my xterm-256tst does seem to demand a specialized colorcube and
according to T846 this is the case elsewhere, so change colorcube to
match, as opposed to one that is more color efficient.
Summary:
- Removed the list of paths in the config.c
- The users selects directly the path of the folder where the
Backgrounds are located, terminology will not show up the Pictures
located in subdirectorys you can easyly go in the subdir with the
fileselector.
- With a double click on a picture you can import the picture to your
user config directory so you can cleanup your Downloads directory for
example without loosing your Background
- If a Background is imported the new Background URL in the config will
point to the imported file.
Test Plan:
- The Bufferting of the Backgrounds is not asyncron, but I have tested it
with a directory with more than 300 pictures and it runs very very fast,
in my eyes no asyncron buffering is needed!
Reviewers: billiob, raster
CC: cedric
Differential Revision: https://phab.enlightenment.org/D471
Summary: This is fixing the Bug which is happening if a symbolic link in a directory is pointing to its root directory.
Reviewers: billiob
Differential Revision: https://phab.enlightenment.org/D463
Summary:
Rewrite colour setting to be more general. Before colour inversion
did not work with 256 colour palette.
Test Plan: echo -e "\e[7;38;5;240;48;5;255m####\e[0m" --> output is invisible
Reviewers: billiob
Reviewed By: billiob
Differential Revision: https://phab.enlightenment.org/D421
Summary:
All the rest of the code expects the autowrap switch to be set at the
end of the line. Now in the case of double width characters the autowrap switch
is set before the line ending.
Reviewers: billiob
Differential Revision: https://phab.enlightenment.org/D433
Test Plan: build some history, vi, Shift+PgUp -> it scrolls when it shouldn't
Reviewers: billiob
Differential Revision: https://phab.enlightenment.org/D422
When we get the evas callback for the termio object being deleted,
then we Do Not want to readd the same callback. Instead we should do
evas_object_event_callback_del_full here.
Signed-off-by: Chris Michael <devilhorns@comcast.net>
Summary: Implemented a little page where you can set your background for your terminology.
Reviewers: raster, billiob
Reviewed By: raster
CC: seoz
Differential Revision: https://phab.enlightenment.org/D362
"Do not let CUU and CUD scroll off screen."
Previously I missed how the default values are implemented.
Reviewers: billiob, cedric
Differential Revision: https://phab.enlightenment.org/D417
Signed-off-by: Cedric BAIL <cedric.bail@samsung.com>
This reverts commit 5ab7a6f827.
This patch did completely break emacs for me. I have no idea what the intent was, but
at least my application/OS is broken.
Summary:
ISO 6429 is a bit vague here but I don't think that the cursor should
be let go off screen.
Test Plan: terminology -g80x30, vttest, 1 -> test failed
Reviewers: billiob
Reviewed By: billiob
Differential Revision: https://phab.enlightenment.org/D409
this makes the colors config work and able to save/load a custom
color palette. when enabled it overrides the theme provided palette
for the "base" 48 colors (normal + bright with intense modes of each).
After we fetch geometry, we are only using w & h for comparison. X/Y
are never used, so remove those two variables.
Signed-off-by: Chris Michael <devilhorns@comcast.net>
'data' is used in this function for terminal assignment, so remove the
EINA_UNUSED from the function declaration.
Signed-off-by: Chris Michael <devilhorns@comcast.net>
Summary:
Mutt does this. It sets the alternate buffer, it disables it, it
disables it again and enables it back.
Test Plan:
Run mutt, quit mutt, run mutt, quit mutt. You should see the
residual background from mutt.
Reviewers: billiob
Differential Revision: https://phab.enlightenment.org/D371
Summary:
After commit 05d091483a screen swap form
alternate buffer to primary buffer is not the same as the reversed. Take this
into account when resizing.
Test Plan: Build history, man echo, resize, quit manual.
Reviewers: billiob
Differential Revision: https://phab.enlightenment.org/D370
This is ugly, makes evas complain… but it's better than nothing and users
might discover the bindings about the "tab" switcher…
And we're in feature-freeze.
this makes the theme selector look much nicer. it's a step in the
right direction, but not 100% there. did this more as an example for
people filling in wallpaper and color selector (colors will need to
have a scheme/theme and preview too and then a way to edit a custom
theme/scheme).
Summary:
simplify code (treat every resize as a single case) and fill the lines
from the bottom up
Reviewers: billiob
Differential Revision: https://phab.enlightenment.org/D355
As characters are deleted, the remaining characters between the cursor and
right margin move to the left. Character attributes move with the characters.
The terminal adds blank spaces with no visual character attributes at the right
margin. DCH has no effect outside the scrolling margins.
Summary:
The circular_offset should not be set to 0 because the first line from
screen2 is copied to the offset line in screen.
Test Plan:
Try "man echo", scroll down and exit. The screen is not restored
properly.
Reviewers: billiob
Reviewed By: billiob
Differential Revision: https://phab.enlightenment.org/D345
This patches uses ctxpopup (ugh, this bitch is awful) to show a
contextual menu about that link with options such Preview (if able to
inline), Open (forces xdg-open) and Copy, then you don't have to
select a link in order to copy it :-)
This fixes T430 (terminology should copy link on right-click and no selection)
A typical case is when you try to force xdg-open on a media (instead
of inline) by pressing "Control". This would trigger single-click box
selection on mouse-down but on mouse up the mouse didn't move yet. By
processing this as selection one also leaves the "didclick ==
EINA_TRUE" from mouse-down, leading the link to not be processed
(_activate_link() wasn't called).
PS: what an awful name is "didclick".
Export new termio_selection_exists() and two signals: "selection,on"
and "selection,off". These will be used to control the state of the
"copy" button and menus (when those exist).
since nothing useful is actually done unless either oldc or newc have
a media char - then try return/skip early. might speed up by 2-5%...
(i see 4.05 vs 4.15 or so cat times for a test case).