Mailing List Archive: MythTV: Dev

Re: Ticket #8141: [PATCH] mythbrowser: Page up/down fix

Index | Next | Previous | View Threaded

leamas.alec at gmail

Mar 7, 2010, 11:21 AM

Post #1 of 7 (1298 views)
Permalink

Re: Ticket #8141: [PATCH] mythbrowser: Page up/down fix

On 03/07/2010 07:02 PM, MythTV wrote:
> Comment:
>
> I think page up/dowm should move a page at a time. If you want finer
> control then use the up or down arrows which move a 1/10 of a page.
>
>
True. But users really expect to be able to read e. g. a long text by
simply using page up/down. And this just dont work as of today, there is
often a line which can't be read unless you use the arrows.

At least Firefox and Chrome implements Page up/down by scrolling less
than a complete page. The same is true for OO writer. Why should not
mythbrowser do the same?

User requirements, the idea to do the expected and the actual
documentation all points to scrolling less than a complete page IMHO.

Cheers!

--alec
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

stuart at tase

Mar 7, 2010, 11:47 AM

Post #2 of 7 (1243 views)
Permalink

Re: Ticket #8141: [PATCH] mythbrowser: Page up/down fix [In reply to]

On Sunday 07 Mar 2010 19:21:16 Alec Leamas wrote:
> On 03/07/2010 07:02 PM, MythTV wrote:
> > Comment:
> >
> > I think page up/dowm should move a page at a time. If you want finer
> > control then use the up or down arrows which move a 1/10 of a page.
>
> True. But users really expect to be able to read e. g. a long text by
> simply using page up/down. And this just dont work as of today, there is
> often a line which can't be read unless you use the arrows.
>
> At least Firefox and Chrome implements Page up/down by scrolling less
> than a complete page. The same is true for OO writer. Why should not
> mythbrowser do the same?
>
> User requirements, the idea to do the expected and the actual
> documentation all points to scrolling less than a complete page IMHO.

Opera scrolls a complete page with PgUp/Down. You are right about Firefox, but
it's about 90%.

I guess you have to ask; should it be consistent with some (but not all)
browsers, or consistent with the rest of mythfrontend where PgUp/PgDown moves
a full 'page' in lists. I'm not voicing an opinion on how it should work, but
laying out some of the facts and possible considerations.
--
Stuart Morgan
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

leamas.alec at gmail

Mar 7, 2010, 12:03 PM

Post #3 of 7 (1242 views)
Permalink

Re: Ticket #8141: [PATCH] mythbrowser: Page up/down fix [In reply to]

On 03/07/2010 08:47 PM, Stuart Morgan wrote:
> On Sunday 07 Mar 2010 19:21:16 Alec Leamas wrote:
>
>> On 03/07/2010 07:02 PM, MythTV wrote:
>>
>>> Comment:
>>>
>>> I think page up/dowm should move a page at a time. If you want finer
>>> control then use the up or down arrows which move a 1/10 of a page.
>>>
>> True. But users really expect to be able to read e. g. a long text by
>> simply using page up/down. And this just dont work as of today, there is
>> often a line which can't be read unless you use the arrows.
>>
>> At least Firefox and Chrome implements Page up/down by scrolling less
>> than a complete page. The same is true for OO writer. Why should not
>> mythbrowser do the same?
>>
>> User requirements, the idea to do the expected and the actual
>> documentation all points to scrolling less than a complete page IMHO.
>>
> Opera scrolls a complete page with PgUp/Down. You are right about Firefox, but
> it's about 90%.
>
> I guess you have to ask; should it be consistent with some (but not all)
> browsers, or consistent with the rest of mythfrontend where PgUp/PgDown moves
> a full 'page' in lists. I'm not voicing an opinion on how it should work, but
> laying out some of the facts and possible considerations.
>
First of all, time to be a little polite. Really. I *do* appreciate
those of you taking time to evaluate and discuss my newbie patches. Thanks!

So, there are more browsers out there! Anyway, the difference is with
the rest of myth is that when you scroll a page in a table, you never
display a partial line at bottom. Eiter you see the line, or not. And if
you can't see a line on current page, it will be on the next as I
understand the situation.

But the browser is different. It just scrolls a number of pixels, and as
a result both the first and last line might be hard to read. And making
a complete page scroll, the line that was hard to read at the bottom of
a page typically becomes equally hard to read at the top of next page.

The question is really if it should be possible to read a long page by
only using page up/down. My own opinion, while not that interesing,
should be obvious :-) OTOH, this is not really a Big Issue.

An even smaller issue is of course if a page scroll should be 75%, 90%
or maybe 95%. I have really no educated guess about this. I kept it
rather low just to make a little closer to the documented 50%.

Cheers

--alec
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

robert.mcnamara at gmail

Mar 7, 2010, 12:46 PM

Post #4 of 7 (1236 views)
Permalink

Re: Ticket #8141: [PATCH] mythbrowser: Page up/down fix [In reply to]

On Sun, Mar 7, 2010 at 12:03 PM, Alec Leamas <leamas.alec [at] gmail> wrote:

> An even smaller issue is of course if a page scroll should be 75%, 90% or
> maybe 95%. I have really no educated �guess about this. I kept it rather low
> just to make a little closer to the documented 50%.

Note that the documentation you quote is for the old version of
MythBrowser, which existed before the bottom-up rewrite of the browser
for .22. Your assistance in updating the documentation to reflect the
new browser would be appreciated.

Robert
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

mythtv at sky

Mar 10, 2010, 2:41 AM

Post #5 of 7 (1173 views)
Permalink

Re: Ticket #8141: [PATCH] mythbrowser: Page up/down fix [In reply to]

Alec Leamas wrote:
> On 03/07/2010 07:02 PM, MythTV wrote:
>> Comment:
>>
>> I think page up/dowm should move a page at a time. If you want finer
>> control then use the up or down arrows which move a 1/10 of a page.
>>
>>
> True. But users really expect to be able to read e. g. a long text by
> simply using page up/down. And this just dont work as of today, there
> is often a line which can't be read unless you use the arrows.
>
> At least Firefox and Chrome implements Page up/down by scrolling less
> than a complete page. The same is true for OO writer. Why should not
> mythbrowser do the same?
>
> User requirements, the idea to do the expected and the actual
> documentation all points to scrolling less than a complete page IMHO.
>
> Cheers!
>
> --alec

Can't say this is really a big deal to me. I think 50 % or 75% is not
enough I'll settle for page scrolls 90% :-)

I've had a quick look at your other patch on ticket #8139 and it does
not look correct. A widget should always pass false back if it doesn't
handle a keypress to allow other handlers like the ones in
MythScreenType to handle them. Things are a little more complicated in
the browser widget because the arrows have a dual function. Sometimes
they are used to scroll in the browser but they are also used to shift
focus to other widgets. The way it works is when you scroll in one
direction and can't scroll any further then the focus is moved, I know
it's a source of frustration for many users but the same problem exists
in other parts of mythui like the buttonlists. In any case that has
nothing to do with the bug you are trying to fix.

Paul H.
_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

leamas.alec at gmail

Mar 11, 2010, 2:48 AM

Post #6 of 7 (1142 views)
Permalink

Re: Ticket #8141: [PATCH] mythbrowser: Page up/down fix [In reply to]

On 03/10/2010 11:41 AM, Paul Harrison wrote:
> Alec Leamas wrote:
>
>> On 03/07/2010 07:02 PM, MythTV wrote:
>>
>>> Comment:
>>>
>>> I think page up/dowm should move a page at a time. If you want finer
>>> control then use the up or down arrows which move a 1/10 of a page.
>>>
>>>
>>>
>> True. But users really expect to be able to read e. g. a long text by
>> simply using page up/down. And this just dont work as of today, there
>> is often a line which can't be read unless you use the arrows.
>>
>> At least Firefox and Chrome implements Page up/down by scrolling less
>> than a complete page. The same is true for OO writer. Why should not
>> mythbrowser do the same?
>>
>> User requirements, the idea to do the expected and the actual
>> documentation all points to scrolling less than a complete page IMHO.
>>
>> Cheers!
>>
>> --alec
>>
> Can't say this is really a big deal to me. I think 50 % or 75% is not
> enough I'll settle for page scrolls 90% :-)
>

90% is fine with me. Over and out on this :-)

> I've had a quick look at your other patch on ticket #8139 and it does
> not look correct. A widget should always pass false back if it doesn't
> handle a keypress to allow other handlers like the ones in
> MythScreenType to handle them. Things are a little more complicated in
> the browser widget because the arrows have a dual function. Sometimes
> they are used to scroll in the browser but they are also used to shift
> focus to other widgets. The way it works is when you scroll in one
> direction and can't scroll any further then the focus is moved, I know
> it's a source of frustration for many users but the same problem exists
> in other parts of mythui like the buttonlists. In any case that has
> nothing to do with the bug you are trying to fix.
>
OK, I'll look into it. Thanks for taking time for this. Be back later.

--alec

_______________________________________________
mythtv-dev mailing list
mythtv-dev [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev

leamas.alec at gmail

Mar 16, 2010, 12:46 AM

Post #7 of 7 (999 views)
Permalink

Re: Ticket #8139: fix for crash when doing "next link" [In reply to]

Hi!

I just saw that you have fixed "my" ticket #8139 together with the old
keyboard focus problem in changeset 23731
<http://svn.mythtv.org/trac/changeset/23731> Nice!

I am definitely a newbie, and on top of that it was some years since I
practised C++. So maybe I should just be quiet, but I'm a litte worried
by some some code here - it doesn't seem to be robust:

void MythWebView::keyPressEvent(QKeyEvent *event)
{
...

if (action == "NEXTLINK")
{
QKeyEvent tabKey(event->type(), ...)
*event = tabKey;
WebView::keyPressEvent(event)
return;
}

...
}

I feel uneasy about the assignment *event = tabKey. Isn't this a
potential risk for memory leaks or worse if event is a plain, non-smart
pointer to a dynamically allocated object? Put in another way, is it
really part of the overall contract for keyPressEvent that a subclass
implementation can make this kind of assignment?

Sorry if I take your time with nonsense, just ignore this in that case.

Cheers!

--alec

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads