Since Scribe launched, the Redirector extension config has needed
occasional attention. Using regular expressions to cover all edge cases
is difficult. After finding out that Scribe's current config can hang
websites, I decided that [LibRedirect] is likely a more robust
solution. It can rely on more than regular expressions, and is less
work to set up.
[LibRedirect]: https://libredirect.github.io/
Instead of providing long detailed instructions for how to configure
the Redirector extension, this provides a single json file that users
can import. I started by making a single file stored in the
`public/assets` directory, but then realized this was a regression since
the instructions were customized to each domain. Instead I can use
Lucky's [data] response to dynamically build the JSON config.
[data]:
https://luckyframework.org/guides/http-and-routing/request-and-response#
handling-responses
Previously the link on the error page was only linking to the path
component of the url, e.g. `/search` but ignoring any query params e.g.
`/search?q=hello`. This uses the HTTP::Request `resource` method which
appears to capture both.
Instead of showing the default Lucky error page, the styles now match
Scribe. In addition, if a URL can't be parsed, Scribe gives some
information as to why this might be (that Scribe can only deal with an
article pages)
This specifies advanced options for configuring the Redirector
extension. If everything is let on (like images) things will break
(like images). It also improves the regular expression a bit to account
for the image CDN
Co-authored-by: Austin Huang <im@austinhuang.me>
This patch further improves the proposed pattern for the Redirector
extension. In contrast to the old pattern, …
* … it will redirect the URL https://medium.com.
* … it will *not* redirect URLs with top-level domains like mediumXcom.
(This point is purely theoretical, but it makes the regular expression
more correct and consistent.)
* … it will *not* redirect URLs like https://link.medium.com/AXEtCilplkb
which Scribe currently cannot handle. These are shortened URLs that
users get when they use the Twitter button on Medium to share a post.
In order to implement the last point (not matching link.medium.com), the
pattern uses negative lookbehind. This feature of regular expressions is
supported by all recent browsers for which Redirector is available
(Firefox, Chrome, Edge, Opera)[^1], including the current version of
Firefox ESR (Extended Stability Release).
[^1]: https://caniuse.com/js-regexp-lookbehind
In the current redirector example, "scribe.rip" is hardcoded as the
destination. This patch simply changes that to use the app_domain
environment variable, so people wanting to use a community instance
aren't mistakenly redirected to the main scribe.rip instance.
The old pattern matches all host names that end with medium.com. The new
pattern matches only medium.com and its sub-domains. For example, the
old pattern would have matched
https://foomedium.com/@user/post-123456abcdef.
The subtitle has been removed because it's difficult to find and error
prone to guess at. It is somewhat accessible from the post's
previewContent field in GraphQL but that can be truncated.
Right now this links to the user's medium page. It may link to an
internal page in the future.
Instead of the Page taking the author as a string, it now takes a
PostResponse::Creator object. The Articles::ShowPage then converts the
Creator (a name and user_id) to an author link.
Finally, I did some refactoring of UserAnchor (which I thought I was
going to use for this) to change it's userId attribute to user_id as is
Crystal convention.
Also wrap the content in an article for semantic formatting
tufte.css requires that content is wrapped in an <article> and at least
one <section>. There's no way of determining new semantic sections so
there is only one.
The API responds with a bunch of paragraphs which the client converts
into Paragraph objects.
This turns the paragraphs in a PostResponse's Paragraph objects into the
form needed to render them on a page. This includes converting flat list
elements into list elements nested by a UL. And adding a limited markups
along the way.
The array of paragraphs is passed to a recursive function. The function
takes the first paragraph and either wraps the (marked up) contents in a
container tag (like Paragraph or Heading3), and then moves onto the next
tag. If it finds a list, it starts parsing the next paragraphs as a list
instead.
Originally, this was implemented like so:
```crystal
paragraph = paragraphs.shift
if list?
convert_list([paragraph] + paragraphs)
end
```
However, passing the `paragraphs` after adding it to the already shifted
`paragraph` creates a new object. This means `paragraphs` won't be
mutated and once the list is parsed, it starts with the next element of
the list. Instead, the element is `shift`ed inside each converter.
```crystal
if paragraphs.first == list?
convert_list(paragraphs)
end
def convert_list(paragraphs)
paragraph = paragraphs.shift
# ...
end
```
When rendering, there is an Empty and Container object. These represent
a kind of "null object" for both leafs and parent objects respectively.
They should never actually render. Emptys are filtered out, and
Containers are never created explicitly but this will make the types
pass.
IFrames are a bit of a special case. Each IFrame has custom data on it
that this system would need to be aware of. For now, instead of trying
to parse the seemingly large number of iframe variations and dealing
with embedded iframe problems, this will just keep track of the source
page URL and send the user there with a link.
The basic idea here is to fetch the post with the medium API, parse the
JSON into types, and then re-display the content. We also have to fetch
each media object as a REST call to get things like embeded iframes.
Edward Loveall
Martin Puppe
Amolith