Pete

I did hear, but haven't checked the facts, that the head of Nokia is an ex-MS guy and through this will become a MS employee again. From the outside and with none of the facts to hand It looks a bit like he went to Nokia to help them out and in the process set up a relatively cheap deal knowing he'd be back at Microsoft. I know it doesn't sound like a cheap deal, but compared to Vodafone's recent transaction it's pocket money

Yeah, but I wanted to take it one step further and hide some of them based on roles. I had totally forgotten about that module though so thanks guys

Hmm yes, I hadn't thought of required fields (I was more thinking about field dependency in that if the user has a role then show that field) - required fields make it a bit trickier as all required fields should be shown of course regardless of the user's role or the page won't save. That might be a useful thing to have in the future though for editing workflow - I guess a simple way of doing it might be something like this: The editor edits the pages they have access to - before the page can be saved, they fill out their fields. If the manager role can see other fields and one or more is required, when the editor saves the page in the step above it simply cannot be published, but it can be saved. When the manager finishes filling out the other required fields then the page can finally be saved. Something like that might get around the issue.

ryan - just a thought, but would it be possible to make it so that fields can be shown depending on whether a user has a certain role, or even ID? Certainly roles would take you one step closer to a system I used to use where there was an editing flow with a website's content - someone would work on content, then their manager would fill out some other fields on the same page, sometimes in another tab in the editor, before finally publishing a page. It's not something that everyone needs, but to my untrained eye it also doesn't look like it would be too difficult to add (*ducks for cover*), would make it even more powerful and could be useful in many different scenarios. I guess the selector for that would be something like user.role=editor|manager as well as user.id=41 maybe for scenarios where the main admin user wants to test some stuff or have hidden fields that other superusers can't see (though that's silly as other superusers could give themselves access anyway - just thinking of odd scenarios!). Can anyone else think of uses for this?

I think I've been getting confused - it must have been in a private beta earlier this year and I'm not as blind as I thought. The actual launch appears to have been the 15th of August so I don't feel so bad now: https://stripe.com/blog/introducing-stripe-uk I think the only drawback is that they don't offer the ability to transfer to your bank account on demand, favouring a 7 day automatic rolling payment schedule (to be honest having the money in their system a few days longer is probably one of the ways they make a profit from the system aside from the fees - which are 1% lower than PayPal). I don't think that that is really a huge deal though as there is usually some waiting period for transfers with most merchants and in this case you can know in advance what day each week money will land in your bank. The fact that they've opened a London office and are working with other European countries as well as Australia also shows their commitment to going global and the fact that they are likely to be around a while. I think they've got the potential to take quite a slice of the market at this rate, or certainly stir it up and kick a few bums into action in terms of the more archaic merchants out there.

OH MY GOD I COULD NOT BE HAPPIER ABOUT THIS NEWS! Yes, my caps lock was on for a reason there, and yes, I'm working on a module for Jason already During my tests, Stripe can be described as nothing short of awesome. It's got the nicest looking backend (ahem) and is a doddle to use! Module on its way soon The great thing when you go to click on a button to make a payment is that stripe's module just slides down the screen in sort of a modal overlay, but the card details being entered are going straight to Stripe's site, so no PCI headaches as whilst it looks like payment details are entered on your site they're not and the customer gets a smooth experience to boot. Basically, other merchants should take note (yes, you too PayPal). EDIT: Oh, it's been around in the UK since March? Why did nobody tell me? Last time I had a look for decent UK merchants was February and didn't notice when doing the module for Jason that it was available in the UK too

Nice, functional, fast site - I like it. That's ProCache!

I think it's more that so many of us hadn't thought about this before but can instantly see the benefits ryan, not that we've necessarily all been waiting for it before now. I think that's testament to how well it's been implemented.

Great addition, thanks ryan or building it and Antti/Avoine for sponsoring it! It can make several of my more complicated forms on some projects a whole lot shorter and more legible now And like many others, I marvel at how well ryan's initial versions of amazing features just seem to be so polished!

It's all possible in ProcessWire, but that looks like a more complex site than most. To my mind it would almost be easier to map out the templates on paper first to work out which fields you need for which templates and how they interact, then we could get a better idea of time involved. The problem from an outsider's perspective is that to me it looks almost over-complicated in places (so many options for searches that might be able to be streamlined using something like $pages->find('template=images|contributors|locations|discussions, keywords%=your search phrase') for example to search a common keyword field across multiple templates, assuming you have that field against many templates).That took me a bit off track, but the rest of the site looks quite large and very bespoke is what I'm getting at - it would be very difficult for an outsider to give you any idea of timescale to reproduce it in ProcessWire. Having said that, if I was going to do anything as complicated as that myself then I would be 100% confident that you could build it all with ProcessWire. The sky is the limit, it just depends how much time you have

I have just one problem with this approach. It's perfect for what I need for a site where customers can buy things/services as every email address is unique, however a username might not be! For example, if I have two customers called John Smith sign up, I have to do a check during registration to make sure that if there is a John Smith already that a number gets appended to their username (so john-smith-1 or something like that). Feels a bit messy, but there really isn't any way around that I guess? I'm being too picky though I think as even when a user edits their profile they don't need to know that behind the scenes their user page is stored with a name of john-smith-1 Okay, I'll shut up now since I've talked myself out of actually having a problem!

Cool Errm... many Mac apps aren't, and vice-versa.

I always forget about that - another good reason!

That's the best way to go about it - you can then take it further if you ever need to by doing searched on pages with a certain colour etc. Might not be that likely in your case, but if it was for a clothing website for example then something that lets the user filter by red clothes if it's their favourite colour might be desirable. Sometimes it can feel a little overkill to do it this way, but it always leaves you with future possibilities rather than just static values

Yes. You can create a textarea field and select to use Markdown as the textformatter. You can also use BBCode or Textile, but Markdown is a core module.

That script is basically just iterating through the files and folders and parsing, much as you would do with the script I linked. What we need to see to be able to help further is maybe a screenshot of the folder structure of the site (is it many levels deeo or are all pages in one folder for example) and perhaps a test page or two - without those I'm not sure we can suggest anything else! Iterating through the folders and using a script like I linked to or even a regular expression is as hands-off as you'll get, but if we can have the information I just mentioned we can probably hook you up with some code

Here's a link to a conversation about parsing HTML pages - I used the script in question and it is very good, as long as each HTML page follows sensible structures: http://processwire.com/talk/topic/569-php-simple-html-dom

I try and back it all up to Brain 1.0, but I think Brain needs an upgrade

It's whichever version is under the Dev branch on Github - the verisons in the readme or anywhere else in the Dev branch are irrelevant. The point about the Dev branch on Github is that it will always contain the very latest code regardless of whether a version number has changed or not - if ryan changed the version number every time he committed a batch of new code we'd be on version 2.3.712 as he has uploaded something new to the dev branch 712 separate times since v2.3 was created (or maybe since 2.2... can't remember if this was a new Git repo for 2.2 or 2.3). These might be a few updated files or just a single line of code, but every time new code is pushed to Github it's a separate "commit". Long story short, version numbers don't change with every commit to Github - they change when those commits are merged with the stable (Master on Github) branch. The dev version isn't really mentioned on the PW site itself as it's for developers who wish to tinker and other curious parties and is not generally recommended for live sites - it's "beta" really, at least the changes are - but some of us can't help ourselves

I think I've been relying on pen and paper too much and need to look up a lot of these

It will read in anything from the mailbox and delete the emails once it's done, so be careful with that!

I've done what horst suggested before to be honest - you don't have to worry too much about the headaches of duplicate data as all you're doing is creating a dummy PW user account and the way I did it was have one extra integer field in the PW user template to store the ID of the website account - simple as that You could then take that further if different user groups from the intranet system have different access rights on the website. I would personally make it do some checks during PW login so that: If someone enters a username and password, it checks the intranet DB and if it finds the details there but on in PW then create an account If someone is in the intranet DB but is now a disabled account, disable them in PW (unpublish) If their email has changed in the intranet DB, change their PW email address Definitely do the same with password Basically you're always authenticating against the intranet database first, and you can take it further by mapping different user groups or more as you see fit. It's the only way I'm slowly able to migrate sections of a large intranet I custom-coded several years ago to ProcessWire - so I can regain my sanity and roll out new features in a fraction of the time

Hi Alan It should work with a Gmail account, though I've not tried yet - all mine have rather too much mail in to give it a go without consuming vast amounts of server memory

Thanks - I'll make a note to reactivate it tomorrow with the new version

Well I still like MODx though I don't use it any more