Stephen Harris

 The Loop

Where humor doesn’t work.
Oct 28, 2016 15:00
I see your point though - if you're allowing a user to inject script tags on the front-end, that's generally speaking no less a security issue than if there were able to inject it the wp-admin. The principle being explained was that 'escaping' data doesn't make sense without context - escaping data for an SQL statement, rendering, displaying in a textarea all mean different things. The text jumping out of a textarea was a simple illustration of that.
Oct 28, 2016 14:54
@MarkKaplun True (though iframes, and script tags are usually a big no-no). But even if you want the user to enter unfiltered HTML into a widget. You don't want them to enter unfiltered HTML into wp-admin
Oct 28, 2016 12:44
@MarkKaplun ping me in chat if you want to continue our discussion data sanitisation. Comments is not the right place for it.
Aug 9, 2016 10:02
ah, brilliant. Missed that. thanks :)
Aug 9, 2016 10:01
Sorry, I mean SE. I'd like to see what kind of interactions the perspective mods have had with users
Aug 9, 2016 09:56
Is there any means of viewing comments left by a particular user? (i.e. a list of recent comments)
Jun 29, 2016 11:42
@GhostToast - do you still hang around these parts? If so, you should consider creating a separate ticket for core.trac.wordpress.org/ticket/17115 - that said, there are some issues with your patch, what if my post type doesn't support title, content or excerpt?
Jun 26, 2016 13:17
@toscho why, oh why, did you not use the photo booth at the ball last night? @kaiser rather suits a pink cow-boy hat.
Jun 20, 2016 22:39
@cjbj You going to WCEU? If so, there's a few of us around that could go for that beer :)
Jun 8, 2016 15:46
Me too. I think Leiden was the last time I saw everyone. I don't think we've met properly either @kraftner.
Jun 8, 2016 15:38
@kraftner I was suggesting that those 3465 votes were from his other accounts ;)
Jun 8, 2016 15:33
@kraftner He has multiple accounts.
Jun 1, 2016 14:40
@Rarst mustache.php does, but for mustache.js it looks like you have to do something a bit more... hacky...
Jun 1, 2016 14:37
But then having written that I found this: github.com/janl/mustache.js/issues/277 so it does kind of support helpers
Jun 1, 2016 14:37
@Rarst That's what I started with, and moved on from them to Handlebars. I really liked it, but you have a limited concept of 'escape' - i.e. you can escape HTML, but not URLs. PHP version has helpers, but JS version does not.
Jun 1, 2016 14:29
Also looked at Handlebars & LightnCandy (php port) - but that screams at you to cache compiled PHP (fine for client work, less so for public distribution). LightnCandy boasts on its performance but I imagine they are measuring with cache.
Jun 1, 2016 14:27
Trying to find a workable solution to duplication of templates for PHP and JavaScript. Twig looks most promising so far (it has a .js port).
Jun 1, 2016 14:25
@Rarst That's what I thought (about the filesystem writes) :)
Jun 1, 2016 14:21
@Rarst does Meadow cache the compiled php templates or just regenerate them on each page load?
May 5, 2016 09:13
@CoderSte If you were cynical you might conclude that it is to help the government's statistics: "X% of school leavers find employment"
Apr 21, 2016 08:01
@toscho "What are you expecting to happen?" "For it to work" "What is happening when you run the code?" "It doesn't work"
Apr 19, 2016 11:47
They deleted sites. And their backups. Oops.
Apr 15, 2016 20:11
@kraftner Nope.,. Honestly, didn't expect to.
Apr 6, 2016 08:30
@toscho Who else is going to be using the namespace 'WordPress'? :P
Mar 22, 2016 10:26
@MarkKaplun Never ran wp on php7, but the tests at least pass.
Mar 21, 2016 10:39
yes and no: Someone followed up on my e-mail for clarification, but I haven't heard anything since...
Mar 21, 2016 10:38
@MarkKaplun
Mar 18, 2016 14:34
@userabuser yup
Mar 18, 2016 14:31
Off topic: anyone else noticed that the pseudo-markdown support in the editor has borked shortcode attributes with an underscore....
Mar 16, 2016 17:08
As far as Markdownify is concerned it might just be worth manually fixing the php7 - constructor issue (assuming that's all there is...)
Mar 16, 2016 17:07
and
Mar 16, 2016 17:06
@lkraav That's the one. Though a lot has changed (and looking at the history of WP-Markdown, I think some changes were made to that version to ensure compatability with WP).
Mar 16, 2016 15:51
@lkraav Its hard-coded in: composer is used only for dev dependencies. wp.org doesn't support composer dependencies.
Mar 16, 2016 13:29
I'm not really proactive at maintaining it at the moment, as there's always other stuff in the way. I still actively use it (just not on php7 severs ;) )
Mar 16, 2016 13:27
@lkraav ah, is this the Markdown_Parser dependency? I've not tested it on php7. Happy to accept a PR using a new version if it passes the tests. Just updated GitHub.
Mar 16, 2016 11:35
@userabuser Are you suggesting its only a matter of time before US foreign policy turns its attention to WordPress plug-ins ;)
Mar 16, 2016 11:34
@MarkKaplun It's more a case of I work in £ and that's the price in £. How much it costs in $ will vary depending on much the payment gateway wants to charge you :)
Mar 16, 2016 11:32
@MarkKaplun Yeah... and its not unique to WordPress...
Mar 16, 2016 10:41
@kraftner I haven't had chance to yet (and probably won't have time before 4.5 is due to be released), but seems fairly clear from the code. But I'll report it to the security team anyway rather than updating the trac
Mar 16, 2016 10:24
(yes, this is WP and yes this is the same vulnerability we discussed a few days ago... in a public chat room :) )
Mar 16, 2016 10:23
after all, it comes with the health warning of "dont use on production sites"
Mar 16, 2016 10:22
@all - responsible disclosure does not apply to vulnerabilities that only exist in beta-versions, right...?
Mar 11, 2016 09:57
@MarkKaplun In my experience developers (I include myself) can get a bit touchy when someone critiques their work. They probably interpreted it as more derisive then was intended. Still, sad that they would sink to name calling.
Mar 10, 2016 22:13
@MarkKaplun what's been said?
Mar 10, 2016 12:31
> PHP has added a new configuration variable max_input_vars, that can limit the number of parameters... This can be easier to use than limiting the length of the HTTP request, because application programmers usually know, how many parameters they need, while the maximum length of the request is harder to predict. However, if a single request parameter contains a data structure, that is later parsed and put into a hash table, the application might still be vulnerable.
Mar 10, 2016 12:26
The security vulnerability that time forgot....