rbbt-util 5.8.7 → 5.8.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 1a0b0f59e7a2c5b61fd8ab0fc16b19008f534231
-  data.tar.gz: 590b7d23d0f33022de18241f12b221f640336a75
+  metadata.gz: 79182b6a00b36773864861817a0994c8b89b466a
+  data.tar.gz: 64eda216950e3ea231a67d5ebf463ddfd6f99855
 SHA512:
-  metadata.gz: 77577ab31857ac920b06346463a5f46ed85238a2d5fc0c55070b488dd2cf80cefd08d7ec6062d6cb967eb0aaafbe2875bd865ae8404effbbd2c4bc9795decffe
-  data.tar.gz: a935d916adc432ef32d2d254f9d518a7cd27c087ff8f01980233183a430d1f196442fe4ebe4c41aa62436f8f212ab8e5a222796d5ab68977bdeb099d2cfb6002
+  metadata.gz: 77f978c48f2f87df1b39e200d02ddec6b2b9e99413281935c4e95d8c8320e8f1ad3b4ec30e43677af723b19386152f86b10a0faee2e466fdd6d745c65d1d2c83
+  data.tar.gz: 0144b36d0f85345835b9abba865aa866e7d98df42f73f3594c967b8d5dfb8cb07f19719fe11c1bea3f03b5871779fac3146966b19a6e6edf2f10f21e55d28521

data/lib/rbbt/util/misc.rb

@@ -32,19 +32,31 @@ Lockfile.refresh = false if ENV["RBBT_NO_LOCKFILE_REFRESH"] == "true"
 module Misc
 
   def self.format_paragraph(text, size = 80, indent = 0, offset = 0)
-    words = text.gsub(/\s+/, "\s").split(" ")
-    lines = []
-    line = " "*offset
-    word = words.shift
-    while word
-      while word and line.length + word.length < size - indent
-        line << word << " "
-        word = words.shift
-      end
-      lines << ((" " * indent) << line[0..-2])
-      line = ""
-    end
-    (lines * "\n")
+    i = 0
+    re = /((?:\n\s*\n\s*)|(?:\n\s*(?=\*)))/
+    text.split(re).collect do |paragraph|
+      i += 1
+      str = if i % 2 == 1
+              words = paragraph.gsub(/\s+/, "\s").split(" ")
+              lines = []
+              line = " "*offset
+              word = words.shift
+              while word
+                word = word[0..size-indent-offset-4] + '...' if word.length >= size - indent - offset
+                while word and line.length + word.length <= size - indent
+                  line << word << " "
+                  word = words.shift
+                end
+                lines << ((" " * indent) << line[0..-2])
+                line = ""
+              end
+              (lines * "\n")
+            else
+              paragraph
+            end
+      offset = 0
+      str
+    end*""
   end
 
   def self.format_definition_list_item(dt, dd, size = 80, indent = 20, color = :yellow)
@@ -67,7 +79,7 @@ module Misc
   def self.format_definition_list(defs, size = 80, indent = 20, color = :yellow)
     entries = []
     defs.each do |dt,dd|
-      text = format_definition_list_item(dt,dd,size,indent, color)
+      text = format_definition_list_item(dt,dd,size,indent,color)
       entries << text
     end
     entries * "\n\n"

data/lib/rbbt/workflow/usage.rb

@@ -72,6 +72,32 @@ module Workflow
       dependencies = self.rec_dependencies(task_name).collect{|dep_name| self.tasks[dep_name.to_sym]}
 
       task.doc(dependencies)
+
+      if self.libdir.examples[task_name].exists?
+        self.libdir.examples[task_name].glob("*").each do |example_dir|
+          example = File.basename(example_dir)
+
+          puts Log.color(:magenta, "Example " << example) + " -- " + Log.color(:blue, example_dir)
+
+          inputs = {}
+
+          task.input_types.each do |input,type|
+            if example_dir[input].exists?
+              case type
+              when :tsv, :array, :text
+                head = example_dir[input].read.split("\n")[0..5].compact * "\n\n"
+                head = head[0..500]
+                puts Misc.format_definition_list_item(input, head).gsub("\n\n","\n")
+              else
+                puts Misc.format_definition_list_item(input, example_dir[input].read)
+              end
+            end
+          end
+
+          puts
+        end
+
+      end
     end
   end
 end

data/share/rbbt_commands/tsv/unzip

@@ -22,7 +22,7 @@ else
     file = file.dup
     Path.setup(File.expand_path(file))
   end
-  tsv = TSV.open(file)
+  tsv = TSV.open(file, :type => :double)
 end
 
 puts tsv.unzip(field, merge).to_s

data/share/rbbt_commands/workflow/task

@@ -37,7 +37,7 @@ def usage(workflow = nil, task = nil, exception=nil)
   puts SOPT.doc
   puts
   if workflow.nil?
-    puts "No workflow specified"
+    puts "No workflow specified. Use `rbbt workflow list` to list available workflows."
     exit -1 
   end
 
@@ -118,9 +118,9 @@ def fix_options(workflow, task, job_options)
               when TSV
                 value
               when '-'
-                TSV.open(STDIN, :unnamed => true)
+                TSV.open(STDIN, :unnamed => true, :sep => $field_separator, :sep2 => $array_separator || "|")
               else
-                TSV.open(value, :unnamed => true)
+                TSV.open(value, :unnamed => true, :sep => $field_separator, :sep2 => $array_separator || "|")
               end
             else
               value
@@ -135,6 +135,7 @@ options = SOPT.get <<EOF
 -h--help Show this help:
 -wd--workdir* Change the working directory of the workflow:
 -as--array_separator* Change the character that separates elements of Arrays, ',', '|', or '\\n' by default:
+-fs--field_separator* Change the character that separates fields of TSV files '\\t' by default:
 -jn--jobname* Job name to use. The name 'Default' is used by default:
 -pn--printname Print the name of the job and exit without starting it:
 -pf--printpath Print the path of the job result:
@@ -164,6 +165,7 @@ recursive_clean = !!options.delete(:recursive_clean)
 out = options.include?(:output) ? File.open(options[:output], 'wb') : STDOUT
 
 $array_separator = options.delete(:array_separator)
+$field_separator = options.delete(:field_separator) || "\t"
 
 # Get workflow
 

data/test/rbbt/util/test_misc.rb

@@ -6,20 +6,24 @@ require 'rbbt/entity'
 
 class TestMisc < Test::Unit::TestCase
 
-  def _test_format_paragraph
+  def test_format_paragraph
     p = <<-EOF
-
 Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor
 incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis
 nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.
+
 Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu
-    fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt
-    in culpa qui officia deserunt mollit anim id est laborum.
+fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt
+in culpa qui officia deserunt mollit anim id est laborum.
 
 
+    * one
+    * two
+    * three
+
     EOF
 
-    puts Misc.format_paragraph p, 70, 10, 5
+    assert Misc.format_paragraph(p, 70, 10, 5) =~ /\n\s*\* two/sm
   end
 
   def test_format_dl
@@ -38,27 +42,27 @@ Sed ut perspiciatis unde omnis iste natus error sit voluptatem accusantium
 doloremque laudantium, totam rem aperiam, eaque ipsa quae ab illo inventore
 veritatis et quasi architecto beatae vitae dicta sunt explicabo. Nemo enim
 ipsam voluptatem quia voluptas sit aspernatur aut odit aut fugit, sed quia
-consequuntur magni dolores eos qui ratione voluptatem sequi nesciunt. Neque
-porro quisquam est, qui dolorem ipsum quia dolor sit amet, consectetur,
+consequuntur magni dolores eos qui ratione voluptatem sequi nesciunt. 
+
+Neque porro quisquam est, qui dolorem ipsum quia dolor sit amet, consectetur,
 adipisci velit, sed quia non numquam eius modi tempora incidunt ut labore et
 dolore magnam aliquam quaerat voluptatem. Ut enim ad minima veniam, quis
 nostrum exercitationem ullam corporis suscipit laboriosam, nisi ut aliquid ex
 ea commodi consequatur? Quis autem vel eum iure reprehenderit qui in ea
 voluptate velit esse quam nihil molestiae consequatur, vel illum qui dolorem
 eum fugiat quo voluptas nulla pariatur?"
-
     EOF
 
-    puts Misc.format_definition_list({:paragraph_first => p1, :paragraph_second => p2})
+    assert Misc.format_definition_list({:paragraph_first => p1, :paragraph_second => p2}) =~ /       /
   end
 
-  def _test_parse_cmd_params
+  def test_parse_cmd_params
     assert_equal ["workflow", "task", "Translation", "translate", "-f", "Associated Gene Name", "-l", "-"],
       Misc.parse_cmd_params("workflow task Translation translate -f 'Associated Gene Name' -l -")
   end
 
 
-  def _test_fixutf8
+  def test_fixutf8
     string = "abc\xffdef"
     string = string.force_encoding("UTF-8") if string.respond_to? :force_encoding
     assert(! string.valid_encoding?) if string.respond_to? :valid_encoding?
@@ -67,37 +71,37 @@ eum fugiat quo voluptas nulla pariatur?"
     assert( Misc.fixutf8(string).valid_encoding) if string.respond_to? :valid_encoding
   end
 
-  def _test_colors_for
+  def test_colors_for
     colors, used = Misc.colors_for([1,2,2,1,2,1,2,2,3,3,2,3,2])
     assert_equal Misc::COLOR_LIST[1], used[2]
   end
 
-  def _test_total_length
+  def test_total_length
     ranges = [(0..100), (50..150), (120..160)]
     ranges = [(0..100), (50..150), (120..160), (51..70)]
     assert_equal 161, Misc.total_length(ranges)
   end
 
-  def _test_id_filename?
+  def test_id_filename?
     TmpFile.with_file("") do |file|
       assert Misc.is_filename?(file)
       assert ! Misc.is_filename?("TEST STRING")
     end
   end
 
-  def _test_merge_sorted_arrays
+  def test_merge_sorted_arrays
     assert_equal [1,2,3,4], Misc.merge_sorted_arrays([1,3], [2,4])
   end
 
-  def _test_intersect_sorted_arrays
+  def test_intersect_sorted_arrays
     assert_equal [2,4], Misc.intersect_sorted_arrays([1,2,3,4], [2,4])
   end
 
-  def _test_sorted_array_matches
+  def test_sorted_array_matches
     assert_equal [1,3], Misc.sorted_array_hits(%w(a b c d e), %w(b d))
   end
 
-  def _test_binary_include?
+  def test_binary_include?
     a = %w(a b c d e).sort
     assert Misc.binary_include?(a, "a")
     assert(!Misc.binary_include?(a, "z"))
@@ -106,24 +110,24 @@ eum fugiat quo voluptas nulla pariatur?"
     assert(Misc.binary_include?(a, "d"))
   end
 
-  def _test_process_to_hash
+  def test_process_to_hash
     list = [1,2,3,4]
     assert_equal 4, Misc.process_to_hash(list){|l| l.collect{|e| e * 2}}[2]
   end
 
-#  def _test_pdf2text_example
+#  def test_pdf2text_example
 #    assert PDF2Text.pdf2text(datafile_test('example.pdf')).read =~ /An Example Paper/i
 #  end
 #
-#  def _test_pdf2text_EPAR
+#  def test_pdf2text_EPAR
 #    assert PDF2Text.pdf2text("http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Scientific_Discussion/human/000402/WC500033103.pdf").read =~ /Tamiflu/i
 #  end
 #
-#  def _test_pdf2text_wrong
+#  def test_pdf2text_wrong
 #    assert_raise CMD::CMDError do PDF2Text.pdf2text("http://www.ema.europa.eu/docs/en_GB#").read end
 #  end
 
-  def _test_string2hash
+  def test_string2hash
     assert(Misc.string2hash("--user-agent=firefox").include? "--user-agent")
     assert_equal(true, Misc.string2hash(":true")[:true])
     assert_equal(true, Misc.string2hash("true")["true"])
@@ -134,12 +138,12 @@ eum fugiat quo voluptas nulla pariatur?"
     assert_equal(:j, Misc.string2hash("a=b#c=d#:h=:j")[:h])
   end
   
-  def _test_named_array
+  def test_named_array
     a = NamedArray.setup([1,2,3,4], %w(a b c d))
     assert_equal(1, a['a'])
   end
 
-  def _test_path_relative_to
+  def test_path_relative_to
     assert_equal "test/foo", Misc.path_relative_to('/test', '/test/test/foo')
 
     Misc.profile do
@@ -149,8 +153,8 @@ eum fugiat quo voluptas nulla pariatur?"
     end
   end
 
-#  def _test_chunk
-#    _test =<<-EOF
+#  def test_chunk
+#    test =<<-EOF
 #This is an example file. Entries are separated by Entry
 #-- Entry
 #1
@@ -165,7 +169,7 @@ eum fugiat quo voluptas nulla pariatur?"
 #    assert_equal "1\n2\n3", Misc.chunk(test, /^-- Entry/).first.strip
 #  end
 
-  def _test_hash2string
+  def test_hash2string
     hash = {}
     assert_equal hash, Misc.string2hash(Misc.hash2string(hash))
 
@@ -183,14 +187,14 @@ eum fugiat quo voluptas nulla pariatur?"
  
   end
 
-  def _test_merge
+  def test_merge
     a = [[1],[2]]
     a = NamedArray.setup a, %w(1 2)
     a.merge [3,4]
     assert_equal [1,3], a[0]
   end
 
-  def _test_indiferent_hash
+  def test_indiferent_hash
     a = {:a => 1, "b" => 2}
     a.extend IndiferentHash
 
@@ -200,7 +204,7 @@ eum fugiat quo voluptas nulla pariatur?"
     assert_equal 2, a[:b]
   end
 
-  def _test_lockfile
+  def test_lockfile
 
     TmpFile.with_file do |tmpfile|
       pids = []
@@ -224,7 +228,7 @@ eum fugiat quo voluptas nulla pariatur?"
     end
   end
 
-  def _test_positions2hash
+  def test_positions2hash
     inputs = Misc.positional2hash([:one, :two, :three], 1, :two => 2, :four => 4)
     assert_equal 1, inputs[:one]
     assert_equal 2, inputs[:two]
@@ -232,18 +236,18 @@ eum fugiat quo voluptas nulla pariatur?"
     assert_equal nil, inputs[:four]
   end
 
-  def _test_mean
+  def test_mean
     assert_equal 2, Misc.mean([1,2,3])
     assert_equal 3, Misc.mean([1,2,3,4,5])
   end
 
-  def _test_zip_fields
+  def test_zip_fields
     current = [[:a,1], [:b,2]]
     assert_equal [[:a, :b],[1,2]], Misc.zip_fields(current)
     assert_equal current, Misc.zip_fields(Misc.zip_fields(current))
   end
 
-  def _test_add_zipped
+  def test_add_zipped
     current = [[:a,1], [:b,2]]
     new = %w(A B)
     Misc.append_zipped current, new
@@ -254,20 +258,20 @@ eum fugiat quo voluptas nulla pariatur?"
     assert_equal Math.sqrt(2), Misc.sd([1,3])
   end
 
-  def _test_divide
+  def test_divide
     assert_equal 2, Misc.divide(%w(1 2 3 4 5 6 7 8 9),2).length
   end
 
-  def _test_ordered_divide
+  def test_ordered_divide
     assert_equal 5, Misc.ordered_divide(%w(1 2 3 4 5 6 7 8 9),2).length
   end
 
-  def _test_collapse_ranges
+  def test_collapse_ranges
     ranges = [(0..100), (50..150), (51..61),(200..250), (300..324),(320..350)]
     assert_equal [(0..150),(200..250), (300..350)], Misc.collapse_ranges(ranges)
   end
 
-  def _test_humanize
+  def test_humanize
     str1 = "test_string"
     str2 = "TEST_string"
     str3 = "test"
@@ -279,22 +283,22 @@ eum fugiat quo voluptas nulla pariatur?"
     assert_equal "mutation_enrichment", Misc.snake_case("MutationEnrichment")
   end
 
-  def _test_snake_case
+  def test_snake_case
     str1 = "ACRONIMTest"
     str2 = "ACRONIM_test"
     assert_equal "ACRONIM_test", Misc.snake_case(str1)
     assert_equal "ACRONIM_test", Misc.snake_case(str2)
   end
 
-  def _test_correct_vcf_mutations
+  def test_correct_vcf_mutations
     assert_equal [737407, ["-----", "-----G", "-----GTTAAT"]], Misc.correct_vcf_mutation(737406, "GTTAAT", "G,GG,GGTTAAT")
   end
 
-  def _test_fingerprint
+  def test_fingerprint
     assert_equal '{a=>1}', Misc.fingerprint({:a => 1})
   end
 
-  def _test_tarize
+  def test_tarize
     path = File.expand_path('/home/mvazquezg/git/rbbt-util/lib')
     stream = Misc.tarize(path)
     TmpFile.with_file do |res|
@@ -304,7 +308,7 @@ eum fugiat quo voluptas nulla pariatur?"
     end
   end
 
-  def _test_camel_case
+  def test_camel_case
     assert_equal "DbSNP", Misc.camel_case("db_SNP")
     assert_equal "D3Js", Misc.camel_case("D3Js")
     assert_equal "Structure", Misc.camel_case("Structure")
@@ -312,7 +316,7 @@ eum fugiat quo voluptas nulla pariatur?"
     assert_equal "COSMIC", Misc.camel_case("COSMIC")
   end
 
-  def _test_pipe
+  def test_pipe
     t = 5
     stream = Misc.open_pipe do |sin|
       t.times do |i|

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rbbt-util
 version: !ruby/object:Gem::Version
-  version: 5.8.7
+  version: 5.8.8
 platform: ruby
 authors:
 - Miguel Vazquez
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-03-15 00:00:00.000000000 Z
+date: 2014-03-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake