Pete

@Robin S thank you for fixing the issue I reported so quickly on Github - I've not looked at new modules in a while and several of yours are going to be real time-savers for me on a current project. I have a feature request but I'm not sure how difficult it would be. When you have the field limited to only one item, I'd like to be able to just click on the image rather than have to delete it via trash icon to select a different image. I think in terms of UI, instead of seeing the "drag handle" icon on hover for a single image it could show some other symbol and a tooltip perhaps stating that it would replace it? My use case is for a website where they are selecting an image to use in search results and categories on the site and it's just a nice easy visual field to do that with, but if they want to play around selecting the best one they need two clicks to delete and select a different photo instead of one, and there are >100 pages where they might want to do this.

I don't think you can rely on Omnipay for all but it may be okay for some. A warning sign for me is the age of some of the open issues in the Omnipay PayPal repo: https://github.com/thephpleague/omnipay-paypal/issues and the date of the last commit being quite a while back. A few other Omnipay modules have similar issue lists. PayPal and Stripe I imagine are going to be the two most common and have pretty good docs and vaguely similar interfaces (they both suggest the JS checkout - PayPal are starting to direct developers that way and are using wording suggesting the old ways of initiating the payment are becoming legacy though I can't remember the exact docs). Both have the PHP libraries for manipulating payments after the transaction is in their systems which is fine as I believe they're all trying to make the PCI server compliance issue go away by the card data never being on your server. Even Worldpay has gone JS + PHP now with a similar implementation from what I've briefly seen of their docs. I would think that if any were to be officially supported as part of the project it would be PayPal and Stripe as the two most common. PayPal is still going to be more common than Stripe sadly so probably that as the first one. I do agree with the Omnipay approach of a common interface, so perhaps if these were the first two and @kongondo you took inspiration from how they've done it in terms of the common interface, then those devs capable of writing additional payment modules can use the same approach. Or perhaps because this is such a critical piece of the puzzle it needs collaboration right from the start with those with a lot of experience with payment libraries - I'm not sure - but that wouldn't be me (I just follow the provider docs and if it works it works ? ).

I still use it. I do have one workflow thing I do where I add a "keyword list" textarea field below the keywords as the SEO tool we use to generate lists of keywords for subjects spits them out one per line, so we copy and paste that into that textarea and on save I turn it into a CSV string and put it in the proper keywords field. Sounds a bit lazy but one site had hundreds of articles. I may also need to build a Y0ast-like page score module at some point soon too, though I think those things can be misleading as much as they can be helpful.

Yes my colleague tested the demo site and Apple watch worked as did iPhone ? I may get rid of my Yubikey in future as fingerprint auth on phone is as good in my opinion, and I'll have my phone closer than my keys usually. My face is also usually nearby when I'm on my laptop ?

It's worth noting, on Windows where I have Windows Hello set up, it was setup with my face to login, but also allows the other connected options such as my Windows pin and hardware key though I've only had to set it up with my face - so any linked login methods under the same umbrella become available which is great if I'm wearing a balaclava during the usual cold British weather ?

Perfect - works on live. This is so good ? I had my phone plugged into my computer, so added Windows Hello (face recognition/pin) then added my Pixel phone and I can login with my fingerprint on that and face on my laptop. Absolutely awesome - love it! I'll point it out to Ryan but so far it works great!

Ah I'm willing to bet it's because I'm in a local dev environment or something - will try on a live site shortly.

Oh wow, thanks Adam! I'm stumbling at the first hurdle it seems:

I'll admit to not having looked at the PW auth modules in terms of code at all. I might ping Ryan and see what he thinks. Can't hurt to have more brains on the case and his is bigger than mine ? I think for all my logins, since I use a laptop, I would use Windows Hello... though actually then if I want to check something on my mobile without grabbing my laptop I can't can I? ? I see the problem now. Perhaps multiple instances of the same module somehow and it lets you choose on the login page? Otherwise as you say it just uses the one you set it up with by default.

Thanks for looking at this Adam. I just got a colleague who uses Apple for everything to test and it allowed tapping her Apple watch to login and Face ID. She's got another device to try later with fingerprint but it does sound like it does all devices which is nice, and gives plenty of options, so seems like where things are going in the future. If there are changes required to the base TFA class to help in any way let me know and I can ask Ryan ? Sometimes things won't be hookable that it would be helpful to be hookable etc. I think with WebAuthn It would be more useful if the base class allowed for more keys to be stored per module right, so then per-website I could set it to allow login via facial recognition and fingerprint on the same app - that sort of thing - rather than just allowing one key per module which I think is how it works now? Also happy to help/test in any way I can on this one. As I say for me it's not a huge thing the current module stopping working as I have alternatives, but WebAuthn has me a little excited as it seems like a more natural way to allow people to login.

@Adam this looks like a good potential option - it's one I'd noticed a while back: https://github.com/lbuchs/WebAuthn?ref=https://githubhelp.com but it may also be the one you're referring to with the long example code ? There's a good demo - I was impressed. On my laptop it allowed facial recognition via Windows, PIN via Windows and I had my Google phone plugged in so that was a third option using the fingerprint scanner! It says it supports the physical keys like Yubikeys too and is a PHP/JS library that's been updated reasonably recently. I may look into it myself at some point but like you I wouldn't be able to soon.

@kongondo can you remind me on Monday and I'll get a forum set up? Not sure if we should call it Padloper Pre-Sales or something and have a pinned topic at the top saying for paid support to request access to the other forum or something like that?

Ah nuts, well Chrome is here to stay and Edge is based on it too, so for business environments where this is more likely to be used I'll most likely be one of those two browsers. It's not a big issue for me - I can switch to one of the other auth options for now so whenever you get time. Hopefully now this new standard is out it won't change again for a long time after that.

@Adam For the last few months, using this module shows this warning in Chrome: Is it a big task to change the module to use that? It seems to suggest the module will simply not work after February ?

I'm interested - getting more requests for ecommerce now and would prefer to be using ProcessWire.

I sometimes add a repeater matrix type that u only want to use under a certain template or part of the page tree - is it possible to limit that already or should I put in a feature request? Love this post by the way as we also heavily use repeater matrix to build out content using blocks.

I think we need to setup ElasticSearch for the search index as it's partly a limitation of the MySQL setup but also so searches don't overload the forum itself when people search for "the" etc (imagine if someone just wanted to slow the whole system down - you can do it fairly easily with 3 letter searches as it'll find a lot more matches even with flood control systems in place). I'd like to get it setup eventually as Horst's suggestion works for public forums but for other parts like the Pro module forums it doesn't work there. I thought we had it in place already actually but apparently not - I'll look into where we got up to with it.

Sorry I'm not sure what that could even be. I don't think anyone else has reported that before.

I'm very much looking forward to the Snapshots module ?

Reputation count is now back next to post count ?

I'll see what I can do about the rep count - in the meantime I've done something someone else asked for (and my older eyes needed) - increased font size on mobile. I haven't gone too far with it but it's a lot better than 13px ?

Not sure why they removed it from the design flow, but you can easily hover any member's avatar to see it:

I found the setting for this and turned it off: I get why they added it, but for some communities it makes more sense than others. Personally I think they could have added a setting to control the truncate length to be a "best of both worlds" but switching it off is fine for this forum.

We'd have to submit the forum to their app to get it included. I'm not sure what the process is exactly but you do lose navigation to the rest of the site - it basically just shows the common parts of the forum and throws out any theme customisations from what I can tell. No idea. Maybe when you add stuff to the basket it allows you to complete the purchase from anywhere, but you'd think even if you paid via a modal it would just load when the modal loads ??‍♂️ There is JS and CSS hooked in from the main site too in a very basic fashion to make theme upgrades simple (like we're including all of UIKit for the header) so not sure if it's from there but I don't imagine it is. Like I say, it largely "is what it is" with each upgrade because they like to change things every so often.

Ah well unfortunately I won't be able to do anything about that. They do now have an app but I'm not sure if there's value in adding us to it - theoretically that would run faster but probably wouldn't solve your issue.