;ELC
;;; Compiled
;;; in Emacs version 27.1
;;; with all optimizations.
;;; This file contains utf-8 non-ASCII characters,
;;; and so cannot be loaded into Emacs 22 or earlier.
(and (boundp 'emacs-version)
(< (aref emacs-version (1- (length emacs-version))) ?A)
(string-lessp emacs-version "23")
(error "`%s' was compiled for Emacs 23 or later" #$))
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
(defvar html2text-format-single-element-list '(("hr" . html2text-clean-hr)))
#@344 The map of entity to text.
This is an alist were each element is a dotted pair consisting of an
old string, and a replacement string. This replacement is done by the
function `html2text-substitute' which basically performs a
`replace-string' operation for every element in the list. This is
completely verbatim - without any use of REGEXP.
(defvar html2text-replace-list '(("´" . "`") ("&" . "&") ("'" . "'") ("¦" . "|") ("¢" . "c") ("ˆ" . "^") ("©" . "(C)") ("¤" . "(#)") ("°" . "degree") ("÷" . "/") ("€" . "e") ("½" . "1/2") (">" . ">") ("¿" . "?") ("«" . "<<") ("&ldquo" . "\"") ("‹" . "(") ("‘" . "`") ("<" . "<") ("—" . "--") (" " . " ") ("–" . "-") ("‰" . "%%") ("±" . "+-") ("£" . "£") (""" . "\"") ("»" . ">>") ("&rdquo" . "\"") ("®" . "(R)") ("›" . ")") ("’" . "'") ("§" . "§") ("¹" . "^1") ("²" . "^2") ("³" . "^3") ("˜" . "~")) (#$ . 486))
#@577 A list of removable tags.
This is a list of tags which should be removed, without any
formatting. Note that tags in the list are presented *without*
any "<" or ">". All occurrences of a tag appearing in this
list are removed, irrespective of whether it is a closing or
opening tag, or if the tag has additional attributes. The
deletion is done by the function `html2text-remove-tags'.
For instance the text:
"Here comes something big ."
will be reduced to:
"Here comes something big."
If this list contains the element "font".
(defvar html2text-remove-tag-list '("html" "body" "p" "img" "dir" "head" "div" "br" "font" "title" "meta") (#$ . 1540))
#@472 An alist of tags and processing functions.
This is an alist where each dotted pair consists of a tag, and then
the name of a function to be called when this tag is found. The
function is called with the arguments p1, p2, p3 and p4. These are
demonstrated below:
" This is bold text "
^ ^ ^ ^
| | | |
p1 p2 p3 p4
Then the called function will typically format the text somewhat and
remove the tags.
(defvar html2text-format-tag-list '(("b" . html2text-clean-bold) ("strong" . html2text-clean-bold) ("u" . html2text-clean-underline) ("i" . html2text-clean-italic) ("em" . html2text-clean-italic) ("blockquote" . html2text-clean-blockquote) ("a" . html2text-clean-anchor) ("ul" . html2text-clean-ul) ("ol" . html2text-clean-ol) ("dl" . html2text-clean-dl) ("center" . html2text-clean-center)) (#$ . 2243))
#@201 Another list of removable tags.
This is a list of tags which are removed similarly to the list
`html2text-remove-tag-list' - but these tags are retained for the
formatting, and then moved afterward.
(defvar html2text-remove-tag-list2 '("li" "dt" "dd" "meta") (#$ . 3126))
#@63 Replace FROM-STRING with TO-STRING in region from MIN to MAX.
(defalias 'html2text-replace-string #[(from-string to-string min max) "b\210\306 !\306\n!Z\307 \310\n
\311#\203 \312 !\210 \f\\\202
*\207" [min to-string from-string change delta max string-width 0 search-forward t replace-match] 4 (#$ . 3405)])
#@35 Get value of ATTRIBUTE from LIST.
(defalias 'html2text-attr-value #[(list attribute) "\302 \"A@\207" [attribute list assoc] 3 (#$ . 3725)])
(defalias 'html2text-get-attr #[(p1 p2) "b\210\303\304 \305#\210\306\303\307 \305#\203> \310\311!\312\313!\2057 \311\225b\210\314\306w\210\312\315!\2040 \312\316!\2057 \311\225b\210\310\311!D\nB\202 \n)\207" [p1 p2 attr-list re-search-forward "\\s-+" t nil "[-a-z0-9._]+" match-string 0 looking-at "\\s-*=" "[:space:]" "\"[^\"]*\"\\|'[^']*'" "[-a-z0-9._:]+"] 4])
(defalias 'html2text-clean-list-items #[(p1 p2 list-type) "b\210\305\211\306\307 \310#\203 T\202 b\210\n W\205H \nT\306\307d\310#\210\f\311\230\2033 \312c\210\202 \f\313\230\203B \314\315\n\"c\210\202 \316c\210\202 *\207" [p1 items item-nr p2 list-type 0 search-forward "" t "ul" " o " "ol" format " %s: " " x "] 4])
(defalias 'html2text-clean-dtdd #[(p1 p2) "b\210\306\211\307\310 \311#\203 \nT\202 b\210 \nW\205\202 T\312\313d\311#\210\314\315!\2037 ``\316\314\315!!Z|\210`\306 \312\317d\311#\210\314\315!\203o \316\314\315!! \316\314\320!! \\ `Zb\210``\\|\210`+\202w `\316\314\320!!Z\321
\f\322\323$\210*\202 *\207" [p1 item-nr items p2 def-p2 def-p1 0 search-forward "" t re-search-forward "\\([ ]*\\)" match-string 1 string-width "\\([ ]*\\)\\(\\|\\)" 2 put-text-property face bold mw1 mw2 mw] 5])
(defalias 'html2text-delete-tags #[(p1 p2 p3 p4) " |\210\n ZZ ZZ|\207" [p1 p2 p3 p4] 4])
(defalias 'html2text-delete-single-tag #[(p1 p2) " |\207" [p1 p2] 2])
(defalias 'html2text-clean-hr #[(p1 p2) "\303 \"\210b\210\304\305!\210\306\n\307\"c\207" [p1 p2 fill-column html2text-delete-single-tag newline 1 make-string 45] 3])
(defalias 'html2text-clean-ul #[(p1 p2 p3 p4) "\304 \n $\210\305\n ZZ\306#\207" [p1 p2 p3 p4 html2text-delete-tags html2text-clean-list-items "ul"] 5])
(defalias 'html2text-clean-ol #[(p1 p2 p3 p4) "\304 \n $\210\305\n ZZ\306#\207" [p1 p2 p3 p4 html2text-delete-tags html2text-clean-list-items "ol"] 5])
(defalias 'html2text-clean-dl #[(p1 p2 p3 p4) "\304 \n $\210\305\n ZZ\"\207" [p1 p2 p3 p4 html2text-delete-tags html2text-clean-dtdd] 5])
(defalias 'html2text-clean-center #[(p1 p2 p3 p4) "\304 \n $\210\305\n ZZ\"\207" [p1 p2 p3 p4 html2text-delete-tags center-region] 5])
(defalias 'html2text-clean-bold #[(p1 p2 p3 p4) "\304 \305\306$\210\307\n $\207" [p2 p3 p1 p4 put-text-property face bold html2text-delete-tags] 5])
(defalias 'html2text-clean-title #[(p1 p2 p3 p4) "\304 \305\306$\210\307\n $\207" [p2 p3 p1 p4 put-text-property face bold html2text-delete-tags] 5])
(defalias 'html2text-clean-underline #[(p1 p2 p3 p4) "\304 \305\306$\210\307\n $\207" [p2 p3 p1 p4 put-text-property face underline html2text-delete-tags] 5])
(defalias 'html2text-clean-italic #[(p1 p2 p3 p4) "\304 \305\306$\210\307\n $\207" [p2 p3 p1 p4 put-text-property face italic html2text-delete-tags] 5])
(defalias 'html2text-clean-font #[(p1 p2 p3 p4) "\304 \n $\207" [p1 p2 p3 p4 html2text-delete-tags] 5])
(defalias 'html2text-clean-blockquote #[(p1 p2 p3 p4) "\304 \n $\207" [p1 p2 p3 p4 html2text-delete-tags] 5])
(defalias 'html2text-clean-anchor #[(p1 p2 _p3 p4) "\305 \"\306\n\307\"\f|\210 \205- b\210\310\311 \"\203$ \312 \313\314#\202% c\210\315`\316\317$*\207" [p1 p2 attr-list href p4 html2text-get-attr html2text-attr-value "href" string-match "\\`['\"].*['\"]\\'" substring 1 -1 put-text-property face bold] 5])
(defalias 'html2text-fix-paragraph #[(p1 p2) "b\210\304\211\305\306 \307#\203< b\210\305\310 \307#\203< \311 \210` b\210\312\310\n\307#\210\313y\210\304\210`\314\315\316\n`$\\\317\n \"\210*\314\315\316 $\207" [p1 refill-stop refill-start p2 nil re-search-forward "
$" t ".+[^<][^b][^r][^>]$" beginning-of-line re-search-backward 1 html2text-replace-string "
" "" fill-region] 6])
#@149 This _tries_ to fix up the paragraphs - this is done in quite a ad-hook
fashion, quite close to pure guess-work. It does work in some cases though.
(defalias 'html2text-fix-paragraphs #[nil "eb\210\301\302\303\304#\203 \305\224\305\225|\210\202 eb\210m?\2057 `\306\307!\210\310`S\"\210b\210m\2043 \306\307!\210)\202 \207" [p1 re-search-forward "^
$" nil t 0 forward-paragraph 1 html2text-fix-paragraph] 4 (#$ . 7568) nil])
#@107 Removes the tags listed in the list `html2text-remove-tag-list'.
See the documentation for that variable.
(defalias 'html2text-remove-tags #[(tag-list) "\303\211\205* \n@eb\210\304\305\306 \"d\307#\203\" \310\224\310\225|\210\202 \nA\211\204 \303*\207" [tag-list tag --dolist-tail-- nil re-search-forward format "\\(]*>\\)" t 0] 5 (#$ . 8007) nil])
#@65 See the variable `html2text-format-tag-list' for documentation.
(defalias 'html2text-format-tags #[nil "\306\211\205n \n@\211@ A eb\210\307\310\311\f\"d\312#\203e \306`\306\211 \313\314e\312#\210`\315\310\316\f\"d\312#\204H b\210\310\316\f\"c\210`\313\317e\312#\210`
$\210b\210,\202 *\nA\211\204 \306*\207" [html2text-format-tag-list tag-and-function --dolist-tail-- function tag p4 nil re-search-forward format "\\(<%s\\( [^>]*\\)?>\\)" t search-backward "<" search-forward "" "]*\\)?>\\)" t search-backward "<" p1] 5 nil nil])
#@51 Convert HTML to plain text in the current buffer.
(defalias 'html2text #[nil "\212\304\305\306\n!\210\307 \210\306 !\210\310 \210\311 \210\312 +\207" [buffer-read-only case-fold-search html2text-remove-tag-list html2text-remove-tag-list2 t nil html2text-remove-tags html2text-format-tags html2text-substitute html2text-format-single-elements html2text-fix-paragraphs] 2 (#$ . 9591) nil])
(provide 'html2text)