rbbt-util 5.25.46 → 5.25.47

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: d7a6bfb0c182d36d1323ddefe70c93f57e173c55
-  data.tar.gz: fff96627b902d989f53d92a4fe6f327502f75c89
+  metadata.gz: 07dd5ab637d8ab626487a17775efe56830ba49d3
+  data.tar.gz: 83b9ca45962531d0a320c31fe6624b2dceb939fe
 SHA512:
-  metadata.gz: 9cd88404da356142456ea01f3efdb78ec784613a4b7670a4826ca40a4a8d74275632c7cf7ba521437b1554332e713b02904da2735fc30489bdaed9cc858f5e9e
-  data.tar.gz: 26d5b2fc59d6cd3446dfb7de5094c5dec73b44ab9e438e451f160bee3b9a3eddbdb3b49f5614bfc4fb40048bafe7b745360185a203147c0a9d9f2b23b441ae15
+  metadata.gz: e92de8699285124584a83188e1475dd957e5bf6e20b568aebb2c7c06104e763ec2063275e09c8273778e2d5f111c8793f0fc242a0a64ff14797714a0b106a23e
+  data.tar.gz: c0dd02ec8593f146a2b06a9bfd66360e2e30769e45f5dd8445865fc981672d97243699076b74abbece10197a1eaf2923a1c3a741392945278a81d0a9f4be5141

data/lib/rbbt/util/config.rb

@@ -110,7 +110,7 @@ module Rbbt::Config
       end
     end
 
-    value = priorities.empty? ? default : priorities.sort_by{|p,v| p}.first.last.first
+    value = priorities.empty? ? default : priorities.collect{|p| p }.reverse.sort_by{|p,v| p}.first.last.first
     value = false if value == 'false'
 
     Log.debug "Value #{value.inspect} for config key '#{ key }': #{tokens * ", "}"

data/lib/rbbt/util/misc/inspect.rb

@@ -374,6 +374,19 @@ module Misc
     obj2digest(obj)
   end
 
+  def self.file2md5(file)
+    if File.exists?(file + '.md5')
+      Open.read(file + '.md5')
+    else
+      md5 = CMD.cmd("md5sum '#{file}'").read.strip.split(" ").first
+      begin
+        Open.write(file + '.md5', md5)
+      rescue
+      end
+      md5
+    end
+  end
+
   def self.get_filename(obj)
     if obj.respond_to? :filename
       obj.filename

data/lib/rbbt/util/misc/omics.rb

@@ -328,6 +328,8 @@ module Misc
   end
 
   def self.chr_cmp_strict(chr1, chr2)
+    chr1 = chr1.sub(/^chr/)
+    chr2 = chr2.sub(/^chr/)
     if (m1 = chr1.match(/(\d+)$/)) && (m2 = chr2.match(/(\d+)$/))
       m1[1].to_i <=> m2[1].to_i
     elsif chr1 =~ /\d+$/
@@ -339,6 +341,10 @@ module Misc
     end
   end
 
+  def self.chr_cmp_contigs(chr1, chr2, contigs)
+    contigs.index(chr1) <=> contigs.index(chr2)
+  end
+
   def self.genomic_location_cmp(gpos1, gpos2, sep = ":")
     chr1, _sep, pos1 = gpos1.partition(sep)
     chr2, _sep, pos2 = gpos2.partition(sep)
@@ -363,6 +369,18 @@ module Misc
     end
   end
 
+  def self.genomic_location_cmp_contigs(gpos1, gpos2, contigs, sep = ":")
+    chr1, _sep, pos1 = gpos1.partition(sep)
+    chr2, _sep, pos2 = gpos2.partition(sep)
+    iii [gpos1, gpos2, chr1, chr2]
+    cmp = chr_cmp_contigs(chr1, chr2, contigs)
+    case cmp
+    when 0
+      pos1.to_i <=> pos2.to_i
+    else
+      cmp
+    end
+  end
   def self.intersect_streams_cmp_chr(chr1, chr2)
     chr1 <=> chr2
   end

data/lib/rbbt/workflow/accessor.rb

@@ -57,6 +57,10 @@ class Step
     File.join(dir, '.' << filename)
   end
 
+  def self.md5_file(path)
+    path.nil? ? nil : path + '.md5'
+  end
+
   def self.pid_file(path)
     path.nil? ? nil : path + '.pid'
   end

data/lib/rbbt/workflow/provenance.rb

@@ -1,16 +1,16 @@
 class Step
   def self.prov_status_msg(status)
     color = case status.to_sym
-            when :error, :aborted, :missing, :dead
+            when :error, :aborted, :missing, :dead, :unsync
               :red
             when :streaming, :started
               :cyan
-            when :done
+            when :done, :noinfo
               :green
-            when :noinfo
-              :blue
-            when :dependencies, :waiting, :setyp
+            when :dependencies, :waiting, :setup
               :yellow
+            when :notfound
+              :blue
             else
               if status.to_s.index ">"
                 :cyan
@@ -27,7 +27,7 @@ class Step
     task = Log.color(:yellow, parts.pop)
     workflow = Log.color(:magenta, parts.pop)
     if status.to_s == 'noinfo' and parts.last != 'jobs'
-      task, status, workflow = Log.color(:yellow, info[:task_name]), Log.color(:blue, "file"), Log.color(:magenta, "-")
+      task, status, workflow = Log.color(:yellow, info[:task_name]), Log.color(:green, "file"), Log.color(:magenta, "-")
     end
 
     path_mtime = begin
@@ -73,7 +73,8 @@ class Step
     status = info[:status] || :missing
     status = "remote" if Open.remote?(path)
     name = info[:name] || File.basename(path)
-    status = :unsync if status == :done and not Open.exist? path
+    status = :unsync if status == :done and not Open.exist?(path)
+    status = :notfound if status == :noinfo and not Open.exist?(path)
     str = " " * offset
     str << prov_report_msg(status, name, path, info)
     step.dependencies.each do |dep|
@@ -83,7 +84,7 @@ class Step
         seen << path
         str << prov_report(dep, offset + 1, task, seen)
       else
-        str << Log.color(:blue, Log.uncolor(prov_report(dep, offset+1, task)))
+        str << Log.color(:green, Log.uncolor(prov_report(dep, offset+1, task)))
       end
     end if step.dependencies
     str

data/lib/rbbt/workflow/step.rb

@@ -296,6 +296,7 @@ class Step
   def self.clean(path)
     info_file = Step.info_file path
     pid_file = Step.pid_file path
+    md5_file = Step.md5_file path
     files_dir = Step.files_dir path
 
     if ! (Open.writable?(path) && Open.writable?(info_file))
@@ -309,11 +310,11 @@ class Step
       @pid = nil
 
       Misc.insist do
-        Open.rm info_file if Open.exists? info_file
+        Open.rm info_file if Open.exists?(info_file)
+        Open.rm md5_file if Open.exists?(md5_file)
         Open.rm path if (Open.exists?(path) or Open.broken_link?(path))
-        Open.rm path if (Open.exists?(path) or Open.broken_link?(path))
-        Open.rm_rf files_dir if Open.exists? files_dir
-        Open.rm pid_file if Open.exists? pid_file
+        Open.rm_rf files_dir if Open.exists?(files_dir)
+        Open.rm pid_file if Open.exists?(pid_file)
       end
     end
   end

data/share/rbbt_commands/system/status

@@ -171,7 +171,7 @@ workflows.sort.each do |workflow,tasks|
       str = file.dup
       if options[:quick] and i[:done]
         status = 'done'
-        str << " #{ status_msg status }"
+        str << " #{ Step.prov_status_msg status }"
         if inputs and inputs.any? 
           str << input_msg(file, inputs)
         end

data/share/rbbt_commands/workflow/trace

@@ -56,7 +56,7 @@ jobs.each do |job|
   next unless job.done?
   started = job.info[:started]
   ddone = job.info[:done]
-  data[job.name] = [job.workflow.to_s, job.task_name,started,ddone]
+  data[[job.task_name.to_s, job.clean_name] * "."] = [job.workflow.to_s, job.task_name,started,ddone]
   if min_start.nil?
     min_start = started
   else
@@ -70,21 +70,73 @@ jobs.each do |job|
   end
 end
 
-#data.process "Start" do |value|
-#  value - min_start
-#end
-#
-#data.process "End" do |value|
-#  value - min_start
-#end
+data.add_field "Start.second" do |k,value|
+  value["Start"] - min_start
+end
+
+data.add_field "End.second" do |k,value|
+  value["End"] - min_start
+end
 
 Log.severity = 0
-data.R_interactive <<-EOF
-  rbbt.require('timevis')
+#data.R_interactive <<-EOF
+#  rbbt.require('timevis')
+#  data$id = rownames(data)
+#  data$content = data$Task
+#  data$start = data$Start
+#  data$end = data$End
+#
+#  timevis(data)
+#EOF
+
+
+data.R <<-EOF, [:svg]
+  rbbt.require('tidyverse')
+  rbbt.require('ggplot2')
+
+  names(data) <- make.names(names(data))
   data$id = rownames(data)
   data$content = data$Task
   data$start = data$Start
   data$end = data$End
-
-  timevis(data)
+  data$Project = data$Workflow
+
+  tasks = data
+
+  #theme_gantt <- function(base_size=11, base_family="Source Sans Pro Light") {
+  theme_gantt <- function(base_size=11, base_family="Sans Serif") {
+    ret <- theme_bw(base_size, base_family) %+replace%
+      theme(panel.background = element_rect(fill="#ffffff", colour=NA),
+            axis.title.x=element_text(vjust=-0.2), axis.title.y=element_text(vjust=1.5),
+            title=element_text(vjust=1.2, family="Source Sans Pro Semibold"),
+            panel.border = element_blank(), axis.line=element_blank(),
+            panel.grid.minor=element_blank(),
+            panel.grid.major.y = element_blank(),
+            panel.grid.major.x = element_line(size=0.5, colour="grey80"),
+            axis.ticks=element_blank(),
+            legend.position="bottom", 
+            axis.title=element_text(size=rel(0.8), family="Source Sans Pro Semibold"),
+            strip.text=element_text(size=rel(1), family="Source Sans Pro Semibold"),
+            strip.background=element_rect(fill="#ffffff", colour=NA),
+            panel.spacing.y=unit(1.5, "lines"),
+            legend.key = element_blank())
+    
+    ret
+  }
+
+  tasks.long <- tasks %>%
+  gather(date.type, task.date, -c(Project, Task, id, Start.second, End.second)) %>%
+  arrange(date.type, task.date) %>%
+  mutate(id = factor(id, levels=rev(unique(id)), ordered=TRUE))
+
+  x.breaks <- seq(length(tasks$Task) + 0.5 - 3, 0, by=-3)
+
+  timeline <- ggplot(tasks.long, aes(y=id, yend=id, x=Start.second, xend=End.second, colour=Task)) + 
+    geom_segment() + 
+    geom_vline(xintercept=x.breaks, colour="grey80", linetype="dotted") + 
+    guides(colour=guide_legend(title=NULL)) +
+    labs(x=NULL, y=NULL) + 
+    theme_gantt() + theme(axis.text.x=element_text(angle=45, hjust=1))
+
+  rbbt.png_plot('#{plot}', 'timeline', width=2000)
 EOF

data/test/rbbt/util/test_misc.rb

@@ -620,5 +620,11 @@ EOF
     end
     iii Misc.remove_long_items(tsv)
   end
+
+  def test_digest_file
+    TmpFile.with_file("FILE CONTENT") do |file|
+      assert_match /^[0-9a-z]{10,}$/, Misc.file2md5(file)
+    end
+  end
 end
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rbbt-util
 version: !ruby/object:Gem::Version
-  version: 5.25.46
+  version: 5.25.47
 platform: ruby
 authors:
 - Miguel Vazquez
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-04-12 00:00:00.000000000 Z
+date: 2019-04-14 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake