Class: Dratools::RunRecordCollector

Inherits:
Object
  • Object
show all
Defined in:
lib/dratools/run_record_collector.rb

Overview

BioProject などの上位レコードから DDBJ sra-run レコードを集める。

Constant Summary collapse

XREF_URL_PATTERN =
%r{/(?:resource|search/entry)/([^/]+)/([^/?#.]+)}
TRAVERSABLE_XREF_TYPES =
[
  DdbjRecordFields::SRA_RUN_RESOURCE_TYPE,
  DdbjRecordFields::SRA_EXPERIMENT_RESOURCE_TYPE,
  DdbjRecordFields::SRA_SAMPLE_RESOURCE_TYPE,
  DdbjRecordFields::SRA_STUDY_RESOURCE_TYPE,
  DdbjRecordFields::SRA_SUBMISSION_RESOURCE_TYPE,
  DdbjRecordFields::BIOPROJECT_RESOURCE_TYPE,
  DdbjRecordFields::BIOSAMPLE_RESOURCE_TYPE
].freeze

Instance Method Summary collapse

Constructor Details

#initialize(client:) ⇒ RunRecordCollector

Returns a new instance of RunRecordCollector.



23
24
25
# File 'lib/dratools/run_record_collector.rb', line 23

def initialize(client:)
  @client = client
end

Instance Method Details

#collect_run_records(ddbj_record, seen_keys = Set.new) ⇒ Object



27
28
29
# File 'lib/dratools/run_record_collector.rb', line 27

def collect_run_records(ddbj_record, seen_keys = Set.new)
  explore(ddbj_record, seen_keys: seen_keys).run_records
end

#explore(ddbj_record, seen_keys: Set.new, relation: TraversalNode::ROOT_RELATION, tolerant: false, direct_run_fetch_limit: nil) ⇒ Object



31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
# File 'lib/dratools/run_record_collector.rb', line 31

def explore(ddbj_record, seen_keys: Set.new, relation: TraversalNode::ROOT_RELATION,
            tolerant: false, direct_run_fetch_limit: nil)
  node = node_from_record(ddbj_record, relation: relation)
  return node if run_record?(ddbj_record)

  xrefs = ddbj_record.fetch(DdbjRecordFields::DB_XREFS_KEY, [])
  run_xrefs = xrefs.select { |xref| sra_run_xref?(xref) }
  if (lightweight_children = lightweight_direct_run_nodes(run_xrefs, direct_run_fetch_limit))
    node.children.concat(lightweight_children)
    return node
  end

  direct_children = explore_edges(run_xrefs, TraversalNode::DB_XREF_RELATION, seen_keys,
                                  tolerant: tolerant,
                                  direct_run_fetch_limit: direct_run_fetch_limit)
  if direct_children.any? { |child| child.run? || child.run_records.any? }
    node.children.concat(direct_children)
    return node
  end

  recursive_xrefs = xrefs.select { |xref| traversable_xref?(xref) }
  validate_recursive_non_run_xref_count!(ddbj_record, recursive_xrefs)
  db_xref_edges = explore_edges(
    recursive_xrefs,
    TraversalNode::DB_XREF_RELATION,
    seen_keys,
    tolerant: tolerant,
    direct_run_fetch_limit: direct_run_fetch_limit
  )
  child_edges = explore_edges(
    child_bioprojects(ddbj_record),
    TraversalNode::CHILD_BIOPROJECT_RELATION,
    seen_keys,
    tolerant: tolerant,
    direct_run_fetch_limit: direct_run_fetch_limit
  )
  node.children.concat(db_xref_edges + child_edges)
  node
end