Commit graph

63 commits

Author SHA1 Message Date
ee5bc046ae UI-Coordinates: draft how the builder manipulates content 2017-10-02 00:38:22 +02:00
ebed6fff1a UI-Coordinates: draft structure of the builder-API 2017-10-01 23:25:23 +02:00
5097637f0d UI-Coordinates: basic unit test PASS 2017-10-01 21:54:35 +02:00
1079d51c7e UI-Coordinates: implement named component access 2017-10-01 21:37:04 +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
372512006f UI-Coordinates: use a recursive implementation layout
this is a more or less arbitrary guess regaring performance requirements
2017-09-25 00:26:19 +02:00
5e1c25aaf5 UI-Coordinates: extract PathArray base abstraction into a library class 2017-09-24 22:50:42 +02:00
4082526ec6 UI-Coordinates: stub basic path element iteration 2017-09-24 21:14:26 +02:00
6073dbfcaf UI-Coordinates: stub basic access operations (WIP) 2017-09-24 17:20:47 +02:00
feb8414016 UI-Coordinates: stub to pass compilation 2017-09-23 02:25:52 +02:00
afda9e0a69 UI-coordinates: also need to define a topological addressing scheme (#1106) 2017-09-10 00:32:31 +02:00
fef0a812c1 DockAccess: start implementation draft for the DSL 2017-09-09 23:30:44 +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
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
1ed4bd2c4c UI-top-level: the former Application Object (GtkLumiera) is obsoleted now
All the backbone functionality has been factored out into dedicated facilities
2017-05-19 16:57:16 +02:00
5cb5ad3507 Menu-Actions: just log unimplemented actions for now (closes #1085)
this topic is basically settled by now.
Mostly by delegating to other entites not-yet-implemented :-D
2017-04-23 18:46:58 +02:00
e29a31a32a Menu-Actions: relocate some further global operations
...which do not really belong into the Actions class,
but can be delegated within the global context
2017-04-23 18:28:32 +02:00
1ccf54fa41 CmdAccess: draft some further command invocations
...which opens more questions than it solves at the moment.
Especially note #1096, the question how to refer to object-IDs
Maybe we need to enable sending EntryIDs via GenNode?

Anyway, the magic spell is broken now: we have a way how to
establish commands and how to issue them from the UI, with full integration
of UI-Bus, layer separation facade, instance management and ProcDispatcher

Looks like a stepping stone
2017-04-17 23:16:57 +02:00
4e7c64bf39 CmdAccess: **Big Bang** wire the first real command from UI down into the Session
...just the implementation within SessManagerImpl is not yet there P-)
So everything remains as defunct as before, but at least: "yes we can"
2017-04-17 21:54:36 +02:00
6a80053395 CmdAccess: reworked draft for context-bound commands and resolver expressions 2017-04-17 21:20:51 +02:00
10c2e4b9a9 CmdAccess: rename the front-end to CmdContext to clarify the purpose 2017-04-17 20:00:07 +02:00
82d66cef73 CmdAccess: discard the InvocationTrail concept
after extended analysis, it turned out to be a "placeholder concept"
and introduces an indirection, which can be removed altogether

- simple command invocation happens at gui::model::Tangible
- it is based on the command (definition) ID
- instance management happens automatically and transparently
- the extended case of context-bound commands will be treated later,
  and is entirely self-contained
2017-04-17 18:21:52 +02:00
5f6854621e Command-Cycle: remove the separate 'bang!' message
as it turns out, we can always trigger commands right away,
the moment all arguments are known. Thus it is sufficient to
send a single argument binding message, which allows us to
get rid of a lot or ugly complexities (payload visitor).
2017-04-14 23:45:35 +02:00
35a4e7705b CmdAccess: expand on the DSL draft 2017-04-14 03:22:08 +02:00
22c1a1d189 Commands: rename some of the planned components for command access
...to make the names more handy
2017-04-08 16:24:36 +02:00
9c21164ae6 Doxygen Fixes (#1062)
This changeset fixes a huge pile of problems, as indicated in the
error log of the Doxygen run after merging all the recent Doxygen improvements

unfortunately, auto-linking does still not work at various places.
There is no clear indication what might be the problem.
Possibly the rather unstable Sqlite support in this Doxygen version
is the cause. Anyway, needs to be investigated further.
2017-04-02 04:22:51 +02:00
ff42530f25 push on the topic of global action definitions (#1085)
...because this topic serves as a vehicle to elaborate various core concepts
of the UI backbone, especially how to access, bind and invoke Proc-Layer commands
2017-03-14 04:30:02 +01:00
789246fc3a draft a concept for command instantiation (#1070) 2017-03-08 04:25:33 +01:00
41ea59176c UI-top-level: include global help controller ("wizzard") 2017-03-02 23:49:23 +01:00
8d27585976 Menu-Actions: add stubs to forward session operations to InteractionDirector 2017-03-02 23:08:01 +01:00
198ccff396 UI-top-level: install presentation state recording service (#1081) 2017-02-19 04:46:13 +01:00
60adaa5639 UI-top-level: simplify name and namespace
the (Presentation)StateManager interface and implementation
seems to fit in more into the ctrl package
2017-02-19 04:27:09 +01:00
4f1def6366 UI-top-level: setup wiring of interacton control core components 2017-02-19 03:46:00 +01:00
cddc5afe41 UI-top-level: establish top-level model and control structure 2017-02-19 02:50:55 +01:00
f1f7b06d90 UI-top-level: introduce new entities for Interaction Control (#979, #1078, #1080)
In fact this also introduces various new concepts and represents
a fundamental decision regarding the organisation of the UI
2017-02-18 02:54:50 +01:00
3e6b2ae51f UI-top-level: rearrange various file locations
"workspace" is no longer the de-facto backbone of the UI,
rather, we got a global context residing in "ctrl"
2017-02-17 21:16:42 +01:00
b52ab62caf SessionCommand: define function test for message based invocation
the intention is to cover more of the full invocation path,
without running all of the application infrastructure. So this
second test cases simulates how messages are handled in CoreService,
where the CommandHandler (visitor) actually invokes the SessionCommand
facade
2017-01-13 08:26:41 +01:00
cfbbb750f8 considerations regarding the integration of commmand invocations (#1046)
not quite sure how to get the design straight.
Also a bit concerned because we'll get this much indirections;
the approach to send invocations via the UI-Bus needs to prove its viability
2017-01-09 01:22:43 +01:00
d7d90bf491 Element protocol: broadcast of state reset messages unit test PASS
This basically finishes definition of the fundamental
UI-Element and Bus protocol -- with one notable exception:
how to mutate elements by diff.

This will be the next topic to address
2016-02-14 05:03:08 +01:00
5bbf08adcb implement deleting of individual property state data 2016-02-14 04:29:40 +01:00
4da75dd4d3 bus protocol change: special handling for reset state marks
- suppres sending redundant stat mark messages from MockElm
- emit a "reset" state mark when an actual reset happens
- let the PresentationStateManager discard recorded special state
  when receiving a "reset" mark for a given element
2016-02-13 23:48:34 +01:00
d57af50ad6 state manager storage implemented and covered by unit test
sigh.
If you want to feel slick and cool,
never dare to write any unit test....
2016-02-13 22:55:59 +01:00
f80982b52b gen-node: fix insidious data conssitency problem
I assumed that, since GenNode is composed of copyable and
assignable types, the standard implementation will do.
But I overlooked the run time type check on the opaque
payload type within lib::Variant. When a type mismatch
is detected, the default implementation has already
assigned and thus altered the IDs.

So we need to roll our own implementation, and to add
insult to injury, we can't use the copy-and-swap idiom either.
2016-02-13 22:55:59 +01:00
121cd41408 ouch: GCC-4.9 doesn't yet support the C++14 transparent comparators
This is actually a STL library feature, and was added precisely
for the reason encountered here: if we want logarithmic search,
we'll have to construct a new GenNode object, just to have something
for the set to invoke the comparison operator.

C++14 introduced the convention that the Comparator of the set
may define a marker type `is_transparent` alongside with a generic
comparison operator. But, as is obvious from the source code of
our GNU Standard library implementation, our std::set has no such
overload to make use of that feature

http://en.cppreference.com/w/cpp/container/set/find
http://stackoverflow.com/questions/20317413/what-are-transparent-comparators

The only good thing is that, just 10 minutes ago, I felt like
a complete moron because I'm writing a unit test for such a simple
storage class. ;-)
2016-02-13 22:55:59 +01:00
94fc160525 implementation of storage for state manager 2016-02-13 22:55:59 +01:00
94576af4df finialise simple state manager implementation
...and rearrange storage interface to suit
2016-02-13 22:55:59 +01:00
92d9a7323f implement state manager through stubbed storage functions
...push down the frontier by one level
2016-02-13 22:55:58 +01:00
071f49027f change presentation state manager API
...based on elementIDs rather, to avoid any
tangling and trickery with reconstructing IDs
2016-02-13 22:55:58 +01:00