#! /usr/bin/env guile
!#

(use-modules (rnrs programs)
             (rnrs io ports)
             (srfi srfi-1)
             (srfi srfi-26)
             (ice-9 getopt-long)
             (ice-9 match)
             (ice-9 regex)
             (transform strings)
             (transform sql)
             (transform triples)
             (transform special-forms))


(define-transformer gn:set->gn:dataset
  (tables (Species
           (inner-join InbredSet "ON InbredSet.SpeciesId = Species.Id")
           (inner-join PublishFreeze "ON PublishFreeze.InbredSetId = InbredSet.Id"))
          "WHERE PublishFreeze.public > 0 AND Species.Name != 'monkey' GROUP BY Species.Name, PublishFreeze.ShortName")
  (triples (string->identifier "set" (field InbredSet Name InbredSetName) #:separator "_")
    (multiset gnt:has_phenotype_data
              (map (cut string->identifier "dataset" <> #:separator "_")
                   (string-split
                    (field ("GROUP_CONCAT(PublishFreeze.Name SEPARATOR ',')"
                            dataset_name))
                    #\,)))))

(define-transformer gn:dataset->gn:set
  (tables (Datasets
           (inner-join InfoFiles "ON InfoFiles.DatasetId = Datasets.DatasetId")
           (inner-join InbredSet "ON InbredSet.Id = InfoFiles.InbredSetId")
           (inner-join PublishFreeze "ON PublishFreeze.InbredSetId = InbredSet.Id"))
          "WHERE PublishFreeze.public > 0 GROUP BY Datasets.DatasetId")
  (triples (string->identifier "dataset" (field PublishFreeze Name) #:separator "_")
    (set gnt:has_strain (string->identifier "set" (field InbredSet Name InbredSetName) #:separator "_"))))

(define-transformer gn:dataset->metadata
  (tables (PublishXRef
           (inner-join InbredSet "ON InbredSet.InbredSetId = PublishXRef.InbredSetId")
           (inner-join Species "ON InbredSet.SpeciesId = Species.Id")
           (inner-join PublishFreeze "ON PublishFreeze.InbredSetId = InbredSet.Id")
           (inner-join Publication "ON Publication.Id = PublishXRef.PublicationId")
           (inner-join Phenotype "ON Phenotype.Id = PublishXRef.PhenotypeId"))
          "WHERE InbredSet.public > 0 GROUP BY Species.Name, PublishFreeze.Name")
  (triples (string->identifier "dataset" (field PublishFreeze Name) #:separator "_")
    (set dct:created (annotate-field (field PublishFreeze CreateTime) '^^xsd:datetime))
    (set gnt:has_strain (string->identifier "set" (field InbredSet Name InbredSetName) #:separator "_"))))

(define-transformer gn:dataset->gn:trait
  (tables (PublishXRef
           (inner-join InbredSet "ON InbredSet.InbredSetId = PublishXRef.InbredSetId")
           (inner-join Species "ON InbredSet.SpeciesId = Species.Id")
           (inner-join PublishFreeze "ON PublishFreeze.InbredSetId = InbredSet.Id")
           (inner-join Publication "ON Publication.Id = PublishXRef.PublicationId")
           (inner-join Phenotype "ON Phenotype.Id = PublishXRef.PhenotypeId"))
          "WHERE InbredSet.public > 0")
  (triples (string->identifier "dataset" (field PublishFreeze Name) #:separator "_")
    (set gnt:has_phenotype_trait
         (let ((post-abbrev (blank-p (field Phenotype Post_publication_abbreviation)))
               (pre-abbrev (blank-p (field Phenotype Pre_publication_abbreviation)))
               (post-desc (blank-p (field Phenotype Post_publication_description)))
               (pre-desc (blank-p (field Phenotype Post_publication_description))))
           (string->identifier
            "trait"
            (format #f "~a_~a" (field PublishFreeze Name)
                    (or post-abbrev pre-abbrev post-desc pre-desc))
            #:separator "_")))
    (set dct:created (annotate-field (field PublishFreeze CreateTime) '^^xsd:datetime))
    (set gnt:has_strain (string->identifier "set" (field InbredSet Name InbredSetName) #:separator "_"))))


(let* ((option-spec
        '((settings (single-char #\s) (value #t))
          (output (single-char #\o) (value #t))
          (documentation (single-char #\d) (value #t))))
       (options (getopt-long (command-line) option-spec))
       (settings (option-ref options 'settings #f))
       (output (option-ref options 'output #f))
       (documentation (option-ref options 'documentation #f))
       (%connection-settings
        (call-with-input-file settings
          read)))
  (with-documentation
   (name "Phenotype Datasets")
   (connection %connection-settings)
   (table-metadata? #f)
   (prefixes
    '(("dcat:" "<http://www.w3.org/ns/dcat#>")
      ("dct:" "<http://purl.org/dc/terms/>")
      ("gn:" "<http://rdf.genenetwork.org/v1/id/>")
      ("gnc:" "<http://rdf.genenetwork.org/v1/category/>")
      ("gnt:" "<http://rdf.genenetwork.org/v1/term/>")
      ("rdf:" "<http://www.w3.org/1999/02/22-rdf-syntax-ns#>")
      ("rdfs:" "<http://www.w3.org/2000/01/rdf-schema#>")
      ("owl:" "<http://www.w3.org/2002/07/owl#>")
      ("skos:" "<http://www.w3.org/2004/02/skos/core#>")
      ("xkos:" "<http://rdf-vocabulary.ddialliance.org/xkos#>")
      ("xsd:" "<http://www.w3.org/2001/XMLSchema#>")))
   (inputs
    (list
     gn:set->gn:dataset
     gn:dataset->gn:set
     gn:dataset->metadata
     gn:dataset->gn:trait))
   (outputs
    `(#:documentation ,documentation
      #:rdf ,output))))