1 ;;;; -*- Mode: Lisp; Syntax: ANSI-Common-Lisp; Base: 10 -*-
2 ;;;; *************************************************************************
3 ;;;; FILE IDENTIFICATION
6 ;;;; Purpose: Library to access PubMed web application
7 ;;;; Programmer: Kevin M. Rosenberg
8 ;;;; Date Started: Jun 2001
10 ;;;; $Id: pubmed.lisp,v 1.3 2002/10/31 19:42:26 kevin Exp $
12 ;;;; This file, part of cl-pubmed, is Copyright (c) 2002 by Kevin M. Rosenberg
14 ;;;; cl-pubmed users are granted the rights to distribute and use this software
15 ;;;; as governed by the terms of the GNU Lesser General Public License
16 ;;;; (http://www.gnu.org/licenses/lgpl.html)
17 ;;;; *************************************************************************
19 (declaim (optimize (debug 3) (speed 3) (safety 1) (compilation-speed 0)))
21 (in-package #:cl-user)
24 (:use #:common-lisp #:kmrcl)
43 ;; pm-article-set accessors
61 #:article-mesh-headings
67 (defparameter +pubmed-host+ "www.ncbi.nlm.nih.gov")
68 (defparameter +pubmed-query-url+ "/entrez/utils/pmqty.fcgi")
69 (defparameter +pubmed-fetch-url+ "/entrez/utils/pmfetch.fcgi")
72 (define-condition pubmed-condition ()
76 (define-condition pubmed-server-error (error pubmed-condition)
77 ((response :initarg :response
79 :reader pubmed-server-error-response))
80 (:report (lambda (c stream)
81 (format stream "A PubMed server error occurred.")
82 (awhen (pubmed-server-error-response c)
83 (format stream " The server response was:~&~S" it)))))
85 ;;; Article-Set and Article Classes
87 (defclass pm-article-set ()
88 ((query :type string :initarg :query :accessor articles-query)
89 (articles :type list :initarg :articles :accessor articles)
90 (total :type fixnum :initarg :total :accessor articles-total)
91 (count :type fixnum :initarg :count :accessor articles-count)
92 (start :type fixnum :initarg :start :accessor articles-start))
93 (:documentation "Pubmed Article Set Class")
94 (:default-initargs :total 0 :start 0 :count 0
95 :query nil :articles nil))
97 (defclass pm-article ()
99 (pmid :type integer :accessor article-pmid)
100 (title :type string :accessor article-title)
101 (authors :type list :accessor article-authors)
102 (affiliation :type string :accessor article-affiliation)
103 (journal :type string :accessor article-journal)
104 (date :type string :accessor article-date)
105 (volume :type string :accessor article-volume)
106 (issue :type string :accessor article-issue)
107 (pages :type string :accessor article-pages)
108 (abstract :type string :accessor article-abstract)
109 (mesh-headings :type list :accessor article-mesh-headings))
110 (:documentation "Pubmed Article Class"))
112 (defmethod print-object ((obj pm-article-set) (s stream))
113 (print-unreadable-object (obj s :type t :identity t)
114 (format s "~d total articles, ~d articles starting at #~d"
120 (defmethod print-object ((obj pm-article) (s stream))
121 (print-unreadable-object (obj s :type t :identity t)
122 (format s "pmid: ~d" (article-pmid obj))))
124 (defun article-equal-p (a b)
125 (check-type a pm-article)
126 (check-type b pm-article)
127 (eql (article-pmid a) (article-pmid b)))
129 (defun article-ref (art)
130 "Return a string of publication data for an article"
132 (awhen (article-date art)
133 (string-append ref (format nil "~a; " it)))
134 (awhen (article-volume art)
135 (string-append ref it))
136 (awhen (article-issue art)
137 (string-append ref (format nil "(~a)" it)))
138 (awhen (article-pages art)
139 (string-append ref (format nil ":~a" it)))
142 (defmethod print-article-set ((artset pm-article-set) &key (os *standard-output*) (format :text)
143 (complete nil) (print-link t))
144 "Display an article set to specified stream in specified format"
145 (dotimes (i (articles-count artset))
146 (print-article (nth i (articles artset)) :os os :format format
147 :complete complete :print-link print-link)))
149 (defmethod print-article ((art pm-article) &key (os *standard-output*) (format :text)
150 (complete nil) (print-link t))
152 (if (eql format :text)
153 (format os "~a~%~a~%~a~a ~a~%~a~%"
155 (list-to-delimited-string (article-authors art) ", ")
156 (aif (article-affiliation art)
157 (format nil "~a~%" it) "")
158 (article-journal art) (article-ref art)
159 (aif (article-abstract art)
162 "Abstract available")
163 "No abstract available")
165 (format os "~a~%" (article-mesh-headings art))))
167 (let ((has-link (or (article-abstract art) (article-mesh-headings art))))
168 (when (and print-link has-link)
169 (format os "<a href=\"~a?key=~a\">" (make-url "print-article") (article-pmid art)))
170 (format os "<div class=\"article-title\">~a</div>~%" (article-title art))
171 (when (and print-link has-link)
173 (format os "<div class=\"article-authors\">~a</div>~%"
174 (list-to-delimited-string (article-authors art) ", "))
175 (format os "<div class=\"article-reference\">~a ~a</div>~%"
176 (article-journal art) (article-ref art))
177 (when (and complete (article-abstract art))
178 (format os "<div class=\"article-abstract\">~a</div>~%"
179 (article-abstract art)))
180 (when (and complete (article-mesh-headings art))
181 (format os "<div class=\"mesh-heading-title\">Mesh Headings:</div>")
182 (dolist (mh (article-mesh-headings art))
183 (format os "<div class=\"mesh-heading\">~a</div>~%" mh)))
184 (format os "<p/>~%"))))
187 ;;; PubMed Query Functions
189 (defun pm-query (searchstr &key maximum start)
190 "Performs PubMed query and fetch and returns article-set structure"
193 (pubmed-search-xml searchstr :maximum maximum :start start)
194 (when (xml-tag-contents "Count" status)
195 (let ((as (make-instance 'pm-article-set)))
197 (articles-total as) (parse-integer (xml-tag-contents "Count" status))
198 (articles-query as) searchstr
199 (articles-start as) (parse-integer (xml-tag-contents "DispStart" status))
200 (articles-count as) (parse-integer (xml-tag-contents "DispMax" status))
201 (articles as) (extract-article-set results))
204 (defun pm-fetch-ids (pmids)
205 "Fetchs list of Pubmed ID's and returns pm-article-set class"
206 (setq pmids (mklist pmids))
207 (let ((results (pubmed-fetch-pmids-xml pmids)))
208 (unless (xml-tag-contents "Error" results)
209 (let ((as (make-instance 'pm-article-set)))
211 (articles-total as) (length pmids)
212 (articles-query as) (list-to-delimited-string pmids #\,)
213 (articles-start as) 0
214 (articles-count as) (length pmids)
215 (articles as) (extract-article-set results))
219 (defun pubmed-search-tree (searchstr &key maximum start)
220 "Performs a pubmed search and returns two values:
221 tree of PubMed search results and tree of PubMed search status"
223 (xml-search-results xml-search-status)
224 (pubmed-search-xml searchstr :maximum maximum :start start)
225 (if xml-search-results
226 (values (parse-xml-no-ws xml-search-results)
227 (parse-xml-no-ws xml-search-status))
228 (values nil (parse-xml-no-ws xml-search-status)))))
230 (defun pubmed-search-xml (searchstr &key maximum start)
231 "Performs a Pubmed search and returns two values:
232 XML string of PubMed search results and XML search status"
234 (pmids search-status)
235 (pubmed-query-xml searchstr :maximum maximum :start start)
236 (values (pubmed-fetch-pmids-xml pmids) search-status)))
238 (defun pubmed-query-xml (searchstr &key maximum start)
239 "Performs a Pubmed search and returns two values:
240 list of PubMed ID's that match search string and XML search status"
241 (let ((search-results (pubmed-query-status searchstr :maximum maximum :start start)))
242 (values (extract-pmid-list search-results) search-results)))
244 (defun pubmed-query-status (searchstr &key start maximum)
245 "Performs a Pubmed search and returns XML results of PubMed search
246 which contains PubMed ID's and status results"
247 (let ((query-alist `(("db" . "m") ("term" . ,searchstr) ("mode" . "xml"))))
248 (when maximum (push (cons "dispmax" maximum) query-alist))
249 (when start (push (cons "dispstart" start) query-alist))
250 (net.aserve.client:do-http-request
251 (format nil "http://~a~a" +pubmed-host+ +pubmed-query-url+)
253 :query query-alist)))
255 (defun pubmed-fetch-pmids-xml (pmids)
256 "Fetch articles for a list of PubMed ID's and return XML string"
257 (setq pmids (mklist pmids)) ;; Ensure list
259 (net.aserve.client:do-http-request
260 (format nil "http://~a~a" +pubmed-host+ +pubmed-fetch-url+)
263 `(("db" . "PubMed") ("report" . "xml") ("mode" . "text")
264 ("id" . ,(list-to-delimited-string pmids #\,))))))
266 ;;; XML Extraction Routines
268 (defun extract-article-set (results)
269 "Extract article set from PubMed XML string, return results in pm-article-set class"
270 (multiple-value-bind (as-start as-end as-next)
271 (positions-xml-tag-contents "PubmedArticleSet" results)
272 (declare (ignore as-end as-next))
279 (a-start a-end a-next)
280 (positions-xml-tag-contents "PubmedArticle" results pos)
283 (push (extract-article results a-start a-end) articles)
287 (nreverse articles)))))
289 (defun extract-article (xmlstr a-start a-end)
290 "Extract article contents from PubMed XML string and return results in pm-article class"
291 (let ((article (make-instance 'pm-article)))
293 (article-pmid article) (parse-integer (xml-tag-contents "PMID" xmlstr a-start))
294 (article-title article) (xml-tag-contents "ArticleTitle" xmlstr a-start)
295 (article-journal article) (xml-tag-contents "MedlineTA" xmlstr a-start)
296 (article-pages article) (xml-tag-contents "MedlinePgn" xmlstr a-start)
297 (article-affiliation article) (xml-tag-contents "Affiliation" xmlstr a-start)
298 (article-abstract article) (xml-tag-contents "AbstractText" xmlstr a-start))
299 (multiple-value-bind (ji-start ji-end ji-next)
300 (positions-xml-tag-contents "JournalIssue" xmlstr a-start a-end)
301 (declare (ignore ji-next))
303 (article-volume article) (xml-tag-contents "Volume" xmlstr ji-start ji-end)
304 (article-issue article) (xml-tag-contents "Issue" xmlstr ji-start ji-end))
305 (aif (xml-tag-contents "MedlineDate" xmlstr ji-start ji-end)
306 (setf (article-date article) it)
307 (setf (article-date article)
308 (concatenate 'string (xml-tag-contents "Year" xmlstr ji-start ji-end)
309 (aif (xml-tag-contents "Month" xmlstr ji-start ji-end)
310 (format nil " ~a" it)
313 (multiple-value-bind (al-start al-end al-next)
314 (positions-xml-tag-contents "AuthorList" xmlstr a-start a-end)
315 (declare (ignore al-next))
316 (setf (article-authors article)
323 (au-start au-end au-next)
324 (positions-xml-tag-contents "Author" xmlstr pos al-end)
327 (push (extract-author xmlstr au-start au-end) authors)
330 (nreverse authors)))))
332 (multiple-value-bind (mhl-start mhl-end mhl-next)
333 (positions-xml-tag-contents "MeshHeadingList" xmlstr a-start a-end)
334 (declare (ignore mhl-next))
335 (setf (article-mesh-headings article)
342 (mh-start mh-end mh-next)
343 (positions-xml-tag-contents "MeshHeading" xmlstr pos mhl-end)
346 (push (extract-mesh-heading xmlstr mh-start mh-end) mesh-headings)
350 (nreverse mesh-headings)))))
354 (defun extract-author (xmlstr start end)
355 "Extract author name from XML string"
356 (let ((last-name (xml-tag-contents "LastName" xmlstr start end))
357 (initials (xml-tag-contents "Initials" xmlstr start end)))
358 (concatenate 'string last-name " " initials)))
360 (defun extract-mesh-heading (xmlstr start end)
361 "Extract and format mesh headings from XML string"
362 (let ((desc (xml-tag-contents "DescriptorName" xmlstr start end))
363 (sh (xml-tag-contents "SubHeading" xmlstr start end)))
365 (format nil "~a(~a)" desc sh)
368 (defun extract-pmid-list (results)
369 "Returns list of PubMed ID's from XML result string"
370 (if (or (search "<ERROR>" results)
371 (search "<H1>Server Error</H1>" results))
372 (error 'pubmed-server-error :response results)
373 (awhen (xml-tag-contents "Id" results)
374 (delimited-string-to-list it #\space))))