Mailinglist Archive: opensuse (1092 mails)

< Previous Next >
Re: [opensuse] Re: Question re why part of a magazine article does not print
  • From: "Carlos E. R." <robin.listas@xxxxxxxxxxxxxx>
  • Date: Sun, 12 Aug 2018 14:46:09 +0200 (CEST)
  • Message-id: <alpine.LSU.2.21.1808121436461.8868@Telcontar.valinor>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Content-ID: <alpine.LSU.2.21.1808121441320.8868@Telcontar.valinor>


On Saturday, 2018-08-11 at 17:05 +1000, Basil Chupin wrote:
On 6/8/18 8:57 pm, Carlos E. R. wrote:
On Monday, 2018-08-06 at 15:31 +1000, Basil Chupin wrote:

On 05/08/18 16:55, jdd@xxxxxxxxx wrote:
Le 05/08/2018 à 07:49, Basil Chupin a écrit :

in may be in length.  If Screengrab can do it why can't Spectacle do
same?

because a *screen* capture do what is said, capture the screen. How
can it know what is there out? spectacle now nothing about Firefox

Now, I know nothing about how these things work but it seems to me that
a screengrab program must look for some marker which "tells" it where
the 'screen' to capture begins and where it ends. Right? Or put another
way, the program looks for the marker showing the beginning of a "file"
(SOF) and the end of file (EOF) and captures what is between the SOF and
the EOF.

To me, something tells Spectacle to consider the EOF to be the last
pixel of the displayed screen, while Screengrab! actually looks for the
EOF of, say, an article when one selects the 'Complete Page' option.


No, Spectacle does not look for the end. It only sees what is displayed in
the display, nothing more. It simply reads the graphic card memory, or the
X representation of the same.


I am trying hard to understand this, but surely the complete "page" is stored in the gpu memory or the X representation of it, as you put it, because the invisible part, which is not visible on the screen, is instantly displayed when you either hit the 'down' key or use the scroll wheel on the mouse?

Nope.

The application may have the invisible part of the page prepared or even rendered, but the image memory doesn't know anything about that.



What is beyond the display is only known to each application. Spectacle
does not have permission to read each application own memory,


Then give it "permission" to do so.

Can't. Huge security hole. Plus spectacle would have to be programmed to know internals of every application that paints the screen. There would be then a list of supported apps and not. And then supported versions and sub-versions.

Huge programming job. Hundreds of people needed, for ages. Any modification in any of the supported applications, and spectacle would have to be re-compiled and possibly re-designed.



even if it knew how to interpret it.

And to me, the description of what each of the "Capture Mode" (here:
https://docs.kde.org/stable5/en/kdegraphics/spectacle/using.html#idm46336202854880
) means may just as well been been written in Aramaic :-).

The Full Screen (All Monitors)

The Current Screen

The Active Window

The Window Under Cursor

The Rectangular Region option allows you to select a rectangular region of
your desktop with your mouse


They seem pretty obvious to me :-?  I don't even need to read the
descriptions.


OK, then please tell me what each is supposed to mean -- apart from the "The Current Screen" and "The Rectangular Region" which even I can understand :-).

The current screen is exactly and only what is displayed by the current monitor. Plain simple. I don't understand what you don't understand :-?

If there are more monitors, then the Full Screen is all of them, joined. As if a single huge monitor of the appropriate shape was used instead.

Active Window, the Thunderbird window frame where you are reading this. You click on another window, that becomes the active window. If you move the mouse to another window but don't click on it, that is the window under cursor.

- -- Cheers,
Carlos E. R.
(from openSUSE 42.3 x86_64 "Malachite" at Telcontar)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iEYEARECAAYFAltwLBgACgkQtTMYHG2NR9UPKQCfQyW66ezZCqaqULKIcd/KMNga
sjIAnRX6X1qqBeXPlq/nQN8QHyL1lWiq
=uhkO
-----END PGP SIGNATURE-----
< Previous Next >