ceberlin

+1 for a filtering out all ".json" files. I do not need to check here all those language translation files and files that the proflle exporter uses.

WHOW!. - looks like it ! I could not figure out how to change the preset of a page field that appears as a select. (I would love being able to change the preselected page id). That tiny checkboxes/ selections is, where editors overlook something easily and presets would be very helpful. Maybe the feature is not implemented everywhere yet.

DiagnoseFilesystem Root directory Exists, is readable, is writable, 0750 Failure Folder should not be writable for webserver user That reads like DRAMA. Are you sure? What I would like to check frequently is that the config.php is read only (400). Unfortunately this gets reset by GIT updates to 755 now and then and a warning would be great here. Maybe with a "fix it"-link? - That possible feature was one of the reasons for me to try Diagnostics in the first place. (Drupal7 is very cool about this - it resets the settings files to 400 automatically.) htaccess file Exists, is readable, is writable, 0664 Warning .htaccess must exist, must be readable, should not be writable. If I set htaccess to read only (400), I receive a 403 error when going to the procache settings at ADMIN.

I investigated more. This time also with a production site which is live: It DOES work RIGHT on the LIVE server, I just found out. Only on localhost (MAMP PRO) on my testing ground it does not optimize - which does not matter much. So ... a false alarm, regarding the live site.

Login_persist InnoDB / utf8_general_ci Rows: 5 Data: 16.00 KB Index: 0 bytes Overhead: 295.00 MB Warning Optimize table to reclaim space and reorganize indexes. Do you think that "Overhead 295 MB" makes sense here? Or is this a calculation error? Clicking on the "optimize" link has no effect, by the way.

Hi Nico, great idea, this spares quite some time to set this up individually. One question about "Keywords". Do you mean the Keywords Meta Tags? Google says they don't use it any more: http://www.mattcutts.com/blog/keywords-meta-tag-in-web-search/ Some of our clients had a wrong idea of SEO when spending a lot of time end efforts with littering the SEO keywords meta tag with all sorts of words, instead of using their energy on other SEO more relevant topics, like a carefully chosen title and a good summary. So we stopped offering that field. This is (in German) what Meta Tags Google uses: http://support.google.com/webmasters/bin/answer.py?hl=de&answer=79812 EDIT: A given "Focus Keyword" for a page might be a good source for your planned suggestion tool, the WP Plugin SEO YOAST seems to use it that way.

Hi Adrian, great idea. I will try this out on my new project. Just curious: Isn't there a need to also sanitize the password somehow (or is PW handling that?) Like a max-length, trim whitespaces (for user convenience)

There are a number of field settings that can be overwritten by templates which use the field. It the default value was overwritable too, that would help reusing fields a lot, giving editors a good usage-experience.

interesting... didn't realize they have a short url now. Much less typing

To check from outside if server is down (you probably know that link): http://www.downforeveryoneorjustme.com German provider? We use http://www.df.eu (domainfactory) since over 10 years and we are very satisfied.

Thank you for your comment totoff... Well, I believe this client has checked that part of the website with the laws thoroughly. The recommendation system is embedded in cooperation with a specialized service for German clinics and doctors. (We only restyled that content so it fit better to the rest of the website.) I found this story on the web (about the revised version of that law from 2012): http://projectworkers.de/marketingblog/116-mehr-vorteile-fuer-aerzte-und-heilberufe-durch-das-neue-heilmittelwerbegesetz They say: Publication of recommendations by patients is ok - as long as the text does not contain drug-/treatment-misuse or repulsive or misleading content. (Hope I got the translation right ) In case the law is applicable at all (because they offer only diagnosis and treatments prescribed by doctors), there is no problem as long as the content of a recommendation is within certain limits (which means, of course, every single entry needs to be checked individually). General speaking: None of the content of the website should seduce a patient to a wrong medical self-diagnosis and self-treatment.

Good idea to change the menu calendar-button behavior (for appointments). Thank you for the feedback.

Hi, great points you made about typography. In this case this is content and up to the clients. Maybe one day we could replace the textarea fields with the new PW pro table fields and render the output ourselves with any fancy typography we want. (If you want to take it further, the n-dash is also surrounded by tiny 1/4 spaces left and right, if I recap that correctly). Phone number formatting is another interesting scenario. Maybe modules like the SuperSmartyPants Typographer become smart enough one day to assist in cases like this. - I wished we had more clients caring for typography that much that they would hire an editorial office ("Lektorat") to have content, spelling and typography checked. Also the point with the navigational overkill is valid. - If you have a client with 11 competing centers and 50 decision makers, who even discuss the shape of a button sometimes, you get there quickly, believe me. Client is king. On the other hand, those special requests were one of the reasons, why the project was very interesting and exciting for us. To get back to the essence: This presentation here was not about design and not about content but what Processwire can help solving. I wanted to showcase what you can do with Processwire, if you have tons of navigational elements as a fact, that behave context-related completely differently, depending from where they were called. And this only with the power of Processwire selectors. We could not do that with Drupal or Wordpress easily. EDIT: Actually I liked your comment and fixed that n-dashes quickly myself. I admit, that looks better now.

http://www.radiologische-allianz.de Step1. Homepage uses a classic design. Purpose was to make complex information look easy for patients. If you use the homepage with a mobile, you get a special view with only the things you need when you are on the road. (The most time-consuming action was organizing all this data and find an easy to maintain ways to link this all together.) We have 11 centers. This homepage is a portal page for all of them and includes individual homepages of each center at the same time: If you access things for only one center through their homepage, you get a filtered view with only items offered there. There is also an individually built center-finder which uses Google maps to find the right center for your therapy, using Processwire technology. Processwire was chosen mainly because of the enormous flexibility to build links between all different sorts of information (without it being rocket science), especially because we were able to handle even non-hierachical data (e.g. a child is allowed to have 2 parents - doctors think that's normal ), and because we could offer the client a very clean and easy to use backend, hiding all the complexity of the homepage away from him: They started using it with only 2 hours of training. Not sure you could do this with Wordpress... (This is the first homepage we did with UIkit, which was a recommendation from this forum.) (Step2 will add some speed optimizations...)

I am really looking forward to some documentation for this apparently great module.

strong point, Adrian !!!! Yes the hack script kiddies, I forgot. Your are 100% right with your concern. If I would build such a script I definitely would need to use a filter, like - show only links with 5 hits and more (links with less hits would neither bother Google nor me, I think, and script kiddies do not repeat their urls) - delete link after 1 week if there were less than 5 hits and no target set yet by the admin And I would have a different table for those "candidates"... and if a target is set, the link moves to the redirects list table. This would avoid some of the mess i.m.h.o. Actually I would use such an option only some weeks after a site relaunch and then switch it off. (Most of the important 404 errors show up within the first days after a relaunch.) Showing when the last hit was, would be also great, so it is easy to decide which links are candidates for deletion. This would also help to have the list clean and up-to-date. Why I am writing this at all: Yes I know how to collect 404 errors and throw correct redirects into the htaccess file for a while. But I would love to delegate this all so one of the editors, since this is can be a lot of work if there are many links.

Just dreaming: Would't it be nice if there was a way to catch 404 errors and list them each for applying a redirect with this module? This would spare some time studying PIWIK pages for 404 errors and copying links over to PW, right?

Hi Luis, great module! Thank you much! I see a small display glitch using PW 2.5 and the standard admin theme. The tabs are half overlapping this text string: Data from Site ID: 1Also, instead of SideID, which is meaningless to my client, can this instead output the site name for that ID from the PIWIK panel?

We have the same experience as Pete. This is my scenario with larger teams: "draft" means: needs to be finished and approved: the editor sets the page up but has no right to change the status. The supervisor approves by changing the status. So, "draft" clearly says: not ready, work needed. (As a workaround I use a checkbox field for that at the moment.) "unpublished" means then: content is ready but for other reasons the page is not live (like a campaign/news with a start date and an end date).

@ nico ( "How to add custom translations for e.g. modules?") That is a really good question and maybe another reason for a module. What about the possibility to put individual files to the LANGUAGES (copied by the click of the mouse over from the module) which overwrite the standards so that you can maintain individual translations if needed. Advantages: your individual strings would survive language updates from the module - that eliminates the risk of updating languages the number of files that need individual treatment are probably not so many (opposed to the 500+ files right now) (This all is a loud thinking - not a finished concept yet)

(This discussion is moved here from the GitHub issues board) The discussion ended by considering of moving the admin language files to a module (maybe even aware if the translations are for the regular version or the DEV version of ProcessWire). You comments and ideas are highly appreciated! --- DISCUSSION SUMMARY FROM GITHUB: tbba (ceberlin) wrote Today I had the honour to remove admin language translation files from the default language to have that back to English. That ment clicking the trashcan a 500 times while paying attention not to remove anything else but those admin language files. (Well I ended up doing this in the database directly as a workaround.) For the Wishlist: Why not adding a button where all instance that start with "wire--" have the trashcan seleced automatically. Or better: I would prefer at least to have them in a separate repeater list so that I find my own frontend translations and those settings by site--modules like supersmartypans easier. Or even better: why not separating Admin translation files to a module somehow so that installing/de-installiing, updating(via ModulesManager!) and switching them on and off is much easier? ryancramerdesign commented Double click the trash can and it'll mark them all for deletion. tbba (ceberlin) commented Thats cool. I did not know that. There is still the possibility of deleting too much (site-- stuff). What about the idea to separate the site-- files from the wire-- files? Wouldn't that be much easier? EDIT: And a module for the wire files would make the updates over the modules manager so easy. No more dropping of tons of files - and alerts when they are outdated. NicoKnoll commented I think it probably really makes more sense to change language packages to modules which are requiring "Language Support" to be installed. Pros: easy updatable makes my module "LanguageInstantInstall" obsolet because you just could list al of the languages which are in the module directory with "module manager" It's more the ProcessWire way of being modular Cons: How to add custom translations for e.g. modules?Maybe that's something for 2.6. tbba (ceberlin) commented A module would have another advantage: There could be something implemented that matches the translations to the PW version. Right now, using the DEV version could mean: not matching translations. But - thinking loud - there could be a DEV version of the language files on GIT that then matches with the PW DEV version - and the module selects the right one. ryancramerdesign commented I agree, language packs as modules would be nice for the future. Like Carl mentioned in splitting site and wire files, perhaps language pack modules would be for wire, and the existing storage system could be exclusively for site.

I wonder if PW becomes vulnerable to the XML Quadratic Blowup Attack (as Drupal and Wordpress were), in connection with XML-RPC... src: http://mashable.com/2014/08/06/wordpress-xml-blowup-dos/

There is another issue, but only on one installation. Here the module does not install properly this time.... Notice: Undefined variable: parent in /Users/carl/WWW_TESTSERVER/ullewaeh-git/site/modules/ProcessPageDelete/ProcessPageDelete.module on line 187 As I said the older versions of the module worked flawlessly.

ok, de-installation worked with commenting out the uninstall function. I got a message that hooks were removed. There was no "delete page" subpage in admin/setup/pages/. There was one in admin/pages/ - hope that's the one to delete Then I re-installed - de-installed and re-installed again. Works! - Can I invite you to a beer?

Hi Nico, that would explain that newly installed versions work, old installs not. I use ModuleManager do keep Modules up-to-date, your module is the current version. Since this is a standard module I use often when setting up things, this is a very old install, with various updates. So if you changed something with the install routine in the meantime, this might not be compatible with old installs. I wonder how to de-install this best - the brute-force way. Trashing the files and removing the entry from the DB table "modules" is probably not enough...