ismrm-utils.el

;;; to download all abstracts: 
;;; parallel wget --no-clobber --no-parent -r "https://submissions.mirasmart.com/ISMRM2022/Itinerary/Files/PDFFiles/{}.html" ::: $(seq -f "%04g" 00 10000)
;;; Make a simple index so I can search through abstracts in emacs
;;; So that other people can use it, I have put this into an index.html and use a js search framework that mimics completing-read.
;;; See for example https://cheatcode.co/tutorials/how-to-implement-client-side-search-with-fuse-js
;;;
;; List and open any file on eww
(defvar ismrm-abstract-dir "/Users/dg/ismrm2022/PROGRAM/")

(defun ismrm--find-abstract-file ()
  (interactive)
  (let ((default-directory ismrm-abstract-dir))
    (concat "file://"
            (expand-file-name (read-file-name "Choose abstract:")))))

(defun ismrm-find-abstract ()
  (interactive)
  (eww-browse-url (call-interactively #'ismrm--find-abstract-file)))

;; Extract title from html. Title is the content of the first span with class submissionTitle
(defun read-html-into-dom (html)
  (with-temp-buffer
    (insert-file-contents html)
    (libxml-parse-html-region (point-min) (point-max))))

;; same as above, but before parsing the region, narrow it to the div with class
;; col-lg-8, excluding anything beyond the end of that div
(defun read-html-into-dom-narrow (html)
  (with-temp-buffer
    (insert-file-contents html)
    (goto-char (point-min))
    (search-forward "<div class=\"col-lg-8")
    (narrow-to-region (point) (search-forward "        </div>"))
    (libxml-parse-html-region (point-min) (point-max))))

(defun ismrm-cleanup-whitespace (string)
  (replace-regexp-in-string "[ \t\n]+" " " string))

(defun ismrm-extract-title (html)
  (let ((title (car (dom-by-class (read-html-into-dom-narrow html) "submissionTitle"))))
    (if title
        (ismrm-cleanup-whitespace (replace-regexp-in-string "\n" " " (dom-text title)))
      "No title")))

(defun ismrm-extract-author (html)
  (let ((abstract (car (dom-by-class (read-html-into-dom-narrow html) "AffiliationBlockContainer"))))
    (if abstract
        (ismrm-cleanup-whitespace
         (replace-regexp-in-string "<[^>]*>" "" (replace-regexp-in-string "\n" " " (dom-text abstract))))
      "No author")))

(use-package dash)
(defun construct-abstract-file-alist ()
  (let ((files (mapcar #'expand-file-name (directory-files ismrm-abstract-dir t ".*\\.html$"))))
    (-zip (mapcar #'ismrm-extract-title files) files)))

(setq conf-abstract-file-alist (construct-abstract-file-alist))

;; Cleanup whitespace in string
(defun ismrm-cleanup-whitespace (string)
  (replace-regexp-in-string "[ \t\n]+" " " string))

;; Convert the alist to json and save to file called /tmp/abstracts.json
(defun ismrm-save-abstracts-json ()
  (interactive)
  (let ((json-encoding-pretty-print t))
    (with-temp-file "/tmp/abstracts.json"
      (insert (json-encode conf-abstract-file-alist)))))

(defun ismrm-abstract-alist-to-json (alist)
  (let ((json-encoding-pretty-print t))
    (json-encode alist)))


(defun abstract-file-name (title)
  (cdr (assoc title conf-abstract-file-alist)))

(defun ismrm-open-abstract (title)
  "Open abstract in eww. TITLE is the title of the abstract."
  (interactive (list (completing-read "Choose abstract: " conf-abstract-file-alist)))
  (let ((file (abstract-file-name title))
        (eww-header-line-format (concat (abstract-file-name title) " - " title)))
    (if file
        (eww-browse-url (concat "file://" file))
      (message "No abstract for %s" title))))


;; Create a json object from all abstracts, with title, author and filename as properties
(defun ismrm-abstracts-to-json ()
  (interactive)
  (let ((json-encoding-pretty-print t))
    (with-temp-file "/tmp/abstracts.json"
      (insert (json-encode (mapcar (lambda (x) (list (car x) (ismrm-extract-author (cdr x)) (abstract-file-name (car x)))) conf-abstract-file-alist))))))


(defun ismrm-abstracts-to-json-list ()
  (interactive)
  (let ((json-encoding-pretty-print t))
    (with-temp-file "/tmp/abstracts.json"
      (insert (json-encode-plist (mapcar (lambda (x) (list "" :name (car x) :author (ismrm-extract-author (cdr x)) :filename (abstract-file-name (car x)))) conf-abstract-file-alist))))))

(defun ismrm-abstracts-to-json-list-with-id ()
  (interactive)
  (let ((ismrm-abstract-id -1)
        (json-encoding-pretty-print t))
    (with-temp-file "/tmp/abstracts.json"
      (insert (json-encode-plist (mapcar (lambda (x) (list "" :id (cl-incf ismrm-abstract-id) :name (car x) :author (ismrm-extract-author (cdr x)) :filename (abstract-file-name (car x)))) conf-abstract-file-alist))))))

(defun ismrm-remove-duplicate-tr-tags ()
  (interactive)
  (save-excursion
    (goto-char (point-min))
    (while (re-search-forward "</tr>\n<tr>" nil t)
      (replace-match ""))))