szabesz

And time for https://2020.stateofjs.com/ :) And here is more stats with broader view: https://almanac.httparchive.org/en/2021/

Related, so I did NOT want to create a new topic for it. Note: I am especially happy to see enums, as a think implementing state machines / statecharts is the way to go when developing "expandable systems". Not that personally I am into some large systems or anything like that, but ProcessWire is large enough from benefiting it, I think. For example, Java seems to have been having it for years: https://www.google.com/search?q=enum+for+state+machines Still, when I find the time, I will experiment with enums to see if I can use it to improve my coding. I hope I can :)

Maybe money? I mean according to the article, that's the easiest way to do it (apart from spreading the message): https://opencollective.com/phpfoundation I think that currently there is no way to do it together but correct me if I'm wrong. Maybe Ryan could set up a way to pay a small extra for PHP foundation support. What I mean is that there could be an option in the Pro module checkout process for paying an extra $5 for PHP foundation support, so that he could sign up "ProcessWire as a supporter". But the $100/month option would probably be too much, and the OPTIONAL $5 per purchase would never cover it. If there was a $30/month option for small communities like us, that might work out well. Ryan could ask them for such a support option, theoretically...

I used to "play" with VMs but I did not find them to be particularly useful, but that is a subjective topic, of course, as it depends on one's needs. Besides, I do not have Windows since I am not fond of it either. Anyway, for anyone who likes MAMP, I think saving some money on it cannot hurt ;)

For my Mac there are no too many alternatives, but there are some, of course: https://alternativeto.net/software/mamp/?platform=mac

https://www.mamp.info/mamp-friday/ I'm a MAMP Pro + Nano user, I can recommend them. Especially MAMP Pro saves me a lot of time, with features like: 1 click site cloning 1 click site spanhost & restore 1 click SSL config zero configuration Mailhog and more...

Nikita Popov is about to leave JetBrains, which triggered the birth of the PHP Foundation in order to make sure there will be enough knowledgable and paid developers behind the PHP project: https://blog.jetbrains.com/phpstorm/2021/11/the-php-foundation/ You can even apply for such a position ;) (just bad joke...)

I agree that some sort of on-the-fly option to opt out of autosave would be essential. One option is what you suggest ("PageAutoSave O N L Y when in PreviewMode?") but maybe an on/off switch would be nice to have too (next to the page title or something like that). So I am thinking of not just one way to opt out temporarily but more than one way to suit "all needs". I do not have the time to test the autosave feature, so I cannot be more specific but the issue you are facing sounds quite reasonable and that is why I commented on it ;)

It is a bit "hidden" but somewhat documented over here: https://processwire.com/docs/security/file-permissions/#securing-your-site-config.php-file Quote: "This includes /site/assets/ and everything within it, as well as /site/modules/ and everything in it (if you want to have a writable modules file system). Read the next section for details on how to do this." Also, the PW installer is also informative (sure, these are the conditions required by the installer, so later on /site/modules and config.php can be locked down, as also explained in the docs):

As soon as one starts relying on HTMX, one needs to support the project, for one's own sake, I think: https://github.com/sponsors/bigskysoftware?frequency=recurring or: https://github.com/sponsors/bigskysoftware?frequency=one-time Welcome to the ProcessWire forums!

I think so. It is only used with passwords: https://processwire.com/talk/topic/6629-config-httphosts-and-passwords/?p=64888 Quoting Ryan: "That particular value is generated randomly when you first install ProcessWire. It is forever tied to the passwords as a secondary salt. It's not technically necessary to have it, and passwords are already blowfish'd, but I've always felt better having one part of the salt disconnected from the database itself. If that salt changes or is lost, then all the passwords are forever broken."

Dear Ryan, I think you have not been "trying" to..., instead, you have been working hard on them (not just big features you mentioned today, but smaller yet still useful ones, as always). Thank you so much! Maybe it is worth taking a look at Unpoly as well: https://unpoly.com/ Various opinions: https://groups.google.com/g/unpoly/c/aKfjcIZKi4w https://groups.google.com/g/unpoly/c/w5mVyqA5zmg Demo site created by the author: https://demo.unpoly.com/

Sir, please count me in ;) Waiting patiently for your Google Form thingy. And have a nice weekend, BTW!

Hello, thanks for sharing! Have you forgotten to link to the module perhaps? eg.: https://tech-c.net/posts/login-fail-notifier-for-processwire/ https://github.com/techcnet/LoginFailNotifier

You might be interested in my "cheat sheet" posted over here: https://processwire.com/talk/topic/23641-markup-regions-pw-vs-data-pw-different-behavior/?tab=comments#comment-201538

The weakest point of WordPress has always been not relying on well established design patterns but on custom, ad-hoc solutions they made up as they went. Also, there is a great belief in not starting from scratch but trying to refactor the whole system somehow to get rid fo the technical debt, even thought it rarely succeeds, if ever. Maybe it might work for small projects but not for substantial ones. The devs of WordPress certainly failed in that. Another team trying to make a gradual but complete refactoring to make this idea come true is the devs of PrestaShop. They advertise it as the only sensible way to go, as this way – according to their reasoning – merchants can keep their investments in paid modules and other customizations applied to their shop(s), yet, when the next major PrestaShop version upgrade comes along, merchants are expected to start from scratch, somehow importing their old (and "incompatible") shop data into a freshly installed PrestaShop! They also state that module developers can better support their modules because they can keep them compatible with several mayor versions of PrestaShop without writing and maintaining their modules for each major versions separately. However, the the thing is that they do it by writing extremely long if-else statements all over the place to take account into all the differences between the systems (and there are many!), so you can imagine how fragile that style of coding is. Does that really help the module developers? I don't think so. ProcessWire used to have a v1 version and Ryan decided to start from scratch, and there must have been a good reason behind it. When something fails to deliver what it was supposed to deliver, then – in the long run – the only way to move forward is it to go back to square one. Two more "success stories" regarding starting form the "beginning": Ashley Hewson's interview : https://youtu.be/ABM60-0fbuM?t=344 Construct: https://www.construct.net/en/about Even Adobe has failed in this regard, as their applications are just somewhat similar to each other and they are still not yet rewritten, they just keep growing in size, never really bringing real UX enchantments along they way by turning into one, actually integrated suite like the Affinity apps. I have already switched from InDesign and Illustrator to Publisher and Designer. Affinity Photo has yet to deliver smart objects and pixel wrapping so that I can leave Photoshop as well, but I hope they are already working on these features. Fingers crossed ? Sorry for partially being off-topic.

I wish Apple's development practices were in line with such a feat, but sadly quite the contrary is happening these days: https://eclecticlight.co/2021/06/14/are-macos-updates-easier-than-ever/ Form Bug Sur and up, they can't even provide a Safari quick fix with a relatively small installer, but a big system updater of few gigs! https://eclecticlight.co/2021/05/03/apple-has-released-big-sur-11-3-1-update/ "Although it apparently addresses just these two vulnerabilities, it’s still 3.3 GB for an M1 Mac and 2.4 GB for an Intel model." Yikes ?

For this to work, one needs to modify PW's root .htaccess at "# 4. Protect from XSS with Apache headers" of the "loaded site", something like: <IfModule mod_headers.c> # prevent site from being loaded in an iframe on another site # you will need to remove this one if you want to allow external iframes # Header always append X-Frame-Options SAMEORIGIN # Lets the listed domains load content into an iFrame Header always append Content-Security-Policy "frame-ancestors 'self' example.com example.test www.anotherexample.com;" also, it is probably a good idea to turn the iframe into a responsive container: https://github.com/davidjbradshaw/iframe-resizer

Great work! Can't wait ?

Thanks Bernhard and Ryan! I agree. Especially regarding .../wire/modules/AdminTheme/AdminThemeUikit/uikit/src/scss which is never used. However, I disagree with "forcing" us to use node.js. There should be no need for node.js just to use core features. Maybe an additional module which is not bundled with the core but still maintained "as if it was" a core module could solve this. A lot of times there is no need for admin design customizations, for example when just installing PW for "clean install" testing purposes. What if there were a set of "additional core modules" which can still be installed with a few clicks but kept separately from the main zip file of PW? Ideas for such modules: https://github.com/ryancramerdesign/FieldtypeTextUnique https://github.com/ryancramerdesign/FieldtypeEvents https://processwire.com/talk/topic/2708-module-process-changelog https://processwire.com/modules/module-release-notes/

There used to be a "solved" feature we do not have anymore but with this last upgrade somehow it sneaked in again ? These Invision Community members love experimenting...

When working with ProcessWire always think "pages first", there are loads of creative ways to utilize them, they form the bases of the ecosystem. @Robin S and @adrian and others have lots of useful free modules to make it even easier to work with them. Just a few examples, randomly selected: https://processwire.com/talk/topic/25340-dynamic-options/ https://processwire.com/modules/process-page-field-select-creator/ https://processwire.com/modules/admin-page-field-edit-links/ https://processwire.com/talk/topic/24136-lister-selector https://github.com/ryancramerdesign/FieldtypeTextUnique https://processwire.com/modules/fieldtype-events/ https://github.com/kixe/FieldtypeSelectExtOption https://processwire.com/modules/process-admin-actions/ https://github.com/Toutouwai/AdminActionsUnorderedListToPages and more to discover: https://processwire.com/modules/ BTW, welcome to ProcessWire ?

You are right ? It has been, a lot of times. It is a know shortcoming we have to live with, so to speak.

Great question! I am also interested. I've been browsing the Forum with Dark Reader for years but a native solution is usually better.

Thanks! Yes, the first option sets the URL to the same one as the URL of the XYZ hour ago link. Thank you very much, Pete! Some examples for the "right column" feature: https://processwire.com/talk/topic/17068-markupsitemap/page/7/ https://processwire.com/talk/topic/12208-tracy-debugger/page/1 screenshots: Looks like it is the long topics which get these stats feature added automatically perhaps? I am not sure how useful these stats are. This is more of a social media like feature you talked about, which is kinda irrelevant in our case, I think too. It just takes up space. I vote for turning it off, if possible. Thanks in advance, Pete! EDIT: It is not possible to use the scroll wheel of the mouse to get to "under" the spoiler, I had to position the cursor in the text above and use the arrow keys of the keyboard to edit anything below. I wish the forum's developers spent more time on improving the RTE instead of anything else. Using the RTE is still a pita sometimes. I know that a JavaScript based RTE in the browser is the hardest thing to implement ever, but still... ?