Show form validation errors
Categories
(GeckoView :: General, defect, P2)
Tracking
(Webcompat Priority:P3, firefox104 affected, firefox105 affected, firefox106 affected, firefox107 affected)
Webcompat Priority | P3 |
People
(Reporter: snorp, Assigned: m_kato, NeedInfo)
References
(Blocks 1 open bug)
Details
(Whiteboard: [geckoview:2022h2?] [geckoview:m107] [geckoview:m108] [geckoview:m109])
Attachments
(4 files)
Fennec shows form validation messages[0], but GeckoView does not. Most of this is in FormAssistant.jsm in Fennec. [0] https://developer.mozilla.org/en-US/docs/Web/API/Constraint_validation
Comment 1•6 years ago
|
||
James can you provide a bit more context? Hsin-Yi does the DOM team have an opinion on the importance of this bug?
Reporter | ||
Comment 2•6 years ago
|
||
Fennec has some glue to provide Java UI in order to show the validation errors. We don't have this in GeckoView currently, and would prefer to have an in-content solution rather than showing Java UI (which is very error prone when it comes to placement and scrolling, etc).
Comment 3•6 years ago
|
||
(In reply to David Bolter [:davidb] (NeedInfo me for attention) from comment #1) > James can you provide a bit more context? Hsin-Yi does the DOM team have an > opinion on the importance of this bug? Hi David, not really, though it's always good to have spec-aligned behavior. P3 looks good to me. :)
Updated•5 years ago
|
Comment 5•5 years ago
|
||
(In reply to James Willcox (:snorp) (jwillcox@mozilla.com) (he/him) from comment #2)
Fennec has some glue to provide Java UI in order to show the validation
errors. We don't have this in GeckoView currently, and would prefer to have
an in-content solution rather than showing Java UI (which is very error
prone when it comes to placement and scrolling, etc).
James, does "in-content solution" mean GV would display the validation errors instead of notifying the app with a GV API?
Makoto filed bug 1539308 suggesting that GV might need something like Fennec's FormAssistant.jsm. If Fennec can show validation errors, do we need to fix this in GV before we migrate Fennec users to Fenix?
Reporter | ||
Comment 6•5 years ago
|
||
I don't think it blocks Fenix, IMHO, but would be good to fix. I think most sites are still doing their own form validation/messaging anyway. But yeah, we would need to add something similar to FormAssistant.jsm
or tell the app so it can show its own UI.
Comment 10•3 years ago
|
||
Is someone able and willing to point vaguely to the part of the code where you'd ideally add the code to make this work? I would very much prefer not to roll my own validation yet. I might have the energy to wade in and see whether I could help.
Comment 11•3 years ago
|
||
I'm also waiting for this to be added to Geckoview.
All browsers currently support HTML5 Form Validation method, so it's irritating me I have to create another solution on my website just for Geckoview.
Comment 12•3 years ago
|
||
Updated•3 years ago
|
Updated•2 years ago
|
Updated•2 years ago
|
Updated•2 years ago
|
Updated•2 years ago
|
Assignee | ||
Updated•2 years ago
|
Updated•2 years ago
|
Comment 14•2 years ago
|
||
Someone filed a new Fenix webcompat bug for this issue this week: https://github.com/mozilla-mobile/fenix/issues/26713
Since this issue is still causing webcompat problems, we should try to fix it soon, maybe in 107.
Updated•2 years ago
|
Updated•2 years ago
|
Updated•2 years ago
|
Comment 17•1 year ago
|
||
Seeing that this is tracked as affecting version 104 through 107, I'm wondering if the browser-compat-data project should reflect the partial implementation. What's your take on this?
Comment 18•1 year ago
|
||
I noticed that the issue affects latest Nightly 109.0a1, making Firefox Android the only main browser not fully supporting cliend-side hml5 built-in validation. This bug report is open from 2018, so I believe it's hard to fix. Are there any plans/dates for the fix?
Assignee | ||
Comment 20•1 year ago
|
||
I would like to use same logic of notifyInvalidSubmit in GeckoView, so
move a part of notifyInvalidSubmit to other module.
Assignee | ||
Comment 21•1 year ago
|
||
I would like to expose form validation APIs to add geckoview-junit, and
for web browser that can customize this notification.
Depends on D164987
Assignee | ||
Comment 22•1 year ago
|
||
Unfortunately, Android has no bubble style pop up notification. As
default, I add native form validation pop up in GeckoView default.
But since its API is exposed, web browser app can add custom
implementation.
This resources are from Fennec's code base, so this design is same as
Fennec.
Depends on D164988
Comment 24•1 year ago
|
||
(In reply to Chris Peterson [:cpeterson] from comment #23)
Enhancements should have severity N/A.
Is this an enhancement or a bug? It does not currently even scroll back up to the error field.
IMO this is a bug.
Comment 25•1 year ago
|
||
(In reply to jake from comment #24)
Is this an enhancement or a bug? It does not currently even scroll back up to the error field.
IMO this is a bug.
Thanks for catching that. You're right that this is definitely a bug causing webcompat problems for some sites. I'm not sure why this bug was categorized as an enhancement before.
Comment 26•9 months ago
|
||
5 years later, and we're still not seeing form validation errors on Firefox mobile. There's no pop-up or emphasized border. Interestingly, it works perfectly on Firefox in Windows and Chrome on Android. Thank you for your continued efforts on this; it's much appreciated. Could you please provide an update on the current status of the issue?
Comment 28•5 months ago
|
||
Hi, is it possible to have some feedback about this issue?
If you are working on it, if there is a possible date to deploy the fix, if is not in your radrs for some reason...
Thanks!
Comment 30•4 months ago
|
||
Yes, sorry about that. I looked but could not locate reports of this bug elsewhere on Bugzilla. (Shoud have cast a wider net; would have found reports on Reddit, GitHub, Stackoverflow, yada yada. Oh, well.) This thing has been hanging around for five years or more -- is nobody addressing it? This particular bug is seriously affecting my efforts with one client in particular. Must I bribe someone to get some action?
Comment 32•3 months ago
|
||
Makoto Kato [:m_kato], given that your pull request hasn’t received a response from the reviewer in a year, do you think it is reasonable now to ping them again?
Comment 33•1 month ago
|
||
It took so long.
Description
•