Skipping URLs in verbatim elements didn't take nested
elements into consideration, which were not verbatim.
For instance, the following HTML snippet would yield
`https://example.com` in non-verbatim mode, even if
it is nested inside a verbatim `<pre>` element:
```html
<pre><a href="https://example.com">link</a></pre>
```
This commit fixes the behavior for both `html5gum` and
`html5ever`.
Note that nested verbatim elements of the same kind
still are not handled correctly.
For instance, the following HTML snippet would still yield
`https://example.com`:
```html
<pre>
<pre></pre>
<a href="https://example.com">link</a>
</pre>
```
The reason is that we currently only keep track of a single
verbatim element and not a stack of elements, which we
would need to unwind and resolve the situation.
Fixes https://github.com/lycheeverse/lychee/issues/986.
I don't see the reason why publish-check has to wait for the other jobs to complete. After all, it's yet another "lint".
Running it in parallel might shave of a significant amount of time from our CI builds.
crates.io sometimes takes a bit until new releases
are available through the API.
This can break the binary release because it depends
on the lychee binary. Therefore let's wait a bit
instead of immediately publishing to reduce
the risk that this is happening.
Since github.repository_owner refers to the base repository on pull request events, while secrets need to exist on the head repository, the Docker Hub login fails for pull requests opened from forks. This commit assures that this step in case of pull request events only runs for internal pull requests, i.e. when head and base repository are the same.
For the actual Docker Hub upload, no change is required: The build can run in every case, the upload is not done on pull request events in general.
Signed-off-by: MichaIng <micha@dietpi.com>
* Cache `absolute_path` to decrease allocations
While profiling local file handling, I noticed that resolving paths was taking a
significant amount of time. It also caused quite a few allocations.
By caching the path and using a constant value for the current
directory, we can reduce the number of allocs by quite a lot.
For example, when testing on the sentry documentation, we do 50,4%
less allocations in total now. That's just a single test-case of course,
but it's probably also helping in many other cases as well.
* Defer to_string for attr.value to reduce allocs
* Use Tendrils instead of Strings for parsing (another ~1.5% less allocs)
* Move option parsing code into separate module
* Handle base dir more correctly
* Temporarily disable dry run
