(define-module (gn data species)
  #:use-module (json)
  #:use-module (ice-9 match)
  #:use-module (ice-9 format)
  #:use-module (ice-9 iconv)
  #:use-module (ice-9 receive)
  #:use-module (ice-9 string-fun)
  #:use-module (gn db sparql)
  #:use-module (gn data group)
  #:use-module (web gn-uri)

  #:export (
            gnid-species
            get-species
            get-species-meta
            get-species-data
            get-species-shortnames
            get-species-binominal-names
            get-expanded-species
            get-expanded-taxon-meta
            get-expanded-taxon-data
            ))

(define (gnid-species short-name)
  "Find the GN identifier from shortname, e.g. Mus_musculus"
  (let ([rec (get-expanded-taxon-data short-name)])
    (url-parse-id (assoc-ref rec "gnid"))
  ))

(define (get-species)
  (receive (names res) (memo-sparql-species-meta)
    (let* ([table (get-rows names res)]
           [recs '()]
           [h (compile-species recs table)])
      (species-digest h))
    ))

;; result should be a vector of list of pair
(define (species-digest recs)
  (map (lambda (r)
	 (let* ([k (car r)]
		[v (cdr r)])
	   ; with key use (cons k (map (lambda (i) (cons (car i) (car (cdr i)))) v))
	   (map (lambda (i) (cons (url-parse-id (car i)) (car (cdr i)))) v)
	   ))
	 recs  )
  )

(define (expand-species rec)
  (let ([wd-id (url-parse-id (assoc-ref rec "22-rdf-syntax-ns#isDefinedBy"))]
	[short-name (normalize-id (assoc-ref rec "shortName"))])
    (if (string=? wd-id "unknown")
	rec
                                        ; wikidata query:
	(receive (names row) (tsv->scm (memo-sparql-wd-species-info wd-id))
	  (match (pk (car row))
	    ((taxonomy-name ncbi descr)
	     (let ([ncbi-id (strip-lang ncbi)]
		   [taxonomy-lnk (string-replace-substring (strip-lang taxonomy-name) " " "_")])
	        (cons `("id" . ,short-name)
		(cons `("wikidata" . ,wd-id)
		(cons `("taxonomy-id" . ,ncbi-id)
		(cons `("ncbi-url" . ,(string-append "https://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?mode=Info&id=" ncbi-id))
		(cons `("uniprot-url" . ,(string-append "https://www.uniprot.org/taxonomy/" ncbi-id))
		(cons `("wikidata-url" . ,(string-append "http://www.wikidata.org/entity/" wd-id))
		(cons `("wikispecies-url" . ,(string-append "https://species.wikimedia.org/wiki/" taxonomy-lnk))
		(cons `("taxonomy-name" . ,(strip-lang taxonomy-name))
		(cons `("meta" . ,(mk-meta short-name))
		(cons `("description" . ,(strip-lang descr))
		      rec))))))))))))
		)
	   )))
  )


(define (get-expanded-taxon-data short-name)
  "Here we add information related to one taxonomy species"
  (call/cc (lambda (return) ; use call/cc to be able to return early
             (for-each (lambda (rec)
                         (if (string=? (assoc-ref rec "shortName") short-name)
                             (return (expand-species rec))))
                       (get-species))
             (return #f)
)))


(define (get-expanded-species)
  "Here we add information related to each species"
  (map (lambda (rec)
         (expand-species rec)
         ) (get-species)))

(define (get-species-api-str)
  (scm->json-string #("https://genenetwork.org/api/v2/mouse/"
                      "https://genenetwork.org/api/v2/rat/")))

(define (get-species-shortnames recs)
  (map (lambda r (assoc-ref (car r) "shortName")) recs))

(define (get-species-binominal-names recs)
  (map (lambda r (url-parse-id (car (car r)))) recs))

(define (get-species-meta2 recs)
  "Return a list of short names and expand them to URIs"
  (map (lambda r
	 (let ([shortname (assoc-ref (car r) "shortName")])
	   (cons shortname (mk-meta shortname)))) recs)
  )

(define (get-species-links recs)
  "Return a list of short names and expand them to URIs"
  (map (lambda r
	 (let ([shortname (assoc-ref (car r) "shortName")])
	   (cons shortname (mk-data shortname)))) recs)
  )

(define (get-species-data)
  (list->vector (get-expanded-species)))

(define (get-species-meta)
  (let ([recs (get-expanded-species)])
    `(("info" . "Get information on species by visiting the data link or one of the individual links")
      ("doc" . ,(mk-doc "species"))
      ("meta" . ,(mk-meta "species"))
      ("data" . ,(mk-data "species"))
      ("up" . ,(string-append (prefix) "/"))
      ("meta-links" . ,(get-species-meta2 recs))
      ("links" . ,(get-species-links recs)))))


(define (get-expanded-taxon-meta id)
  "Get information on a specific species, e.g. mouse"
  `(("info" . ,id)
    ("doc" . ,(mk-doc id))
    ("meta" . ,(mk-meta id))
    ("data" . ,(mk-data id))
    ("up" . ,(mk-meta "species"))
    ("meta-links" . ,(list->vector (get-group-links (gnid-species id)
                                                    (lambda (r) (mk-meta (url-parse-id r))))))
    ("links" . ,(list->vector (get-group-links (gnid-species id)
                                               (lambda (r) (mk-data (url-parse-id r))))))
  ))