Jump to content


  • Posts

  • Joined

  • Last visited

  • Days Won


Everything posted by alan

  1. Sorry for the silence @teppo and @netcarver, life changes here have pulled me, for a while, away from PW dev :/ Not bad changes, but just taking my time so I cannot PW. I pay $30 CDN per month to host around 10 sites on AWS. AWS HW is "On Demand Linux t2.micro" w/ 64GB SSD added. Hope all is good w/ you guys here in PW-land : )
  2. THANK you @dadish — this; SPA + PW as an API + GraphQL for PW, for me is a milestone exciting proposition. I hope I make it the biggest thing to happen to my PW world since I read the word "ProcessWire" on the bottom generous Marty Walker's web site and first visited processwire.com. First steps for me are reading up on PWAs and finally trying to finish a Wes Bos course on ES6, then I'm off to add in PW as an API & GraphQL. Exciting!
  3. Wow o_O @dadish I am humbled by your excellent work here. I stumbled on this thread researching PWAs and then was transfixed for the full 37 minutes of your excellent video intro. Thank you very much indeed for all of this, I am yet again thankful for the ProcessWire community's generous and excellent work. I've had Vue.js noted down as "something I want to use", along with using ProcessWire as/via an API, and most recently, PWAs. I am hoping, if I've not misunderstood, now I may be able to use all these together for a project. Brilliant! \o/
  4. Oo, didn't notice how long that gap was o_O (maybe I'll be the first to post back if I get it working )
  5. Over six months into the project and I have, touch wood, had a great experience moving from traditional hosting to Amazon EC2 for hosting around 10 sites. I still plan to write this up, but just wanted to share one little tip I found, probably something any DBA would instantly know but that I had to learn the hard way. I modified a script to export all databases and set it via cron to run at midnight. This worked AOK for 6 months, then I found service `httpd` was down one morning. The reason was simply that the day before I had added one more database to my EC2 and that had tipped the backup process over the brink as far as server load went. Each db was backed up, one after the other, and cumulatively this added to too much intense load. The solution I used was simply to add a 5min pause in the loop in the backup script, between databases. Fixed. Hope this helps someone 👍
  6. @justb3a Thank you very much for this, 100% fixed the problem, cheers!
  7. NO! Chat sometimes results in a solution, even if one doesn't know about a thing Thanks for trying and Happy Christmas in Austria \o/
  8. Yes, can do that, thanks again @bernhard but it feels like this is something the SCF module does already, e.g. it supports setting a success message in a set of JSON options, and indeed it then shows the message. So all looks 100% ok, unless I nav to some other page, then go back. Maybe a bug in the module (doubt it, more likely a bug in me )
  9. Thanks @bernhard by coincidence I am on that exact page now I've tried some of this before but not this exact page. It looks like I can 100% solve it by redirecting to a new page (the module allows me to do this), but if I do then the module's nice support for a `Success` message is lost, so no error but also no message. Gonna keep experimenting to see if I can get the best of both worlds.
  10. still unable to fix my form resubmission warning, just in case anyone has a clever thought to spare that might give me a clue to what I am missing.
  11. @Thomasaurus Always worth checking the anti-spam settings to make sure it's not one of those tripping over. The only other thing I've had that I think was similar was when I had two tabs open, signed in as admin - got weird stuff happening nothing to do with SCF specifically but it affected session info - closed one tab, instant fix. Good luck, I am using this Module more now I am getting familiar with it, it's super useful.
  12. I have two forms, each on different pages, each with `submitName` values that are unique, in case that helps, and I am finding after a succsessful form submission, if I navigate to a new page, then let the page load then go back in history, to the page where the form was sent from, my browser is complaining `Confirm Form Resubmission`. I can't see anything in the docs on how to clear POST data on a succsessful submission, assuming it's that. Any pointers/help would be much appreciated Thanks in advance 👍
  13. Hi @Donald, I recently used this Module and may have some tips, but I don't know it as well as others: I was able to send as many tests as I liked, as long as I had settings: Minimum time: 1 Maximum time: 300 (i.e. large) Restrict submissions: 300 (i.e. large) Exclude IP: Sorry if you have tried exactly this and it doesn't help. In the GUI Admin, click Setup > Logs > simplecontactform-log This is a bit more involved, but essentially read the docs where it tells you how to use an 'options' array, populated with settings. One of the settings in naming the form and naming the fields that form is able to use. Then elsewhere in your template or another template, you do the same but the contents of the 'options' array is different, naming different fields from those available listed in the Module (where you add them). Essentially the Module GUI allows you to insert any fields that will be used across one or more forms and it's the calling of the forms in your code that determines which fields are rendered and used. I think I have got this correct and I hope it's helpful, very sorry if not. Good luck!
  14. Sorry if this is not too relevant to this thread, but, in case anyone is here trying to solve why they get `?it=` added to the URL and anchor redirects don't work properly, for my install that always failed I finally found a workaround.
  15. Dunno if the problem I have had is exactly the same, but, FWIW, I have a site with ProCache and regardless of whether I delete and rebuild the cache, if I add a redirect in `.htaccess` to a page with an anchor tag, then the site actually ends up at a page with `?it=` and the redirect only gets as far as the page, the scroll to the anchor is ignored. For me the fix was to use this syntax: Redirect 301 /my-short-cut http://example.com/the/long/page/to/go-to/?it=qik/my-short-cut#theAnchorIwant The rule I *would* have used until now was Redirect 301 /my-short-cut http://example.com/the/long/page/to/go-to/#theAnchorIwant but that fails as described. Now I just amend to insert the `/?it=qik/` followed by the text to the right of `301` i.e. `my-short-cut` followed by a `/` and finally the `#anchor`. This was a hard error I could NOT get rid of and I just had to accept not being able to do this. SO glad I found a workaround, hope this helps others 😊 PS: I don't think this is to do with ProCache, I think it's maybe to do with my hoster(?) - I'm not sure, but this one's been *such* a pain I really don't care, just glad to have got past it
  16. THANK you @BitPoet for the answer and @Neo for posting this Q — totally saved me from a frustrating end to a day of coding. I suppose standard practice for uploading a site is to not transfer the contents of the /assets/cache/ folder? If so, I'll bet this is documented and I missed it…
  17. @adrian thank you very much for the tips and that link, off to go and bookmark—brilliant!
  18. o_O s t u n n e d to see what I have been missing by not using Tracy. THANK you again @adrian for the post and for Tracy!
  19. @adrian not at all! I love being reminded of Tracy. Off to go look it up again. I am embarrassed to say I looked briefly when you released Tracy and I think I saw some dependancy of mine that meant a slight question and I never went further. I am very impressed to see that you can do per your screenshot, clearly I am a total newb' in some respects, when it comes to debugging, I will turn on, er, `debug` and that's about as far as I get, then it's adding `echo` here and there to see stuff. tl;dr I am eXCITEd to go look at your module! Thanks! 👍
  20. Oh nos... I was convinced I had ruled out any chance it could be session data, browser cache, etc, for example opening the same URL in a browser I don't use for dev. But I think it was somehow polluted session from using the CodeKit url `http://myMacName.local:5757/` rather than the LAMP url that I tell CodeKit is where my site is running in one tab, and the LAMP (non-CodeKit) url in another tab. tl;dr: $user may not behave as expected when previewing on a CodeKit URL, or if you do and another tab is open at the non-CodeKit URL. Takeaway, check this first and/or if I insist on keeping both Admin and end-user view open, at different URLs, use two different browsers.
  21. Thanks @tpr, I wondered if I had too and so previously had done a `print_r($user)` and got a dump that looked convincingly like an array of user info, so I assumed I had not overwritten it. But I now see I get the same dumped output whether logged in or not... So that seems to suggest it's not null, is an array but looks like maybe it's an array of stuff that is not the $user object! Thanks for the pointer, I will go back and look harder at that now and report back here, cheers!
  22. Thanks @Mustafa-Online - I was pretty certain it was good, now the worse bit, working out what else I have here that's causing this
  23. Thanks @Soma for confirming that. I'm confused in that case as I have a site where this test is failing. I'll go back and re-re-check to see if I am missing something else.., thanks again for the sanity check 👍
  • Create New...