Pelican-kappa-nero is forked from pelican-bootstrap4 and customized. It is a dark theme. This theme is not aimed to be a complete pelican-bootstrap3 rewrite to support Bootstrap 4. You are welcome to clone it or reference it create a generic pelican-bootstrap4 theme.
My previous theme is called pelican-bootstrap3-nero. I created this theme so that I may use Bootstrap 4.
Use it at your own risk.
-
Known problems
- If main content column uses col-xl-auto, the tables or converted code block with line numbers by default are not table responsive
- DISABLE_SIDEBAR_TITLE_ICONS would not work because of code changes
- This theme uses Bootstrap 4. But the codes was copied from old theme, which uses Bootstrap 3. So some parts of the theme may not work
- DISPLAY_BREADCRUMBS is not working. Related code was not migrated to Bootstrap 4
- Changed from Font Awesome to Feather Icons. Some icons like Reddit are misssing. Some codes that use the old icons are not updated to use the new icons.
-
Limitation
- For the tags, I had replaced a tag with spaces by dashes for the filename when generating the html. It may cause problem if you have a same tag with dash. Like 'flash storage' and 'flash-storage'.
-
Other points to note
- rel=prev/next implementation. Looks like the official method for Google Search engine is to add it in the head section. But could be a problem for Pelican to get the URL of previous/next page. The current rel=prev/next is added inside the hyperlink tag '<a>' during pagination link generation. It is done according to the HTML specification.
-
Options that must be set in pelicanconf.py:
- PLUGINS = ['i18n_subsites'] (i18n is required since pelican-bootstrap3)
- JINJA_ENVIRONMENT = {'extensions': ['jinja2.ext.i18n', 'jinja2.ext.do']}
-
No need to set BOOTSTRAP_THEME, bootstrap.min would be used.
-
We have new settings specific to this theme, also check CHANGELOG.rst
-
Setup Cookie Consent 2
- COOKIE_CONSENT2 = True
- Set your message to display in COOKIE_CONSENT2_MESSAGE
- Set your URL to privacy policy with COOKIE_CONSENT2_PRIVACY_URL
-
Setup your account id from IntenseDebate
- INTENSEDEBATE_ACCOUNT = 'Your account id from IntenseDebate'
-
Enable InteseDebate comment system and Google Analytics only if Cookie Consent 2 was consented and accepted
- GOOGLE_ANALYTICS_UNIVERSAL_CHECK_CONSENT = True
- Also need to configure parameters according to your setting
- GOOGLE_ANALYTICS_UNIVERSAL = 'UA-XXXXXX-Y'
- GOOGLE_ANALYTICS_UNIVERSAL_PROPERTY = 'auto'
-
Options for sidebar bottom
- OPTIONAL_SIDEBAR_BOTTOM = True
- OPTIONAL_SIDEBAR_BOTTOM_TITLE = 'Disclaimer'
- OPTIONAL_SIDEBAR_BOTTOM_ID= 'disclaimer'
- OPTIONAL_SIDEBAR_BOTTOM_TEXT = 'Replace by your text for disclaimer'
-
Display at bottom of articles. If you have ad-block software it may not be displayed Enable it in pelicanconf.py, TWITTER_CUSTOM = True
-
JSON-LD
- JSON_LD = True
- Based on existing work from Flex. Thanks to Alexandre
- Based on existing work from Emoji-gen. Thanks to Emoj-Gen and Pine.
- For base / includes/json_ld.html, If enabled, there would be JSON-LD blocks for
- Type Website and WebPage for all html
- Type BlogPosting for all articles
- A limitation of Pelican: currently there is no easy way to get the current URL of the page. So some pages generated for tags and etc could not put the current URL into the JSON-LD. If Pelican provides a Jinja2 variable for this, we would try to update the JSON-LD template.
-
JSON-LD, article image and publisher logo are required by Google schema checking
- SITE_LOGO = 'images/your-image.jpg'
- PUBLISHER_LOGO = 'images/your-image.jpg'
- Recommended to set 'CATEGORY_URL' explictly. The include/json_ld.html refers to that variable. If it is not set, it use a default value which may not be same as your system setting
-
JSON_LD_BREADCRUMBLIST = True
- If both JSON_LD is also enabled:
- Add type BreadcrumbList for all html
- Note: there is an option DISPLAY_BREADCRUMBS and have visible breadcrumbs and actually added some schema for breadcrumbs. Google search engine would take the value from that. I do not know clearly what is performed by search engine if both Breadcrumb support is enabled in both JSON-LD BREADCRUME and DISPLAY_BREADCRUMS are enabled
-
JSON_LD_AUTHOR_SAME_AS = '"https://URL-1", "https://URL-2"'
- Update it to your links for the sameAs attribute for author. Assume there is only one author for this blog and multiple authors are not supported
-
JSON_LD_PUBLISHER_SAME_AS = '"https://URL-1", "https://URL-2"
- Similar to JSON_LD_AUTHOR_SAME_AS
-
Copy button on the top right corner of all "div.highlight pre" elements
- SPHINX_COPYBUTTON = True
- Adopted from sphinx copybutton
- Credits for the copy button goes to Chris Holdgraf
-
Added an option OTHER_PAGE_METATAG_DESCRIPTION
- It is for html file that are not page or article
- That would a meta tag description to the head section
-
Include a KMP search js for tipue
- TIPUESEARCH_KMP = True
- Credit goes to Psychoinformatics.de
-
Meilisearch
- MEILISEARCH = True
- MEILISEARCH_SERVER_URL = 'https://your meilisearch app url'
- MEILISEARCH_API_KEY = 'KEY', note public key is enough
- MEILISEARCH_INDEX_UID = 'docs', replace by your index uid
-
New option to set meta tag for robots or crawlers
- Typical setting may be like 'noindex,noimageindex,nosnippet,follow'
- METATAG_ROBOTS_AUTHOR , for author.html and authors.html
- METATAG_ROBOTS_TAG , for tag.html and tags.html
- METATAG_ROBOTS_SEARCH , for meilisearch.html and tipuesearch.html
- METATAG_ROBOTS_CATGORY , for individual category
- METATAG_ROBOTS_CATGORIES_PAGE , for categories.html only
- METATAG_ROBOTS_ARCHIVES_PAGE , for archives.html only
-
URL for cover image, for JSON-LD, Facebook OG and Twitter Cards. This is special case for my blog
- For articles, I use this as URL https://hostname/posts/category/YYYY/MM/DD/slug/
- Also, I put the cover image of an article in the same directory as the source rst file. In that case, I add this
- :article_cover_filename: your-article-cover-image.jpg' metadata in the rst
- So I could use JINJA2 template to generate the correct URL for the cover image for JSON-LD, Facebook OG and Twitter cards
- If the URL format is different (like https://hostname/category/YYYY/MM/DD/slug.html), then we should use another method (like article.og_image metadata) to specify the URL for cover image
- Added :display_cover_in_summary: metadata/boolean checking, used must use with 'article_cover_filename" together