| Commit message (Collapse) | Author |
|
This adds the responsive-image plugin directly to the project, including
its source code, documentation, and tests, allowing it to be maintained
locally.
It also configures responsive image sizes, updates gallery images to use
srcset, and preserves original image URLs for full-resolution viewing.
Markdown images gain lazy loading while retaining compatibility with the
existing image viewer.
|
|
This adds a native dialog for enlarging images throughout the site,
including gallery images with previous and next navigation.
The viewer supports keyboard navigation, image captions, and responsive
layouts, while preserving the existing behaviour of linked images and
interactive maps.
|
|
This adds navigation links between adjacent blog posts, allowing readers
to continue browsing without returning to the blog index. The navigation
uses a responsive two-column layout and is excluded from Pagefind
indexing.
|
|
This adds a project-local Lektor plugin that estimates reading time from
article text and images, using NLTK for word tokenization and decreasing
time weights for successive images.
The plugin exposes a `reading_time` template function and displays the
estimated reading time alongside other blog post metadata.
|
|
This standardizes indentation and line wrapping across the 404 page,
gallery, archive, category, and tag templates, and simplifies stylesheet
declarations.
It also replaces HTML entities with literal ellipsis characters in
pagination controls.
|
|
This adds a robots.txt file that allows all crawlers to access the site
and advertises the XML sitemap for easier discovery.
|
|
This removes redundant `type="text/css"` attributes from stylesheet
links in the base template and consolidates the main stylesheet
declaration onto a single line.
|
|
This adds a dedicated 404 page with a playful message that reflects the
blog's title and tagline, along with a colourful gradient heading and
animated thinking dots.
It also provides reduced-motion support, excludes the page from search
engine and Pagefind indexing, and directs visitors to the site's
navigation and search.
|
|
This adds a sitemap template that recursively lists site pages and their
URLs, with child pages sorted by path for consistent output.
It also registers `/sitemap.xml` as a non-discoverable Lektor page and
advertises the sitemap in the site's document head.
|
|
This removes hardcoded timestamps and timezone offsets from publication
and modification dates in Open Graph and JSON-LD metadata.
Dates are now formatted as `YYYY-MM-DD`, avoiding the need to specify an
arbitrary time or timezone for date-only information.
|
|
This applies Jinja's `tojson` filter to dynamically generated text
fields in the site's JSON-LD templates, including titles, descriptions,
keywords, and categories.
This ensures special characters are escaped correctly and prevents
invalid JSON in the generated structured data.
|
|
This adds publication dates, categories, and tags to Pagefind metadata,
enabling search results to be sorted by date and filtered by category or
tag.
It also excludes tables of contents, post metadata, and footer sections
from search indexing to keep results focused on the actual article
content.
|
|
This replaces the flex-based article header layout with floats, allowing
article content to flow alongside longer tables of contents instead of
leaving unused space beneath the header.
It also adjusts the responsive layout, clears floats before the article
footer, and removes the unused blog-specific stylesheet and its drop cap
styling.
|
|
This adds an email reply link to every blog post, allowing readers to
respond directly with a prefilled subject line. The email address is
encoded using HTML entities to discourage simple address harvesting.
It also reorganizes the article footer into separate sections for tags,
replies, and changelogs, with consistent typography and spacing.
|
|
This replaces the manual `updated_date` field with a changelog flow
field so post updates can be recorded as dated entries.
The blog template now derives the "Last updated" metadata from the
latest changelog entry and renders the full changelog in the article
footer, with dedicated styling for compact change notes.
|
|
This adds a project-local RSS plugin that generates RSS 2.0 feeds from
configured Lektor sources, including feed titles, item dates, authors,
body content, and stable permalink IDs.
It also adds latest and all-post RSS feed configurations and advertises
both feeds from the base template alongside the existing Atom feeds.
|
|
|
|
This adds the Pagefind component UI assets and wires the search modal
into the base template with a compact header trigger.
The integration marks searchable page bodies explicitly, exposes page
titles to Pagefind metadata, and adds site-specific styling so the
search modal, trigger, focus states, and keyboard-selected results match
the blog’s colour scheme.
|
|
This adds a project-local gallery plugin that builds deterministic rectangular
image layouts from attachment aspect ratios, with both in-process and disk-backed
layout caching.
It also adds gallery flow blocks, templates, and CSS so posts can include
responsive galleries with lazy-loaded images, configurable gaps, target ratios,
and the existing breakout/centering utilities.
|
|
This adds a trail flow block so posts can include attached GPX files as
interactive maps with route display, elevation charts, activity
summaries, and start/finish markers.
The trail assets are loaded only when needed, keeping the shared
templates clean while vendoring the Leaflet dependencies required for
offline, self-hosted map rendering.
|
|
This moves the “Tagged with” label into the tag list itself, so the
footer can wrap naturally like a paragraph instead of splitting the
label and tags onto separate alignment contexts.
The CSS now treats each tag item as inline text while keeping later
items together, which avoids awkward line breaks after the label.
|
|
This adds JSON-LD for the site, author, blog, and individual blog posts, so
search engines and other consumers can understand the blog structure more
clearly.
It also adds Microformats2 classes to blog posts and the about page, keeping
machine-readable author, post, category, date, and content data attached to the
visible HTML.
|
|
This adds a scalable SVG favicon alongside the existing PNG favicon and
exposes both through explicit icon links in the base template.
The SVG is used for browsers that support it, while the 32×32 PNG
remains as a fallback for clients that still expect a raster favicon.
|
|
This adds a description field to blog posts and uses it for the HTML
meta description and Open Graph description, with the blog tagline
remaining as the fallback.
Archive, category, and tag pages now define their own descriptions so
previews are more specific instead of all sharing the same generic site
tagline.
|
|
This replaces the placeholder Open Graph image with a dedicated 1200×630
asset and adds the SVG source for future edits.
The base metadata now describes the site and preview image more
accurately, while blog post timestamps use the local publishing timezone
instead of UTC.
|
|
This adds the Lektor project structure, models, templates, styling,
feeds, categories, tags, archives, custom Markdown heading handling, and
the initial content needed to make the blog usable.
|