Re: [openstreetmap/openstreetmap-website] Change naming of boundaries based on admin_level=* (Issue #5450)

[Minh Nguyễn via rails-dev]

For Overpass queries, there’s currently a special case that pulls in a 
dedicated string based on the `admin_level=*` if it’s present. Maybe instead we 
could use the generic “Administrative Boundary” label, using the normal code 
path, but add a suffix like “(Level 3)”?

https://github.com/openstreetmap/openstreetmap-website/blob/7565f5aa4aad13bb45929ba26e82a1ab3ca28210/app/assets/javascripts/index/query.js#L73-L78

For Nominatim queries, if a `place=*` tag is present on either the boundary or 
the `label` member, we show that as the label instead of anything related to 
the boundary:

https://github.com/openstreetmap/openstreetmap-website/blob/7565f5aa4aad13bb45929ba26e82a1ab3ca28210/app/controllers/geocoder_controller.rb#L101-L107

For example, searching for 
“[Wapakoneta](https://www.openstreetmap.org/search?query=Wapakoneta#map=13/40.56551/-84.19397)”
 returns [this `boundary=administrative` `admin_level=8` 
relation](https://www.openstreetmap.org/relation/182636) but labels it as a 
“Town” based on the `place=town` tag on [this `label` 
member](https://www.openstreetmap.org/node/153976720). The label doesn’t call 
it a “boundary” of any sort, even if that’s what gets highlighted when you 
select the result. This is pretty reasonable, because the boundary represents a 
populated place that we’d tag as a `place=town` regardless of the boundary or 
legal designation. However, for something that doesn’t represent a populated 
place, such as a U.S. county or a French department, a `label` wouldn’t be as 
relevant or appropriate, so there still needs to be a “boundary” label for the 
search result.

I think it would be possible to use a suffix here too. It would need some 
additional translatable format strings.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5450#issuecomment-2565817707
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Change naming of boundaries based on admin_level=* (Issue #5450)

[Minh Nguyễn via rails-dev]

Fortunately, Nominatim does expose `border_type` in the `extratags` field of 
the response, and both Overpass queries (for nearby and enclosing features) 
include all the element’s tags.

Unfortunately, `border_type=*` values are unlocalized and un-namespaced. We’d 
only be able to pretty-print the snake_case keywords in English. The forum 
thread discusses some potential solutions. The simple solution would be to 
define `geocoder.search_osm_nominatim.border_types.#{border_type}` strings for 
the most common values of `border_type=*`. This would require every 
localization on Translatewiki.net to come up with a single word that 
corresponds to each English word. In some languages, this can be quite 
difficult for generic words like 
“[town](https://en.wiktionary.org/wiki/town#Translations)”. I’m not sure it 
would be received by the community any better than the current `admin_level=*` 
labels.

Fortunately, the Nominatim result comes with a `country_code` field, so we 
could prefer more specific string keys like 
`geocoder.search_osm_nominatim.border_types.#{country_code}.#{border_type}`, 
falling back to a more generic 
`geocoder.search_osm_nominatim.border_types.global.#{border_type}`. We could 
either introduce the per-country keys as needed or come up with a more 
comprehensive list of common `border_type=*` values per country. The latter 
would put much more of a burden on translators. I’ve seen that in the iD 
project, where translators have to localize each part of an address field for 
every country.

Unfortunately, the Overpass queries execute independently of each other, so the 
results for nearby features won’t have any context about the containing 
country. Since the response is handled on the client side in JavaScript, we 
could parse the enclosing features results for a country boundary, whenever it 
arrives, and patch up the labels on the enclosing features. In the meantime, 
these labels would use the global strings. Kind of icky, but I guess it could 
work.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5450#issuecomment-2566049746
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 pushed 1 commit.

e7f1141fc1d3c78ce808db55ee53b13ec770444c  Label boundaries by regional official 
designations

-- 
View it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493/files/1b9dcca4a9fca8ac670188bf313c8ef78173aeec..e7f1141fc1d3c78ce808db55ee53b13ec770444c
You are receiving this because you are subscribed to this thread.

Message ID: 

___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

[openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

Search results for Nominatim and Overpass queries now label administrative 
boundaries according to the real-world designation according to the 
`border_type=*` tag. If the value of this tag is unrecognized, the label falls 
back to a more generic “Administrative Boundary (Level _n_)”, where _n_ is the 
`admin_level=*` value. The English localization no longer attempts to translate 
these administrative levels to a fictional country’s administrative hierarchy.




Before
After







[](https://www.openstreetmap.org/query?lat=39.252971&lon=-84.281251#map=18/39.253109/-84.280527)




[](http://0.0.0.0:3000/query?lat=39.252971&lon=-84.281251#map=18/39.253109/-84.280527)








[](https://www.openstreetmap.org/query?lat=43.068068&lon=-79.019396#map=18/43.068339/-79.018591)




[](http://0.0.0.0:3000/query?lat=43.068068&lon=-79.019396#map=18/43.068339/-79.018591)







[](https://www.openstreetmap.org/search?query=Hamilton#map=11/39.1673/-84.5384)




[](http://0.0.0.0:3000/search?query=Hamilton#map=11/39.1673/-84.5384)






There are [many tagging 
schemes](https://wiki.openstreetmap.org/wiki/Key:border_type#Alternatives) for 
the real-world designation, each with their advantages and disadvantages, but I 
focused on `border_type=*` for now because it’s the most common scheme globally.

Fixes #5450.
You can view, comment on, or merge this pull request online at:

  https://github.com/openstreetmap/openstreetmap-website/pull/5493

-- Commit Summary --

  * Label administrative boundaries by level
  * Level 2 is always an international boundary
  * Label boundaries by regional official designations

-- File Changes --

M app/assets/javascripts/index/query.js (8)
M app/controllers/geocoder_controller.rb (6)
M config/locales/en.yml (47)

-- Patch Links --

https://github.com/openstreetmap/openstreetmap-website/pull/5493.patch
https://github.com/openstreetmap/openstreetmap-website/pull/5493.diff

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +arrondissement: "Arrondissement Boundary"
+borough: "Borough Boundary"
+cercle: "Cercle Boundary"
+city: "City Boundary"
+comarca: "Comarca Boundary"
+county: "County Boundary"
+departement: "Department Boundary"
+department: "Department Boundary"
+district: "District Boundary"
+distrito: "District Boundary"
+freguesia: "Freguesia Boundary"
+local_authority: "Local Authority Boundary"
+municipality: "Municipal Boundary"
+municipi: "Municipal Boundary"
+município: "Municipal Boundary"
+nation: "International Boundary"
+national: "International Boundary"
+neighbourhood: "Neighborhood Boundary"
+parish: "Parish Boundary"
+province: "Provincial Boundary"
+região: "Regional Boundary"
+region: "Regional Boundary"
+state: "State Boundary"
+town: "Town Boundary"
+township: "Township Boundary"
+village: "Village Boundary"

For starters, I added a translatable string for each of the 30 or so 
`border_type=*` values that occur most frequently with 
`boundary=administrative`. Only one translatable string is provided per 
`border_type=*` value globally. This will likely present a problem for some 
languages that would translate English words like “district”, “county”, or 
“town” differently depending on the country. I think we’ll need to add some 
more specific strings qualified by country code, as explained in 
https://github.com/openstreetmap/openstreetmap-website/issues/5450#issuecomment-2566049746,
 but it’ll require a bit more research to find out which values are important 
in which countries.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#pullrequestreview-2545317289
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Change naming of boundaries based on admin_level=* (Issue #5450)

[Minh Nguyễn via rails-dev]

#5493 implements the enhancements above except for distinguishing a given 
`border_type=*` value that needs to be translated differently depending on the 
country.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5450#issuecomment-2585517181
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Copyedit English localization as American English (PR #5029)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +level2: "International Border"
+level3: "Regional Administrative Boundary"
+level4: "Regional Administrative Boundary"
+level5: "Regional Administrative Boundary"
+level6: "Local Administrative Boundary"
+level7: "Local Administrative Boundary"
+level8: "Local Administrative Boundary"
+level9: "Sublocal Administrative Boundary"
+level10: "Sublocal Administrative Boundary"
+level11: "Sublocal Administrative Boundary"

This is superseded by #5493.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5029#discussion_r1912377698
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +level3: "Administrative Boundary (Level 3)"
+level4: "Administrative Boundary (Level 4)"
+level5: "Administrative Boundary (Level 5)"
+level6: "Administrative Boundary (Level 6)"
+level7: "Administrative Boundary (Level 7)"
+level8: "Administrative Boundary (Level 8)"
+level9: "Administrative Boundary (Level 9)"
+level10: "Administrative Boundary (Level 10)"
+level11: "Administrative Boundary (Level 11)"

I noticed that other boundary types don’t say “Boundary”. For example, 
`boundary=electoral` is just “Electoral”. We could shorten “Administrative 
Boundary (Level 4)” to just “Administrative (Level 4)”, but I’m ambivalent 
about similarly truncating the `border_type=*`-based labels, such as “City 
~~Boundary~~”. That would easily get confused with the `place=*`-based labels, 
which follow different criteria.

Another option would be to drop the level suffix, since @tomhughes doesn’t seem 
enamored with including this information anyways. To the extent that this 
introduces ambiguity in the search results, the solution would be to tag 
boundaries with `border_type=*`, which is a good idea anyways.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#discussion_r1912562705
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +level3: "Administrative Boundary (Level 3)"
+level4: "Administrative Boundary (Level 4)"
+level5: "Administrative Boundary (Level 5)"
+level6: "Administrative Boundary (Level 6)"
+level7: "Administrative Boundary (Level 7)"
+level8: "Administrative Boundary (Level 8)"
+level9: "Administrative Boundary (Level 9)"
+level10: "Administrative Boundary (Level 10)"
+level11: "Administrative Boundary (Level 11)"

Not a bad idea overall, though the numbers probably aren’t as intuitive for 
many mappers as designations coming from either `border_type=*` or `place=*`. I 
had forgotten about the code to transform an unrecognized tag into a label. As 
long as we have labels, we could fall back to a format string like “_Border 
Type_ Boundary” if `border_type=*` is present but set to an unrecognized value.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#discussion_r1912585645
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  border_type = t 
> "geocoder.search_osm_nominatim.border_types.#{extratag.attributes['value']}", 
> :default => prefix_name if extratag.attributes["key"] == "border_type"
+  place_type = t 
"geocoder.search_osm_nominatim.prefix.place.#{extratag.attributes['value']}", 
:default => prefix_name if extratag.attributes["key"] == "linked_place" || 
extratag.attributes["key"] == "place"

Yes, good catch. I originally had two loops that set `prefix_name` directly. 
Now we probably don’t need a `:default` at all.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#discussion_r1912575079
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +level3: "Administrative Boundary (Level 3)"
+level4: "Administrative Boundary (Level 4)"
+level5: "Administrative Boundary (Level 5)"
+level6: "Administrative Boundary (Level 6)"
+level7: "Administrative Boundary (Level 7)"
+level8: "Administrative Boundary (Level 8)"
+level9: "Administrative Boundary (Level 9)"
+level10: "Administrative Boundary (Level 10)"
+level11: "Administrative Boundary (Level 11)"

Would the number be clearer than the locally common word for the type of 
boundary? Mappers who specialize in boundaries would know the local 
`admin_level=*` scale by heart, but others would find words more intuitive and 
would already know which are larger than others. Also, the Overpass results are 
ranked by size.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#discussion_r1913602838
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add OpenMapTiles vector map (PR #4042)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -9,6 +14,43 @@ L.extend(L.LatLngBounds.prototype, {
   }
 });
 
+if (OSM.MAPTILER_KEY) {
+  maplibregl.setRTLTextPlugin(
+
"https://api.mapbox.com/mapbox-gl-js/plugins/mapbox-gl-rtl-text/v0.2.3/mapbox-gl-rtl-text.js";,

To avoid this external dependency on a Mapbox API, install the 
`@mapbox/mapbox-gl-rtl-text` package and have a build script copy 
node_modules/@mapbox/mapbox-gl-rtl-text/dist/mapbox-gl-rtl-text.js into 
app/assets/javascripts/. Then this becomes just `mapbox-gl-rtl-text.js`.

> +var mainLocale = I18n.locale.slice(0, 2);
+var localeIndex = supportedLanguages.findIndex(function (locale) {
+  return locale === mainLocale;
+});

This website and maplibre-gl-omt-language both support some locales that have 
three-character-long language codes, such as [Western 
Punjabi](https://github.com/openstreetmap/openstreetmap-website/blob/e206dd527ef7c6d1829f131533d722e536e4bfb5/config/locales/pnb.yml)
 (pnb). I don’t think this code will match such locales. If I’m not mistaken, 
the intention is to match on only the language code and ignore any script or 
country code. If so, this would be more robust:

```suggestion
var mainLocale = I18n.locale.split('-')[0];
var localeIndex = supportedLanguages.findIndex(function (locale) {
  return locale === mainLocale;
});
```

Note that this is still quite a rudimentary locale matching heuristic. It won’t 
match `zh-Hans` or `zh-Hant`, which the plugin also supports.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/4042#pullrequestreview-2603380276
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Contribution heatmap is off by a month (Issue #5802)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5802)

I can reliably reproduce this issue:

* In Firefox 137.0b6 and Safari 18.3.1 on macOS 15.3.2, and Safari on iOS 16.7
* With `I18n.locale` evaluating to `en` or `vi`
* With a screen 1,728 pixels wide or 375 pixels wide

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5802#issuecomment-2726768075
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Contribution heatmap is off by a month (Issue #5802)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5802)

>  Not sure if timezone would affect this, but I'm in UTC -4.

Ah, that’s a good insight. I’m in UTC−7. Changing my system clock to UTC works 
around the issue:

https://github.com/user-attachments/assets/ea4c16eb-c893-48d8-b085-4a2c72e61ef8";
 width="960" alt="No contributions on March 13, 2025, with a label March above 
it.">

So this issue affects anyone in the Western Hemisphere.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5802#issuecomment-2726796685
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

[openstreetmap/openstreetmap-website] Missing translations should fall back to English (Issue #5819)

[Minh Nguyễn via rails-dev]

1ec5 created an issue (openstreetmap/openstreetmap-website#5819)

Fall back to English when a string is missing from the localization or the 
translation refers to a nonexistent interpolation argument.

## Rationale

Sometimes a localizable string has to get changed to a different ID whenever 
the interpolated placeholders change. Otherwise, translations coming from 
Translatewiki.net can get out of sync with the new string’s set of 
interpolations, causing the `I18n.t` function to raise an 
`I18n::MissingInterpolationArgument` exception. Changing the string ID can be a 
good signal that the translator needs to slow down and potentially retranslate 
the message. But often, the content change is minor yet it’s difficult to track 
down the old string to ensure consistency. As a translator, I feel this every 
time a nontrivial change occurs on the Copyright page.

Over in the OpenHistoricalMap fork of this project, we’ve had to customize many 
messages that have interpolation arguments. For example, we had a couple 
miscues merging #4572, which renamed a string on the homepage at the same time 
it changed its interpolation arguments. Git detected the change as a merge 
conflict in en.yml but not in the source files that used the string. As a 
result, many locales wound up with a raw ID in the UI.

I thought OpenHistoricalMap/ohm-website#264 would fix the problem on the 
homepage by aligning with the ID from upstream. Unfortunately, my change 
instead led to 500 errors like OpenHistoricalMap/issues#972, either because 
translators hadn’t gotten around to updating the string or because they’d 
accept the translation memory suggestion coming from openstreetmap-website’s 
localization. We can rename the string again to avoid further problems, but 
it’s just one of many lurking issues along these lines. Another example is 
OpenHistoricalMap/issues#537.

## Suggestion

Translatewiki.net assumes the software behaves like MediaWiki, which is 
resilient to strings that include nonexistent interpolation arguments. For the 
I18n library we’re using, [this blog 
post](https://kinduff.com/2022/02/07/an-i18n-coding-adventure/) explores 
setting `missing_interpolation_argument_handler` to suppress the error in favor 
of something gentler, such as a “missing key” string. Even better, I suspect we 
could implement a handler that falls back to English, the base localization, by 
passing the same string into `t` but passing `en` as the `locale`. If we need 
to retain the ability to spot when a translation is missing, maybe we can limit 
this handler to production code.

I might be missing a simpler solution, but in general it seems like the 
translation process is too fragile for a 500 error to be the right response to 
a malformed translation.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5819
You are receiving this because you are subscribed to this thread.

Message ID: ___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Missing interpolations in translations should fall back to English (Issue #5819)

[Minh Nguyễn via rails-dev]

Closed #5819 as not planned.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5819#event-16852904738
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Missing interpolations in translations should fall back to English (Issue #5819)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5819)

Right, I’ve definitely found this test useful. I’ll have to look into why it 
isn’t running on Translatewiki.net PRs like OpenHistoricalMap/ohm-website#276. 
I recall you’ve periodically had to go into osm-community-index to report 
issues where someone overtranslated the name of the interpolation variable. If 
you’re OK with that kind of issue blocking a deployment, then what I described 
is mainly an issue for when forks have to merge from upstream. But with the low 
number of highly customized forks, I suppose that’s something we can handle 
downstream.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5819#issuecomment-2735157309
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] pretty formatting of OSM URLS in notes, changeset comment, and comments (Issue #5780)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5780)

GitHub gets around this problem by requiring a `#` in front of the number, or 
the full URL. Even just detecting and linkifying the URL would be a significant 
improvement.

By the way, the URL format for elements used to be slightly different, for 
example, https://www.openstreetmap.org/browse/way/1249366504.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5780#issuecomment-2746261182
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

[openstreetmap/openstreetmap-website] Contribution heatmap is off by a month (Issue #5802)

[Minh Nguyễn via rails-dev]

1ec5 created an issue (openstreetmap/openstreetmap-website#5802)

The horizontal axis on the user contribution heatmap is off by a month. Each 
label across the top appears to be centered over the following month. For 
example, if you’ll excuse me showing off my own profile:

https://github.com/user-attachments/assets/bf959db2-5dc1-416c-9c8b-23f0987d7c85";
 width="960" alt="The tooltip says March 04, 2025, but the label above it says 
February.">

Not sure if this is a bug in cal-heatmap upstream or in this repository, since 
[the 
template](https://github.com/openstreetmap/openstreetmap-website/blob/037be97bf1fd758dfacd7861276f13a629261a59/app/views/users/show.html.erb#L245-L253)
 only contains an explicit list for the day labels along the side.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5802
You are receiving this because you are subscribed to this thread.

Message ID: ___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Missing interpolations in translations should fall back to English (Issue #5819)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5819)

Sorry, the title of the issue was misleading; I changed it. This is 
specifically about when a string in the YAML file includes a `%{foo}` but `foo` 
isn’t one of the arguments passed into `I18n.t`.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5819#issuecomment-273454
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] pretty formatting of OSM URLS in notes, changeset comment, and comments (Issue #5780)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5780)

> Also osm.wiki could be considered another URL variant, but idk how official 
> that one is.

The OSMF owns the osm.wiki domain.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5780#issuecomment-2746967396
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Enable native share if available (PR #5882)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5882)

> I'm not sure if it's very useful because you can just share the current page 
> url from the browser menu.

In Safari on macOS, a share sheet appears with a variety of options, including 
shortcuts to people I’ve chatted with recently and actions provided by 
third-party extensions. Ultimately it’s based on a URL, but it can save users 
some steps.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5882#issuecomment-2788591689
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] pretty formatting of OSM URLS in notes, changeset comment, and comments (Issue #5780)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5780)

Longer comments are more likely to prompt respondents to quote individual 
passages. In changeset and note discussions, this is usually a sign that the 
discussion should move outside the confines of the sidebar, to either direct 
messages or the forum. Other places like diary comments already support 
Markdown, so there’s a little less urgency there.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5780#issuecomment-2762152123
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] pretty formatting of OSM URLS in notes, changeset comment, and comments (Issue #5780)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5780)

That reminds me of another detail about GitHub’s pretty-printing: if you copy 
the pretty-printed text, some rich text formatting get paired with it, so that 
pasting the text will result in a working link in any rich text editor or 
Markdown-enabled field. Originally, they didn’t have this functionality, but # 
links would get linkified anyways, so it wasn’t a big deal unless you pasted 
outside of GitHub.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5780#issuecomment-2761808349
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Make contribution heat map opt in (Issue #5804)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5804)

Who is this widget’s intended audience and what’s its purpose? 
https://github.com/openstreetmap/openstreetmap-website/issues/5373#issue-2714677416
 frames it this way:

> The current user profile page provides a detailed summary of contributions 
> but lacks a visual representation of activity trends. A calendar heatmap, 
> similar to GitHub's user contribution calendar, would make it easier to 
> understand user engagement over time at a glance.

Going back further, 
https://github.com/openstreetmap/openstreetmap-website/issues/5356#issue-2691128594
 opens with this problem statement:

> We want to enhance the OSM profile pages with more engaging and informative 
> features, similar to HDYC.

I’m not sure I agree with the premise that the profile page had a detailed 
summary in the first place. But putting that aside, all I see is a desire to 
match the visual attractiveness of GitHub profile pages and Pascal Neis’s How 
Did You Contribute? tool, and to track “user engagement”. As user engagement 
isn’t a meaningful metric for OSM as a project, I think this feature could’ve 
benefited from more detailed consideration of use cases. For example:

* If it’s for the user to track their own edits, it probably belongs on their 
dashboard, which is visible only to them. That should mitigate the privacy 
concerns expressed here, even if we extend the feature to make it easier to 
reminisce about a changeset by date.

* If it’s for the user to show off their editing streak to others, as a light 
form of gamification, then there should be an option about that. At the very 
least, some community members don’t view their changeset history as a fair 
portrayal of their value to the community. I hear griping about that almost 
every year when OSMF board candidates’ heatmaps get posted on the wiki for all 
to see (even nonmembers).

* If it’s for others to gauge at a glance whether the user is actively editing 
these days, then I don’t think this level of granularity helps very much. I 
don’t need to know whether the user edits on weekends in order to predict 
whether they’re likely to respond to a direct message promptly or complain 
about their work getting reverted, or whether I should be concerned that my 
~~friend~~ follow is drifting away from the project. A simpler all-time line 
graph would convey this information more effectively (irrespective of any 
privacy considerations).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5804#issuecomment-2726753639
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add social profile links (PR #5439)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +
+class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => %r{\Ahttps?://.+\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {
+:bluesky => %r{\Ahttps?://(?:www\.)?bsky\.app/profile/([a-zA-Z0-9\._-]+)},
+:discord => %r{\Ahttps?://(?:www\.)?discord\.com/users/(\d+)},
+:facebook => %r{\Ahttps?://(?:www\.)?facebook\.com/([a-zA-Z0-9.]+)},
+:github => %r{\Ahttps?://(?:www\.)?github\.com/([a-zA-Z0-9_-]+)},
+:gitlab => %r{\Ahttps?://(?:www\.)?gitlab\.com/([a-zA-Z0-9_-]+)},
+:instagram => %r{\Ahttps?://(?:www\.)?instagram\.com/([a-zA-Z0-9._]+)},
+:linkedin => %r{\Ahttps?://(?:www\.)?linkedin\.com/in/([a-zA-Z0-9_-]+)},
+:line => %r{\Ahttps?://(?:www\.)?line\.me/ti/p/([a-zA-Z0-9_-]+)},
+:mastodon => 
%r{\Ahttps?://(?:(?:www\.)?mastodon\.social|en\.osm\.town)/@([a-zA-Z0-9_]+)},

Bluesky also uses a leading at sign, but only Mastodon uses two at signs as far 
as I know.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5439#discussion_r1995945136
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Contribution heatmap is off by a month (Issue #5802)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5802)

Ah sorry, I’m [Minh Nguyen](https://www.openstreetmap.org/user/Minh%20Nguyen) 
on OSM, though I’m seeing this on everyone’s heatmap, such as 
[Steve](https://www.openstreetmap.org/user/Steve) and [Mateusz Konieczny - bot 
account](https://www.openstreetmap.org/user/Mateusz%20Konieczny%20-%20bot%20account).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5802#issuecomment-2726089781
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Remove u-turn side decision (PR #5777)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5777)

> and the `driving_side` is wrong more often.

As noted in 
https://github.com/Project-OSRM/osrm-backend/issues/2269#issuecomment-357154269,
 a GeoJSON of [left-hand-drive 
regions](https://github.com/Project-OSRM/osrm-backend/blob/4ee9968e3b585ee8ef28aec4a7fe3cfdea1f9ea2/data/driving_side.geojson)
 needs to be passed into `osrm-extract`. Otherwise, `driving_side` will always 
fall back to the default value of `right`. This seems to be the current 
behavior of the FOSSGIS OSRM server.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5777#issuecomment-2788636408
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add kilometers/miles switch to directions (PR #5915)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



>  } else {
-  return OSM.i18n.t("javascripts.directions.distance_km", { distance: 
Math.round(m / 1000) });
+  const ft = m / ftSize;
+  const mi = m / miSize;
+  if (ft < 1000) {

> You mean at 0.1 miles?

Yep, that was a typo, fixed. The difference between metric and non-metric in 
these SDKs is due to a large quantity like 999 meters still being 
comprehensible compared to 5,280 feet. Fewer digits helps too.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5915#discussion_r2039879862
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add kilometers/miles switch to directions (PR #5915)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



>  } else {
-  return OSM.i18n.t("javascripts.directions.distance_km", { distance: 
Math.round(m / 1000) });
+  const ft = m / ftSize;
+  const mi = m / miSize;
+  if (ft < 1000) {

That would seem more reasonable to me. 👍

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5915#discussion_r2040013366
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Can we avoid inlining svgs into erb templates? (Issue #5880)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5880)

Conceivably the project could have SVGs that aren’t just paths, for example the 
larger decorative image on the login screen or the occasional banner 
advertisement on the homepage, in order to make it scalable or more easily 
translatable. I suspect contributors would be more comfortable editing such 
images in a full-fledged editor like Inkscape, even if it would be overkill for 
the simpler icons.

(As a point of reference, the OSM Americana project has [a suggested 
workflow](https://github.com/osm-americana/openstreetmap-americana/blob/c4829f2c91099c3b5619fff829a390133ea08a53/CONTRIBUTING.md#highway-shield-contributors-guide)
 involving Inkscape and an automated build script that optimizes the exported 
SVG, stripping out the excessive metadata that Inkscape adds.)

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5880#issuecomment-2813582540
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Replace CC by-sa license link with ODbL in history feeds (PR #5933)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -10,9 +10,7 @@ atom_feed(:language => I18n.locale, :schema_date => 2009,
   feed.logo image_url("mag_map-rss2.0.png")
 
   feed.rights :type => "xhtml" do |xhtml|
-xhtml.a :href => "https://creativecommons.org/licenses/by-sa/2.0/"; do |a|
-  a.img :src => image_url("cc_button.png"), :alt => "CC by-sa 2.0"
-end
+xhtml.a "Open Data Commons Open Database License", :href => 
"https://opendatacommons.org/licenses/odbl/";

I’m fixing the license in OpenHistoricalMap/ohm-website#288. (I hadn’t noticed 
the problem until this PR of yours.) In local testing, the feed does seem to 
pull from at least en.yml, but I haven’t tested with other locales.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5933#discussion_r2046420644
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Replace CC by-sa license link with ODbL in history feeds (PR #5933)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -10,9 +10,7 @@ atom_feed(:language => I18n.locale, :schema_date => 2009,
   feed.logo image_url("mag_map-rss2.0.png")
 
   feed.rights :type => "xhtml" do |xhtml|
-xhtml.a :href => "https://creativecommons.org/licenses/by-sa/2.0/"; do |a|
-  a.img :src => image_url("cc_button.png"), :alt => "CC by-sa 2.0"
-end
+xhtml.a "Open Data Commons Open Database License", :href => 
"https://opendatacommons.org/licenses/odbl/";

That’s true, it’s probably not as useful as synchronizing with the overall site 
setting. I only inspected the source code to see that it had pulled it from the 
YAML file instead of leaving a raw string key. 🤷‍♂️ Anyways, this isn’t a huge 
deal, just curious if there could’ve been a way to keep this bug from lurking 
for so long.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5933#discussion_r2046460047
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] What is the license of the images on the OSM website? (Issue #5966)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5966)

> I also wouldn't say it's that common for open source projects that are 
> primarily code based to go to the trouble of specifying a different license 
> for artwork. I'm sure it does happen but I can't think of any examples off 
> the top of my head.

It happens with some map style repositories, such as [this fork of OSM 
Carto](https://imagico.de/blog/en/on-open-map-design-licensing/) and [Mapbox’s 
original vector 
styles](https://github.com/mapbox/mapbox-gl-styles/blob/c4b34ab336fdb70be489fa018f2fe9e6f341b4f5/LICENSE.md),
 but these projects are known primarily for their visual output. Some vector 
tile schemas license “cartographic design” separately from the code, such as 
[OpenMapTiles](https://github.com/openmaptiles/openmaptiles/blob/f70ae783b286e362431efc53a07dfa35472be2bd/LICENSE.md),
 in order to require attribution in the finished product. Occasionally I’ve 
seen website projects separately license graphical elements within, but I don’t 
know of any OSM-related examples off the top of my head.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5966#issuecomment-2845086226
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Please add NATO coordinates, Here is why. (Issue #5984)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5984)

> aren't you guys sort of competing with Google?

Only in the sense of providing geographic data about on-the-ground stuff that 
others could use to create services that compete with Google. You’ve basically 
asked the plumber to bake you a New York pizza. 😉

Taking a step back, perhaps you could post to [our help 
forum](https://community.openstreetmap.org/c/help-and-support/7) explaining 
what it is you’d like to do with grid reference systems, practically speaking. 
Folks in the community might know of existing software that provides this 
capability in conjunction with OSM data, or they might be able to suggest 
suitable workarounds that you could implement yourself. (As with anywhere on 
the Internet, you might want to focus on technical problem-solving without so 
much geopolitical context to ensure constructive responses.)

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5984#issuecomment-2851781334
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] [Feature request] Changelog (Issue #5950)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5950)

Yes, I’m looking into starting a series of posts on the forum periodically 
highlighting noteworthy changes as they go in, similar to what I’ve been doing 
for OpenHistoricalMap. These won’t technically be release notes, since this 
project uses a continuous deployment model.

Separately, some OSM-related projects [maintain running 
“changelogs”](https://keepachangelog.com/en/1.0.0/) that are more like draft 
release notes. In principle, the document could be divided by month rather than 
release. In iD and related projects, we’ve been discussing the best way to 
manage these documents. iD’s changelog has gotten so long that GitHub sometimes 
breaks down trying to render it. Drafting updates at release time slows down 
the release process, sometimes by days, but the running changelog concept 
hasn’t really worked out either, since most contributors don’t bother writing 
their own blurbs (and there’s a whole issue around merge conflicts if they do).

All of this is to say: let’s start with high-level updates outside the repo for 
now, and see how it goes. Look for an announcement probably next week tied to 
this discussion. 🙂

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5950#issuecomment-2830897286
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Replace CC by-sa license link with ODbL in history feeds (PR #5933)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -10,9 +10,7 @@ atom_feed(:language => I18n.locale, :schema_date => 2009,
   feed.logo image_url("mag_map-rss2.0.png")
 
   feed.rights :type => "xhtml" do |xhtml|
-xhtml.a :href => "https://creativecommons.org/licenses/by-sa/2.0/"; do |a|
-  a.img :src => image_url("cc_button.png"), :alt => "CC by-sa 2.0"
-end
+xhtml.a "Open Data Commons Open Database License", :href => 
"https://opendatacommons.org/licenses/odbl/";

Should this use `Settings.license_url`?

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5933#pullrequestreview-2769525094
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add kilometers/miles switch to directions (PR #5915)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



>  } else {
-  return OSM.i18n.t("javascripts.directions.distance_km", { distance: 
Math.round(m / 1000) });
+  const ft = m / ftSize;
+  const mi = m / miSize;
+  if (ft < 1000) {

For what it’s worth, the [MapLibre Navigation 
SDK](https://github.com/maplibre/maplibre-navigation-ios/blob/f91faffa1aa03be16fc588dc5076b7811d1c5166/MapboxCoreNavigation/DistanceFormatter.swift#L146-L160)
 switches at 1,000 meters = 1 kilometer, 176 yards = 1 mile 
for the UK, or 528 feet = 1 mile for the U.S. (ignoring the rounding 
increments that are applied for readability at a glance). 
[Ferrostar](https://github.com/stadiamaps/ferrostar/blob/f10b01c03bec9554a7cf827327795d253ad4c1f0/apple/Sources/FerrostarCarPlayUI/Measurement/MKDistanceFormatterUnits.swift#L36-L54)
 switches at 1,000 meters = 1 kilometer, 300 yards = 
0.17 miles for the UK, or 289 feet = 0.16 miles for the U.S. 
These navigation SDKs don’t aim for equivalent thresholds between measurement 
systems, since most users aren’t switching back and forth between them.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5915#discussion_r2038424247
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Missing American English localization (Issue #4975)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#4975)

As the original post and #5029 illustrate, `en` is currently a mix of British 
English, American English, OSM English, and some outright mistakes. Regardless, 
Translatewiki.net assumes `en` is American English, as does pretty much every 
user agent (operating system or browser). It is possible to create our own 
custom mapping from Translatewiki.net’s locale codes to YAML files. However, 
Translatewiki.net contributors probably wouldn’t abide by any unusual mapping 
from `en-CA` to `en-US`, especially not any Canadian translators there, who 
would be rightfully proud of their country code.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/4975#issuecomment-2707699299
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  border_type = t 
> "geocoder.search_osm_nominatim.border_types.#{extratag.attributes['value']}", 
> :default => prefix_name if extratag.attributes["key"] == "border_type"
+  place_type = t 
"geocoder.search_osm_nominatim.prefix.place.#{extratag.attributes['value']}", 
:default => prefix_name if extratag.attributes["key"] == "linked_place" || 
extratag.attributes["key"] == "place"

Fixed.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#discussion_r1984987788
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 pushed 1 commit.

7869631e83a7686c06d92aaefb8a23245e930b57  Label boundaries by regional official 
designations

-- 
View it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493/files/84ca7f4d897b8361d2209968a3abeb4f78dfb86b..7869631e83a7686c06d92aaefb8a23245e930b57
You are receiving this because you are subscribed to this thread.

Message ID: 

___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -101,9 +101,13 @@ def search_osm_nominatim
   if klass == "boundary" && type == "administrative"
 rank = (place.attributes["address_rank"].to_i + 1) / 2
 prefix_name = t 
"geocoder.search_osm_nominatim.admin_levels.level#{rank}", :default => 
prefix_name
+border_type = false
+place_type = false

Fixed.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#discussion_r1984987577
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

@1ec5 pushed 3 commits.

ad5d67931fa1cbfe0049db3cb5a0a910f351c07f  Label administrative boundaries by 
level
7006aa983947fde2a48ea4143059668d50d1cb16  Level 2 is always an international 
boundary
84ca7f4d897b8361d2209968a3abeb4f78dfb86b  Label boundaries by regional official 
designations

-- 
View it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493/files/e7f1141fc1d3c78ce808db55ee53b13ec770444c..84ca7f4d897b8361d2209968a3abeb4f78dfb86b
You are receiving this because you are subscribed to this thread.

Message ID: 

___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Label administrative boundaries by real-world designation (PR #5493)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5493)

Thanks for your patience. I’ve addressed the code-level feedback. 
https://github.com/openstreetmap/openstreetmap-website/pull/5493#discussion_r1912562048
 is unresolved; there are a couple options for making it briefer (in English at 
least), but not without losing some detail.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5493#issuecomment-2706367281
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Missing American English localization (Issue #4975)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#4975)

> Isn't Canadian English similar to US English?

For the kinds of strings being translated in this repository, Canadian English 
spelling is closer to Australian English than American English, for example, 
“cancelled”, “centre”, and “licence” (versus AmEng “canceled”, “center”, and 
“license”, respectively). But as I mentioned in the PR, I’m more concerned 
about vocabulary than spelling, which most users can simply get used to. 
Unfortunately, there are also some differences in terminology that could cause 
confusion, such as “parkade” (versus “parking garage”) and “washroom” (versus 
“restroom”).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/4975#issuecomment-2708631689
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Missing American English localization (Issue #4975)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#4975)

No, currently there is no U.S. English localization. This issue was closed 
almost immediately, and #5029 is only a draft for discussion about what a U.S. 
English localization would contain, if we could find a 
Translatewiki.net-compatible code for it.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/4975#issuecomment-2707604491
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Modify the way "Friends" are added (PR #5261)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5261)

I just noticed this renaming as I went to update translations over on 
Translatewiki.net. Like others here, I was also struck by the oddity of 
describing an individual as “a following” and a set of individuals as 
“followings”. It’s basically a made-up word at this point. Worse, as 
https://github.com/openstreetmap/openstreetmap-website/pull/5261#issuecomment-2613978828
 points out, it’s completely backwards: a following is the people who are 
following _you_, not the people whom _you_ are following.

As it’s used in social media, “follow” is a figure of speech, a 
[skeuomorph](https://en.wikipedia.org/wiki/Skeuomorph), for the act of chasing 
after someone. The opposite of a _follower_ is a _leader_, and I think that 
underscores how poorly the social media model fits this website. (And, let’s be 
honest, the opposite of a _stalker_ is a _target_.)

In Vietnamese, I’ve already been translating “subscribe (to a changeset)” as 
“follow”, so I’d probably do that for this as well, but “a follow” 
unfortunately has no translation. The closest I can come up with would be “a 
followed user”, which might as well be what we say in English. Do we even need 
a user-facing word for this concept, apart from the internal data model? The 
dashboard has plenty of space to say “Users you are following” (above “Other 
nearby users”), and the notifications can say “[User name] is now following 
you”.

As far as I know, all this function currently does is list a user on your 
dashboard: 
https://github.com/openstreetmap/openstreetmap-website/issues/3310#issuecomment-912283262.
 Is the intention to eventually make this functionality behave more like a 
social media feed? If not, there must be some word that better describes it 
than either “friend” or “follow”. Maybe a better mental model would be “pinning 
a pinned user”.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5261#issuecomment-2708463482
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] allow markdown in changeset comments (#844)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#844)

Many websites that accept Markdown end up pretty-printing raw URLs. For 
example, GitHub turns 
`https://github.com/openstreetmap/openstreetmap-website/issues/844#issuecomment-2705479356`
 into 
https://github.com/openstreetmap/openstreetmap-website/issues/844#issuecomment-2705479356.
 It isn’t the most discoverable syntax: you’d probably see people occasionally 
trying to replicate the presentation by hand as 
```[`informal=yes`](https://wiki.openstreetmap.org/wiki/Tag:informal%3Dyes)```, 
but I don’t think that would be a huge problem.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/844#issuecomment-2708470716
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add OpenMapTiles vector map (PR #4042)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +
+  var copyright = I18n.t("javascripts.map.copyright_text", { 
copyright_link: copyright_link });
+
+  var openmaptiles_link = I18n.t("javascripts.map.openmaptiles", {
+openmaptiles_url: "https://openmaptiles.org/";,
+maptiler_url: "https://www.maptiler.com/";
+  });
+
+  var terms = $("", {
+href: "https://wiki.osmfoundation.org/wiki/Terms_of_Use";,
+text: I18n.t("javascripts.map.website_and_api_terms")
+  }).prop("outerHTML");
+
+  this.baseLayers.push(new L.OpenMapTiles({
+attribution: copyright + ". " + openmaptiles_link + ". " + terms,
+code: "V",

I think it’s just based on the name of the layer, using whatever letter is 
available and unlikely to clash with the others. OpenHistoricalMap’s fork has 
also assigned [its own 
codes](https://github.com/OpenHistoricalMap/ohm-website/blob/ad4ea829a7f794a456ef5c1484e977653cd9f7c3/app/assets/javascripts/leaflet.map.js#L81),
 currently J, O, R, and W, but there hasn’t been a conflict because some of 
OSM’s featured layers have been removed. The only risk from reusing a 
previously assigned letter is that some old links to the old layer will now 
point to the new layer, but these links have been falling back to the Standard 
layer anyways.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/4042#discussion_r1976457225
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add social profile links (PR #5439)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +
+class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => %r{\Ahttps?://.+\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {
+:bluesky => %r{\Ahttps?://(?:www\.)?bsky\.app/profile/([a-zA-Z0-9\._-]+)},
+:discord => %r{\Ahttps?://(?:www\.)?discord\.com/users/(\d+)},
+:facebook => %r{\Ahttps?://(?:www\.)?facebook\.com/([a-zA-Z0-9.]+)},
+:github => %r{\Ahttps?://(?:www\.)?github\.com/([a-zA-Z0-9_-]+)},
+:gitlab => %r{\Ahttps?://(?:www\.)?gitlab\.com/([a-zA-Z0-9_-]+)},
+:instagram => %r{\Ahttps?://(?:www\.)?instagram\.com/([a-zA-Z0-9._]+)},
+:linkedin => %r{\Ahttps?://(?:www\.)?linkedin\.com/in/([a-zA-Z0-9_-]+)},
+:line => %r{\Ahttps?://(?:www\.)?line\.me/ti/p/([a-zA-Z0-9_-]+)},
+:mastodon => 
%r{\Ahttps?://(?:(?:www\.)?mastodon\.social|en\.osm\.town)/@([a-zA-Z0-9_]+)},

This is very restrictive. A lot of OSM contributors have accounts in other 
places, such as but not limited to [Mapstodon](https://mapstodon.space/). Sites 
that accept Mastodon accounts typically accept them in one of the following 
formats:

* `map...@example.net`
* `@map...@example.net`
* `https://example.net/@mapper`

There’s no way to validate a Mastodon profile URL or Mastodon address by 
domain, as that would defeat the purpose of the Fediverse.

If the purpose of this regex is only to assign a Mastodon icon, that’s no big 
deal. But it looks like the link in the profile only displays the local part of 
the Mastodon address. This can get confusing when multiple servers have 
accounts with the same name. The standard display format is 
`@map...@example.com`.

> +#
+# Foreign Keys
+#
+#  fk_rails_...  (user_id => users.id)
+#
+
+class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => %r{\Ahttps?://.+\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {
+:bluesky => %r{\Ahttps?://(?:www\.)?bsky\.app/profile/([a-zA-Z0-9\._-]+)},
+:discord => %r{\Ahttps?://(?:www\.)?discord\.com/users/(\d+)},
+:facebook => %r{\Ahttps?://(?:www\.)?facebook\.com/([a-zA-Z0-9.]+)},
+:github => %r{\Ahttps?://(?:www\.)?github\.com/([a-zA-Z0-9_-]+)},

Flickr might be a nice one to add. I’ve seen lots of examples of people using 
their Flickr accounts to host photos taken during field surveys, and our forum 
has special handling for Flickr photos.

> +#
+# Indexes
+#
+#  index_social_links_on_user_id  (user_id)
+#
+# Foreign Keys
+#
+#  fk_rails_...  (user_id => users.id)
+#
+
+class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => %r{\Ahttps?://.+\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {

Some suggestions for additional sites that users might want to display on their 
profiles in the same manner, in case the list isn’t strictly limited to social 
media accounts:

* OSM Community Forum (though it should be automatic)
* OpenGeofiction and OpenHistoricalMap
* Wikipedia, Wikidata, Wikivoyage, and Wikimedia Commons

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5439#pullrequestreview-2683519315
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Source of translations of future dates (Issue #5769)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5769)

CLDR has string formats for [future/past relative 
years/months/etc.](https://www.unicode.org/cldr/charts/47/summary/sk.html#38e7d288d00bc475)
 However, it doesn’t have any provisions for “about”. As far as I know, “about” 
timestamps are relatively rare these days, since most software would write out 
the specific date past a certain point.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5769#issuecomment-2723097448
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Calendar heatmap of user contributions on user profile (Issue #5373)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5373)

This landed in #5402. This issue didn’t close automatically because “addresses” 
isn’t one of [GitHub’s closing 
keywords](https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/using-keywords-in-issues-and-pull-requests).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5373#issuecomment-2722827109
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Source of translations of future dates (Issue #5769)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5769)

Are rails-i18n and homegrown translation strings the only options for 
formatting dates? On the client side, JavaScript has [a robust API for 
formatting relative 
dates](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat)
 according to a given locale. It’s powered by CLDR, which is the industry 
standard for this sort of thing. If we can find a Rails library that pulls data 
from CLDR, then we don’t need to burden translators with every permutation that 
their language doesn’t need. Or we could leave it as an absolute timestamp and 
rely on JavaScript to relativize it on the fly.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5769#issuecomment-2722906195
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add OpenMapTiles vector map (PR #4042)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -9,6 +14,43 @@ L.extend(L.LatLngBounds.prototype, {
   }
 });
 
+if (OSM.MAPTILER_KEY) {
+  maplibregl.setRTLTextPlugin(
+
"https://api.mapbox.com/mapbox-gl-js/plugins/mapbox-gl-rtl-text/v0.2.3/mapbox-gl-rtl-text.js";,

See 
https://github.com/openstreetmap/openstreetmap-website/pull/4042#discussion_r1966816765,
 sorry for the delay.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/4042#discussion_r1966833648
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Company Name on Profile Page (PR #5987)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5987)

There’s already a fairly strong disincentive to claim corporate affiliation in 
at least some parts of the OSM community. 😅 Maybe optimistically linking to a 
matching wiki page could help by giving the user a way to verify for 
themselves. Do we have a way to link to only certain categories of wiki pages 
like [Organised Editing 
Teams](https://wiki.openstreetmap.org/wiki/Category:Organised_Editing_Teams)? 
If spam is a concern, I can probably rig something to alert wiki administrators 
when a page gets added to that category (or a more specific category of your 
choosing), so we can add it to our watchlists.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5987#issuecomment-2859962047
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] API versions, capabilities and permissions resources (PR #5979)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5979)

I [updated the 
documentation](https://wiki.openstreetmap.org/wiki/Special:Diff/2852642) to 
reflect the removal.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5979#issuecomment-2871275388
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] API versions, capabilities and permissions resources (PR #5979)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5979)

Oh! My bad. I [reverted the 
change](https://wiki.openstreetmap.org/wiki/Special:Diff/2852676).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5979#issuecomment-2871750842
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Change language for welcome mat (Issue #5991)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5991)

The welcome mat for corporate and institutional editing teams is a separate 
website. Unfortunately it doesn’t appear to be translatable yet: 
osmfoundation/welcome-mat#8.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5991#issuecomment-2861807347
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Improve statistucs numbers (PR #5994)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.

This will make the stats much more readable at a glance.

Do we know of anything that scrapes this page and possibly expects these 
numbers to remain unformatted? Does this also affect [the statistics page on 
the planet site](https://planet.openstreetmap.org/statistics/data_stats.html) 
or only on the main domain? I’m no fan of scraping pages for stuff like this, 
but we don’t have an equivalent API endpoint, so I might need to spread the 
word about this change.

> @@ -23,11 +29,11 @@ begin
 relation_count = Relation.where(:visible => true).count
 
 puts ""
-puts "Number of users#{user_count}"
-puts "Number of uploaded GPS 
points#{tracepoint_count}"
-puts "Number of nodes#{node_count}"
-puts "Number of ways#{way_count}"
-puts "Number of relations#{relation_count}"
+puts "Number of users#{user_count.format()}"
+puts "Number of uploaded GPS 
points#{tracepoint_count.format()}"
+puts "Number of nodes#{node_count.format()}"
+puts "Number of ways#{way_count.format()}"
+puts "Number of 
relations#{relation_count.format()}"

If you’re open to it, an alternative approach would be to add a small amount of 
JavaScript that [formats the 
number](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Number/toLocaleString)
 on the fly.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5994#pullrequestreview-2825679494
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add OpenLocationCode to OSM website (#1807)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#1807)

> My suggestion for next steps is to adopt the syntax `geo:$x,$y;u=$u` as a 
> Latitude/Longitude input. For example, `geo:-23.55,-46.63;u=15`.

#1123 tracks accepting a geo: URI as search input.

The topic of standing up a companion website probably deserves a separate 
discussion, perhaps in [osm-planning](https://github.com/osmlab/osm-planning/) 
or [operations](https://github.com/openstreetmap/operations/) (if you foresee 
the OWG or OSMF taking an active role in this), or better yet, somewhere like 
[the forum](https://community.openstreetmap.org/t/plus-codes/84969) where 
people are used to discussions going off in many very interesting tangents.

Same goes with the topic of grid references inside geo: URIs. If there’s 
already a standard affordance for grid references inside geo: URIs, then we 
could track providing that URI as part of the Share panel alongside the 
existing geo: URI, but that needs to be a separate issue. Otherwise, it 
probably wouldn’t be very useful for this site to unilaterally accept grid 
reference geo: URIs that no one else produces yet, so there’s some groundwork 
to do outside of this project.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/1807#issuecomment-2867089812
You are receiving this because you commented.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add social profile links (PR #5439)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +#
+# Foreign Keys
+#
+#  fk_rails_...  (user_id => users.id)
+#
+
+class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => %r{\Ahttps?://.+\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {
+:bluesky => %r{\Ahttps?://(?:www\.)?bsky\.app/profile/([a-zA-Z0-9\._-]+)},
+:discord => %r{\Ahttps?://(?:www\.)?discord\.com/users/(\d+)},
+:facebook => %r{\Ahttps?://(?:www\.)?facebook\.com/([a-zA-Z0-9.]+)},
+:github => %r{\Ahttps?://(?:www\.)?github\.com/([a-zA-Z0-9_-]+)},

This isn’t critical. Maybe I’ll open a separate issue for it later on, 
depending how this overall feature is received.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5439#discussion_r2081994482
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add social profile links (PR #5439)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +#
+# Indexes
+#
+#  index_social_links_on_user_id  (user_id)
+#
+# Foreign Keys
+#
+#  fk_rails_...  (user_id => users.id)
+#
+
+class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => %r{\Ahttps?://.+\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {

By the way, I think these can be tail work. It could be a good opportunity for 
someone less involved with the codebase to contribute something small.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5439#discussion_r2081993571
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add locale selector (PR #5201)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +
+  <% Locale.available
+   .select { |locale| I18n.exists? 
"shared.language_selector.#{locale}" }
+   .sort_by { |locale| t(".#{locale}") }
+   .each do |locale| %>
+><%= t(".#{locale}") %>
+  <% end %>
+

[`Intl.DisplayNames.prototype.of()`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/DisplayNames/of)
 returns undefined if the language code is unrecognized, so we could fall back 
to the raw code. It only throws an exception on a malformed language code. But 
if we already have a string that we can access at build time, that would be 
good too. In that case, we wouldn’t need all those added lines in the YAML 
file, one for each language…

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5201#discussion_r2087048210
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add locale selector (PR #5201)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +
+  <% Locale.available
+   .select { |locale| I18n.exists? 
"shared.language_selector.#{locale}" }
+   .sort_by { |locale| t(".#{locale}") }
+   .each do |locale| %>
+><%= t(".#{locale}") %>
+  <% end %>
+

https://translatewiki.net/w/i.php?title=Special:Translations&message=Osm%3ASite.copyright.native.native+link
 shows that most localizations are correctly translating into their own 
language, though a few have `THIS_LANGUAGE_NAME_HERE` untranslated.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5201#discussion_r2087050998
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add locale selector (PR #5201)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +
+  <% Locale.available
+   .select { |locale| I18n.exists? 
"shared.language_selector.#{locale}" }
+   .sort_by { |locale| t(".#{locale}") }
+   .each do |locale| %>
+><%= t(".#{locale}") %>
+  <% end %>
+

Do we need to generate this menu statically? If we can rely on JavaScript to 
generate the labels, then 
[`Intl.DisplayNames`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/DisplayNames)
 would save our translators the burden of translating all these language names 
yet again. (Or you could use the same API to preserve each language’s name for 
itself.) For example, [here’s how OpenStreetMap 
Americana](https://github.com/osm-americana/openstreetmap-americana/blob/7fe7232d102e309e2398c719eedcc294220c39cc/src/js/language_label.js#L39-L114)
 generates its language selector. If JavaScript is disabled or the browser 
doesn’t support this API, we could fall back to ISO 639 codes or a text 
field.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5201#pullrequestreview-2835595164
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

[openstreetmap/openstreetmap-website] HTTP 500 error on user profile pages (Issue #6023)

[Minh Nguyễn via rails-dev]

1ec5 created an issue (openstreetmap/openstreetmap-website#6023)

### URL

https://www.openstreetmap.org/user/Minh%20Nguyen

### How to reproduce the issue?

Open certain user profile pages (can’t figure out a pattern yet).

### Screenshot(s) or anything else?

![Application error: The OpenStreetMap server encountered an unexpected 
condition that prevented it from fulfilling the request (HTTP 
500)](https://github.com/user-attachments/assets/feb8ce84-b92f-4cd8-88f4-15f0f4c1c8eb)

Reproduced so far in:

* https://www.openstreetmap.org/user/Minh%20Nguyen
* https://www.openstreetmap.org/user/TomH
* https://www.openstreetmap.org/user/SomeoneElse
* https://www.openstreetmap.org/user/woodpeck
* https://www.openstreetmap.org/user/NE2

but not:

* https://www.openstreetmap.org/user/Steve
* https://www.openstreetmap.org/user/TIGERcnl

Reproduces when logged in and logged out.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/6023
You are receiving this because you are subscribed to this thread.

Message ID: ___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Company Name on Profile Page (PR #5987)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -153,6 +153,14 @@
   
   <%= @user.home_location_name %>
 <% end %>
+<% if @user.company&.strip.present? %>
+  
+http://www.w3.org/2000/svg"; width="16" height="16" 
fill="currentColor" class="bi bi-suitcase-lg-fill" viewBox="0 0 16 16">
+  
+
+  
+  <%= @user.company %>

> Maybe optimistically linking to a matching wiki page could help by giving the 
> user a way to verify for themselves.

For what it’s worth, I think a verification system (whether based on the wiki 
or backlinks or whatever) can be separate from this PR. The existing reporting 
system can probably keep things from getting out of hand in the meantime. If 
we’re concerned about impersonation in the meantime, we could add a tooltip to 
this box indicating that the company affiliation is self-reported.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5987#pullrequestreview-2852066222
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Company Name on Profile Page (PR #5987)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -153,6 +153,14 @@
   
   <%= @user.home_location_name %>
 <% end %>
+<% if @user.company&.strip.present? %>
+  
+http://www.w3.org/2000/svg"; width="16" height="16" 
fill="currentColor" class="bi bi-suitcase-lg-fill" viewBox="0 0 16 16">
+  

Similar to the location item above, the path should have a `title` containing 
alternative text for accessibility.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5987#pullrequestreview-2852069343
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add social profile links (PR #5439)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +#
+# Indexes
+#
+#  index_social_links_on_user_id  (user_id)
+#
+# Foreign Keys
+#
+#  fk_rails_...  (user_id => users.id)
+#
+
+class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => %r{\Ahttps?://.+\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {

Thank you, this looks pretty good. Yeah, OHM definitely needs a more adaptable 
logo for situations like this, but that’s outside of the PR’s scope. 😉

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5439#discussion_r2091884846
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add social profile links (PR #5439)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +class SocialLink < ApplicationRecord
+  belongs_to :user
+
+  validates :url, :format => { :with => 
%r{\A(https?://.+|@([a-zA-Z0-9_]+)@([\w\-\.]+))\z}, :message => 
:http_parse_error }
+
+  URL_PATTERNS = {
+:bluesky => %r{\Ahttps?://(?:www\.)?bsky\.app/profile/([a-zA-Z0-9\._-]+)},
+:discord => %r{\Ahttps?://(?:www\.)?discord\.com/users/(\d+)},
+:facebook => %r{\Ahttps?://(?:www\.)?facebook\.com/([a-zA-Z0-9.]+)},
+:flickr => %r{\Ahttps?://(?:www\.)?flickr\.com/people/([a-zA-Z0-9@._-]+)},
+:github => %r{\Ahttps?://(?:www\.)?github\.com/([a-zA-Z0-9_-]+)},
+:gitlab => %r{\Ahttps?://(?:www\.)?gitlab\.com/([a-zA-Z0-9_-]+)},
+:instagram => %r{\Ahttps?://(?:www\.)?instagram\.com/([a-zA-Z0-9._]+)},
+:linkedin => %r{\Ahttps?://(?:www\.)?linkedin\.com/in/([a-zA-Z0-9_-]+)},
+:line => %r{\Ahttps?://(?:www\.)?line\.me/ti/p/([a-zA-Z0-9_-]+)},
+:mastodon => 
%r{\Ahttps?://(?:(?:www\.)?(mastodon\.social|en\.osm\.town))/([@a-zA-Z0-9_]+)},

Yes, a leading `@` is required. Technically the `@` can still appear inside the 
handle in order to load a profile remotely from a different server, for 
example, https://mapstodon.space/@t...@en.osm.town. But I think it’s perfectly 
fine if we don’t handle that case, since the user should provide a direct URL 
anyways.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5439#discussion_r2091887989
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add lang attribute to Nominatim results from CJK languages (PR #6079)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -12,6 +14,14 @@ def create
   # ask nominatim
   response = fetch_xml(nominatim_reverse_query_url(:format => "xml"))
 
+  # add lang attribute for frontend in certain regions
+  addressparts = response.elements["reversegeocode/addressparts"]
+  lang = nil
+  if addressparts
+region_code = addressparts.elements["ISO3166-2-lvl3"]&.text == "CN-HK" 
? "hk" : addressparts.elements["country_code"]&.text
+lang = region_code ? LANGUAGE_CODES[region_code] : nil

Does this assume the displayed name is always in Chinese if the result is from 
China? That could result in suboptimal font selection and unexpected screen 
reader behavior if the name is in a different language due to the interface 
language. Based on `extratags`, could we determine whether the display name is 
based on one of the preferred interface languages or falling back to `name=*`?

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6079#pullrequestreview-2886672465
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Make "W" in osm wiki social icon more readable (PR #6065)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6065)

And with my wiki hat on, the difference is very minor with the low contrast. I 
don’t expect this small deviation from the wiki’s current full-size logo to be 
a problem. For those who squint, it’s just a little more legible, so it can’t 
hurt.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6065#issuecomment-2923855702
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Use only differences to OSM logo in internal social link icons (PR #6072)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6072)

What if we reduce the OHM icon to the outline of the map with the H centered 
inside of it, similar to [the Public Domain Map 
logo](https://commons.wikimedia.org/wiki/File:Public_Domain_Map_logo.svg)? This 
would just be an icon anyways, not necessarily the project’s official logo 
(which obviously needs work). The wiki icon could be something similar with a 
W, not sure about the others.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6072#issuecomment-2950101341
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add hide/show selection for contribution heatmap on edit profile page (PR #5853)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5853)

> Has there been any policy decision yet?

By the way, I think this is in reference to the question in 
https://github.com/openstreetmap/openstreetmap-website/pull/5706#issuecomment-2744397402
 about whether the terms of service apply to everyone automatically or only 
those who’ve affirmatively checked a box on the website. We’re still waiting 
for the outcome of that discussion. In other tickets, there was also a 
discussion about whether hiding the heatmap would disadvantage other users who 
need to inspect contribution history to weed out misuse.

Regardless, do we anticipate that the website will ultimately have an option to 
show or hide the heatmap? We could always fine-tune the behavior tied to this 
option later based on either discussion. (For example: “Display a heatmap of 
your contributions on your profile page to eligible logged-in users.”)

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5853#issuecomment-2956198714
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6127)

> Copy and pasting certainly isn't the way we want to be interacting with an 
> upstream repository...

If this becomes a built-in feature, I assume it would obviate the Chrome and 
Firefox plugins, right @ZeLonewolf? So probably no need to keep them in sync.

> Obviously a strong point, but I'd say right now Wikimedia has more OSM 
> integration as the other way around.

Another way of looking at it is that editors like iD have long had this same 
functionality. Obviously a different set of consequences for any downtime, but 
some mappers would probably appreciate seeing the same kind of preview that 
they do while editing, as with the color swatches for `colour=*` and such.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#issuecomment-2997166190
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Source of translations of future dates (Issue #5769)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5769)

> Or we could leave it as an absolute timestamp and rely on JavaScript to 
> relativize it on the fly.

Another advantage of client-side relative dates is that we can make them update 
dynamically as you’re viewing the page. For example, on this page (and also on 
the wiki if you have the Comments in Local Time gadget installed), the time on 
a recent comment updates live. However, a timestamp younger than a minute or 
older than than a week remains as an absolute date, so that the user doesn’t 
have to do the math in their head.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5769#issuecomment-2920723208
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Make "W" in osm wiki social icon more readable (PR #6065)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6065)

> It is the logo in the wiki, idk why it isn't the favicon.

Tracking in openstreetmap/operations#1245.

> @1ec5 was involved in the discussion about it not showing up in the vector 
> 2022 skin.

In [that 
discussion](https://wiki.openstreetmap.org/wiki/Talk:Wiki#Wordmark_in_Vector_New_version),
 I think we were going for a logo plus a wordmark, so there wasn’t as much need 
to deface the logo more legibly.

For the social media bar, I wonder if we even need to repeat the OSM logo motif 
on OSM-internal sites. Maybe we could just use a generic 📝 icon or something 
for the wiki, 💬 for the forum, etc., in the same stroked style as the 🌐 for 
unrecognized websites.

> Though the label is just "iki" and the W has a label of "W" which seems quite 
> silly so may something added that automatically?

Huh, that’s cute. Inkscape automatically adds a `` when you convert text 
to paths, but some other tool might add `aria-title`. I’m not confident that 
any screen reader will actually do anything with this label, since it’s 
embedded inside an `` tag (with `osm_wiki` as the alt text).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6065#issuecomment-2923348520
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Make "W" in osm wiki social icon more readable (PR #6065)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6065)

Something like an ⌛ or 🕐 or just an **H**? Otherwise, with my OHM hat on, I’d 
say you can do whatever you want with the OHM logo, such as recoloring it to 
sepia or defacing it with a spike to represent abandoned railways. (For what 
it’s worth, the project is formally if indirectly affiliated to OSM. Otherwise 
OHM wouldn’t be allowed to use anything resembling the OSM logo. But this 
situation is a great example of why OHM needs a new logo.)

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6065#issuecomment-2923493824
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Can't see the itinerary's duration (Issue #6094)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6094)

Finnish, French, Guadaloupean Creole, and Punjab have long translations for 
[km](https://translatewiki.net/wiki/Special:Translations?message=Osm%3ADirections.show.button+km&namespace=1222)
 and/or 
[mi](https://translatewiki.net/wiki/Special:Translations?message=Osm%3ADirections.show.button+mi&namespace=1222).

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/6094#issuecomment-2965124137
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] use JOSM/tag2link to linkify tag values (PR #6197)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6197)

Or at least preferring `osmwiki:P8` and `wikidata:P1630` as sources before 
falling back to `wikidata:P3303`.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6197#issuecomment-3084722278
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Remove u-turn side decision (PR #5777)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5777)

I contacted the FOSSGIS routing server’s e-mail address to ask them to look 
into the missing parameter when running `osrm-extract`.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5777#issuecomment-3093218990
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Move Intl.DisplayNames inside renderWikidataResponse (PR #6208)

[Minh Nguyễn via rails-dev]

@1ec5 approved this pull request.





-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6208#pullrequestreview-3035481036
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Support configurable coordinate format (lat, lon vs lon, lat) when displaying addresses (Issue #6165)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6165)

> This would make copying coordinates into development tools and code snippets 
> much more convenient and reduce confusion or errors.

If one of the internal result’s labels is just regurgitating what the user 
entered into the search bar, it seems unlikely to me that the user would be 
interested in copying the label.

For what it’s worth, iD’s Measurement panel is intended for copying, so it does 
show coordinates in two different formats: decimal and DMS, [always putting the 
latitude 
first](https://github.com/openstreetmap/iD/blob/f5b82daf24de0bbc01bea2f82bc7a55596403a97/modules/util/units.js#L180-L185).

https://github.com/user-attachments/assets/7326a4b6-25ca-41a9-8de6-a248b3d8d3ac";
 width="250" alt="Location: 39°16′58″N, 84°17′7″W; 39.2826582, -84.2851886">

For search input, openstreetmap/iD#10805 added a few formats that are rather 
specific to OSM, just for convenience when copy-pasting URL fragments and such.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/6165#issuecomment-3033405275
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + data.icon)
+.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
"Special:FilePath/" + data.icon, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  $(`<${data.label ? "sup" : "div"}>`)

OK, that’s fair. I do see a lot of second- and third-party Wikidata tools 
labeling links like “Wikipedia [es]” as well, with the language code wrapped in 
an `` tag that provides a tooltip for the human-readable language name. 
We don’t have any precedent for that in this codebase as far as I know, but 
personally I find it self-explanatory.

On that note, we should probably surround any foreign-language text with an 
element that specifies a `lang=` attribute for proper font selection and screen 
reader support.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#discussion_r2186117806
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Please help with pull requests! (Issue #3815)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#3815)

> A big part of that, I think, comes down to differing review criteria between 
> contributors and maintainers. That’s fair to some extent, but it becomes 
> frustrating when a maintainer fast-tracks their PR that renders an earlier, 
> unreviewed PR obsolete — without any discussion or acknowledgment of the 
> overlap. It feels dismissive to those who spent time contributing and 
> reviewing.

I haven’t been closely following this repository as long as you, but I didn’t 
notice a “queue” pattern that skips what contributors are working on. The 
closest thing would be that we were trying to respond to Microsoft’s 
contributions while we had their attention.

You have quite a few PRs open, so it must be frustrating to have to juggle them 
all, keeping them all up to date with the master branch as it changes. Would 
you be interested in self-triaging these PRs, commenting on them to point out 
which ones are more fragile (difficult to keep in sync) or which ones need to 
be staged first? Or I could take a look and make an educated guess…

> No, but going out of their way to submit a compact PR that is easily 
> reviewable (and mergeable - not self-merge though) to undermine a PR, forcing 
> their point with a necessary rebase rather than simply having a discussion.

Would you mind elaborating on the situation on your PR or privately, so we can 
figure out how to handle it better next time without calling anyone out in this 
pinned issue?

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/3815#issuecomment-3058535926
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Can we avoid inlining svgs into erb templates? (Issue #5880)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#5880)

Pardon my ignorance, but could someone remind me why we wound up using inline 
SVGs all over in the first place? Was there a particular benefit we were aiming 
for, or is it just convenient? I see some potential tangible benefits of 
separate source files:

* We can visually inspect any changes to the image directly inside GitHub’s 
code review interface, just as with a PNG or JPEG. @tomhughes [doesn’t know 
much about 
SVG](https://github.com/openstreetmap/openstreetmap-website/pull/5884#issuecomment-2898778060),
 but he should be able to take one look at the change and tell whether it’s 
going to change the image size, mess with its colors, or introduce esoteric 
syntax that popular SVG renderers will mangle.
* We could more easily integrate something like [SVGO](https://svgo.dev/) to 
lint and optimize the SVG, minimizing the things the reviewer would have to 
nitpick about more manually.
* It would promote reuse. Projects like iD have seen their icons get reused in 
other projects because it’s easier to find and extract them from the project.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/5880#issuecomment-3063135535
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + data.icon)
+.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
"Special:FilePath/" + data.icon, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  $(`<${data.label ? "sup" : "div"}>`)

The superscript suffix style probably comes from Wikidata’s former 
practice of glossing any fallback label with its fallback language, for 
example, “openstreetmap-website English” when you view 
[Q115259353](https://www.wikidata.org/wiki/Q115259353) in a language other than 
English. Otherwise, a user might not easily intuit that a suffix like “enwiki” 
means the title came from the English Wikipedia.

> +.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
> "Special:FilePath/" + data.icon, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  $(`<${data.label ? "sup" : "div"}>`)
+.append($("")
+  .attr("href", `https://${data.article.site.slice(0, 
-4)}.wikipedia.org/wiki/` + encodeURIComponent(data.article.title))

The Wikidata API can return the item’s label in the relevant language. I think 
the label would be more relevant to OSM than the Wikipedia article, which often 
contains disambiguators and substitute characters for technical reasons. The 
label is designed to pair well with the description, which we’re using below. 
(Ideally, we’d fall back to the `mul` code to improve language coverage: 
ZeLonewolf/osm-wikidata-greasemonkey#9.)

If the idea is that the Wikidata item is already linked above, maybe we could 
link the label to the Wikidata item redundantly, then have a separate Wikipedia 
link right after it, reusing app/assets/images/social_link_icons/wikipedia.svg.

> +items
+  .filter(qid => entities[qid])
+  .map(qid => getLocalizedResponse(entities[qid]))
+  .filter(data => data.label || data.icon || data.description || 
data.article)
+  .map(data => renderWikidataResponse(data, $btn))
+  );
+  })
+  .catch(() => $btn.removeClass("disabled").addClass("wdt-preview"));
+  }
+
+  function getLocalizedResponse(entity) {
+const localizedProperty = (property, langs) => langs.reduce((out, lang) => 
out ?? entity[property][lang], null);
+const data = {
+  qid: entity.id,
+  label: localizedProperty("labels", langs)?.value,
+  icon: ["P8972", "P154", "P14"].reduce((out, prop) => out ?? 
entity.claims[prop]?.[0]?.mainsnak?.datavalue?.value, null),

For reference:

* [small logo or icon 
(P8972)](https://www.wikidata.org/wiki/Property:P8972)
* [logo image 
(P154)](https://www.wikidata.org/wiki/Property:P154)
* [traffic sign 
(P14)](https://www.wikidata.org/wiki/Property:P14)

P14 surprised me at first, but then I remembered that this would probably 
result in displaying a route marker alongside a `wikidata=*` tag when viewing a 
route relation. That does make some sense to me.

> +items
+  .filter(qid => entities[qid])
+  .map(qid => getLocalizedResponse(entities[qid]))
+  .filter(data => data.label || data.icon || data.description || 
data.article)
+  .map(data => renderWikidataResponse(data, $btn))
+  );
+  })
+  .catch(() => $btn.removeClass("disabled").addClass("wdt-preview"));
+  }
+
+  function getLocalizedResponse(entity) {
+const localizedProperty = (property, langs) => langs.reduce((out, lang) => 
out ?? entity[property][lang], null);
+const data = {
+  qid: entity.id,
+  label: localizedProperty("labels", langs)?.value,
+  icon: ["P8972", "P154", "P14"].reduce((out, prop) => out ?? 
entity.claims[prop]?.[0]?.mainsnak?.datavalue?.value, null),

In case the item has multiple statements associated with one of these 
properties, the `claims[prop]` array will contain multiple items, but the first 
one isn’t guaranteed to be the best. For example, [Deutsche Post 
(Q157645)](https://www.wikidata.org/wiki/Q157645#P154) has four 
different P154 statements. The last is the most current, so the `mainsnak` has 
a `rank` of `preferred` rather than `normal`.

> +  label: localizedProperty("labels", langs)?.value,
+  icon: ["P8972", "P154", "P14"].reduce((out, prop) => out ?? 
entity.claims[prop]?.[0]?.mainsnak?.datavalue?.value, null),
+  description: localizedProperty("descriptions", langs)?.value,
+  article: localizedPr

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  icon: ["P8972", "P154", "P14"].reduce((out, prop) => out ?? 
> entity.claims[prop]?.[0]?.mainsnak?.datavalue?.value, null),
+  description: localizedProperty("descriptions", langs)?.value,
+  article: localizedProperty("sitelinks", wikis)
+};
+return data;
+  }
+
+  function renderWikidataResponse(data, $btn) {
+const cell = $("")
+  .attr("colspan", 2)
+  .addClass("bg-body-tertiary");
+
+if (data.icon && OSM.WIKIMEDIA_COMMONS_URL) {
+  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + data.icon)
+.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
"Special:FilePath/" + data.icon, height: "32" }))

> Did Wikimedia ever reacted to how we use their images in iD, or was it just 
> in wiki? Or are they already rate-limiting us in iD too but it's not very 
> noticeabe because images are small?

Only on the wiki, because the requests to Wikimedia Commons are centralized 
through the wiki’s server. This client-side usage is different in that each 
individual user will request only the images they’re interested in.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#discussion_r2183556357
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Allow to agree to new terms if already agreed to old terms (PR #5706)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



>  not_agreed_with_pd: "You haven't declared that you consider your 
> edits to be in the Public Domain."
-pd_link_text: "declare"
+review: "Review the Terms"
+review_and_accept: "Review and accept the Terms"
+consider_pd: "Consider Public Domain"
+  terms_of_use:
+heading: "Terms of Use"
+agreed: "You have agreed to the Terms of Use on %{date}."

Minor grammatical nit (also above):

```suggestion
agreed: "You agreed to the Terms of Use on %{date}."
```

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5706#pullrequestreview-2983456736
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Allow to agree to new terms if already agreed to old terms (PR #5706)

[Minh Nguyễn via rails-dev]

@1ec5 approved this pull request.

The logic looks correct to me. Just to confirm, does this PR result in actively 
presenting pre-2019 users with some sort of notice about the new terms, without 
having to visit the account settings first? If not, we still need to do that as 
tail work based on the decision at last month’s LWG meeting.



-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/5706#pullrequestreview-2983655979
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + data.icon)
+.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
"Special:FilePath/" + data.icon, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  $(`<${data.label ? "sup" : "div"}>`)

I guess that works, though I’m not sure it’s necessary to qualify the Wikipedia 
link with the language, since the URL already hints at the language. The main 
reason Wikidata appends the language name to labels is to avoid confusion when 
different languages have false cognates and such, but this isn’t a risk for a 
link with a fixed label or icon.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#discussion_r2184586554
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + 
encodeURIComponent(data.icon))
+.append($("").attr({ src, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  const lang = data.article.site.slice(0, -4);

I think this is fine, since lots of Wikimedia-related software makes the same 
assumption. It just needs a comment explaining the magic number −4.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#discussion_r2184073867
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> +  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + data.icon)
+.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
"Special:FilePath/" + data.icon, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  $(`<${data.label ? "sup" : "div"}>`)

I guess I was thinking of something like this (for English), based on how 
Wikidata formats labels on its own site:

* [Bench](https://www.wikidata.org/wiki/Q816714) 
([Wikipedia](https://en.wikipedia.org/wiki/Bench_(British_clothing_brand)))
* [Shell](https://www.wikidata.org/wiki/Q110716465)
* [Q111384397](https://www.wikidata.org/wiki/Q111384397)

The QID would be redundant to what’s in the normal tag view, though…

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#discussion_r2184105261
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -1,10 +1,16 @@
 (function () {
+  let abortController = null;
+  const langs = [...new Set([...OSM.preferred_languages.map(l => 
l.toLowerCase()), "mul", "en"])];
+  const wikis = [...new Set(langs.map(l => l.split("-")[0] + "wiki"))];

Wikimedia wiki codes don’t quite conform to ISO 639. For example, the 
response will include `be_x_oldwiki` for what this repository calls 
`be-Tarask`. But this will affect only a handful of locales that we have 
localizations for, so it probably isn’t critical.

> @@ -101,4 +110,97 @@
   scrollableList.scrollLeft = scrollableList.scrollWidth - 
scrollableList.offsetWidth;
 }
   }
+
+  function previewWikidataValue($btn) {
+if (!OSM.WIKIDATA_API_URL) return;
+const items = $btn.data("qids");
+if (!items?.length) return;
+$btn.addClass("disabled").removeClass("wdt-preview");
+fetch(OSM.WIKIDATA_API_URL + "?" + new URLSearchParams({
+  action: "wbgetentities",
+  format: "json",
+  origin: "*",
+  ids: items.join("|"),
+  props: "labels|sitelinks|claims|descriptions",

We can request `sitelinks/urls` here, so that each object in `sitelinks` will 
contain a `url` property set to the absolute URL to the article.

> +  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + 
encodeURIComponent(data.icon))
+.append($("").attr({ src, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  const lang = data.article.site.slice(0, -4);

Is this trimming “wiki” from the end of the sitename?

> +  $("")
+.attr("href", OSM.WIKIMEDIA_COMMONS_URL + "File:" + data.icon)
+.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
"Special:FilePath/" + data.icon, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  $(`<${data.label ? "sup" : "div"}>`)

This feature hooks into something labeled `wikidata` with a Wikidata icon, so 
there should probably be an explicit reference to Wikipedia to counter the 
assumption that the link goes to Wikidata. The reference to Wikipedia would be 
more relevant than the language name, since most of the time it’s going to be 
the user’s preferred language anyways. If we don’t want to bother with a 
Wikipedia icon, “(Wikipedia)” in regular text would communicate the link target 
pretty well.

> +.append($("").attr({ src: OSM.WIKIMEDIA_COMMONS_URL + 
> "Special:FilePath/" + data.icon, height: "32" }))
+.addClass("float-end mb-1 ms-2")
+.appendTo(cell);
+}
+if (data.label) {
+  $btn
+.siblings(`a[href*="wikidata.org/entity/${data.qid}"]`)
+.clone()
+.text(data.label)
+.addClass("me-1")
+.appendTo(cell);
+}
+if (data.article) {
+  $(`<${data.label ? "sup" : "div"}>`)
+.append($("")
+  .attr("href", `https://${data.article.site.slice(0, 
-4)}.wikipedia.org/wiki/` + encodeURIComponent(data.article.title))

Oh, I think I misread this part of the diff. I didn’t realize we’re always 
showing the label if available, followed by the superscript link to Wikipedia. 
I thought it was all one link based on an earlier screenshot.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#pullrequestreview-2984838528
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Stop Commons from redirecting outside CSP (PR #6202)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.

This is another subtle difference between Special:Redirect and the MediaWiki 
API, which is why the initial request for the item details doesn’t run into 
this problem.

> @@ -171,8 +171,8 @@
   .addClass("bg-body-tertiary");
 
 if (icon && OSM.WIKIMEDIA_COMMONS_URL) {
-  let src = OSM.WIKIMEDIA_COMMONS_URL + "Special:Redirect/file/" + 
encodeURIComponent(icon);
-  if (!icon.endsWith(".svg")) src += "?width=128";
+  let src = OSM.WIKIMEDIA_COMMONS_URL + "Special:Redirect/file/" + 
encodeURIComponent(icon) + "?mobileaction=toggle_view_desktop";

In cursory testing, this doesn’t cause subsequent user-initiated page loads on 
Commons to switch back to desktop view, which is good. Still a bit 
counterintuitive in this code, so a comment explaining the parameter would be a 
good idea.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6202#pullrequestreview-3031368828
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Add Cross-Origin-Opener-Policy header (2ff4d6a)

[Minh Nguyễn via rails-dev]

A fix is pending in facebook/Rapid#1724. See the temporary workaround at 
https://github.com/facebook/Rapid/issues/1722#issuecomment-3060409827.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/commit/2ff4d6a4e633e479568572090eb6a16074103cd9#commitcomment-162322980
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Integrate human-readable wikidata browser plugin (PR #6127)

[Minh Nguyễn via rails-dev]

@1ec5 commented on this pull request.



> @@ -1,10 +1,18 @@
 (function () {
+  let abortController = null;
+  const languagesToRequest = [...new Set([...OSM.preferred_languages.map(l => 
l.toLowerCase()), "mul", "en"])];

@lucaswerkmeister 
[suggested](https://wikis.world/@LucasWerkmeister/114876146945706671) passing 
in `languagefallback=1` instead of manually requesting a list of languages. I 
believe it would cause the response to use MediaWiki’s internal [language 
fallback 
chain](https://meta.wikimedia.org/wiki/Wikidata/Notes/Language_fallback) rather 
than the OSM setting or browser setting. @hlfan would it be a better match for 
the intended behavior here?

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/pull/6127#pullrequestreview-3039419713
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Including the smooth-zoom of the Qwant-map (#2283)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#2283)

As explained in 
https://github.com/openstreetmap/openstreetmap-website/issues/2283#issuecomment-506391221,
 the lack of smooth zooming is just a symptom of using Leaflet. We’re still 
using Leaflet, only with MapLibre GL JS drawing off-screen behind the scenes 
and relying on the 
[maplibre-gl-leaflet](https://github.com/maplibre/maplibre-gl-leaflet/) plugin 
as a compatibility shim. #2686 and gravitystorm/openstreetmap-website#289 
discuss migrating to MapLibre GL JS for real. We aren’t going to migrate to 
[Erdapfel](https://github.com/Qwant/erdapfel/) as implied by the original post, 
so we should track any migration to GL JS in a separate issue rather than this 
one.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/2283#issuecomment-3104898458
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev

Re: [openstreetmap/openstreetmap-website] Shortbread vector tiles >=z24 causes strange panning behavior (Issue #6222)

[Minh Nguyễn via rails-dev]

1ec5 left a comment (openstreetmap/openstreetmap-website#6222)

Possibly related to maplibre/maplibre-gl-leaflet#73.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/6222#issuecomment-3105082844
You are receiving this because you are subscribed to this thread.

Message ID: 
___
rails-dev mailing list
rails-dev@openstreetmap.org
https://lists.openstreetmap.org/listinfo/rails-dev