Locale Display Name algorithm for time zones

[Kip Cole]

I have mostly completed implementation of the locale display 
name algorithm but am unable to identify the resolution of 
two of the test cases in testData/localeIdentifiers/localeDisplayName.txt.

As always, hints, suggestions and pointers welcome.

Test case: "en-u-tz-gblon"
Expected Result: "English (Time Zone: United Kingdom Time)"
Actual Result: "English (Time Zone: London Time)"

The first alias for “gblon” is “Europe/London” which has
an available exemplar city localisation of “London”. 

Is is possible that in the past there was no tz code 
for “gblon” and the ICU implementation falls back to 
using the region code “gb” and its localisation of 
“United Kingdom” (then formatted in “en”
as “United Kingdom time”)?

TR35 (https://unicode-org.github.io/cldr/ldml/tr35-general.html#locale_display_name_algorithm <https://unicode-org.github.io/cldr/ldml/tr35-general.html#locale_display_name_algorithm>)
doesn’t have a specific reference to resolving the
“tz” key that I can see.


Test case: "en-u-tz-chzrh"
Expected Result: "English (Time Zone: Zurich Time)"
Actual Result: "English (Time Zone: Switzerland Time)”

Would appear to the be same issue and perhaps the same
underlying cause? The only available alias for “chzrh”
is “Switzerland/Zurich” which has a translation to “Zurich”
as the exemplar city.

Thanks, —Kip

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://corp.unicode.org/pipermail/cldr-users/attachments/20210630/0aecc65e/attachment.htm>