Every time I remove the word “Hebrew” from the following line, the number flips to being on the opposite site of the “עברית” rtl word.
What is the usual way of mixing english and hebrew in a list? (I was going to say ‘proper’ but I suspect there is no proper way to mix RTL and LTR.)
Discourse behaves slightly differently:
and renders as:
עברית Hebrew 24 MiB PDF
changes to
עברית 24 MiB PDF
: [https://gramps-project.org/wiki/index.php/File:Gramps5.1UserManual_he.pdf עברית Hebrew 24 MiB PDF]
Which RTL override method works for this? My experiments with an LTR environment are confusing.
I’ll demonstrate:
In this case you just need to add LRM before the “24” so it’ll align to the right direction.
Interesting. Looking at the posting with Discourse’s editor shows no “LRM”. Is it invisible after being insert?
It has no visual representation other than the logical ordering of the letters.
I converted these both to hexadecimal code units (using this):
| Description | Text | UTF-8 Hex code units |
|---|---|---|
| 24 flipped version | עברית 24 MiB PDF | D7 A2 D7 91 D7 A8 D7 99 D7 AA 20 32 34 20 4D 69 42 20 50 44 46 |
| 24 with LRM | עברית 24 MiB PDF | D7 A2 D7 91 D7 A8 D7 99 D7 AA 20 E2 80 8E 32 34 20 4D 69 42 20 50 44 46 |
The changes are marked, although this is a Unicode definition the different rendering methods are being handled by the rendering library (QT, pango, harfbuzz, VTE, etc.)
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
