Merge pull request #2100 from getnikola/feed-previewimage

New option FEED_PREVIEWIMAGE includes post previewimage meta
getnikola · Sep 20, 2015 · 24c86aa · 24c86aa
2 parents a390487 + b417c82
commit 24c86aa
Show file tree

Hide file tree

Showing 10 changed files with 137 additions and 68 deletions.
diff --git a/CHANGES.txt b/CHANGES.txt
@@ -4,10 +4,17 @@ New in master
 Features
 --------
 
+* New option ``FEED_PREVIEWIMAGE`` includes the ``post.meta.previewimage``
+  image in Atom and RSS feeds. (Issue #2095)
+
 Bugfixes
 --------
 
 * Fix reST post list date formatting error (Issue #2104)
+* Deprecated ``RSS_TEASERS``, ``RSS_PLAIN``, ``RSS_READ_MORE_LINK``, and
+  ``RSS_LINKS_APPEND_QUERY`` in favor of ``FEED_TEASERS``, ``FEED_PLAIN``,
+  ``FEED_READ_MORE_LINK``, and ``FEED_LINKS_APPEND_QUERY`` for both Atom
+  and RSS feeds. (Issue #2095)
 * /robots.txt was never being built (Issue #2098)
 
 New in v7.7.1

diff --git a/docs/manual.txt b/docs/manual.txt
@@ -515,8 +515,8 @@ In Markdown (or basically, the resulting HTML of any format):
 By default all your RSS feeds will be shortened (they'll contain only teasers)
 whereas your index page will still show complete posts. You can change
 this behaviour with your ``conf.py``: ``INDEX_TEASERS`` defines whether index
-page should display the whole contents or only teasers. ``RSS_TEASERS``
-works the same way for your RSS feeds.
+page should display the whole contents or only teasers. ``FEED_TEASERS``
+works the same way for your Atom and RSS feeds.
 
 By default, teasers will include a "read more" link at the end. If you want to
 change that text, you can use a custom teaser:

diff --git a/nikola/conf.py.in b/nikola/conf.py.in
@@ -425,9 +425,6 @@ HIDDEN_AUTHORS = ['Guest']
 # output / TRANSLATION[lang] / RSS_PATH / rss.xml
 # RSS_PATH = ""
 
-# Number of posts in RSS feeds
-# FEED_LENGTH = 10
-
 # Slug the Tag URL. Easier for users to type, special characters are
 # often removed or replaced as well.
 # SLUG_TAG_PATH = True
@@ -677,18 +674,18 @@ IMAGE_FOLDERS = {'images': 'images'}
 
 # 'Read more...' for the index page, if INDEX_TEASERS is True (translatable)
 INDEX_READ_MORE_LINK = ${INDEX_READ_MORE_LINK}
-# 'Read more...' for the RSS_FEED, if RSS_TEASERS is True (translatable)
-RSS_READ_MORE_LINK = ${RSS_READ_MORE_LINK}
+# 'Read more...' for the RSS_FEED, if FEED_TEASERS is True (translatable)
+FEED_READ_MORE_LINK = ${FEED_READ_MORE_LINK}
 
-# Append a URL query to the RSS_READ_MORE_LINK in Atom and RSS feeds. Advanced
+# Append a URL query to the FEED_READ_MORE_LINK in Atom and RSS feeds. Advanced
 # option used for traffic source tracking.
 # Minimum example for use with Piwik: "pk_campaign=feed"
 # The following tags exist and are replaced for you:
 # {feedRelUri}                  A relative link to the feed.
 # {feedFormat}                  The name of the syndication format.
 # Example using replacement for use with Google Analytics:
 # "utm_source={feedRelUri}&utm_medium=nikola_feed&utm_campaign={feedFormat}_feed"
-RSS_LINKS_APPEND_QUERY = False
+FEED_LINKS_APPEND_QUERY = False
 
 # A HTML fragment describing the license, for the sidebar.
 # (translatable)
@@ -892,22 +889,32 @@ MARKDOWN_EXTENSIONS = ['fenced_code', 'codehilite', 'extra']
 # them. Generate Atom for tags by setting TAG_PAGES_ARE_INDEXES to True.
 # Atom feeds are built based on INDEX_DISPLAY_POST_COUNT and not FEED_LENGTH
 # Switch between plain-text summaries and full HTML content using the
-# RSS_TEASER option. RSS_LINKS_APPEND_QUERY is also respected. Atom feeds
+# FEED_TEASER option. FEED_LINKS_APPEND_QUERY is also respected. Atom feeds
 # are generated even for old indexes and have pagination link relations
 # between each other. Old Atom feeds with no changes are marked as archived.
 # GENERATE_ATOM = False
 
+# Only inlclude teasers in Atom and RSS feeds. Disabling include the full
+# content. Defaults to True.
+# FEED_TEASERS = True
+
+# Strip HTML from Atom annd RSS feed summaries and content. Defaults to False.
+# FEED_PLAIN = False
+
+# Number of posts in Atom and RSS feeds.
+# FEED_LENGTH = 10
+
+# Inclue preview image as a <figure><img></figure> at the top of the entry.
+# Requires FEED_PLAIN = False. If the preview image is found in the content,
+# it will not be included again. Image will be included as-is, aim to optmize
+# the image source for Feedly, Apple News, Flipboard, and other popular clients.
+# FEED_PREVIEWIMAGE = True
+
 # RSS_LINK is a HTML fragment to link the RSS or Atom feeds. If set to None,
 # the base.tmpl will use the feed Nikola generates. However, you may want to
 # change it for a FeedBurner feed or something else.
 # RSS_LINK = None
 
-# Show teasers (instead of full posts) in feeds? Defaults to True.
-# RSS_TEASERS = True
-
-# Strip HTML in the RSS feed? Default to False
-# RSS_PLAIN = False
-
 # A search form to search this site, for the sidebar. You can use a Google
 # custom search (https://www.google.com/cse/)
 # Or a DuckDuckGo search: https://duckduckgo.com/search_box.html

diff --git a/nikola/nikola.py b/nikola/nikola.py
@@ -80,7 +80,7 @@
 
 # Default "Read more..." link
 DEFAULT_INDEX_READ_MORE_LINK = '<p class="more"><a href="{link}">{read_more}…</a></p>'
-DEFAULT_RSS_READ_MORE_LINK = '<p><a href="{link}">{read_more}…</a> ({min_remaining_read})</p>'
+DEFAULT_FEED_READ_MORE_LINK = '<p><a href="{link}">{read_more}…</a> ({min_remaining_read})</p>'
 
 # Default pattern for translation files' names
 DEFAULT_TRANSLATIONS_PATTERN = '{path}.{lang}.{ext}'
@@ -450,16 +450,17 @@ def __init__(self, **config):
             'PRETTY_URLS': False,
             'FUTURE_IS_NOW': False,
             'INDEX_READ_MORE_LINK': DEFAULT_INDEX_READ_MORE_LINK,
-            'RSS_READ_MORE_LINK': DEFAULT_RSS_READ_MORE_LINK,
-            'RSS_LINKS_APPEND_QUERY': False,
             'REDIRECTIONS': [],
             'ROBOTS_EXCLUSIONS': [],
             'GENERATE_ATOM': False,
+            'FEED_TEASERS': True,
+            'FEED_PLAIN': False,
+            'FEED_PREVIEWIMAGE': True,
+            'FEED_READ_MORE_LINK': DEFAULT_FEED_READ_MORE_LINK,
+            'FEED_LINKS_APPEND_QUERY': False,
             'GENERATE_RSS': True,
             'RSS_LINK': None,
             'RSS_PATH': '',
-            'RSS_PLAIN': False,
-            'RSS_TEASERS': True,
             'SASS_COMPILER': 'sass',
             'SASS_OPTIONS': [],
             'SEARCH_FORM': '',
@@ -543,7 +544,7 @@ def __init__(self, **config):
                                       'EXTRA_HEAD_DATA',
                                       'NAVIGATION_LINKS',
                                       'INDEX_READ_MORE_LINK',
-                                      'RSS_READ_MORE_LINK',
+                                      'FEED_READ_MORE_LINK',
                                       'INDEXES_TITLE',
                                       'POSTS_SECTION_COLORS',
                                       'POSTS_SECTION_DESCRIPTIONS',
@@ -608,6 +609,38 @@ def __init__(self, **config):
         for i1, i2, i3 in self.config['PAGES']:
             self.config['post_pages'].append([i1, i2, i3, False])
 
+        # RSS_TEASERS has been replaced with FEED_TEASERS
+        # TODO: remove on v8
+        if 'RSS_TEASERS' in config:
+            utils.LOGGER.warn('The RSS_TEASERS option is deprecated, use FEED_TEASERS instead.')
+            if 'FEED_TEASERS' in config:
+                utils.LOGGER.warn('FEED_TEASERS conflicts with RSS_TEASERS, ignoring RSS_TEASERS.')
+            self.config['FEED_TEASERS'] = config['RSS_TEASERS']
+
+        # RSS_PLAIN has been replaced with FEED_PLAIN
+        # TODO: remove on v8
+        if 'RSS_PLAIN' in config:
+            utils.LOGGER.warn('The RSS_PLAIN option is deprecated, use FEED_PLAIN instead.')
+            if 'FEED_PLAIN' in config:
+                utils.LOGGER.warn('FEED_PLIN conflicts with RSS_PLAIN, ignoring RSS_PLAIN.')
+            self.config['FEED_PLAIN'] = config['RSS_PLAIN']
+
+        # RSS_LINKS_APPEND_QUERY has been replaced with FEED_LINKS_APPEND_QUERY
+        # TODO: remove on v8
+        if 'RSS_LINKS_APPEND_QUERY' in config:
+            utils.LOGGER.warn('The RSS_LINKS_APPEND_QUERY option is deprecated, use FEED_LINKS_APPEND_QUERY instead.')
+            if 'FEED_TEASERS' in config:
+                utils.LOGGER.warn('FEED_LINKS_APPEND_QUERY conflicts with RSS_LINKS_APPEND_QUERY, ignoring RSS_LINKS_APPEND_QUERY.')
+            self.config['FEED_LINKS_APPEND_QUERY'] = utils.TranslatableSetting('FEED_LINKS_APPEND_QUERY', config['RSS_LINKS_APPEND_QUERY'], self.config['TRANSLATIONS'])
+
+        # RSS_READ_MORE_LINK has been replaced with FEED_READ_MORE_LINK
+        # TODO: remove on v8
+        if 'RSS_READ_MORE_LINK' in config:
+            utils.LOGGER.warn('The RSS_READ_MORE_LINK option is deprecated, use FEED_READ_MORE_LINK instead.')
+            if 'FEED_READ_MORE_LINK' in config:
+                utils.LOGGER.warn('FEED_READ_MORE_LINK conflicts with RSS_READ_MORE_LINK, ignoring RSS_READ_MORE_LINK')
+            self.config['FEED_READ_MORE_LINK'] = utils.TranslatableSetting('FEED_READ_MORE_LINK', config['RSS_READ_MORE_LINK'], self.config['TRANSLATIONS'])
+
         # DEFAULT_TRANSLATIONS_PATTERN was changed from "p.e.l" to "p.l.e"
         # TODO: remove on v8
         if 'TRANSLATIONS_PATTERN' not in self.config:
@@ -1280,10 +1313,12 @@ def generic_rss_renderer(self, lang, title, link, description, timeline, output_
 
         for post in timeline[:feed_length]:
             data = post.text(lang, teaser_only=rss_teasers, strip_html=rss_plain,
-                             rss_read_more_link=True, rss_links_append_query=feed_append_query)
+                             feed_read_more_link=True, feed_links_append_query=feed_append_query)
             if feed_url is not None and data:
                 # Massage the post's HTML (unless plain)
                 if not rss_plain:
+                    if self.config["FEED_PREVIEWIMAGE"] and 'previewimage' in post.meta[lang] and post.meta[lang]['previewimage'] not in data:
+                        data = "<figure><img src=\"{}\"></figure> {}".format(post.meta[lang]['previewimage'], data)
                     # FIXME: this is duplicated with code in Post.text()
                     try:
                         doc = lxml.html.document_fromstring(data)
@@ -1835,8 +1870,6 @@ def atom_link(link_rel, link_type, link_href):
         nslist = {}
         if context["is_feed_stale"] or "feedpagenum" in context and (not context["feedpagenum"] == context["feedpagecount"] - 1 and not context["feedpagenum"] == 0):
             nslist["fh"] = "http://purl.org/syndication/history/1.0"
-        if not self.config["RSS_TEASERS"]:
-            nslist["xh"] = "http://www.w3.org/1999/xhtml"
         feed_xsl_link = self.abs_link("/assets/xml/atom.xsl")
         feed_root = lxml.etree.Element("feed", nsmap=nslist)
         feed_root.addprevious(lxml.etree.ProcessingInstruction(
@@ -1881,31 +1914,45 @@ def atom_link(link_rel, link_type, link_href):
         feed_generator.text = "Nikola"
 
         feed_append_query = None
-        if self.config["RSS_LINKS_APPEND_QUERY"]:
-            feed_append_query = self.config["RSS_LINKS_APPEND_QUERY"].format(
+        if self.config["FEED_LINKS_APPEND_QUERY"]:
+            feed_append_query = self.config["FEED_LINKS_APPEND_QUERY"].format(
                 feedRelUri=context["feedlink"],
                 feedFormat="atom")
 
-        for post in posts:
-            data = post.text(lang, teaser_only=self.config["RSS_TEASERS"], strip_html=self.config["RSS_TEASERS"],
-                             rss_read_more_link=True, rss_links_append_query=feed_append_query)
-            if not self.config["RSS_TEASERS"]:
+        def atom_post_text(post, text):
+            if not self.config["FEED_PLAIN"]:
+                if self.config["FEED_PREVIEWIMAGE"] and 'previewimage' in post.meta[lang] and post.meta[lang]['previewimage'] not in text:
+                    text = "<figure><img src=\"{}\"></figure> {}".format(post.meta[lang]['previewimage'], text)
+
                 # FIXME: this is duplicated with code in Post.text() and generic_rss_renderer
                 try:
-                    doc = lxml.html.document_fromstring(data)
+                    doc = lxml.html.document_fromstring(text)
                     doc.rewrite_links(lambda dst: self.url_replacer(post.permalink(lang), dst, lang, 'absolute'))
                     try:
                         body = doc.body
-                        data = (body.text or '') + ''.join(
+                        text = (body.text or '') + ''.join(
                             [lxml.html.tostring(child, encoding='unicode')
                                 for child in body.iterchildren()])
                     except IndexError:  # No body there, it happens sometimes
-                        data = ''
+                        text = ''
                 except lxml.etree.ParserError as e:
                     if str(e) == "Document is empty":
-                        data = ""
+                        text = ""
                     else:  # let other errors raise
                         raise(e)
+            return text
+
+        for post in posts:
+            summary = atom_post_text(post, post.text(lang, teaser_only=True,
+                                                     strip_html=self.config["FEED_PLAIN"],
+                                                     feed_read_more_link=True,
+                                                     feed_links_append_query=feed_append_query))
+            content = None
+            if not self.config["FEED_TEASERS"]:
+                content = atom_post_text(post, post.text(lang, teaser_only=self.config["FEED_TEASERS"],
+                                                         strip_html=self.config["FEED_PLAIN"],
+                                                         feed_read_more_link=True,
+                                                         feed_links_append_query=feed_append_query))
 
             entry_root = lxml.etree.SubElement(feed_root, "entry")
             entry_title = lxml.etree.SubElement(entry_root, "title")
@@ -1922,14 +1969,19 @@ def atom_link(link_rel, link_type, link_href):
             entry_root.append(atom_link("alternate", "text/html",
                               post.permalink(lang, absolute=True,
                                              query=feed_append_query)))
-            if self.config["RSS_TEASERS"]:
-                entry_summary = lxml.etree.SubElement(entry_root, "summary")
-                entry_summary.text = data
+            entry_summary = lxml.etree.SubElement(entry_root, "summary")
+            if not self.config["FEED_PLAIN"]:
+                entry_summary.set("type", "html")
             else:
+                entry_summary.set("type", "text")
+            entry_summary.text = summary
+            if content:
                 entry_content = lxml.etree.SubElement(entry_root, "content")
-                entry_content.set("type", "xhtml")
-                entry_content_nsdiv = lxml.etree.SubElement(entry_content, "{http://www.w3.org/1999/xhtml}div")
-                entry_content_nsdiv.text = data
+                if not self.config["FEED_PLAIN"]:
+                    entry_content.set("type", "html")
+                else:
+                    entry_content.set("type", "text")
+                entry_content.text = content
             for category in post.tags_for_language(lang):
                 entry_category = lxml.etree.SubElement(entry_root, "category")
                 entry_category.set("term", utils.slugify(category))
@@ -1980,8 +2032,7 @@ def generic_index_renderer(self, lang, posts, indexes_title, template_name, cont
         kw['indexes_prety_page_url'] = self.config["INDEXES_PRETTY_PAGE_URL"]
         kw['demote_headers'] = self.config['DEMOTE_HEADERS']
         kw['generate_atom'] = self.config["GENERATE_ATOM"]
-        kw['feed_link_append_query'] = self.config["RSS_LINKS_APPEND_QUERY"]
-        kw['feed_teasers'] = self.config["RSS_TEASERS"]
+        kw['feed_link_append_query'] = self.config["FEED_LINKS_APPEND_QUERY"]
         kw['currentfeed'] = None
 
         # Split in smaller lists
@@ -2072,6 +2123,9 @@ def generic_index_renderer(self, lang, posts, indexes_title, template_name, cont
                 context["currentfeedlink"] = kw["currentfeed"]
                 context["feedpagenum"] = i
                 context["feedpagecount"] = num_pages
+                kw['feed_teasers'] = self.config['FEED_TEASERS']
+                kw['feed_plain'] = self.config['FEED_PLAIN']
+                kw['feed_previewimage'] = self.config['FEED_PREVIEWIMAGE']
                 atom_task = {
                     "basename": basename,
                     "name": atom_output_name,

diff --git a/nikola/plugins/command/check.py b/nikola/plugins/command/check.py
@@ -227,7 +227,7 @@ def analyze(self, fname, find_sources=False, check_remote=False):
                 d = lxml.etree.parse(filename)
                 link_elements = lxml.html.fromstring('<html/>')
                 for elm in d.findall('*//{http://www.w3.org/2005/Atom}link'):
-                    feed_link = elm.attrib['href'].split('?')[0].strip()  # strip RSS_LINKS_APPEND_QUERY
+                    feed_link = elm.attrib['href'].split('?')[0].strip()  # strip FEED_LINKS_APPEND_QUERY
                     link_elements.append(lxml.etree.Element('a', href=feed_link))
                 link_elements = list(link_elements.iterlinks())
             elif filename.endswith('sitemap.xml') or filename.endswith('sitemapindex.xml'):

diff --git a/nikola/plugins/command/init.py b/nikola/plugins/command/init.py
@@ -41,7 +41,7 @@
 import tarfile
 
 import nikola
-from nikola.nikola import DEFAULT_TRANSLATIONS_PATTERN, DEFAULT_INDEX_READ_MORE_LINK, DEFAULT_RSS_READ_MORE_LINK, LEGAL_VALUES, urlsplit, urlunsplit
+from nikola.nikola import DEFAULT_TRANSLATIONS_PATTERN, DEFAULT_INDEX_READ_MORE_LINK, DEFAULT_FEED_READ_MORE_LINK, LEGAL_VALUES, urlsplit, urlunsplit
 from nikola.plugin_categories import Command
 from nikola.utils import ask, ask_yesno, get_logger, makedirs, STDERR_HANDLER, load_messages
 from nikola.packages.tzlocal import get_localzone
@@ -71,7 +71,7 @@
     'CATEGORY_OUTPUT_FLAT_HIERARCHY': False,
     'TRANSLATIONS_PATTERN': DEFAULT_TRANSLATIONS_PATTERN,
     'INDEX_READ_MORE_LINK': DEFAULT_INDEX_READ_MORE_LINK,
-    'RSS_READ_MORE_LINK': DEFAULT_RSS_READ_MORE_LINK,
+    'FEED_READ_MORE_LINK': DEFAULT_FEED_READ_MORE_LINK,
     'POSTS': """(
     ("posts/*.rst", "posts", "post.tmpl"),
     ("posts/*.txt", "posts", "post.tmpl"),
@@ -210,10 +210,10 @@ def prepare_config(config):
     """Parse sample config with JSON."""
     p = config.copy()
     p.update({k: json.dumps(v, ensure_ascii=False) for k, v in p.items()
-             if k not in ('POSTS', 'PAGES', 'COMPILERS', 'TRANSLATIONS', 'NAVIGATION_LINKS', '_SUPPORTED_LANGUAGES', '_SUPPORTED_COMMENT_SYSTEMS', 'INDEX_READ_MORE_LINK', 'RSS_READ_MORE_LINK')})
+             if k not in ('POSTS', 'PAGES', 'COMPILERS', 'TRANSLATIONS', 'NAVIGATION_LINKS', '_SUPPORTED_LANGUAGES', '_SUPPORTED_COMMENT_SYSTEMS', 'INDEX_READ_MORE_LINK', 'FEED_READ_MORE_LINK')})
     # READ_MORE_LINKs require some special treatment.
     p['INDEX_READ_MORE_LINK'] = "'" + p['INDEX_READ_MORE_LINK'].replace("'", "\\'") + "'"
-    p['RSS_READ_MORE_LINK'] = "'" + p['RSS_READ_MORE_LINK'].replace("'", "\\'") + "'"
+    p['FEED_READ_MORE_LINK'] = "'" + p['FEED_READ_MORE_LINK'].replace("'", "\\'") + "'"
     # fix booleans and None
     p.update({k: str(v) for k, v in config.items() if isinstance(v, bool) or v is None})
     return p