On 12/1/25 4:15 PM, Per Bothner wrote:
On 12/1/25 15:26, Raymond Toy wrote:
On 12/1/25 3:11 PM, Per Bothner wrote:
On 12/1/25 14:05, Raymond Toy wrote:
When I try this on a simulated iPhone SE in portrait, I get [no
wrapping]
I don't think that's what we want.
On Gnome Web (which uses webkitgtk) when the window is too narrow
I get a horizontal scrollbar rather than wrapping. Tolerable
but not ideal.
Oh. I was using Firefox.
My understanding is that until recently "Firefox" on IPhone was
just a wrapper over Apple's WebKit browser engine, due to Apple app
restrictions.
Those restrictions have recently been disallowed, at least in Europe
Chrome was the same way. Perhaps that's changed now too?
WebKit-based browsers (which includes Safari, the IPhone browser),
and Gnome Web - formerly Epiphany) are known for being less
current-standards compliant than Firefox or Chrome.
According to https://caniuse.com/flexbox Safari has supported flexbox
for a long time. Not sure what that implies about putting a
scrollbar, though.
​
That does not mean it does so correctly. Flexbox is a huge specification,
and there are a lot of features to implement. There may also be genuine
differences in how the spec is interpreted. However, if Chrome and
Firefox
agree and WebKit/Safari differs, I would consider the former more
likely to be "correct".
Of course, we still want things to work acceptably on WebKit/Safari.
Yes, definitely agree that we should strive to make sure it works on
recent versions of all common browsers.
I think the comments there (provided by Gemini) are really important
to keep.
If the CSS is to be included in each emitted html file, then
no, we don't want to keep the comments. If the CSS will be in a separate
.css file, then we can keep the more non-obvious of the comments.
I guess the comments are only useful to me who can't remember. Certainly
the comments for each element aren't really relevant.
Personally, I dislike the "Lisp style identation" where the amount of
indentation
of overflow lines is based on the length of the initial word/command.
For example,
it makes for different indentation following the "Next:" and
"Previous:" lines.
It's ugly, wastes space, and as far as I know no-one else (outside the
GNU and Lisp
communities) does it that way.
I think that's the beauty of CSS. Each manual can change the style as
seems appropriate as long as the classes are there.
Here is your code changed so each overflow line indents a fixed amount.
It seems to work nicely on both Gnome Web and Firefox.
This looks fine to me and seems like a good default for texinfo. I
wouldn't mind this. I do like that wrapped line for the Previous section
is indented slightly so it's easily separated from the Up section.
​
