;;;; *************************************************************************
;;;; FILE IDENTIFICATION
;;;;
-;;;; Name: parse-2002.lisp
+;;;; Name: parse-rrf.lisp
;;;; Purpose: Parsing and SQL insertion routines for UMLisp which may
;;;; change from year to year
;;;; Author: Kevin M. Rosenberg
(clrhash cuisui-lrl-hash)
(clrhash sab-srl-hash))
(setf
- pfstr-hash (make-hash-table :size 800000)
- cui-lrl-hash (make-hash-table :size 800000)
- lui-lrl-hash (make-hash-table :size 1500000)
- sui-lrl-hash (make-hash-table :size 1500000)
- cuisui-lrl-hash (make-hash-table :size 1800000)
+ pfstr-hash (make-hash-table :size 1300000)
+ cui-lrl-hash (make-hash-table :size 1300000)
+ lui-lrl-hash (make-hash-table :size 4600000)
+ sui-lrl-hash (make-hash-table :size 5100000)
+ cuisui-lrl-hash (make-hash-table :size 2000000)
sab-srl-hash (make-hash-table :size 100 :test 'equal))))
(defun ensure-preparse (&optional (force-read nil))
(when (and *preparse-hash-init?* (not force-read))
(return-from ensure-preparse 'already-done))
(make-preparse-hash-table)
- (with-umls-file (line "MRCONSO.RRF")
- (let ((cui (parse-ui (nth 0 line)))
- (lui (parse-ui (nth 3 line)))
- (sui (parse-ui (nth 5 line)))
- (sab (nth 11 line))
- (srl (parse-integer (nth 15 line))))
- (unless (gethash cui pfstr-hash) ;; if haven't stored pfstr for cui
- (if (and (string-equal (nth 1 line) "ENG") ; LAT
- (string-equal (nth 2 line) "P") ; ts
- (string-equal (nth 4 line) "PF")) ; stt
+ (let ((counter 0))
+ (declare (fixnum counter)
+ (ignorable counter))
+ (with-umls-file (line "MRCONSO.RRF")
+ (let ((cui (parse-ui (nth 0 line)))
+ (lui (parse-ui (nth 3 line)))
+ (sui (parse-ui (nth 5 line)))
+ (sab (nth 11 line))
+ (srl (parse-integer (nth 15 line))))
+ #+sbcl
+ (when (= 0 (mod (incf counter) 100000)) (sb-ext:gc :full t))
+
+ ;; pfstr deprecated by KPKENG field in MRCONSO
+ #+nil
+ (unless (gethash cui pfstr-hash) ;; if haven't stored pfstr for cui
+ (when (and (string-equal (nth 1 line) "ENG") ; LAT
+ (string-equal (nth 2 line) "P") ; ts
+ (string-equal (nth 4 line) "PF")) ; stt
(setf (gethash cui pfstr-hash) (nth 14 line))))
- (set-lrl-hash cui srl cui-lrl-hash)
- (set-lrl-hash lui srl lui-lrl-hash)
- (set-lrl-hash sui srl sui-lrl-hash)
- (set-lrl-hash (make-cuisui cui sui) srl cuisui-lrl-hash)
- (multiple-value-bind (val found) (gethash sab sab-srl-hash)
- (declare (ignore val))
- (unless found
- (setf (gethash sab sab-srl-hash) srl)))))
+ (set-lrl-hash cui srl cui-lrl-hash)
+ (set-lrl-hash lui srl lui-lrl-hash)
+ (set-lrl-hash sui srl sui-lrl-hash)
+ (set-lrl-hash (make-cuisui cui sui) srl cuisui-lrl-hash)
+ (multiple-value-bind (val found) (gethash sab sab-srl-hash)
+ (declare (ignore val))
+ (unless found
+ (setf (gethash sab sab-srl-hash) srl))))))
(setq *preparse-hash-init?* t)
t)
- (defun pfstr-hash (cui) (gethash cui pfstr-hash))
+ #+nil (defun pfstr-hash (cui) (gethash cui pfstr-hash))
(defun cui-lrl (cui) (gethash cui cui-lrl-hash))
(defun lui-lrl (lui) (gethash lui lui-lrl-hash))
(defun sui-lrl (sui) (gethash sui sui-lrl-hash))
(defparameter +col-datatypes+
'(("AV" sql-f) ("BTS" sql-i) ("CLS" sql-i) ("COF" sql-i) ("CUI1" sql-u)
("AUI" sql-u) ("AUI1" sql-u) ("AUI2" sql-u) ("PCUI" sql-u)
- ("PLUI" sql-u) ("PAUI" sql-u)
+ ("PLUI" sql-u) ("PAUI" sql-u) ("RUI" sql-u)
("CUI2" sql-u) ("CUI" sql-u) ("CXN" sql-s) ("FR" sql-i) ("LRL" sql-s)
("LUI" sql-u) ("MAX" sql-s) ("MIN" sql-s) ("RANK" sql-s) ("REF" sql-c)
+ ("PTR" sql-c)
("RNK" sql-s) ("RWS" sql-i) ("SRL" sql-s) ("SUI" sql-u) ("TUI" sql-u)
("MAPRANK" sql-s)
;;; Custom columns
(lambda (x) (write-to-string (sab-srl (nth 10 x)))))
("MRRANK.RRF" "KSRL" "SMALLINT" 0
(lambda (x) (write-to-string (sab-srl (nth 1 x)))))
+ ("MRHIER.RRF" "KSRL" "SMALLINT" 0
+ (lambda (x) (write-to-string (sab-srl (nth 4 x)))))
+ ("MRMAP.RRF" "KSRL" "SMALLINT" 0
+ (lambda (x) (write-to-string (sab-srl (nth 1 x)))))
+ ("MRSMAP.RRF" "KSRL" "SMALLINT" 0
+ (lambda (x) (write-to-string (sab-srl (nth 1 x)))))
("MRDEF.RRF" "KSRL" "SMALLINT" 0
(lambda (x) (write-to-string (sab-srl (nth 4 x)))))
#+nil ("MRCXT.RRF" "KSRL" "SMALLINT" 0 (lambda (x) (write-to-string (sab-srl (nth 2 x)))))
(parse-ui (nth 2 x))
(parse-ui (nth 4 x)))))))
- ;; FIXME: For MRREF and MRCOC, add lookups to KPFSTR2 using new MRCONSO index KPFENG
-
#+nil ("MRREL.RRF" "KPFSTR2" "TEXT" 1024 (lambda (x) (pfstr-hash (parse-ui (nth 4 x)))))
#+nil ("MRCOC.RRF" "KPFSTR2" "TEXT" 1024 (lambda (x) (pfstr-hash (parse-ui (nth 2 x)))))
- #+use-mrctx ("MRCXT.RRF" "KCUISUI" "BIGINT" 0 (lambda (x) (write-to-string (make-cuisui (parse-ui (nth 0 x)) (parse-ui (nth 1 x))))))
("MRSAT.RRF" "KCUILUI" "BIGINT" 0
(lambda (x) (write-to-string (make-cuilui (parse-ui (nth 0 x)) (parse-ui (nth 1 x))))))
("MRSAT.RRF" "KCUISUI" "BIGINT" 0
(defparameter +index-cols+
'(("CUI1" "MRCOC") ("CUI" "MRCONSO") ("LUI" "MRCONSO")
("SRL" "MRCONSO") ("AUI" "MRCONSO") ("KPFENG" "MRCONSO")
- ("SUI" "MRCONSO") ("CUI" "MRDEF")
- ("CUI1" "MRREL") ("CUI" "MRSAT") ("LUI" "MRSAT") ("SUI" "MRSAT")
- ("CUI" "MRSTY")
- ("TUI" "MRSTY") ("CUI" "MRXNS_ENG")
- ;; ("CUI" "MRCXT") ("KCUISUI" "MRCXT") ("KSRL" "MRCXT")
- ("AUI" "MRHIER") ("PTR" "MRHIER") ("CUI" "MRHIER") ("CXN" "MRHIER") ("RELA" "MRHIER") ("PAUI" "MRHIER")
+ ("SUI" "MRCONSO") ("SAUI" "MRCONSO") ("CODE" "MRCONSO")
+ ("SCUI" "MRCONSO")
+ ("CUI" "MRDEF")
+ ("CUI1" "MRREL") ("CUI2" "MRREL") ("SAB" "MRREL")
+ ("RUI" "MRREL") ("AUI1" "MRREL") ("AUI2" "MRREL")
+ ("CUI" "MRSAT") ("LUI" "MRSAT") ("SUI" "MRSAT")
+ ("METAUI" "MRSAT") ("ATN" "MRSAT")
+ ("CUI" "MRSTY") ("TUI" "MRSTY") ("CUI" "MRXNS_ENG")
+ ("AUI" "MRHIER") ("CUI" "MRHIER") ("CXN" "MRHIER") ("RELA" "MRHIER") ("PAUI" "MRHIER")
("SAB" "MRHIER")
#+ignore ("NSTR" "MRXNS_ENG" 10)
("CUI" "MRXNW_ENG") ("NWD" "MRXNW_ENG") ("WD" "MRXW_ENG")