org-tag-eldoc-wikipedia.el

   1 ;;; org-tag-eldoc-wikipedia.el --- Web Scraping on Wikipedia -*- lexical-binding: t; -*-
   2 ;; -*- coding: utf-8 -*-
   3
   4 ;; Copyright (C) 2024-2025 Christopher M. Miles, all rights reserved.
   5
   6 ;;; Commentary:
   7
   8 ;; search API: https://wikipedia.org/w/index.php?search=[query]
   9 ;; tag URL: https://en.wikipedia.org/wiki/Computer_science
  10
  11 ;;; Code:
  12 \f
  13 (require 'url)
  14 (require 'url-http) ; for `url-http-end-of-headers'
  15 (require 'request)
  16 (require 'dom)
  17 (require 'elquery nil t) ; optionally use elquery.el to replace dom.el.
  18 (require 'org-tag-eldoc-common)
  19
  20
  21 (defun org-tag-eldoc-wikipedia--request (tag)
  22   "Send HTTP GET request to get TAG explanation data."
  23   (let ((request-backend 'url-retrieve) ; use `url-retrieve' backend for proxy.
  24         (url-proxy-services org-tag-eldoc-request-proxy)
  25         (url (format "https://en.wikipedia.org/w/index.php?title=%s" (capitalize tag))))
  26     (request url
  27       :type "GET"
  28       :parser (lambda ()
  29                 (cond
  30                  ((fboundp 'libxml-parse-html-region) ; convert HTML -> Elisp alist structure
  31                   (libxml-parse-html-region (point-min) (point-max)))
  32                  ((featurep 'elquery) ; convert HTML -> elquery object structure
  33                   (elquery-read-buffer (current-buffer)))))
  34       :success (cl-function
  35                 (lambda (&key data &allow-other-keys)
  36                   ;; DEBUG: (setq request-result data)
  37                   (cond
  38                    ((featurep 'dom)
  39                     ;; for `libxml' parser
  40                     (setq org-tag-eldoc--explanation
  41                           (org-tag-eldoc-common--format-explanation
  42                            (dom-texts
  43                             (dom-search (dom-by-class data "mw-content-ltr\\ mw-parser-output")
  44                                         (lambda (node) (and (eq (cl-first node) 'p) (null (cl-second node)))))))))
  45                    ((featurep 'elquery)
  46                     ;; for `elquery' parser
  47                     (setq org-tag-eldoc--explanation
  48                           (elquery-text (car (elquery-$ "#mw-content-text > div.mw-content-ltr.mw-parser-output > p" data))))))))
  49       :error (cl-function
  50               (lambda (&rest args &key error-thrown &allow-other-keys)
  51                 (message "[org-tag-eldoc] (Wikipedia) request error %s!" error-thrown)
  52                 nil))
  53       :status-code '((404 . (lambda (&rest _) nil))
  54                      (500 . (lambda (&rest _) nil))))
  55
  56     ;; (with-current-buffer (url-retrieve-synchronously url)
  57     ;;   (let ((dom (progn
  58     ;;                (goto-char url-http-end-of-headers)
  59     ;;                (libxml-parse-html-region (point) (point-max)))))
  60     ;;     ;; DEBUG: (setq request-result dom)
  61     ;;     (org-tag-eldoc-common--format-explanation
  62     ;;      (cond
  63     ;;       ((featurep 'dom)
  64     ;;        (dom-texts
  65     ;;         (dom-search (dom-by-class dom "mw-content-ltr\\ mw-parser-output")
  66     ;;                     (lambda (node) (and (eq (cl-first node) 'p) (null (cl-second node)))))))
  67     ;;       ((featurep 'elquery)
  68     ;;        (elquery-text (car (elquery-$ "#mw-content-text > div.mw-content-ltr.mw-parser-output > p" data))))))))
  69     ))
  70
  71 ;;; TEST: https://en.wikipedia.org/w/index.php?title=Computer_science
  72 ;; (org-tag-eldoc-wikipedia--request "computer_science")
  73
  74 ;;; TEST
  75 ;; (let ((url-proxy-services org-tag-eldoc-request-proxy))
  76 ;;   (with-current-buffer
  77 ;;       (url-retrieve-synchronously "https://en.wikipedia.org/w/index.php?title=Computer_science")
  78 ;;     (let ((dom (progn
  79 ;;                  (goto-char url-http-end-of-headers)
  80 ;;                  (libxml-parse-html-region (point) (point-max)))))
  81 ;;       (setq request-result dom)
  82 ;;       (setq org-tag-eldoc--explanation
  83 ;;             (dom-texts
  84 ;;              (dom-search (dom-by-class dom "mw-content-ltr\\ mw-parser-output")
  85 ;;                          (lambda (node) (and (eq (cl-first node) 'p) (null (cl-second node))))))))))
  86
  87 ;;; TEST
  88 ;; (dom-texts
  89 ;;  (dom-search (dom-by-class request-result "mw-content-ltr\\ mw-parser-output")
  90 ;;              (lambda (node) (and (eq (cl-first node) 'p) (null (cl-second node))))))
  91
  92 ;;; TEST
  93 ;;; this example works fine.
  94 ;; (let ((url-proxy-services org-tag-eldoc-request-proxy))
  95 ;;   (with-current-buffer (url-retrieve-synchronously "https://en.wikipedia.org/w/index.php?title=GitHub")
  96 ;;     (let ((dom (libxml-parse-html-region (point-min) (point-max))))
  97 ;;       (let ((paragraph-node (dom-search (dom-by-class dom "mw-content-ltr\\ mw-parser-output")
  98 ;;                                         (lambda (node) (eq (car node) 'p)))))
  99 ;;         (setq node-github paragraph-node)
 100 ;;         (dom-texts paragraph-node)))))
 101 ;;
 102 ;; `node-github' ; looks fine
 103 ;;
 104 ;;; this example failed on `dom-texts'.
 105 ;; (let ((url-proxy-services org-tag-eldoc-request-proxy))
 106 ;;   (with-current-buffer (url-retrieve-synchronously "https://en.wikipedia.org/w/index.php?title=Computer_science")
 107 ;;     (let ((dom (libxml-parse-html-region (point-min) (point-max))))
 108 ;;       (let ((paragraph-node (dom-search (dom-by-class dom "mw-content-ltr\\ mw-parser-output")
 109 ;;                                         (lambda (node) (eq (car node) 'p)))))
 110 ;;         (setq node-computer-science paragraph-node)
 111 ;;         (dom-texts paragraph-node)))))
 112 ;;
 113 ;; `node-computer-science' ; looks fine
 114 ;;
 115 ;;; this examples works by a new `dom-search' rule.
 116 ;; (let ((url-proxy-services org-tag-eldoc-request-proxy))
 117 ;;   (with-current-buffer (url-retrieve-synchronously "https://en.wikipedia.org/w/index.php?title=Computer_science")
 118 ;;     (let ((dom (libxml-parse-html-region (point-min) (point-max))))
 119 ;;       (let ((paragraph-node (dom-search (dom-by-class dom "mw-content-ltr\\ mw-parser-output")
 120 ;;                                         (lambda (node) (and (eq (cl-first node) 'p)
 121 ;;                                                        (null (cl-second node)))))))
 122 ;;         (setq node-computer-science paragraph-node)
 123 ;;         (dom-texts paragraph-node)))))
 124
 125 (defun org-tag-eldoc-wikipedia-query (tag)
 126   "Query TAG on Wikipedia then return a cons cell of tag and explanation."
 127   (if (stringp org-tag-eldoc--explanation)
 128       org-tag-eldoc--explanation
 129     (org-tag-eldoc-wikipedia--request tag)
 130     (sit-for 1.0)
 131     (org-tag-eldoc-database-save tag org-tag-eldoc--explanation)))
 132
 133 \f
 134
 135 (provide 'org-tag-eldoc-wikipedia)
 136
 137 ;;; org-tag-eldoc-wikipedia.el ends here