wbmnfktr

Be careful here @Autofahrn and @psy. @VeiJari is talking about init.php and not _init.php. init.php ready.php finished.php Just to keep this in mind. https://processwire.com/blog/posts/processwire-2.6.7-core-updates-and-more/

I'd support the /site/ready.php solution as well. It seems to be the better place for hooks that don't end in a module. But there is another thing. Your hooks is tied to the published state or the action of publishing it. While testing this you have to unplublish the page first to test your hook. Had this testing issue a while back, too. ?

It could be my eyes but... ? in your second example I can't spot any difference.

That's indeed kind of a strange behaviour. What setup do you use? Laragon, XAMP,... It could be interesting to spot the root of what's causing those problems. I had once problems with Laragon while re-using previous used database names. But that's long time gone now. Regarding that setlocale mistery... there will be an upcoming multilang project and I will keep an eye on it. But for now I really can't remember any issues.

Microsoft Edge arrived in its Chromium-flavour version. For those who want to give it a try already. Right now you can download the DEV (weekly updates) and CANARY (daily updates) versions. https://www.microsoftedgeinsider.com/en-us/download/ Windows 10 needed. Windows 8.1/8/7 and macOS will follow soon.

Quick YouTube test... ProcessWire 3.0.123 VideoOrSocialPostEmbed 1.2.1 So... you might want to check your setup and settings.

First of all I'd limit image size here. You can define max width/height or even pixels in the image field. The image field takes care of that for newly added images. That's sounds like a very broad range of variations. Are they really necessary? If you don't build something like an image library 1920px should be more than enough. Using the code @LostKobrakai mentioned before you will upload one (1) image and ProcessWire will take care of everything else. Or you could use Croppable Image 3 that takes care of some things or everything and modify the img/srcset code accordingly to match the module syntax. And to cite @LostKobrakai again: ?

No... it always depends on the project and website. But there are a few things I don't want to miss. Those are: SSL/https with Let's encrypt SSD webspace because it's faster Admin panels like cPanel are nice but not necessary SFTP/SSH is a benefit Cron of any flavour is highly appreciated GDPR compliance in terms of data-processing-agreements (or however they are called) A good reputation (4.5+ ratings) A good website with all relevant details (Imprint, Privacy, GDPR) A good website that's not powered by a WordPress theme More than 5 years in business Servers in my home country Free or local support hotlines + the things mentioned above Yes... It's more about the company than the webhosting itself but these things are as important as everything else. Maybe even more. Nope... never did anything like that with a ProcessWire project. FileZilla is S/FTP so about 99% of all webhosting companies should fit here. ? I prefer Site Profile Exporter as I create a full copy of everything with only a few clicks. I know others here that prefer Duplicator. So... try everything and decide which way to go. Regarding sync-scripts of any kind... not mine at all. As said before... it really depends on your needs. I never used webp support in a live project so far and therefore never had the need of special modules. But yes... put it on your checklist and later on ask hosting companies if they support those features.

First part... A quick test (ProcessWire 3.0.123 and FI 1.0.0) where I exactly did what you described and... it worked as expected. Added FI language pack to default Logout / Login Admin in finnisch Deleted all language files in default Logout / Login Admin in english again There were no mixed translations in the backend at all. Please try to log out and log in again. That sometimes helps with language hiccups. Or even clear browser cache including cookies / try another browser. Second part... Have had the same feelings about language handling at the beginning. But after ~20 multi-lang installations and lots of hours playing around with it my feelings changed. Works ~95% of the time I need it. Never had any real issues with it. WordPress: total fail Drupal: maybe it changed but a few years back it was quite painful Typo3: never seen a multilanguage setup that worked well Textpattern: mh... nope Silverstripe: can't really remember if it worked that well To be honest... I'm super biased in terms of ProcessWire as it gave me so much options and kind of freedom in the past to get almost everything done right from the start. But nonetheless I can't remember any real issues with multi-language setups, neither in front-end, nor in the backend.

Most of my clients already have some sort of webhosting and only in very few cases it needs an update in terms of more features like SSL and similar things. If they don't have webhosting I recommend a middle-sized shared hosting from trusted providers. An all-in-one package works in almost every situation. Domain + Hosting + E-Mail management = ~15 EUR/month that's it and all they need. In the past I have had root servers and VPS to host client projects. It was a nice option, additional money, and a good way to keep clients close. But as regulations and laws changed over the past years I decided to stop offering hosting services. Hosting isn't that expensive anymore and you find really good options out there. Hosting providers are there for a reason. They take the hassle away from me. If your client already has a website, take a look at their web analytics to see how much traffic there is. You can then decide which way to go and what package might be a good deal if there are performance issues. As far as I can say... a good ProcessWire site doesn't need that much of a high performance hosting. Unless you can't use ProCache even a super cheap $3 shared hosting might work well enough. ProcessWire sites with ProCache, almost no real-time-database interactions, and only a few forms have a nice small footprint.

$settings = $purifier->getConfig(); $def = $settings->getHTMLDefinition(); $def->info_tag_transform['h1'] = new HTMLPurifier_TagTransform_Simple('p'); $def->info_tag_transform['h2'] = new HTMLPurifier_TagTransform_Simple('p'); $def->info_tag_transform['h3'] = new HTMLPurifier_TagTransform_Simple('p'); $def->info_tag_transform['h4'] = new HTMLPurifier_TagTransform_Simple('p'); $def->info_tag_transform['h5'] = new HTMLPurifier_TagTransform_Simple('p'); $def->info_tag_transform['h6'] = new HTMLPurifier_TagTransform_Simple('p'); $clean = $purifier->purify($dirty); That part is awesome. Should bookmark this for later migrations.

In my experience it depends on how you import the content. Via Copy&Paste... just paste as plain text or go the extra route and paste before in a text editor of your choice. While using database import or migrator module you might want to trust the textarea field settings. It takes care of a lot of extra markup already. You might have to iterate via API over all pages to make the clean up happen or push it manually from the settings. You won't get rid of wrong H-tags but most of the other clutter will go.

Kind of related and at least interesting. https://medium.com/@umpox/be-careful-what-you-copy-invisibly-inserting-usernames-into-text-with-zero-width-characters-18b4e6f17b66 https://medium.com/@aidobreen/hidden-text-fingerprints-and-how-to-avoid-them-d0103edd2ce4

That one?

I don't want to undermine your suggestion but in my opinion the presets are fine. There is no one-size-fits-all-needs solution in this case. All websites are different in the matter of design and how it was achieved. I think the provided solutions are a very nice foundation that almost work in most situations already. At least they are a good starting point to build custom solutions later on.

And people just don't want to wait for a website to load.

Why don't you modify or create your own CSS for it?

It's almost the same with writers and (freelance) journalists. They get paid either by (a low) hourly rate or article-based or only a one-time fee. They often sometimes (kind of) sell their copyright and transfer their "possible future profits" based on views or re-prints of their work the big publishers. And with this new foundation publishers' options for bypassing money got even more easier in some cases. Regarding copyright fee... that's already in place since at least 2008. I don't know how it's called in Hungary but here it is called Urheberrechtsabgabe/Pauschalabgabe - see Wikipedia (DE) more more details.

What's happening in your lendout template right after // STUDENT? There is an empty foreach() loop in your snippet. Is that correct?

Is there anything in the real rendered source (not inspector view but CTRL/STRG + U)? Does it render weird while CSS is disabled? Does it render weird while JS is disabled? How does another <h4> above/below gets rendered?

Wow... it took me a while to find that cart button. So... the real magic happens on pages that aren't visible to non-students? As you know who still has something to return you maybe could limit your query to only those students. And maybe it's not even necessary to have those details in real-time quality. Maybe cached pages (hour or day) is enough.

First things first... Why do you load this whole set of module stuff on every page? As far as I can see there is no need for that. As part of this your average page size is ~1.2MB which is way too much for whats happening on the pages. At first I thought you were facing a hosting issue here but static files are loading fast so I guess your querying a lot of data on each and every page load and don't use any caching where possible.

At least for 1Password there was a real good guide. LastPass allows CSV imports which make things easier. There is no real reson for you to move away. I have had license problems and didn't want to pay each month for something I already paid for several times. A while back they changed their pricing model and switched to montly fees. I have had 3 licenses (first one bought in 2007) in total back then plus the iOS apps and then suddenly it became awkward. I didn't feel comfortable being pushed towards monthly fees to use syncing and everything. They were nice and did some good moves for a while but at the end I moved nonetheless. I'm not angry with them and still recommend their solution but no... I'm out. Looked for an alternative solution and found it with LastPass.

A sad day for the internet, the EU, the democracy, and so much more. I don't know anything about any #article13 discussions, protests or media coverage in Spain or anywhere else but at least here in Germany - especially in the last weeks and days - the way of communication from politicians towards protestors, their statements against protests, and media coverage that supports that copyright directive were kind of shocking. They said some really weird things like... Online-protesters were bots. Mailings were from Google (as most of those mails came from gmail.com addresses). Those on the streets were paid [from big US companies]. ... and so on Yet another bad aftertaste especially for Germany and our politicians: https://edition.faz.net/faz-edition/wirtschaft/2019-03-26/f30a5870c08cc1e1b4524c1be19d1faf/?GEPC=s3 Maybe I never fully realized that degree of misleasing information from politicians and media but now I'm not only upset but actually really p*ssed. There seems to be a last option left but I really don't know if this will ever happen but... maybe. Back to your topic: those already big platforms will do fine with that. YouTube already has its Content ID system in place and therefore can offer APIs to those who will need it in the future. At least for video material. But the main issue I have with this copyright directive is this argument "Copyright-holders and creators should earn money with their work!" which is totally fine by all means. But that directive will not help any content creator on any plattform. How could it? Independent writers, musicians, DJs, fotographers, even programmers will still have the problem to find all those pirated copies of their work. They won't earn more money. They still have to prove their copyright claims. But on the other hand publishers got some more power to bypass money from the real creators. And they got their link tax foundation they always asked for. But... there is a positive side-effect. I learned to look closer. I learned to trust less. I learned to think more than twice. I learned who might not be on my side. And I learned who to vote for the next election. Regarding Alex Jones and de-platforming: That was never a real topic here in Germany at all. At least I saw only very few articles or side-notes somewhere or at the bottom of a newsfeed. That's it. And if you find anything about it, they applaud about it because he is some kind of a bad person. This narrative or bias at Twitter, Facebook, YouTube, [name more here] and the banning of people from their platform wasn't invented by those platforms. They would be doing totally fine even with all of those now banned people. Maybe even better. It was cultivated over the last few years.