File.open(path, "r") do |fh|
603 fh.flock(File::LOCK_SH)
604 tmpdir_path = fh.read.chomp.split(";").last
605 return tmpdir_path if File.directory?(tmpdir_path)
611 # Write the Biopiece status to the log file.
613 time1 = Time.new.strftime("%Y-%m-%d %X")
615 script = File.basename($0)
620 File.open(path, "r") do |fh|
621 fh.flock(File::LOCK_SH)
622 time0, args = fh.first.split(";")
625 elap = time_diff(time0, time1)
626 command = [script, args].join(" ")
627 log_file = File.join(ENV["BP_LOG"], "biopieces.log")
629 File.open(log_file, "a") do |fh|
630 fh.flock(File::LOCK_EX)
631 fh.puts [time0, time1, elap, user, exit_status, command].join("\t")
635 # Delete status file.
642 # Path to status file
645 script = File.basename($0)
647 path = File.join(ENV["BP_TMP"], [user, script, pid, "status"].join("."))
652 # Get the elapsed time from the difference between two time stamps.
653 def time_diff(t0, t1)
654 Time.at((DateTime.parse(t1).to_time - DateTime.parse(t0).to_time).to_i).gmtime.strftime('%X')
659 # Set status when 'biopieces' is required.
662 # Clean up when 'biopieces' exists.
664 exit_status = $! ? $!.inspect : "OK"
668 exit_status = "ERROR"
670 exit_status = "INTERRUPTED"
672 exit_status = "TERMINATED"
678 tmpdir = status.get_tmpdir
679 FileUtils.remove_entry_secure(tmpdir, true) unless tmpdir.nil?
680 status.log(exit_status)