[openstreetmap/openstreetmap-website] Add dir="auto" to Markdown content for bidi (PR #5840)
Nitai Sasson
notifications at github.com
Wed Mar 26 19:21:22 UTC 2025
@NeatNit commented on this pull request.
> + # "li", # child; sometimes container
+
+ "table", # similar to Discourse, where the <div> parent of <table> gets dir="auto" (no parent div in kramdown)
+ # "td", # child
+ # "th", # child; not actually a thing in kramdown but hypothetically might be in the future
+
+ "dl",
+ # "dd", "dt", # child - since it's similar to a list, it has the same behavior as a list
+
+ "math" # don't know how this ends up, but dir="auto" is probably correct
+ ].each do |name|
+ define_method :"convert_#{name}" do |el, indent|
+ attr_bak = el.attr.dup # can't avoid mutating the attr hash, so make a backup
+ el.attr["dir"] = "auto" unless el.attr.key?("dir") # if by some miracle dir is already defined, don't override it
+ ret = super(el, indent)
+ el.attr.replace(attr_bak) # restore backup
> What happens if we don't restore the original attributes? Does the element get reused in a way that breaks things?
Truthfully, nothing. It's only a concern if you ever use a converter other than HtmlBidi. Since we don't, it doesn't matter, and we can mutate the Document freely.
> would just doing `el = el.dup` as the first thing to duplicate the element be a workable alternative?
No, because it would still point to the same `attr` hash, and mutating it would mutate the original. But as I said, it doesn't matter, so the whole backup-restore code can be removed if you wish.
--
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5840#discussion_r2014856583
You are receiving this because you are subscribed to this thread.
Message ID: <openstreetmap/openstreetmap-website/pull/5840/review/2718388322 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/rails-dev/attachments/20250326/52fcd7d2/attachment-0001.htm>
More information about the rails-dev
mailing list