Add published propety

[atom.el.git] / atom.el

diff --git a/atom.el b/atom.el
index 5d3df7b1544872027698f9f6a631905d2a318db0..34111c1d6ebf75f7e93cd3a908dc1d92add1e9ca 100644 (file)
--- a/atom.el
+++ b/atom.el
@@ -1,4 +1,4 @@
-;;; atom.el --- Create an Atom feed
+;;; atom.el --- Create an Atom feed  -*- lexical-binding: t -*-

 
 ;; Copyright (C) 2011  Frédéric Perrin
 
 
 ;; Copyright (C) 2011  Frédéric Perrin
 


@@ -27,8 +27,7 @@
 ;; permanent link and the content of the entry. Text-only, HTML and
 ;; XHTML entries are supported.
 
 ;; permanent link and the content of the entry. Text-only, HTML and
 ;; XHTML entries are supported.
 

-;; A feed is really a Lisp structure as used by the `xml.el' package,
-;; without the parent `feed' element.
+;; It is possible to produce both Atom and RSS feeds.

 
 ;; A typical usage would look like this:
 
 
 ;; A typical usage would look like this:
 


@@ -40,35 +39,25 @@
 ;;    "http://example.org/hello"
 ;;    "Hello the world!")
 ;;
 ;;    "http://example.org/hello"
 ;;    "Hello the world!")
 ;;

-;;   ; A text-only entry, with all the optional pieces of data
-;;   (atom-add-text-entry
-;;    my-atom-feed
-;;    "Bonjour"
-;;    "http://example.org/bonjour"
-;;    "Bonjour à tout le monde !"
-;;    ;; optional: the last modification time
-;;    (date-to-time "2011-01-30 23:40:12")
-;;    ;; optional: an identifier for this entry; a common way to generate it is
-;;    ;; to use the domain name and the creation date of the entry.
-;;    (atom-generate-id "http://example.org"
-;;                  (date-to-time "2011-01-30 10:01:05"))
-;;    ;; optional: a summary for this entry
-;;    "Bonjour, monde.")
-;;

 ;;   (atom-add-xhtml-entry
 ;;    my-atom-feed
 ;;    "An XHTML example"
 ;;    "http://example.org/html-example"
 ;;    "<p>One can also use <acronym>XHTML</acronym> in the entries.</p>")
 ;;   (atom-add-xhtml-entry
 ;;    my-atom-feed
 ;;    "An XHTML example"
 ;;    "http://example.org/html-example"
 ;;    "<p>One can also use <acronym>XHTML</acronym> in the entries.</p>")

-;;   (atom-print my-atom-feed))
+;;
+;;   (atom-print my-atom-feed)
+;;   ;; If you prefer RSS feeds:
+;;   (atom-print-as-rss my-atom-feed))
+
+;; Full documentation is available at <http://tar-jx.bz/code/atom.html>.
+;; See atom-tests.el for usage examples.

 
 ;;; Code:
 
 (require 'xml)
 (require 'url-parse)
 
 ;;; Code:
 
 (require 'xml)
 (require 'url-parse)

-(require 'cl) ; for setf in url-canonalize

 
 

-(defun atom-create (title link &optional subtitle self id author updated)
+(defun atom-create (title link &optional props)

   "Create a new atom structure.
 
 TITLE is the title for the feed, a short, text-only, human
   "Create a new atom structure.
 
 TITLE is the title for the feed, a short, text-only, human


@@ -77,29 +66,34 @@ readable string.
 LINK is the URL of a page responible for the content of this
 feed.
 
 LINK is the URL of a page responible for the content of this
 feed.
 

-SUBTITLE is a subtitle for the feed; it can be a bit longer than
-TITLE, maybe a paragraph long.
+PROPS is an optional plist with the following properties:
+
+- :subtitle is a subtitle for the feed; it can be a bit longer than
+  TITLE, maybe a paragraph long.

 
 

-SELF is the canonical URL to this feed.
+- :self is the canonical URL to this feed. If missing, the resulting
+  feed is non-conforming.

 
 

-ID is a unique identifier for this feed. If not given, it
-defaults to SELF.
+- :id is a unique identifier for this feed. If not given, it
+  defaults to :self.

 
 

-AUTHOR is the author of the feed. See `atom-massage-author' for
-the possible ways to specify it. In particular, `nil' uses
-`user-full-name' and `user-mail-address'.
+- :author is the author of the feed. See `atom-massage-author' for
+the possible ways to specify it. In particular, nil uses
+variable `user-full-name' and `user-mail-address'.

 
 

-UPDATED is the date the feed was last updated. If not given,
+- :updated is the date the feed was last updated. If not given,

 `(current-time)' is used."
   (let ((atom-feed (list (list 'title nil title))))
     (atom-modify-entry atom-feed 'link `(((href . ,link))))
 `(current-time)' is used."
   (let ((atom-feed (list (list 'title nil title))))
     (atom-modify-entry atom-feed 'link `(((href . ,link))))

-    (atom-modify-entry atom-feed 'author (atom-massage-author author))
-    (if subtitle (atom-modify-entry atom-feed 'subtitle subtitle))
-    (if self (atom-modify-entry atom-feed 'link
-                               `(((href . ,self) (rel . "self")
-                                  (type . "application/atom+xml")))))
-    (atom-modify-entry atom-feed 'updated (atom-format-time updated))
-    (atom-modify-entry atom-feed 'id (or id self link))
+    (atom-modify-entry atom-feed 'author (atom-massage-author (plist-get props :author)))
+    (if (plist-member props :subtitle)
+       (atom-modify-entry atom-feed 'subtitle (plist-get props :subtitle)))
+    (if (plist-member props :self)
+       (atom-modify-entry atom-feed 'link
+                          `(((href . ,(plist-get props :self)) (rel . "self")
+                             (type . "application/atom+xml")))))
+    (atom-modify-entry atom-feed 'updated (atom-format-time (plist-get props :updated)))
+    (atom-modify-entry atom-feed 'id (or (plist-get props :id) (plist-get props :self) link))

     atom-feed))
 
 (defun atom-push-entry (atom entry)
     atom-feed))
 
 (defun atom-push-entry (atom entry)


@@ -113,10 +107,10 @@ UPDATED is the date the feed was last updated. If not given,
                (cons name val))))
     (nconc entry (list elem))))
 
                (cons name val))))
     (nconc entry (list elem))))
 

-(defun atom-add-entry (atom title link content
-                           &optional updated id summary)
-  "Add an entry to the atom flux ATOM. Return the newly added
-entry.
+(defun atom-add-entry (atom title link content &optional props)
+  "Add an entry to the atom flux ATOM.
+
+Return the newly adde dentry.

 
 TITLE is a short, text-only, human readable string.
 
 
 TITLE is a short, text-only, human readable string.
 


@@ -126,20 +120,27 @@ may change between successive generations of the atom feed.
 CONTENT is the content of the entry; use `atom-add-html-entry'
 or `atom-add-xhtml-entry' when CONTENT is not text-only.
 
 CONTENT is the content of the entry; use `atom-add-html-entry'
 or `atom-add-xhtml-entry' when CONTENT is not text-only.
 

-If SUMMARY is not given, the entry will not contain any summary.
+PROPS is an optional plist with the following properties:
+
+- :summary, if is not given, the entry will not contain any summary.

 
 

-UPDATED defaults to `(current-time)' if omitted, which is
-probably not a very good default.
+- :updated defaults to `(current-time)'.

 
 

-ID defaults to LINK, which is not optimal; see `atom-generate-id'
-for a way to create good identifiers. For a given entry, it must
-not change between successive generations of the atom feed, even
-when the content of the entry ."
+- :published, if given, is the earliest availability of the
+  entry. It is optional, and shouldn't change even if the entry
+  content (etc.) updated after the initial publication.
+
+- :id is a unique ID for the entry; defaulting to LINK. RFC4287
+  has specific requirements about valid IRI that may be used,
+  which this library does not try to enforce."

   (let ((entry (list (list 'title nil title))))
     (atom-modify-entry entry 'link  (list (list (cons 'href link))))
   (let ((entry (list (list 'title nil title))))
     (atom-modify-entry entry 'link  (list (list (cons 'href link))))

-    (atom-modify-entry entry 'id (or id link))
-    (atom-modify-entry entry 'updated (atom-format-time updated))
-    (if summary (atom-modify-entry entry 'summary summary))
+    (atom-modify-entry entry 'id (or (plist-get props :id) link))
+    (atom-modify-entry entry 'updated (atom-format-time (plist-get props :updated)))
+    (if (plist-member props :published)
+       (atom-modify-entry entry 'published (atom-format-time (plist-get props :published))))
+    (if (plist-member props :summary)
+       (atom-modify-entry entry 'summary (plist-get props :summary)))

     (atom-modify-entry entry 'content content)
     (atom-push-entry atom entry)
     entry))
     (atom-modify-entry entry 'content content)
     (atom-push-entry atom entry)
     entry))


@@ -148,33 +149,31 @@ when the content of the entry ."
   "Add an entry to ATOM, with a textual content. See
 `atom-add-entry' for details.")
 
   "Add an entry to ATOM, with a textual content. See
 `atom-add-entry' for details.")
 

-(defun atom-add-html-entry (atom title link content
-                                 &optional updated id summary)
-  "Add an entry to ATOM, with some HTML content. CONTENT should
-be a string enconding a valid HTML fragment. See `atom-add-entry'
-for additional details."
-  (atom-add-entry atom
-   title link (atom-massage-html content)
-   updated id (and summary (atom-massage-html summary))))
-
-(defun atom-add-xhtml-entry (atom title link content
-                                 &optional updated id summary noconvert)
-  "Add an entry to ATOM, with some XHTML content. CONTENT may be
-given either as a string, or as an XML tree, of a valid XHTML
-fragment. See `atom-add-entry' for additional details.
-
-If CONVERT, translate all links in CONTENT so that they are no
-longer relative to LINK."
-  (let ((xhtml-content (atom-massage-xhtml content)))
-    (unless noconvert
-      (atom-xhtml-convert-links (cadr xhtml-content) link))
-    (atom-add-entry atom
-                   title link xhtml-content
-                   updated id (and summary (atom-massage-xhtml summary)))))
+(defun atom-add-html-entry (atom title link content &optional props)
+  "Add an entry to ATOM, with some HTML content.
+
+TITLE, LINK, PROPS as in `atom-add-entry'. CONTENT should be a string
+enconding a valid HTML fragment. See `atom-add-entry' for
+additional details."
+  (if (plist-member props :summary)
+      (plist-put props :summary (atom-massage-html (plist-get props :summary))))
+  (atom-add-entry atom title link (atom-massage-html content) props))
+
+(defun atom-add-xhtml-entry (atom title link content &optional props)
+  "Add an entry to ATOM, with some XHTML content.
+
+TITLE, LINK, PROPS as in `atom-add-entry'. CONTENT may be given
+either as a string, or as an XML tree, of a valid XHTML fragment.
+See `atom-add-entry' for additional details."
+  (if (plist-member props :summary)
+      (plist-put props :summary (atom-massage-xhtml (plist-get props :summary))))
+  (atom-add-entry atom title link (atom-massage-xhtml content) props))
+
+(defvar atom-xml-declaration "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n")

 
 (defun atom-print (atom)
   "Print the Atom feed ATOM in the current buffer."
 
 (defun atom-print (atom)
   "Print the Atom feed ATOM in the current buffer."

-  (insert "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n")
+  (insert atom-xml-declaration)

   (insert "<feed xmlns=\"http://www.w3.org/2005/Atom\">\n")
   (xml-print atom)
   (insert "\n</feed>"))
   (insert "<feed xmlns=\"http://www.w3.org/2005/Atom\">\n")
   (xml-print atom)
   (insert "\n</feed>"))


@@ -183,14 +182,20 @@ longer relative to LINK."
   "Writes the feed ATOM to FILENAME."
   (with-temp-buffer
     (atom-print atom)
   "Writes the feed ATOM to FILENAME."
   (with-temp-buffer
     (atom-print atom)

-    (write-region (point-min) (point-max) filename)))
+    (write-file filename)))

 
 \f
 
 \f

-(defun atom-to-rss (atom)
-  "Translate an Atom feed into an RSS one, returning the translation.
+(defun atom-to-rss (atom &optional rss-self)
+  "Translate Atom feed ATOM into an RSS one, returning the translation.
+
+If RSS-SELF is given, it is used as self link of the RSS feed.

 
 Some information may be lost or approximated."
   (let ((rss (list (assoc 'title atom))))
 
 Some information may be lost or approximated."
   (let ((rss (list (assoc 'title atom))))

+    (if rss-self
+       (atom-modify-entry rss 'atom:link
+                          `(((href . ,rss-self) (rel . "self")
+                             (type . "application/atom+xml")))))

     (atom-to-rss-translator atom rss '((subtitle . description)
                                       (updated . pubDate)
                                       (link . link)))
     (atom-to-rss-translator atom rss '((subtitle . description)
                                       (updated . pubDate)
                                       (link . link)))


@@ -201,7 +206,7 @@ Some information may be lost or approximated."
     (reverse rss)))
 
 (defun atom-to-rss-item (entry)
     (reverse rss)))
 
 (defun atom-to-rss-item (entry)

-  "Translates an Atom entry into an RSS item."
+  "Translates the Atom entry ENTRY into an RSS item."

   (let ((item (list (assoc 'title entry))))
     (atom-to-rss-translator
      (xml-node-children entry) item
   (let ((item (list (assoc 'title entry))))
     (atom-to-rss-translator
      (xml-node-children entry) item


@@ -214,7 +219,8 @@ Some information may be lost or approximated."
          (setcar (cdr guid) (list (cons 'isPermaLink "false"))))
       (if (and descr
               (equal (xml-get-attribute descr 'type) "xhtml"))
          (setcar (cdr guid) (list (cons 'isPermaLink "false"))))
       (if (and descr
               (equal (xml-get-attribute descr 'type) "xhtml"))

-         (setcar (cddr descr) (xml-node-text descr))))
+         (setcar (cddr descr) (xml-node-as-text descr)))
+      (setcar (cdr descr) nil))

     `(item nil ,@item)))
 
 (defun atom-to-rss-translator (source target translations)
     `(item nil ,@item)))
 
 (defun atom-to-rss-translator (source target translations)


@@ -225,11 +231,6 @@ Some information may be lost or approximated."
       (when data
        (atom-modify-entry target to data)))))
 
       (when data
        (atom-modify-entry target to data)))))
 

-(defun xml-node-text (node)
-  (with-temp-buffer
-    (xml-print (xml-node-children node))
-    (buffer-string)))
-

 (defun atom-to-rss-modify-link (entry)
   (let* ((link (assoc 'link entry))
         (link-addr (xml-get-attribute-or-nil link 'href)))
 (defun atom-to-rss-modify-link (entry)
   (let* ((link (assoc 'link entry))
         (link-addr (xml-get-attribute-or-nil link 'href)))


@@ -237,19 +238,23 @@ Some information may be lost or approximated."
       (setcar (cdr link) nil)
       (setcdr (cdr link) (cons link-addr nil)))))
 
       (setcar (cdr link) nil)
       (setcdr (cdr link) (cons link-addr nil)))))
 

-(defun atom-print-as-rss (atom)
-  (let ((rss (atom-to-rss atom)))
-    (insert "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n")
-    (insert "<rss version=\"2.0\">\n")
+(defun atom-print-as-rss (atom &optional rss-self)
+  "Convert Atom feed ATOM to RSS in the current buffer.
+
+If RSS-SELF is given, it is used as self link of the RSS feed."
+  (let ((rss (atom-to-rss atom rss-self)))
+    (insert atom-xml-declaration)
+    ;; xmlns:atom included in order to allow the atom:link rel=self element
+    (insert "<rss version=\"2.0\" xmlns:atom=\"http://www.w3.org/2005/Atom\">\n")

     (insert "  <channel>\n")
     (xml-print rss "    ")
     (insert "\n  </channel>\n")
     (insert "</rss>")))
 
 (defun atom-to-rss-time (time)
     (insert "  <channel>\n")
     (xml-print rss "    ")
     (insert "\n  </channel>\n")
     (insert "</rss>")))
 
 (defun atom-to-rss-time (time)

-  "Translates a string from the format used by Atom into the
-format used by RSS."
-  ;; Same remark as in `atom-format-time'
+  "Translate TIME from the format used by Atom into the format used by RSS.
+
+TIME is a string."

   (let ((system-time-locale "C"))
     (format-time-string "%a, %d %b %Y %T %z" (atom-parse-time time))))
 
   (let ((system-time-locale "C"))
     (format-time-string "%a, %d %b %Y %T %z" (atom-parse-time time))))
 


@@ -259,18 +264,22 @@ format used by RSS."
     (setcar (cddr pubDate)
            (atom-to-rss-time (car (xml-node-children pubDate))))))
 
     (setcar (cddr pubDate)
            (atom-to-rss-time (car (xml-node-children pubDate))))))
 

-(defun atom-to-rss-write-file (atom filename)
-  "Saves ATOM as a RSS feed into FILENAME."
+(defun atom-to-rss-write-file (atom filename &optional rss-self)
+  "Save ATOM as a RSS feed into FILENAME.
+
+If RSS-SELF is given, it is used as self link of the RSS feed."

   (with-temp-buffer
   (with-temp-buffer

-    (atom-print-as-rss atom)
-    (write-region nil nil filename)))
+    (atom-print-as-rss atom rss-self)
+    (write-file filename)))

 
 \f
 (defvar atom-time-format-string "%Y-%m-%dT%T%z"
   "The format for string representation of dates.")
 
 
 \f
 (defvar atom-time-format-string "%Y-%m-%dT%T%z"
   "The format for string representation of dates.")
 

+(defvar atom-xhtml-namespace "http://www.w3.org/1999/xhtml")
+

 (defun atom-format-time (&optional time)
 (defun atom-format-time (&optional time)

-  "Format a time according to RFC3339."
+  "Format time value TIME according to RFC3339."

   ;; The time zone must be specified in numeric form, but with a colon between
   ;; the hour and minute parts.
   (replace-regexp-in-string
   ;; The time zone must be specified in numeric form, but with a colon between
   ;; the hour and minute parts.
   (replace-regexp-in-string


@@ -278,35 +287,44 @@ format used by RSS."
    (format-time-string atom-time-format-string time)))
 
 (defun atom-parse-time (&optional time)
    (format-time-string atom-time-format-string time)))
 
 (defun atom-parse-time (&optional time)

-  "Parse a time as specified in RFC3339 into Emacs's native format."
+  "Parse string TIME as specified in RFC3339 into Emacs's native format."
+  ;; Same remark as in `atom-format-time': RFC3339 wants a colon between hour
+  ;; and minute parts of the timezome, so remove it before `date-to-time'.

   (date-to-time (replace-regexp-in-string ":\\(..\\)$" "\\1" time)))
 
 (defun atom-massage-html (content)
   (date-to-time (replace-regexp-in-string ":\\(..\\)$" "\\1" time)))
 
 (defun atom-massage-html (content)

-  "Massage CONTENT so it can be used as an HTML fragment in an
-Atom feed. CONTENT must be a string."
+  "Massage CONTENT so it can be used as an HTML fragment in an Atom feed.
+
+CONTENT must be a string."

   (list '((type . "html")) content))
 
 (defun atom-string-to-xml (string)
   "Convert STRING into a Lisp structure as used by `xml.el'."
   (list '((type . "html")) content))
 
 (defun atom-string-to-xml (string)
   "Convert STRING into a Lisp structure as used by `xml.el'."

-  (with-temp-buffer
-    (insert "<div xmlns=\"http://www.w3.org/1999/xhtml\">")
-    (insert string)
-    (insert "</div>")
-    (xml-parse-region (point-min) (point-max))))
+  (require 'xml-xhtml-entities)
+  (let ((xml-entity-alist xml-xhtml-entities)
+       (xml-validating-parser t))
+    (with-temp-buffer
+      (insert "<div xmlns=\"" atom-xhtml-namespace "\">")
+      (insert string)
+      (insert "</div>")
+      ;; `xml-parse-region' returns a list of elements, even though it
+      ;; requires an only root node. We are only interested in the first
+      ;; one, the DIV we just inserted.
+      (car (xml-parse-region (point-min) (point-max))))))

 
 (defun atom-massage-xhtml (content)
 
 (defun atom-massage-xhtml (content)

-  "Massage CONTENT so it can be used as an XHTML fragment in an
-Atom feed."
-  `(((type . "xhtml"))
-    ,@(or (and (stringp content)
-              (atom-string-to-xml content))
-         content)))
+  "Massage CONTENT so it can be used as an XHTML fragment in an Atom feed."
+  (list '((type . "xhtml"))
+       (or (and (stringp content)
+                (atom-string-to-xml content))
+           `(div ((xmlns . ,atom-xhtml-namespace)) ,@content))))

 
 (defun atom-massage-author (author)
   "Return an XML node representing the author. AUTHOR can be:
 
 (defun atom-massage-author (author)
   "Return an XML node representing the author. AUTHOR can be:

-- nil, in which case `user-full-name' and `user-mail-address' are
-  used;
-- a single string, the full name of the author;
+- nil, in which case variables `user-full-name' and `user-mail-address'
+  are used;
+- a single string, the full name of the author; no email address
+  will be included;

 - a list with two elements, the full name and the email address
   of the author;
 - something else, assumed to be a complete `atomPersonConstruct'."
 - a list with two elements, the full name and the email address
   of the author;
 - something else, assumed to be a complete `atomPersonConstruct'."


@@ -327,8 +345,11 @@ absolute, in the context of BASE, an URL."
   (dolist (child (xml-node-children node))
     (when (listp child) (atom-xhtml-convert-links child base))))
 
   (dolist (child (xml-node-children node))
     (when (listp child) (atom-xhtml-convert-links child base))))
 

+\f
+;;; Functions that should probably not be there
+

 (defun url-canonalize (address base)
 (defun url-canonalize (address base)

-  "Make ADRESS an absolute URL, taking it in the BASE context."
+  "Make ADDRESS an absolute URL, taking it in the BASE context."

   ;; I feel such a function should exist in `url-parse'. Did I miss it?
   (let ((url-base (url-generic-parse-url base))
        (url-address (url-generic-parse-url address)))
   ;; I feel such a function should exist in `url-parse'. Did I miss it?
   (let ((url-base (url-generic-parse-url base))
        (url-address (url-generic-parse-url address)))


@@ -339,14 +360,11 @@ absolute, in the context of BASE, an URL."
                              (file-name-directory (url-filename url-base))))
       (url-recreate-url url-base))))
 
                              (file-name-directory (url-filename url-base))))
       (url-recreate-url url-base))))
 

-(defun atom-generate-id (link creation-date)
-  "Generate a string suitable for use as an atom:id element. This
-implements Mark Pilgrom's tag: URI method, using the
-CREATION-DATE of the entry, and the domain part of LINK."
-    (format "tag:%s,%s:/%s"
-           (url-host (url-generic-parse-url link))
-           (format-time-string "%Y-%m-%d" creation-date)
-           (format-time-string "%Y%m%d%H%M%S" creation-date)))
+(defun xml-node-as-text (node)
+  "Return a string representing NODE, an XML structure."
+  (with-temp-buffer
+    (xml-print (xml-node-children node))
+    (buffer-string)))

 
 (provide 'atom)
 ;;; atom.el ends here
 
 (provide 'atom)
 ;;; atom.el ends here