For unknown reasons, it seems some environments emit a `DataChange::Any` event,
rather than specifying content or size, when a file is edited. As an example:

    [src/fs_utils.rs:72:9] &event = DebouncedEvent {
        event: Event {
            kind: Modify(
                Data(
                    Any,
                ),
            ),
            paths: [
                "/home/redacted/content/blog/2024-06-23_example.md",
            ],
            attr:tracker: None,
            attr:flag: None,
            attr:info: None,
            attr:source: None,
        },
        time: Instant {
            tv_sec: 78544,
            tv_nsec: 936740729,
        },
    }

Consequently, because this isn't treated by Zola as a modification, the
site is not rebuilt, which regresses on previous behavior.

This patch fixes this particular case by treating any data modification
events as a modification.

Closes: #2536

* Metainfo: Add 0.19.1

* Metainfo: Fix date for 0.19.0

I entered the wrong date originally.

Fixes 7d612c6

* Update zola.metainfo.xml

---------

Co-authored-by: Vincent Prouillet <balthek@gmail.com>

Co-authored-by: GitHub Action <action@github.com>

…#2569)

When installing `zola` from repositories,
the bash, zsh and fish autocompletion packages
are co-installed as recommended packages.

Package details for all distributions:
https://software.opensuse.org/package/zola?baseproject=ALL

…snippet. (#2572)

…t) (#2574)

Co-authored-by: GitHub Action <action@github.com>

Co-authored-by: GitHub Action <action@github.com>

Co-authored-by: GitHub Action <action@github.com>

* Refine YAML date regex

This commit does a few changes:

- Introduce a new regex
  - it is a bit off-spec (it allows one-digit months and days in date-only mode)
  - uses named groups
  - avoids group duplication
- parses offset once

Fixes #2538

* Fix nanosecond parsing

* Rename variables for brewity

* Add tests

Closes #2563

* Add feature to disable robots.txt and sitemap.xml from the config file.

Addresses feature request #2248

* Add documentation for no_sitemap & no_robots

* Rename no_robots and no_sitemap into generate_robots_txt and generate_sitemap (default to true)

* fix rustfmt issues

* Change documentation to show defaults

* Update documentation for the fields generate_sitemaps (resp. robots_txt), now that the default is true and false is needed to disable

…2575)

- Some error messages refer to `get_taxonomy_term_by_name` instead of
`get_taxonomy_term`
- If a term cannot be found in a taxonomy, the error message
erroneously pretends that the taxonomy itself is unknown instead of
the term.

Cleanup code by using `Option::is_some_and()` and
`Option::as_deref()` instead of recoding them.

* update docs

* add config option, logic, and tests

* remove config option

…ion) (#2606)

* Fix shortcode/continue-reading parsing with inline HTML (smaller version)

* Remove inline <!-- more -->

Describe how to enable 'class' syntax highlighting

Co-authored-by: Keats <680355+Keats@users.noreply.github.com>

Supersedes #3028

Closes #3069

Separate search index item collection and serialization.

Closes #3093

Updated changelog to reflect breaking changes in version 0.22.0.

Co-authored-by: Keats <680355+Keats@users.noreply.github.com>

Signed-off-by: zyxhere💭 <zyx@envs.net>

Co-authored-by: Keats <680355+Keats@users.noreply.github.com>

)

* docs: Add link to TextMate themes in configuration and highlighting docs #3088

* docs: revise TextMate themes link wording

I guess it was a typo.

)

Signed-off-by: SAY-5 <say.apm35@gmail.com>

The documentation (and some doc comments) state that `lower` corresponds
to earlier pages and `higher` to later pages when sorting by date. This
is the opposite of what the code actually does.