As a user I would like to be able to create a ContinuousScalar and have it initialize the VariableData correctly (at the moment it is forced to creating a Pose2-format structure, which causes a BoundsError when it is loaded into the cloud).

TAC

• Add ContinousScalar to the supported VariableTypes
• Parametrize the VariableNodeData so that it initializes correctly for each type
• Add unit tests to validate operation

As a user I would like to be able to create LinearRelative factors.

To use the ipython features should then be something to the effect:

import ipython.markdown
import navabilitysdk

# use nva markdown features

• #9
• Conform naming with other Julia and JS SDK
• Ideally we want to share GQL

Occurs when we do not await a call I believe.

Suggested read: https://gql.readthedocs.io/en/v3.0.0/advanced/async_advanced_usage.html.

Required for ICRA tutorial 1.

As a user I would like to create Prior. This looks to be the same as PriorPose2 but @dehann @Affie can you confirm that we can simply copy PriorPose2 as Prior?

As a user I would like to be able to use the Uniform, Normal, and Rayleigh distributions in Factors.

This is needed in the Mixture factor in the ICRA Tutorial 2 example.

Otherwise it is truly clunky importing.

listNeighbors(clicon..., "x4")

get back the list of factors connected to that variable

["x3x4f_sdr"; "x4x5f_3r3"; "x4l7f_34r"]

And visa versa for the variables connected to a factor, but that case should just alias to equivalent of:

getVariableOrder(dfg, fctlbl)

which gives back

["x3";"x4"]

Thereafter, alias the specific dispatches for list and ls, also lsf.

PS, I'm writing hot and loose wrt String vs Symbol -- should resolve the default there better.

Hi @GearsAD , you might know best. Does Pose2Point2BearingRange exist in this SDK.py yet? I searched for it in the code but did not see it.

This is currently throwing an error, which isn't correct because the data may still be in queue.

E.g. response:

DEBUG:navability.services.variable:Query result: {'USER': [{'name': 'Guest', 'robots': [{'name': 'DemoRobot', 'sessions': []}]}]}

Currently returns the whole payload.

Run readme example, you get the following error:

Traceback (most recent call last):
  File "test.py", line 31, in <module>
    result = navi.solveSession()
TypeError: solveSession() missing 1 required positional argument: 'c'

(converted to issue, someone mentioned this:)

int -> float
I didn't have time to figure python sdk out to see what else is needed.

Point to first instance of same issue:

• NavAbility/NavAbilitySDK.jl#89 (comment)

A lot of info messages from gql.transport.aiohttp, set up logging to minimize this.

Reproduce the error by doing fresh install via readme instructions. Run provided script. Error below occurs:

Traceback (most recent call last):
  File "test.py", line 4, in <module>
    from navability.entities.StatusMessage import StatusMessage
  File "/home/jim/Workspace/NavAbility/NavAbilitySDK/venv/lib/python3.8/site-packages/navability/entities/StatusMessage.py", line 5, in <module>
    from src.navability.entities.Client import Client, ClientSchema
ModuleNotFoundError: No module named 'src'

This looks to be a constructor issue.

Looks the same as the other Z factors.

"{\"label\":\"x0l3f1\",\"_version\":\"0.18.1\",\"_variableOrderSymbols\":[\"x0\",\"l3\"],\"data\":{\"eliminated\":false,\"potentialused\":false,\"edgeIDs\":[],\"fnc\":{\"Z\":{\"_type\":\"IncrementalInference.PackedNormal\",\"mu\":89.44271909999159,\"sigma\":3.0}},\"multihypo\":[],\"certainhypo\":[1,2],\"nullhypo\":0.0,\"solveInProgress\":0,\"inflation\":5.0},\"tags\":[\"FACTOR\"],\"timestamp\":\"2022-03-24T22:17:04.617-05:00\",\"nstime\":\"0\",\"fnctype\":\"Point2Point2Range\",\"solvable\":1}"

As a user of the SDK I would like to be able to optionally set the solve key and whether the solve should be parametric.

These are passed in as SolveOptionsInput in the solveSession call.

As a user I'd like to be able to add blob data to variables in the graph.

Technical acceptance

• Build the entities BlobMetadata and BlobData
• Provide a low-level function that performs the end-to-end upload of a data blob addDataBlob(BlobMetadata, data:(string or bytes))
• Provide a low-level function to link this blob to a variable - need a signature for this

Blocking on:

• #51
• #57

Looks exactly the same as PriorPose2:

"{\"label\":\"l1f1\",\"_version\":\"0.18.1\",\"_variableOrderSymbols\":[\"l1\"],\"data\":{\"eliminated\":false,\"potentialused\":false,\"edgeIDs\":[],\"fnc\":{\"Z\":{\"_type\":\"IncrementalInference.PackedFullNormal\",\"mu\":[10.0,30.0],\"cov\":[1.0,0.0,0.0,1.0]}},\"multihypo\":[],\"certainhypo\":[1],\"nullhypo\":0.0,\"solveInProgress\":0,\"inflation\":5.0},\"tags\":[\"FACTOR\"],\"timestamp\":\"2022-03-24T22:06:41.310-05:00\",\"nstime\":\"0\",\"fnctype\":\"PriorPoint2\",\"solvable\":1}"