Add Scene Management Functionality #179
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This only deals with the device level db, not the modem yet. Handles three variations of config file defintions. Adds the concept of data_1, data_2, and data_3 keys in a definition.
Manually fixed conflicts in: insteon_mqtt/Modem.py insteon_mqtt/device/Base.py
Used to load and save the scenes.yaml file
Migrating code from prior work. Adds the !rel_path directive to config to allow for specifying a relative path in the config file. Reads scenes from the specified file. If a modem controller link is defined without a group ID, one will be assigned. Defined links are then loaded into the db_config for each device.
Will not remove links associated with join or pair. Also won't add them. We somewhat blindly handle pair, since we don't have any list of available groups for each device. This could be fixed by adding in a static list to each object if necessary. Had to change passed object to the device rather than the db. This enables us to use methods on the device, such as find.
Enables running on the entire network. Decrease verbosity of no changes.
Adds or deletes links from the device as defined in scenes.
Ensures that dry_run always matches what would happen. Also decreases namespace pollution. Dry_Run is the default. A user has to specify dry_run=false for an actual sync to occur.
Print log calls from within CommandSeq so messages appear in logical order for user. This required enabling a synchronous function such as LOG.ui which does not accept an on_done argument to be called by CommandSeq. Also run diff after refresh. We want to perform diff with most current data.
It is necessary for the db object to be able to make calls on the base object. Plus when doing an import, we would like to be able to set the db_config as the rhs.
Only added to devices currently, and does not save results to disk. This code is complicated and rather spaghetti like. This is in part because we allow a huge amount of latitude for the users to specify scenes. This process is undoubtedly complex, but the concept is that users will rarely use this feature.
Don't try and import these into the Scenes definition.
Duplicates the functions found on the devices.
Just a sequence initiated by the Modem Object
Scene will be imported with address only.
This still needs work
Redesigned Scenes Class to use SceneEntry Classes for each individual scene and SceneDevice Classes for each device entry in a scene. Adds a lot more code, but seems to be much more readable and logical.
Linking is Hard! Fix some confusion about what group and data values should be pushed to which link Remove unnecessary _load_scenes() command that is no longer used.
Not sure how this got messed up.
Define a weak search that doesn't check for group matches
Wait to do until save, this is moderately faster. Compress N-Way scenes.
For the most part, this can only happen with KPL devices. Imagine a single Modem scene turning on all of the buttons on a KeypadLinc Streatmlined populate scenes to have two less loops. Updated tests to match new design
SceneManager is a better description. Add code documentation to Scenes.py
The fanlinc is similar to a keypadlinc in that it has multiple groups and some dimmer features. Removed a bunch of unnecessary documentation in the link_data functions. Much of it was repetetive.
Write a small dissertation on how it works.
New Feature: Allow users to control the modem virtual scenes via a name defined in the scenes.yaml file.
to Stack Enables returns back to the main loop in between each function call.
Splitting into multiple calls for better handling by Stacks. No need to test items less than or equal to themselves since that test has already been done.
Used the wrong address. This is really complicated, everytime I touch something I seem to make a bug.
Fixes some esoteric bugs. Link_defaults for a multigroup device will report all groups as being default since we feed in the group number to generate the default. This fixes it, so that only group 0 or 1 devices are treated as default. Second, init the group value earlier, as it can affect the link_data names returned. Such as for a fanlinc where there is a ramp_rate for the light, but not for the fan.
The comparision test was backwards.
…etter Adding Log messages to the stack on CommandSeq was a bad idea. Any error in executing a function that threw a TypeError would cause the function to get exectuted again. In order to keep sync messages displayed near the approriate events needed to wrap the add and delete calls in functions that write the log messages. Also fixed a glitch wherein a DB refresh was asked for, but the DB diff was calculated prior to receiving the refresed db data.
The loading of the program still halts. I think this is the right answer. I can't think of a way to handle this and continue that doesn't cause potentially more issues that are harder for the user to diagnose later.
|
Fantastic! Thanks for doing this - that's a huge amount of work. I'll take a look this weekend and get it merged. |
Remotes need multigroup support like KPL for the link_data. Small error in the data_3 value on multigroup controller entries the data_3 is the group controller.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This adds the following new Features:
Syncing:
Importing:
Other:
Known Issues:
Unknown Issues:
Prudent Advice:
Bug Reporting:
Phew, alright good luck!
Fixes #25