Subtle changes, bugs and feedback

My phone isn't rooted, so I don't think that would work. Thanks for the suggestion though.

Is it possible to truncate posts that appear directly in the following feed at a certain length and add a 'read more' link to it? There are some threads, e.g. photography competition where there are regular long posts that can clog up the following page if you get caught in an @ reply. I have no idea how easy it would be to implement but it could be useful.

Hi - quotes have scroll bars underneath them for reasons unknown. On Chrome on a PC running Windows 10.

Screenshot and link to affected comments would be appreciated.

Screenshot and link to affected comments would be appreciated.

https://www.lfgss.com/conversations/131364/?offset=6525#13945094

Edit: This seems to be a Chrome issue I think. Seeing it on other websites as well.

I have the zoom on Chrome on 90% by default, and out of the blue (since yesterday) it's showing a scroll bar for some all quotes:

https://www.lfgss.com/conversations/153782/?offset=75475#13930139

When I zoom in to 100% or out to 80% it disappears.

Browser: Chrome (version 61.0.3163.100 (Official Build) (64-bit)) on Windows 10.

Thanks.

Looks like an odd Chrome thing... there's an overflow:auto on the Quote and this says "let the browser decide what to do when the text is too wide for this space".

Except the browser is deciding to put in a scroll bar when one is not needed.

Removing that instruction fixes it... but I'll need to test whether that breaks things. i.e. if someone put a huge photo in a comment and that were quoted, would it break the page layout if I removed it.

Looks like an odd Chrome thing...

It's funny how the scroll bar even appears on some of Google's own sites, like gmail.com. It even produces an unnecessary vertical one there.

but I'll need to test whether that breaks things. i.e. if someone put a huge photo in a comment and that were quoted

Almost nobody quotes pictures anyway, because you need to manually embed it. 😄

overflow:auto protects a nested HTML tag breaking the parent. So it is used everywhere.

But it looks like, at zoom 90%, the calculation of the width of the nested tag is slightly off (this is Chrome getting it wrong) which then triggers the scroll bar.

Root cause is Chrome and the math for sizing of child HTML elements at 90% zoom.

But I can work around that and remove the scroll bar on here by changing the HTML markup, though that risks allowing quoted things to break the page.

Root cause is Chrome and the math for sizing of child HTML elements at 90% zoom.

Even at 100% zoom. it's not totally right. There seems to be some space underneath the quoted text (see the length of the vertical line next to the text):

Edit: Might be a empty line though, rethinking this. Other quotes do look normal at 100% zoom. Damn you @winnifred0001, and your inaccurate quoting!

Nah... at 100% it is fine.

winnifred quoted like this:

> some text
> > 
> > 

Which results in:

some text

Nah... at 100% it is fine.

winnifred quoted like this:

[...]

I might have accused Winnifred of something that wasn't his fault. Attempting to quote you resulted in why there were probably some empty lines in Winnifred's quote:

This (those extra greater-than signs) is not intentional, right?

Probably not... but markdown WYSIWYG parsing is hard. I'll see if the library we use has a later version that may magically fix that.

But it's a different issue from Chrome handling 90% zoom weirdly.

Bug?

I have a message in my PMs which I have deleted but still showing as an unread icon in the top right of every page, just below my name.

Deleted today?

Message was from ages ago and was read but showing as unread for several months. I deleted the conversation thread just before posting here.

Updates on latest Firefox image uploading issue?

Updates on latest Firefox image uploading issue?

I have docs on the Files API https://developer.mozilla.org/en-US/docs/Web/API/File/Using_files_from_web_applications but need to weigh up whether fixing it will affect more people than not fixing it as https://caniuse.com/#feat=fileapi shows that UK usage is only 60% support it fully, and only 78% support it fully or partially.

In theory the % that do support it is all of the modern browsers, but in practice I bet a load of people are using old work computers to access this site.

Even after I've figured that out... I'd still need to write all of the code.

It will be a little while... the workaround is try to and attach twice, it works the 2nd time (twice before submitting the comment).

And it's still showing unread even though you've deleted it and more than a few moments has passed?

If so, do you object to my viewing your PMs, only the titles, dates, and read status... not the contents. As I'll need to view those parts of the data to figure out where a bug may be.

Workaround it is then :)

thanks!

No worries.

I want to do the "right thing"... which is moving to this new browser supported API that handles the image upload.

But it verges on being too early, and Firefox is handling the old way of doing things badly.

What I don't want to do is to put in an awfully complicated patch that doesn't give us the new thing, nor remove the old thing, and just makes a mess I have to spend ages unpicking later.

So it's a mix of here using the old way and the new Firefox breaking it, but not yet fully supporting the new way.

Isn't there a library that handles these things for you?

Isn't there a library that handles these things for you?

There are several, of varying qualities, of differing interfaces, and supporting different things. This is the awfully complicated patch option.

In the ongoing, slightly odd, battle of people who think that "new page fails" are a thing and those that don't, someone mentioned an idea that didn't seem too bad:

Can the latest page on a thread show the full-page number of comments, rather than the post-pagination remainder?

Basically, the pagination would operate backwards rather than forwards so that the latest page is always "full".

I can see that this could have knock on effects that would break other things, but don't know if it actually would.

What do you think?

I didn't understand the ask.

Can you draw ASCII examples? And/Or describe what the "new page fails" are?

.

/edit was hoping I'd be able to #NPF here.