Commit graph

901 commits

Author SHA1 Message Date
ed76151d14 UI-Coordinates: value representation finished and unit test PASS (#1106) 2017-10-03 00:57:23 +02:00
18d1e7a280 UI-Coordinates: polish test and consider next steps
After completing the self-contained UICoord data elements,
the next thing to consider might be how to resolve UI coordinates
against an actual window topology. We need to define a suitable
command-and-query interface in order to build and verify this
intricate resolution process separated from the actual UI code.
2017-10-02 18:11:21 +02:00
286b1829fe UI-Coordinates: implement path split and appending of multiple components
Unit test passes thus far
2017-10-02 06:49:50 +02:00
835b964e63 UI-Coordinates: implement append / prepend mutation 2017-10-02 06:45:50 +02:00
5097637f0d UI-Coordinates: basic unit test PASS 2017-10-01 21:54:35 +02:00
ac38f0f963 UI-Coordinates: implement string representation 2017-10-01 21:10:18 +02:00
6322f1bc3c UI-Coordinates: define next steps to cover 2017-10-01 20:04:12 +02:00
9378badf6b UI-Coordinates: integrate the initialisation split
...as developed in during the metaprogramming investigation
2017-09-30 00:46:52 +02:00
e5dc7ba2bc Investigation: dissect argument packs
start investigation on generic techniques to dissect an variadic argument pack
2017-09-26 19:23:03 +02:00
6073dbfcaf UI-Coordinates: stub basic access operations (WIP) 2017-09-24 17:20:47 +02:00
08f70c068c UI-Coordinates: dream up some basic properties (WIP)
ZOMG... who is to code up all this stuff...?
2017-09-24 02:04:23 +02:00
78cbf0f57e UI-Coordinates: define basic design 2017-09-23 17:55:40 +02:00
c1f240687b UI-Coordinates: elaborate and simplify DSL draft (WIP) 2017-09-23 01:21:06 +02:00
ff1b22a889 UI-Coordinates: DSL draft (WIP) 2017-09-15 01:38:11 +02:00
8a36327604 DockAccess: elaborate design of the DSL 2017-09-09 15:21:40 +02:00
b27681ec4f DockAccess: design a configuration DSL 2017-09-08 18:50:39 +02:00
a9797e4a4f DockAccess: analysis continued...
exploring the idea of a configuration DSL.
As a first step, this could be a simple internal DSL,
implemented as a bunch of static functor objects, which are internally bound
and thus implemented by the ViewLocator within InteractionDirector
2017-09-08 03:53:52 +02:00
eb04552c88 DockAccess: some analysis regarding Configuration/Strategy for access to views
...we have to face the problem that we need some generic strategy
for access to component views, which possibly will become customisable.
And the allowed patterns of access are quite different for the various
kind of view we know....
2017-09-04 01:21:53 +02:00
753c895035 DockAccess: consider a generic ID addessing scheme for the UI (#1005)
...based on a selection of Model types and an access strategy
2017-09-03 01:04:00 +02:00
0ad387f5c9 DockAccess: create a ViewLocator service
responsible for access and allocation of component views.
Internally wired to the PanelLocator within the global WindowLocator

This setup settles those nasty qeustions of crosswise top-level access
2017-09-02 19:28:57 +02:00
4f77075d9c DockAccess: integrate a PanelLocator intermediary
...into the WindowLocator
This is to become a low-level query and acess front-end
and will be implemented just by traversing the windowList_
2017-09-02 19:03:00 +02:00
0614ca36ca DockAccess: consider how to access view components. Rename WindowList (#1104)
this starts work on a new UI global topic (#1004)

- coin a new term: "view component"
- distinction between veiw component and Panel
- consider how to locate view components
- WindowList becomes WindowLocator
2017-09-02 18:36:58 +02:00
5b445a2361 InfoBox: extract into dedicated widget 2017-09-02 00:51:06 +02:00
ee67e4914c InfoBox: scroll to the last inserted line reliably
need to use a text mark; we can simplify this task somewhat
by exploiting the fact that the text cursor is a default defined text mark
2017-09-01 01:02:48 +02:00
2d2a549341 InfoBox: add a simple text display widget 2017-09-01 00:28:43 +02:00
a91b444ade InfoBox: draft the next preliminary implementation steps (#1002)
...I need a quick-n-dirty temporary solution,
just to have a place to display some messages in the UI asynchronously
2017-08-31 20:32:49 +02:00
45b3a990f2 DemoGuiRoundtrip: add new dock for UI experiments (#1099)
...after investigating problems related to the management of docking pane contents
2017-08-31 20:15:52 +02:00
526fd3ca16 UI-Dispatch: add generic catch-all error handler
...to ensure no exception can escape at top level of the actions
dispatched asynchronously into the UI event thread
2017-08-25 17:38:51 +02:00
937ad64596 DiffMessage: now uniformly plays the role of MutationMessage (closes #1066) 2017-08-13 07:25:32 +02:00
5ea80f39cb DiffMessage: successfully finish extended integration test
now we're able to inject flocks of Borg into the alpha quadrant by diff message
2017-08-13 07:25:32 +02:00
5fbc4b84bf DiffMessage: switch to moving DiffMessage over the bus
basically the opaque-buffer based MutationMessage implementation is obsoleted now
2017-08-12 17:59:02 +02:00
4a2384e242 DiffMessage: add further convenience ctor for varargs
hey... all my dream constructors became true
2017-08-12 03:00:38 +02:00
88b2260496 DiffMessage: draft test steps to drive refactoring 2017-08-11 15:48:28 +02:00
f6baef16c5 DiffMessage: consider to unite the handling of mutation messages (#1066) 2017-08-11 15:23:33 +02:00
f498f3be1b mark next steps to address
- error handling
 - settle MutationMessage
2017-08-11 02:37:29 +02:00
fd0a011ea4 DiffMessage: bold attempt towards a way to produce diffs (#1066)
actually I do not know much regarding the actual situation when,
within the Builder run, we're able to detect a change and generate
a diff description. However, as a first step, I'll pick IterSrouce
as a base interface and use a "generation context", which is to be
passed by shared-ptr
2017-08-11 00:59:10 +02:00
2344d5b03f UI-top-level: implement external UI-Shutdown-Trigger 2017-08-10 20:58:00 +02:00
805e6047ba UI-Dispatch: document some implementation intricacies 2017-08-10 20:57:59 +02:00
768a07f181 UI-Dispatch: draft the dispatcher invocation 2017-08-10 17:14:44 +02:00
07c9ed15e8 UI-Dispatch: how to integrate into the NotificationService
WIP setup of an empty framework
2017-08-10 16:24:36 +02:00
46fc900980 UI-Dispatch: get the multithreded test to work (#1098)
the (trivial) implementation turned out to be correct as written,
but it was (again) damn challenging to get the mulithreaded chaotic
test fixture and especially the lambda captures to work correct.
2017-08-07 05:19:58 +02:00
87dc04f324 UI-Dispatch: verify consistency of argument data handling 2017-08-05 18:44:25 +02:00
3dea3c0fa0 UI-Dispatch: draft basic interface of a queue helper (#1098) 2017-08-05 17:36:32 +02:00
9b285a95c0 UI-Integration: plan the next steps to drive this topic ahead (#1099, #1098)
- concept for a first preliminary implementation of dispatch into the UI thread
 - define an integration effort to build a complete working communication chain
2017-08-05 17:36:32 +02:00
1e81b8d61a Settle proper activation of the external UI interfaces (#1098)
This change was caused by investigation of UI event loop dispatch;
since the GTK UI is designed to run single threaded, any invocation
from other threads need to be diepatched explicitly.

A possible way to achieve this is to use Glib::Dispatcher, which
in turn requires that the current thread (which is in this case the UI thread)
already holds a Glib::MainContext

This prompted me to create a tight link between the external facade interfaces
of the UI and the event loop itself. What remains to be settled is how
to hand over arguments to the action in the main loop
2017-08-05 17:36:32 +02:00
9cf868c9cd after a long break: find my way back into what has to be done.... 2017-08-03 20:34:42 +02:00
c96845ab65 GTK evolution: abandon Gtk::Main and start event loop directly (closes #1032)
After investigation of current GTK and GIO code, I came to the conclusion
that we do *not* want to rely on the shiny new Gtk::Application, which
provides a lot of additional "convenience" functionality we do neither
need nor want. Most notably, we do not want extended desktop integration
like automatically connecting to D-Bus or exposing application actions
as desktop events.

After stripping away all those optional functions and extensions, it turns
out the basic code to operate the GTK main event loop is quite simple.
This changeset extracts this code from the (deprecated) Gtk::Main and
integrates it directly in Lumiera's UI framework object (UiManager).
2017-05-19 23:42:55 +02:00
f089a34934 UI-Lifecycle: tie UI-Bus and UI-Manager directly to GtkLumiera
this is just a tiny change to make things more othogonal.
Now the unwinding and calls to any GTK / Widget dtors happen *after*
emitting the term signal from UI shutdown. Which means, the other subsystems
are shutting down (in their dedicated threads) as well, thus lowering
the probability of some action still using the UI and triggering an exception
2017-05-19 18:12:58 +02:00
5e172ff6a0 UI-top-level: reactivate the updateActionState function as NOP
as it turned out, the former functionality was deactivated in 2009
with changeset 6151415

The whole concept seems to be unfinished, and needs to be reworked
and integrated with "Views and Perspectives" (whatever that is...)

See also #1097
2017-05-19 17:00:41 +02:00
e59e8d0ab5 UI-top-level: consider how to rework the UI main object (#1067)
Gtk::Main is deprecated, but the new solution, instantiating a
Gtk::Application object does not match our use case, since we handle
all application concerns already and just need a Gtk main loop to run.

Anyway, it became clear that the "main object" will be the new UiManager.
As a first step, I've now moved the (deprecated) Gtk::Main object
down there. Next step (planned) will be to inherit from Gio::Application
and clone some functionality from Gtk::Application
2017-05-03 02:37:48 +02:00