www.keisan.nta.go.jp - Firefox is an unsupported browser
Categories
(Web Compatibility :: Site Reports, defect, P3)
Tracking
(Webcompat Priority:P3, Webcompat Score:1)
People
(Reporter: ksenia, Assigned: twisniewski)
References
(Depends on 1 open bug, )
Details
(6 keywords, Whiteboard: [webcompat-source:web-bugs])
User Story
platform:mac impact:blocked configuration:general affects:all diagnosis-team:webcompat branch:release user-impact-score:0 outreach-assignee:birtles outreach-contact-date:2025-09-11
Attachments
(5 files, 2 obsolete files)
Environment:
Operating system: Mac OS X 10.15
Originally reported Firefox version: Firefox 111.0
Last reproduced with the following UA: Mozilla/5.0 (Macintosh; Intel Mac OS X 13.4; rv:126.0) Gecko/20100101 Firefox/126.0
Platforms reproduced on: mac
Steps to reproduce:
- Go to https://www.keisan.nta.go.jp/kyoutu/ky/sm/top_web#bsctrl to file the Japanese Tax
- Click on the first item: スマートフォンを使用してe-Tax
Expected:
Be able to use Firefox to file the tax return.
Actual:
This message which says you need to use Safari.
ご利用のブラウザでは、QRコードを利用したマイナンバーカード方式によるe-Taxをご利用になれません。
ブラウザをSafariに切り替えてください。
推奨環境を確認する方はこちら
Actual Behavior:
Forced to use Safari on macOS
Created from https://github.com/webcompat/web-bugs/issues/118144
Updated•1 year ago
|
Updated•1 year ago
|
Updated•1 year ago
|
Comment 1•1 year ago
|
||
Mass-assigning diagnosis-team to webcompat for "Firefox is not supported" bugs that don't already have a team assigned. This action is done by a script. For your convenience, feel free to filter your bugmail with 559a9604-41ec-11f0-9ec6-f3f21dcd7cf2.
| Assignee | ||
Comment 2•1 year ago
|
||
I can still reproduce the Safari-prompt after clicking on the left (はい/yes) button, then the next, the choosing the left button (スマートフォンを使用する). It seems to not allow me through on MacOS with a Chrome UA override, but rather a Safari one. We might need to spoof as Chrome on other OSes, so we'll need to check on all platforms here for an intervention.
Comment 3•1 year ago
|
||
I can get past the initial check if I spoof as Chrome on Windows. I managed to get through the initial MyNumber login but I can't confirm everything works without actually submitting a 確定申告.
I think we should attempt to contact the NTA to see why they don't include Firefox as a recommended browser.
Comment 4•1 year ago
|
||
Tried to find a suitable point of contact at the NTA today but no luck so far.
Comment 5•1 year ago
|
||
Contacted via https://www.e-tax.nta.go.jp/toiawase/toiawase3.htm
Updated•1 year ago
|
Comment 6•11 months ago
|
||
Can I ask you to make a sitepatch for this? As per comment 2 and comment 3, it looks like we want to spoof Safari on Mac and Chrome (or Edge) on Windows.
| Assignee | ||
Comment 7•11 months ago
|
||
Yes, I'll make a sitepatch for Windows and Mac only, since they don't seem to support Linux or Android even with Chrome.
| Assignee | ||
Comment 8•11 months ago
|
||
Updated•11 months ago
|
Comment 10•11 months ago
|
||
| bugherder | ||
| Assignee | ||
Updated•11 months ago
|
Updated•11 months ago
|
| Assignee | ||
Comment 11•8 months ago
|
||
They've updated their minimum version checks, breaking our intervention. The Safari version we use is now too old, and they also check for Windows 10 in the UA string for Chrome, requiring 11. We'll need to update our code appropriately.
| Assignee | ||
Comment 12•8 months ago
|
||
Comment 13•8 months ago
|
||
Comment 14•8 months ago
|
||
| Assignee | ||
Comment 15•8 months ago
|
||
Updated•8 months ago
|
Comment 16•8 months ago
|
||
Comment 17•8 months ago
|
||
| bugherder | ||
Comment 18•7 months ago
|
||
With sitepatch enabled, the site directed me to install the Chrome version of the Myna Portal app and extension.
Note that the Myna Portal app and extension has a Firefox version, but NTA does not instruct me to install the Firefox version because they do not officially support Firefox and we are spoofing the UA.
Comment 19•7 months ago
|
||
If I already installed the Firefox version of Myna Portal app and extension, I am able to proceed to the next screen.
Comment 21•7 months ago
|
||
(In reply to Thomas Wisniewski [:twisniewski] from comment #7)
Yes, I'll make a sitepatch for Windows and Mac only, since they don't seem to support Linux or Android even with Chrome.
Hi,
I recently opened a Linux compatibility issue (https://github.com/webcompat/web-bugs/issues/208836), which was eventually moved to bug 2019933 and marked as a duplicate of this one.
Since this bug is now the primary landing spot for Linux support, I wanted to reiterate a suggestion from the webcompat thread: would OS spoofing be a viable intervention here? Given that we already use interventions like change_OS_to_MacOSX, perhaps a similar sitepatch could enable support for Linux users as well.
Comment 22•7 months ago
|
||
(In reply to Akihiko Odaki from comment #21)
I recently opened a Linux compatibility issue (https://github.com/webcompat/web-bugs/issues/208836), which was eventually moved to bug 2019933 and marked as a duplicate of this one.
Bug 2019933 has " Platform: Desktop Windows 11". Moreover, bug 2019933 comment #0 says
Operating system: Windows 11 and Mac
Firefox version: Firefox 147.0
Blame @rbucata for poor treatment of the platform.
| Assignee | ||
Comment 23•7 months ago
|
||
(In reply to Akihiko Odaki from comment #21)
(In reply to Thomas Wisniewski [:twisniewski] from comment #7)
would OS spoofing be a viable intervention here? Given that we already use interventions like change_OS_to_MacOSX, perhaps a similar sitepatch could enable support for Linux users as well.
I'm hesitant to do so, as the site blocks Linux specifically. It's a site related to taxes, and I wouldn't want folks to run into weird issues while filing taxes, and I cannot test everything on the site to confirm if it truly works well. That said, I can add a spoof to nightly builds of Firefox for you to test? If you can confirm that the site seems to work to do what it's intended to do, that might help us make a decision.
Comment 24•7 months ago
|
||
(In reply to Thomas Wisniewski [:twisniewski] from comment #23)
(In reply to Akihiko Odaki from comment #21)
(In reply to Thomas Wisniewski [:twisniewski] from comment #7)
would OS spoofing be a viable intervention here? Given that we already use interventions like change_OS_to_MacOSX, perhaps a similar sitepatch could enable support for Linux users as well.I'm hesitant to do so, as the site blocks Linux specifically. It's a site related to taxes, and I wouldn't want folks to run into weird issues while filing taxes, and I cannot test everything on the site to confirm if it truly works well. That said, I can add a spoof to nightly builds of Firefox for you to test? If you can confirm that the site seems to work to do what it's intended to do, that might help us make a decision.
I'm willing to test if it helps. Thanks.
| Assignee | ||
Comment 25•7 months ago
|
||
Updated•7 months ago
|
Comment 26•7 months ago
|
||
Comment 27•7 months ago
|
||
Reverted this because it was causing mochitests failures in browser_ua_helpers.js.
- Revert link
- Push with failures
- Failure Log
- Failure line: TEST-UNEXPECTED-FAIL | browser/extensions/webcompat/tests/browser/browser_ua_helpers.js | test_ua_helpers - getDeviceAppropriateChromeUA({"ua":"Linux","config":{"OS":"windows","noFxQuantum":true},"expected":"Mozilla/5.0 (Windows NT 11.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/143.0.0.0 Safari/537.36"}) - Got "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/143.0.0.0 Safari/537.36",
Comment 29•7 months ago
|
||
Comment 30•7 months ago
|
||
| bugherder | ||
Comment 31•6 months ago
|
||
FWIW, I was able to go through filing my taxes on this site on Linux with nightly. Would it make sense to extend the override to all branches?
Comment 32•6 months ago
|
||
(In reply to Pulsebot from comment #26)
Pushed by twisniewski@mozilla.com:
https://github.com/mozilla-firefox/firefox/commit/3a9e38c433a6
https://hg.mozilla.org/integration/autoland/rev/fec040a378fa
add a nightly-only Linux-only UA override for keisan.nta.go.jp and fix
Windows override; r=webcompat-reviewers,denschub
Perfherder has detected a awsy performance change from push fec040a378fa5998b31d61e74fe73dc2ecbb4fc5.
No action is required from the author; this comment is provided for informational purposes only.
| Improvement | Test | Platform | Options | Absolute values [old vs new] |
|---|---|---|---|---|
| 0.26% | Base Content JS (doc) | windows11-64-24h2-shippable | fission | 1,517,873.33 bytes -> 1,513,960.00 bytes |
Need Help or Information?
If you have any questions, please reach out to fbilt@mozilla.com. Alternatively, you can find help on Slack by joining #perf-help, and on Matrix you can find help by joining #perftest.
Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests.
Updated•6 months ago
|
Comment 33•6 months ago
|
||
Comment 34•6 months ago
|
||
Updated•6 months ago
|
Comment 35•6 months ago
|
||
Comment 36•6 months ago
|
||
Comment 37•6 months ago
|
||
Backed out for causing mochitest failures @test_persistent_contentscripts.py.
| Assignee | ||
Comment 38•6 months ago
|
||
Ah, we will fix the issue and re-land today.
Comment 39•6 months ago
|
||
Comment 40•6 months ago
|
||
| bugherder | ||
Comment 41•3 months ago
|
||
It seems like the intervention is broken here. I'm only able to use Edge on Win11. Even Chrome seems to gets rejected.
| Assignee | ||
Comment 42•3 months ago
|
||
Indeed my automated test seems to be failing now for Linux and Windows, yet still passes on MacOS.
| Assignee | ||
Comment 43•3 months ago
|
||
Ah, this was a bug I introduced in bug 2039461. It isn't running the content-scripts that it generates in the right order for this intervention. Thankfully it seems like this is the only site affected, and the fix is straightforward (I'll do it in bug 2046904).
| Assignee | ||
Comment 44•3 months ago
|
||
@jrmuizel, okay I've landed the fix, so this intervention should start working again on windows and linux on the next nightly. It will make it into 153, and if we feel strongly enough to push it into a 152 dot release, let me know.
Description
•