Class: Cnvrg::Datafiles

Inherits:
Object
  • Object
show all
Defined in:
lib/cnvrg/datafiles.rb

Constant Summary collapse

LARGE_FILE = 
1024*1024*5
MULTIPART_SPLIT = 
10000000

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(owner, dataset_slug, dataset: nil) ⇒ Datafiles

Returns a new instance of Datafiles.



17
18
19
20
21
22
23
 
# File 'lib/cnvrg/datafiles.rb', line 17

def initialize(owner, dataset_slug, dataset: nil)
  @dataset_slug = dataset_slug
  @owner = owner
  @dataset = dataset
  @base_resource = "users/#{owner}/datasets/#{dataset_slug}/"
  @downloader = @dataset.get_storage_client
end

Instance Attribute Details

#base_resource ⇒ Object (readonly)

Returns the value of attribute base_resource.



15
16
17
 
# File 'lib/cnvrg/datafiles.rb', line 15

def base_resource
  @base_resource
end

Instance Method Details

#check_file_sha1(filename, org_sha1, tag: 'conflict') ⇒ Object 



25
26
27
28
29
30
31
 
# File 'lib/cnvrg/datafiles.rb', line 25

def check_file_sha1(filename, org_sha1, tag: 'conflict')
  file_loc = "#{Dir.pwd}/#{filename}"
  sha1 =  OpenSSL::Digest::SHA1.file(file_loc).hexdigest
  return 0 if sha1 == org_sha1
  FileUtils.cp(file_loc, "#{file_loc}.#{tag}")
  return 1
end

#check_files_sha1(files, resolver, tag) ⇒ Object 



49
50
51
52
53
54
55
56
57
 
# File 'lib/cnvrg/datafiles.rb', line 49

def check_files_sha1(files, resolver, tag)
  conflicts = 0
  files.each do |file|
    next if file.ends_with? '/'
    sha1 = resolver.fetch(file, {}).fetch("sha1", nil)
    conflicts += self.check_file_sha1(file, sha1, tag: tag)
  end
  conflicts
end

#clone_in_chunks(commit: 'latest', chunk_size: 1000) ⇒ Object 



934
935
936
937
938
 
# File 'lib/cnvrg/datafiles.rb', line 934

def clone_in_chunks(commit: 'latest', chunk_size: 1000)
  begin

  end
end

#cp_ds(relative: false) ⇒ Object 



73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
 
# File 'lib/cnvrg/datafiles.rb', line 73

def cp_ds(relative: false)
  prefix = @dataset.get_dataset["bucket_prefix"]
  batch_size = 10000
  pbar = ProgressBar.create(:title => "Download Progress",
                                    :progress_mark => '=',
                                    :format => "%b%i| %c Files downloaded",
                                    :starting_at => 0,
                                    :total => nil,
                                    :autofinish => true)
  parallel_options = {
      in_threads: ParallelThreads,
      in_processes: Cnvrg::CLI::ParallelProcesses,
      isolation: true,
      finish: ->(*args) { pbar.progress += 1 }
  }
  finished = false
  while not finished
    current_batch, marker = @downloader.fetch_files(prefix: prefix, marker: marker, limit: batch_size)
    if marker.blank?
      finished = true
    end
    Parallel.map(current_batch, parallel_options) do |file|
      next if file.end_with? "/"
      cutted_key = relative ? @downloader.cut_prefix(prefix, file) : file
      dest_path = File.join(@dataset.local_path, cutted_key)
      @downloader.download(file, dest_path, decrypt: false)
      file
    end
  end
end

#create_dir(absolute_path, relative_path, commit_sha1) ⇒ Object 



577
578
579
580
 
# File 'lib/cnvrg/datafiles.rb', line 577

def create_dir(absolute_path, relative_path, commit_sha1)
  response = Cnvrg::API.request(@base_resource + "create_dir", 'POST', {absolute_path: absolute_path, relative_path: relative_path, commit_sha1: commit_sha1})
  return Cnvrg::CLI.is_response_success(response, false)
end

#delete_commit(commit_sha1) ⇒ Object 



875
876
877
878
879
 
# File 'lib/cnvrg/datafiles.rb', line 875

def delete_commit(commit_sha1)
  response = Cnvrg::API.request("#{base_resource}/commit/#{commit_sha1}", 'DELETE')
  Cnvrg::CLI.is_response_success(response, true)
  return response
end

#delete_commit_files_local(deleted) ⇒ Object 



834
835
836
837
838
839
840
841
 
# File 'lib/cnvrg/datafiles.rb', line 834

def delete_commit_files_local(deleted)
  begin
    FileUtils.rm_rf(deleted) unless (deleted.nil? or deleted.empty?)
    return Cnvrg::Result.new(true, '')
  rescue => e
    return Cnvrg::Result.new(false, '')
  end
end

#delete_dir(absolute_path, relative_path, commit_sha1) ⇒ Object 



572
573
574
575
 
# File 'lib/cnvrg/datafiles.rb', line 572

def delete_dir(absolute_path, relative_path, commit_sha1)
  response = Cnvrg::API.request(@base_resource + "delete_dir", 'DELETE', {absolute_path: absolute_path, relative_path: relative_path, commit_sha1: commit_sha1})
  return Cnvrg::CLI.is_response_success(response, false)
end

#delete_file(absolute_path, relative_path, commit_sha1) ⇒ Object 



567
568
569
570
 
# File 'lib/cnvrg/datafiles.rb', line 567

def delete_file(absolute_path, relative_path, commit_sha1)
  response = Cnvrg::API.request(@base_resource + "delete_file", 'DELETE', {absolute_path: absolute_path, relative_path: relative_path, commit_sha1: commit_sha1})
  return Cnvrg::CLI.is_response_success(response, false)
end

#download_data_file(commit_sha1, dataset_home) ⇒ Object 



752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
 
# File 'lib/cnvrg/datafiles.rb', line 752

def download_data_file(commit_sha1, dataset_home)
  begin
    res = Cnvrg::API.request(@base_resource + "download_data_file", 'POST', {commit_sha1: commit_sha1,new_version:true})
    Cnvrg::CLI.is_response_success(res, false)
    if res["result"]
      download_resp = res
      filename = download_resp["result"]["filename"]

      sts_path = download_resp["result"]["path_sts"]
      if !Helpers.is_verify_ssl
        body = open(sts_path, {ssl_verify_mode: OpenSSL::SSL::VERIFY_NONE}).read
      else
        body = open(sts_path).read
      end
      split = body.split("\n")
      key = split[0]
      iv = split[1]


      access =  Cnvrg::Helpers.decrypt(key, iv, download_resp["result"]["sts_a"])

      secret =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["sts_s"])

      session =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["sts_st"])
      region =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["region"])

      bucket =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["bucket"])
      file_key =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["key"])

      is_s3 = download_resp["result"]["is_s3"]
      if is_s3 or is_s3.nil?
        client = Aws::S3::Client.new(
            :access_key_id =>access,
            :secret_access_key => secret,
            :session_token => session,
            :region => region,
            :http_open_timeout => 60, :retry_limit => 20)
      else
        endpoint = Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["endpoint_url"])
        client = Aws::S3::Client.new(
            :access_key_id =>access,
            :secret_access_key => secret,
            :region => region,
            :endpoint=> endpoint,:force_path_style=> true,:ssl_verify_peer=>false,
            :http_open_timeout => 60, :retry_limit => 20)
      end


      File.open(dataset_home+"/"+filename, 'w+') do |file|
        resp = client.get_object({bucket: bucket,
                              key: file_key }, target: file)
      end
      return filename
    end

  rescue => e
    return false

  end
end

#download_dir(dataset_home, absolute_path) ⇒ Object 



843
844
845
 
# File 'lib/cnvrg/datafiles.rb', line 843

def download_dir(dataset_home, absolute_path)
  FileUtils.mkdir_p("#{dataset_home}/#{absolute_path}")
end

#download_file(absolute_path, relative_path, project_home, conflict = false) ⇒ Object 



813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
 
# File 'lib/cnvrg/datafiles.rb', line 813

def download_file(absolute_path, relative_path, project_home, conflict=false)
  res = Cnvrg::API.request(@base_resource + "download_file", 'POST', {absolute_path: absolute_path, relative_path: relative_path})
  Cnvrg::CLI.is_response_success(res, false)
  if res["result"]
    res = res["result"]
    return false if res["link"].empty? or res["filename"].empty?
    filename = res["filename"]
    file_location = absolute_path.gsub(/#{filename}\/?$/, "")

    FileUtils.mkdir_p project_home + "/" + file_location
    filename += ".conflict" if conflict

    File.open("#{project_home}/#{file_location}/#{filename}", "w+") do |file|
      file.write open(res["link"]).read
    end
  else
    return false
  end
  return true
end

#download_file_s3(absolute_path, relative_path, project_home, conflict = false, commit_sha1 = nil, as_link = false) ⇒ Object 



659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
 
# File 'lib/cnvrg/datafiles.rb', line 659

def download_file_s3(absolute_path, relative_path, project_home, conflict=false, commit_sha1=nil, as_link=false)
  begin
    res = Cnvrg::API.request(@base_resource + "download_file", 'POST', {absolute_path: absolute_path, relative_path: relative_path, commit_sha1: commit_sha1 ,new_version:true, as_link:as_link})
    Cnvrg::CLI.is_response_success(res, false)
      if res["result"]
        file_url = res["result"]["file_url"]

        if as_link
          return res["result"]
        end
        # begin
        #   if !Helpers.is_verify_ssl
        #     tempfile = Down.download(file_url,open_timeout: 60,ssl_verify_mode: OpenSSL::SSL::VERIFY_NONE )
        #
        #   else
        #     tempfile = Down.download(file_url,open_timeout: 60)
        #
        #   end
        #
        #   FileUtils.move(tempfile.path, project_home+"/"+ absolute_path)
        #   return true
        # rescue
        #
        # end
        download_resp = res
        filename = download_resp["result"]["filename"]

        absolute_path += ".conflict" if conflict
        sts_path = download_resp["result"]["path_sts"]
        retries = 0
        success= false
        while !success and retries < 20
          begin
            if !Helpers.is_verify_ssl
              body = open(sts_path, {ssl_verify_mode: OpenSSL::SSL::VERIFY_NONE}).read
            else
              body = open(sts_path).read
            end
            success = true
          rescue => e
            retries +=1
            sleep(5)

          end
        end
        if !success
          return false
        end
        split = body.split("\n")
        key = split[0]
        iv = split[1]

        access =  Cnvrg::Helpers.decrypt(key, iv, download_resp["result"]["sts_a"])

        secret =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["sts_s"])

        session =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["sts_st"])
        region =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["region"])

        bucket =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["bucket"])
        file_key =  Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["key"])

        is_s3 = download_resp["result"]["is_s3"]
        if is_s3 or is_s3.nil?
          client = Aws::S3::Client.new(
              :access_key_id =>access,
              :secret_access_key => secret,
              :session_token => session,
              :region => region,
              :http_open_timeout => 60, :retry_limit => 20)
        else
          endpoint = Cnvrg::Helpers.decrypt(key,iv, download_resp["result"]["endpoint_url"])
          client = Aws::S3::Client.new(
              :access_key_id =>access,
              :secret_access_key => secret,
              :region => region,
              :endpoint=> endpoint,:force_path_style=> true,:ssl_verify_peer=>false,
              :http_open_timeout => 60, :retry_limit => 20)
        end

        File.open(project_home+"/"+absolute_path, 'w+') do |file|
          resp = client.get_object({bucket:bucket,
                                    key:file_key}, target: file)
        end
        return true
      end

  rescue => e
    return false

  end
end

#download_files_chunk(files, conflict: false, progress: nil) ⇒ Object 



966
967
968
969
970
971
972
973
974
975
976
 
# File 'lib/cnvrg/datafiles.rb', line 966

def download_files_chunk(files, conflict: false, progress: nil)
  (1..5).each do |i|
    response = Cnvrg::API.request("users/#{@owner}/datasets/#{@dataset_slug}/download_multi", 'POST', {files: files})
    next unless Cnvrg::CLI.is_response_success(response, false) #trying to api request 5 times.
    files_to_download = response['files']
    data_home = "#{Dir.pwd}/#{response['name']}"
    res = download_multiple_files_s3(files_to_download, data_home, conflict: conflict, read_only: false, progressbar: progress)
    next unless res.is_success? #try again..
    return files_to_download['keys'].length
  end
end

#download_files_in_chunks(files, chunk_size: 1000, conflict: false, commit: 'latest', progress: nil) ⇒ Object 



955
956
957
958
959
960
961
962
963
964
 
# File 'lib/cnvrg/datafiles.rb', line 955

def download_files_in_chunks(files, chunk_size: 1000, conflict: false, commit: 'latest', progress: nil)
  begin
    files.each_slice(chunk_size).each do |files|
      download_files_chunk(files, conflict: conflict, progress: progress)
    end
    return Cnvrg::Result.new(true, "Download Completed")
  rescue Exception => e
    return Cnvrg::Result.new(false, "Can`t download files")
  end
end

#download_image(file_path_to_store, image_slug, owner) ⇒ Object 



428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
 
# File 'lib/cnvrg/datafiles.rb', line 428

def download_image(file_path_to_store, image_slug, owner)


  download_resp = Cnvrg::API.request("users/#{owner}/images/#{image_slug}/" + "download", 'GET')
  path = download_resp["result"]["path"]

  if Cnvrg::CLI.is_response_success(download_resp, false)
    begin
      open(file_path_to_store, 'w+') do |file|
        file << open(path).read
      end

      return true
    rescue => e
      return false
    end

    return true
  else
    return false
  end


end

#download_list_files_in_query(response, dataset_home) ⇒ Object 



581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
 
# File 'lib/cnvrg/datafiles.rb', line 581

def download_list_files_in_query(response, dataset_home)
  sts_path = response["path_sts"]
  if !Helpers.is_verify_ssl
    body = open(sts_path, {ssl_verify_mode: OpenSSL::SSL::VERIFY_NONE}).read
  else
    body = open(sts_path).read
  end
  split = body.split("\n")
  key = split[0]
  iv = split[1]

  access =  Cnvrg::Helpers.decrypt(key, iv, response["sts_a"])

  secret =  Cnvrg::Helpers.decrypt(key,iv, response["sts_s"])

  session =  Cnvrg::Helpers.decrypt(key,iv, response["sts_st"])
  region =  Cnvrg::Helpers.decrypt(key,iv, response["region"])

  bucket =  Cnvrg::Helpers.decrypt(key,iv, response["bucket"])
  is_s3 = response["is_s3"]
  if is_s3 or is_s3.nil?
    client = Aws::S3::Client.new(
        :access_key_id =>access,
        :secret_access_key => secret,
        :session_token => session,
        :region => region,
        :http_open_timeout => 60, :retry_limit => 20)
  else
    endpoint = Cnvrg::Helpers.decrypt(key,iv, response["endpoint_url"])
    client = Aws::S3::Client.new(
        :access_key_id =>access,
        :secret_access_key => secret,
        :region => region,
        :endpoint=> endpoint,:force_path_style=> true,:ssl_verify_peer=>false,
        :http_open_timeout => 60, :retry_limit => 20)
  end
  list_files = response["files"]
  parallel_options = {
      :progress => {
          :title => "Download Progress",
          :progress_mark => '=',
          :format => "%b>>%i| %p%% %t",
          :starting_at => 0,
          :total => list_files.size,
          :autofinish => true
      },
      in_threads: ParallelThreads
  }
  download_count = 0
  Parallel.map((list_files), parallel_options) do |f|
    file_key =  Cnvrg::Helpers.decrypt(key,iv, f["path"])
    begin
    begin
      dir = File.dirname f["fullpath"]
      FileUtils.mkdir_p(dataset_home+"/"+ dir) unless File.exist? (dataset_home+"/"+ dir)
    end

    File.open(dataset_home+"/"+f["fullpath"], 'w+') do |file|
      resp = client.get_object({bucket:bucket,
                                key:file_key}, target: file)
    end
    download_count += 1
    rescue
    end

  end
  if download_count == list_files.size
    return true
  else
    return false
  end




end

#download_multiple_chunks(commit, chunk_size = 1000, progress: nil) ⇒ Object 



978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
 
# File 'lib/cnvrg/datafiles.rb', line 978

def download_multiple_chunks(commit, chunk_size=1000, progress: nil)
  begin
  last_chunk_size = chunk_size
  q = { commit: commit, chunk_size: chunk_size}
  overall = 0
  while last_chunk_size > 0
    response = Cnvrg::API.request("users/#{@owner}/datasets/#{@dataset_slug}/clone", 'POST', q)
    if Cnvrg::CLI.is_response_success(response, false) 
      files = response['files']
      data_home = "#{Dir.pwd}/#{response['name']}"
      last_chunk_size = files['keys'].length
      break if last_chunk_size == 0
      res = download_multiple_files_s3(files, data_home, read_only: false, progressbar: progress)
      overall += last_chunk_size
      q[:latest] = files['latest']
    else
      last_chunk_size = 0
    end
  end
  Cnvrg::Result.new(true, "Cloned #{overall} files!")
  rescue => e
    Cnvrg::Result.new(false, "Cant download chunk", e.message, e.backtrace)

  end
end

#download_multiple_files_s3(files, project_home, conflict: false, progressbar: nil, read_only: false) ⇒ Object 



1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
 
# File 'lib/cnvrg/datafiles.rb', line 1008

def download_multiple_files_s3(files, project_home, conflict: false, progressbar: nil, read_only:false)
begin
  parallel_options = {
      in_threads: ParallelThreads,
      isolation: true
  }
  Parallel.map(files["keys"], parallel_options) do |f|
    begin
    local_path = @dataset.local_path + '/' + f['name']
    Cnvrg::Logger.log_info("Downloading #{local_path}")
    progressbar.progress += 1 if progressbar.present?
    if local_path.end_with? "/"
      @downloader.mkdir(local_path, recursive: true)
      next
    end
      # blob
    local_path = "#{local_path}.conflict" if conflict
    storage_path = f["path"]
    if File.exists? local_path
      Cnvrg::Logger.log_info("Trying to download #{local_path} but its already exists, skipping..")
      next
    end
    resp = @downloader.download(storage_path, local_path)
    Cnvrg::Logger.log_info("Download #{local_path} success resp: #{resp}")
    rescue => e
      Cnvrg::Logger.log_error(e)
    end
  end
  return Cnvrg::Result.new(true,"Downloaded successfully")
rescue => e
    Cnvrg::Logger.log_error(e)
    return Cnvrg::Result.new(false,"Could not download some files", e.message, e.backtrace)
  end
end

#end_commit(commit_sha1, force, success: true, uploaded_files: 0) ⇒ Object 



913
914
915
916
917
918
919
920
921
922
 
# File 'lib/cnvrg/datafiles.rb', line 913

def end_commit(commit_sha1,force, success: true, uploaded_files: 0 )
  begin
    response = Cnvrg::API.request("#{base_resource}/commit/end", 'POST', {commit_sha1: commit_sha1,force:force, success: success, uploaded_files: uploaded_files})
    Cnvrg::CLI.is_response_success(response, true)
    #return response
  rescue => e
    return false
  end

end

#end_commit_tar(commit_sha1, cur_idx) ⇒ Object 



924
925
926
927
 
# File 'lib/cnvrg/datafiles.rb', line 924

def end_commit_tar(commit_sha1, cur_idx)
  response = Cnvrg::API.request("#{base_resource}/commit/end_tar", 'POST', {commit_sha1: commit_sha1, idx: cur_idx})
  return response
end

#generate_parallel_idx ⇒ Object 



1004
1005
1006
 
# File 'lib/cnvrg/datafiles.rb', line 1004

def generate_parallel_idx
  
end

#get_clone_chunk(latest_id: nil, chunk_size: 1000, commit: 'latest') ⇒ Object 



946
947
948
949
950
951
952
953
 
# File 'lib/cnvrg/datafiles.rb', line 946

def get_clone_chunk(latest_id: nil, chunk_size: 1000, commit: 'latest')
  response = Cnvrg::API.request("#{@base_resource}/clone_chunk", 'POST',{commit: commit, chunk_size: chunk_size, latest_id: latest_id})
  unless Cnvrg::CLI.is_response_success(response, false)
    Cnvrg::Logger.log_info("#{{commit: commit, chunk_size: chunk_size, latest_id: latest_id}}")
    return nil
  end
  response['result']['files']
end

#get_commit(commit_sha1) ⇒ Object 



880
881
882
883
884
 
# File 'lib/cnvrg/datafiles.rb', line 880

def get_commit(commit_sha1)
  response = Cnvrg::API.request("#{base_resource}/commit/#{commit_sha1}", 'GET')
  Cnvrg::CLI.is_response_success(response, true)
  return response
end

#get_trees(commit: "latest") ⇒ Object 



940
941
942
943
944
 
# File 'lib/cnvrg/datafiles.rb', line 940

def get_trees(commit: "latest")
  response = Cnvrg::API.request("#{@base_resource}/clone_trees", 'POST',{commit: commit})
  return nil unless Cnvrg::CLI.is_response_success(response, false)
  response['result']['files']
end

#last_valid_commit ⇒ Object 



901
902
903
904
905
906
907
908
909
910
911
 
# File 'lib/cnvrg/datafiles.rb', line 901

def last_valid_commit()
  begin
    #if we are pushing with force or to branch we dont need to send current/next commit cause we want to
    # create a new commit.
    response = Cnvrg::API.request("#{base_resource}/last_valid_commit", 'GET')
    Cnvrg::CLI.is_response_success(response, true)
    return response
  rescue => e
    return false
  end
end

#mark_conflicts(results) ⇒ Object 



59
60
61
62
63
64
65
66
67
68
69
70
71
 
# File 'lib/cnvrg/datafiles.rb', line 59

def mark_conflicts(results)
  begin
    updated = results["updated_on_server"]
    deleted = results["deleted"]
    resolver = results['resolved_files']
    overall = 0
    overall += self.check_files_sha1(updated, resolver, "conflict")
    overall += self.check_files_sha1(deleted, resolver, "deleted")
    overall
  rescue => e
    0
  end
end

#put_commit(commit_sha1) ⇒ Object 



105
106
107
108
109
110
111
112
113
114
 
# File 'lib/cnvrg/datafiles.rb', line 105

def put_commit(commit_sha1)
  response = Cnvrg::API.request("#{@base_resource}/commit/latest", 'PUT', {commit_sha1: commit_sha1})
  if response.present?
    msg = response['result']
  else
    msg =  "Cant save changes in the dataset"
  end

  Cnvrg::Result.new(Cnvrg::CLI.is_response_success(response, false), msg)
end

#revoke_download(tar_files, extracted_files) ⇒ Object 



862
863
864
865
866
867
868
869
870
871
872
873
874
 
# File 'lib/cnvrg/datafiles.rb', line 862

def revoke_download(tar_files, extracted_files)
  begin

    FileUtils.rm_rf(tar_files) unless (tar_files.nil? or tar_files.empty?)
    FileUtils.rm_rf(extracted_files) unless (extracted_files.nil? or extracted_files.empty?)

  rescue => e
    return false
  end

  return true

end

#revoke_download_dir(absolute_path) ⇒ Object 



847
848
849
 
# File 'lib/cnvrg/datafiles.rb', line 847

def revoke_download_dir(absolute_path)
  puts FileUtils.rmtree("#{absolute_path}")
end

#revoke_download_file(absolute_path, filename, conflict = false) ⇒ Object 



851
852
853
854
855
856
857
858
859
860
861
 
# File 'lib/cnvrg/datafiles.rb', line 851

def revoke_download_file(absolute_path, filename, conflict=false)
  begin
    file_location = absolute_path.gsub(/#{filename}\/?$/, "")

    filename += ".conflict" if conflict
    FileUtils.remove("#{file_location}/#{filename}")
    return true
  rescue
    return false
  end
end

#rollback_commit(commit_sha1) ⇒ Object 



929
930
931
932
 
# File 'lib/cnvrg/datafiles.rb', line 929

def rollback_commit(commit_sha1)
  response = Cnvrg::API.request("#{base_resource}/commit/rollback", 'POST', {commit_sha1: commit_sha1})
  Cnvrg::CLI.is_response_success(response, false)
end

#start_commit(new_branch, force = false, delete_commit = nil, chunks: 0, dataset: @dataset, message: nil) ⇒ Object 



886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
 
# File 'lib/cnvrg/datafiles.rb', line 886

def start_commit(new_branch,force=false,delete_commit=nil, chunks: 0, dataset: @dataset, message:nil)
  begin
    #if we are pushing with force or to branch we dont need to send current/next commit cause we want to
    # create a new commit.
    idx = (force || new_branch) ? {} : dataset.get_idx
    commit = idx[:commit]
    next_commit = idx[:next_commit]
    response = Cnvrg::API.request("#{base_resource}/commit/start", 'POST', {dataset_slug: @dataset_slug, new_branch: new_branch,force:force,
                                                                            username: @owner,current_commit: commit, next_commit: next_commit, total_chunks: chunks, message: message})
    Cnvrg::CLI.is_response_success(response, true)
    return response
  rescue => e
    return false
  end
end

#upload_cnvrg_image(absolute_path, image_name, secret) ⇒ Object 



406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
 
# File 'lib/cnvrg/datafiles.rb', line 406

def upload_cnvrg_image(absolute_path, image_name,secret)
  file_name = File.basename relative_path
  file_size = File.size(absolute_path).to_f
  mime_type = MimeMagic.by_path(absolute_path)
  content_type = !(mime_type.nil? or mime_type.text?) ? mime_type.type : "text/plain"
  upload_resp = Cnvrg::API.request("images/#{image_name}/upload_file", 'POST_FILE', {absolute_path: absolute_path, relative_path: absolute_path,
                                                                                 file_name: file_name,
                                                                                 file_size: file_size, file_content_type: content_type,
                                                                                 secret: secret})
  if Cnvrg::CLI.is_response_success(upload_resp, false)
    path = upload_resp["result"]["path"]
    s3_res = upload_large_files_s3(upload_resp, absolute_path)
    if s3_res
      Cnvrg::API.request(@base_resource + "update_s3", 'POST', {path: path, commit_id: upload_resp["result"]["commit_id"],
                                                                blob_id: upload_resp["result"]["id"]})
      return true
    end
  end
  return false

end

#upload_data_log_file(absolute_path, relative_path, data_commit_sha) ⇒ Object 



308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
 
# File 'lib/cnvrg/datafiles.rb', line 308

def upload_data_log_file(absolute_path, relative_path,data_commit_sha)
  file_name = File.basename relative_path
  file_size = File.size(absolute_path).to_f
  content_type = "text/x-log"
  upload_resp = Cnvrg::API.request("/users/#{@owner}/" + "upload_data_log", 'POST_FILE', {absolute_path: absolute_path, relative_path: relative_path,
                                                                                         file_name: file_name, log_date: Time.now,
                                                                                         file_size: file_size, file_content_type: content_type,
                                                                                          data_commit_sha1:data_commit_sha})
  if Cnvrg::CLI.is_response_success(upload_resp, false)
    path = upload_resp["result"]["path"]
    s3_res = upload_small_files_s3(path, absolute_path, "text/plain")
  end
  if s3_res
    return true
  end
  return false

end

#upload_exec_file(absolute_path, image_name, commit_id) ⇒ Object 



328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
 
# File 'lib/cnvrg/datafiles.rb', line 328

def upload_exec_file(absolute_path, image_name, commit_id)
  file_name = File.basename absolute_path
  file_size = File.size(absolute_path).to_f
  content_type = "application/zip"
  begin
    upload_resp = Cnvrg::API.request("users/#{@owner}/images/" + "upload_config", 'POST_FILE', {relative_path: absolute_path,
                                                                                                file_name: file_name,
                                                                                                image_name: image_name,
                                                                                                file_size: file_size,
                                                                                                file_content_type: content_type,
                                                                                                project_slug: @project_slug,
                                                                                                commit_id: commit_id})
    # puts upload_resp
    if Cnvrg::CLI.is_response_success(upload_resp, false)
      if upload_resp["result"]["image"] == -1
        return -1
      end
      path = upload_resp["result"]["path"]
      s3_res = upload_small_files_s3(path, absolute_path, content_type)

    end
    if s3_res
      return upload_resp["result"]["id"]
    end
    return false
  rescue SignalException

    say "\nAborting"
    exit(1)
  end

end

#upload_file(absolute_path, relative_path, commit_sha1) ⇒ Object 



196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
 
# File 'lib/cnvrg/datafiles.rb', line 196

def upload_file(absolute_path, relative_path, commit_sha1)
  file_name = File.basename relative_path
  file_size = File.size(absolute_path).to_f
  mime_type = MimeMagic.by_path(absolute_path)
  content_type = !(mime_type.nil? or mime_type.text?) ? mime_type.type : "text/plain"
  sha1 =  OpenSSL::Digest::SHA1.file(absolute_path).hexdigest
  if (absolute_path.include? "_tags.yml" or absolute_path.include? "_tags.yaml")
    is_valid = false
    begin
    content = open(absolute_path).read()
    hash = YAML.load(open(absolute_path).read())
      # if level 1 keys count is 1
      if hash.keys.count == 1
        if hash["tags"].present?
          is_valid = true
        elsif hash[hash.keys.first].class != Hash
          is_valid = true
        end
      # if level 1 keys count is greater than 1
      elsif hash.keys.count > 1
        if hash["tags"].present? and hash["tags"].class == Hash
          is_valid = false
        else
          is_valid = true
        end
      end
    rescue
      is_valid = false
    end

    if is_valid
      upload_resp = Cnvrg::API.request(@base_resource + "upload_file", 'POST_FILE', {absolute_path: absolute_path, relative_path: relative_path,
                                                                                     commit_sha1: commit_sha1, file_name: file_name,
                                                                                     file_size: file_size, file_content_type: content_type, sha1: sha1, content: content})
    else
      puts("#{absolute_path} is invalid")
      puts("Please check yaml structure.")
    end
  else
    upload_resp = Cnvrg::API.request(@base_resource + "upload_file", 'POST_FILE', {absolute_path: absolute_path, relative_path: relative_path,
                                                                                   commit_sha1: commit_sha1, file_name: file_name,
                                                                                   file_size: file_size, file_content_type: content_type, sha1: sha1})
    puts upload_resp

  end

  if Cnvrg::CLI.is_response_success(upload_resp, false)
    s3_res = upload_large_files_s3(upload_resp, absolute_path)

    return s3_res
  end
end

#upload_image(absolute_path, image_name, owner, is_public, is_base, dpkg, libraries, bash, message, commit_id) ⇒ Object 



362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
 
# File 'lib/cnvrg/datafiles.rb', line 362

def upload_image(absolute_path, image_name, owner, is_public, is_base, dpkg, libraries, bash, message, commit_id)
  file_name = File.basename absolute_path
  file_size = File.size(absolute_path).to_f
  if is_base

    content_type = "application/zip"
  else
    content_type = "application/gzip"
  end
  begin
    upload_resp = Cnvrg::API.request("users/#{owner}/images/" + "upload_cnvrg", 'POST_FILE', {relative_path: absolute_path,
                                                                                              file_name: file_name,
                                                                                              image_name: image_name,
                                                                                              file_size: file_size,
                                                                                              file_content_type: content_type,
                                                                                              is_public: is_public,
                                                                                              project_slug: @project_slug,
                                                                                              commit_id: commit_id,
                                                                                              dpkg: dpkg,
                                                                                              py2: libraries,
                                                                                              py3: libraries,
                                                                                              bash_history: bash,
                                                                                              commit_message: message,
                                                                                              is_base: is_base})
    # puts upload_resp
    if Cnvrg::CLI.is_response_success(upload_resp, false)
      path = upload_resp["result"]["path"]
      s3_res = upload_small_files_s3(path, absolute_path, content_type)
      if s3_res
        commit_resp = Cnvrg::API.request("users/#{owner}/images/#{upload_resp["result"]["id"]}/" + "commit", 'GET')
        if Cnvrg::CLI.is_response_success(commit_resp, false)
          return commit_resp["result"]["image"]
        else
          return false
        end

      end
    end
    return false
  rescue => e
  end

end

#upload_large_files_s3(upload_resp, file_path) ⇒ Object 



453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
 
# File 'lib/cnvrg/datafiles.rb', line 453

def upload_large_files_s3(upload_resp, file_path)
  begin
    sts_path = upload_resp["result"]["path_sts"]

    retries = 0
    success= false
    while !success and retries < 20
      begin
        if !Helpers.is_verify_ssl
          body = open(sts_path, {ssl_verify_mode: OpenSSL::SSL::VERIFY_NONE}).read
        else
          body = open(sts_path).read
        end
        success = true
      rescue => e
        retries +=1
        sleep(5)

      end
    end
    if !success
      return false
    end
    split = body.split("\n")
    key = split[0]
    iv = split[1]
    is_s3 = upload_resp["result"]["is_s3"]
    access =  Cnvrg::Helpers.decrypt(key, iv, upload_resp["result"]["sts_a"])

    secret =  Cnvrg::Helpers.decrypt(key,iv, upload_resp["result"]["sts_s"])

    session =  Cnvrg::Helpers.decrypt(key,iv, upload_resp["result"]["sts_st"])
    region =  Cnvrg::Helpers.decrypt(key,iv, upload_resp["result"]["region"])

    bucket =  Cnvrg::Helpers.decrypt(key,iv, upload_resp["result"]["bucket"])
    server_side_encryption =upload_resp["result"]["server_side_encryption"]
    use_accelerate_endpoint = false
    if is_s3 or is_s3.nil?

      use_accelerate_endpoint =true
      client = Aws::S3::Client.new(
          :access_key_id =>access,
          :secret_access_key => secret,
          :session_token => session,
          :region => region,
          :http_open_timeout => 60, :retry_limit => 20)
    else
      endpoint = Cnvrg::Helpers.decrypt(key,iv, upload_resp["result"]["endpoint"])
      use_accelerate_endpoint = false
      client = Aws::S3::Client.new(
          :access_key_id =>access,
          :secret_access_key => secret,
          :region => region,
          :endpoint=> endpoint,:force_path_style=> true,:ssl_verify_peer=>false,
          :http_open_timeout => 60, :retry_limit => 20)
    end


    if !server_side_encryption
      options = {:use_accelerate_endpoint => use_accelerate_endpoint}
    else
      options = {:use_accelerate_endpoint => use_accelerate_endpoint, :server_side_encryption => server_side_encryption}
    end




    s3 = Aws::S3::Resource.new(client: client)

    resp = s3.bucket(bucket).
        object(upload_resp["result"]["path"]+"/"+File.basename(file_path)).
        upload_file(file_path,options)


    return resp

  rescue => e
    puts e.message
    return false

  end
  return true

end

#upload_log_file(absolute_path, relative_path, log_date) ⇒ Object 



291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
 
# File 'lib/cnvrg/datafiles.rb', line 291

def upload_log_file(absolute_path, relative_path, log_date)
  file_name = File.basename relative_path
  file_size = File.size(absolute_path).to_f
  content_type = "text/x-log"
  upload_resp = Cnvrg::API.request("/users/#{@owner}/" + "upload_cli_log", 'POST_FILE', {absolute_path: absolute_path, relative_path: relative_path,
                                                                                         file_name: file_name, log_date: log_date,
                                                                                         file_size: file_size, file_content_type: content_type})
  if Cnvrg::CLI.is_response_success(upload_resp, false)
    path = upload_resp["result"]["path"]
    s3_res = upload_small_files_s3(path, absolute_path, "text/plain")
  end
  if s3_res
    return true
  end
  return false

end

#upload_multiple_files(commit_sha1, tree, threads: ParallelThreads, force: false, new_branch: false, prefix: '', partial_commit: nil, total: nil) ⇒ Object 



117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
 
# File 'lib/cnvrg/datafiles.rb', line 117

def upload_multiple_files(commit_sha1, tree, threads: ParallelThreads, force: false, new_branch: false, prefix: '', partial_commit: nil, total: nil)
  begin
    Cnvrg::Logger.log_info("Sending Upload Files request")
    error = nil
    upload_resp = nil
    10.times do
      upload_resp = Cnvrg::API.request(@base_resource + "upload_files", 'POST_JSON', {commit_sha1: commit_sha1, tree: tree, force: force, is_branch: new_branch, partial_commit: partial_commit})
      if Cnvrg::CLI.is_response_success(upload_resp, false)
        error = nil
        break
      end
      error = upload_resp
      Cnvrg::Logger.log_method(bind: binding)
      Cnvrg::Logger.log_info("Got an error message from server, #{upload_resp.try(:fetch, "message")}, trying again")
    end
    raise Exception.new("Can't upload data files: #{error["message"]}") if error.present?

    Cnvrg::Logger.log_info("Uploading files")
    results = upload_resp['result'].with_indifferent_access

    if results['files'].blank?
      return 0, []
    end

    if @temp_upload_progressbar.blank?
      @temp_upload_progressbar = ProgressBar.create(:title => "Upload Progress",
                                                    :progress_mark => '=',
                                                    :format => "%b>>%i| %p%% %t",
                                                    :starting_at => 0,
                                                    :total => total,
                                                    :autofinish => true)
    end

    files = results['files']
    upload_error_files = []
    @temp_upload_progressbar.progress += tree.keys.length - files.length if @temp_upload_progressbar.present?
    Parallel.map((files.keys), in_threads: threads) do |k|
      o = tree[k].merge(files[k])
      success = upload_single_file(o)
      if not success
        upload_error_files << {absolute_path: o[:absolute_path]}
        files.except!(k)
        tree.except!(k)
        Cnvrg::Logger.log_error_message("Error while upload single file #{o["path"]}")
      end
      @temp_upload_progressbar.progress += 1 if @temp_upload_progressbar.present?
    end
    blob_ids = files.values.map {|f| f['bv_id']}
    if blob_ids.present?
      dirs = tree.keys.select {|k| tree[k].nil?} || []
      Cnvrg::Logger.info("Sending Upload files save")
      upload_resp = Cnvrg::API.request(@base_resource + "upload_files_save", "POST", {commit: commit_sha1, blob_ids: blob_ids, dirs: dirs})
      unless Cnvrg::CLI.is_response_success(upload_resp, false)
        Cnvrg::Logger.log_method(bind: binding)
        raise Exception.new("Got an error message from server, #{upload_resp.try(:fetch, "message")}")
      end
    end
    Cnvrg::Logger.log_info("Upload Success")
    return files.try(:keys).try(:length), upload_error_files
  rescue => e
    Cnvrg::Logger.log_method(bind: binding)
    Cnvrg::Logger.log_error(e)
    raise e
  end
end

#upload_single_file(file) ⇒ Object 



183
184
185
186
187
188
189
190
191
192
193
194
 
# File 'lib/cnvrg/datafiles.rb', line 183

def upload_single_file(file)
  begin
    file = file.as_json
    Cnvrg::Logger.log_info("Uploading #{file["absolute_path"]}")
    @downloader.safe_upload(file["path"], file["absolute_path"])
    Cnvrg::Logger.log_info("#{file["absolute_path"]} uploaded.")
  rescue => e
    Cnvrg::Logger.log_error_message("Error while upload single file #{file["path"]}")
    Cnvrg::Logger.log_error(e)
    return false
  end
end

#upload_small_files_s3(url_path, file_path, content_type) ⇒ Object 



539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
 
# File 'lib/cnvrg/datafiles.rb', line 539

def upload_small_files_s3(url_path, file_path, content_type)
  url = URI.parse(url_path)
  file = File.open(file_path, "rb")
  body = file.read
  begin
    Net::HTTP.start(url.host) do |http|
      http.send_request("PUT", url.request_uri, body, {
          "content-type" => content_type,
      })
    end
    return true
  rescue Interrupt
    return false
  rescue
    return false
  end
end

#upload_tar_file(absolute_path, relative_path, commit_sha1) ⇒ Object 



248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
 
# File 'lib/cnvrg/datafiles.rb', line 248

def upload_tar_file(absolute_path, relative_path, commit_sha1)
  begin
    file_name = File.basename relative_path
    file_size = File.size(absolute_path).to_f
    mime_type = MimeMagic.by_path(absolute_path)
    content_type = !(mime_type.nil? or mime_type.text?) ? mime_type.type : "text/plain"
    begin
      chunked_bytes = [100, (file_size*0.01)].min
      total_yanked = ""
      open(absolute_path, "rb") do |f|
        total_yanked = f.read(chunked_bytes)
      end
      if !total_yanked.empty?
        sha1 =  OpenSSL::Digest::SHA1.hexdigest(total_yanked)
      else
        sha1 =  OpenSSL::Digest::SHA1.file(absolute_path).hexdigest
      end
    rescue
      sha1 =  OpenSSL::Digest::SHA1.file(absolute_path).hexdigest
    end

    upload_resp = Cnvrg::API.request(@base_resource + "upload_tar_file", 'POST_FILE', {absolute_path: absolute_path, relative_path: relative_path,
                                                                                       commit_sha1: commit_sha1, file_name: file_name,
                                                                                       file_size: file_size, file_content_type: content_type, sha1: sha1,
                                                                                        new_version:true})
    if Cnvrg::CLI.is_response_success(upload_resp, false)
      path = upload_resp["result"]["path"]
      s3_res = upload_large_files_s3(upload_resp, absolute_path)
      if s3_res
        # Cnvrg::API.request(@base_resource + "update_s3", 'POST', {path: path, commit_id: upload_resp["result"]["commit_id"],
        #                                                           blob_id: upload_resp["result"]["id"]})
        return true
      end
    else
      return false
    end
  rescue => e
    puts e.message
    return false
  end
end

#upload_url(file_path) ⇒ Object 



557
558
559
560
561
562
563
564
565
 
# File 'lib/cnvrg/datafiles.rb', line 557

def upload_url(file_path)
  response = Cnvrg::API.request(@base_resource + "upload_url", 'POST', {file_s3_path: file_path})
  if Cnvrg::CLI.is_response_success(response, false)
    return response
  else
    return nil
  end

end

#verify_files_exists(files) ⇒ Object 



33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
 
# File 'lib/cnvrg/datafiles.rb', line 33

def verify_files_exists(files)
  paths = []
  files.each do |file|
    if File.exists? file
      if File.directory? file
        paths += Dir.glob("#{file}/**/*")
      else
        paths << file
      end
      next
    end
    raise SignalException.new(1, "Cant find file #{file}") unless File.exists? "#{Dir.pwd}/#{file}"
  end
  paths.map{|p| p.gsub(/^\.\//, '')}
end