mirror_lkl_layer.md

Mirror Last Known Locations layer (LKL) into a standard feature service

This script will copy data from a Last Known Locations (LKL) layer in a Location Tracking (either Location Tracking Service or Location Tracking View) layer and mirror it into a standard feature service.

As of April 2020, the LKL layer does not support dynamic joins, so the mirroring of data using this script may be useful if the user is trying to relate additional data to their last known locations data. Standard feature services support join features.

Please see the code block below if attempting to use this script in Enterprise.

This script is designed to be run as a scheduled task, at whatever frequency you want to get updated LKL data from your Location Tracking layer. For example, the admin may set up this script to run every 10 minutes in order to get the latest data from the field.

For information on how to set up scheduled tasks, please see this article

and this article

This script requires the use of a point layer that the data can be mirrored into. This layer would ideally share all the same fields as a standard LKL layer, but could have less fields if all data is not requred. You can clone your Location Tracking Service (LTS) using the clone_items functionality in the Python API.

To clone, do:

import arcgis
gis = arcgis.gis.GIS("https://arcgis.com","myusername","mypassword")
item = gis.content.get(gis.properties["helperServices"]["locationTracking"]["id"])
cloned_item = gis.content.clone_items([item], copy_data=False)[0]
# delete Tracks layer from your cloned feature layer collection, so that you're left with only the LKL layer
flc = arcgis.features.FeatureLayerCollection(url=cloned_item.url, gis=gis)
flc.manager.delete_from_definition({ "layers" : [{"id" : 0}]})
# remove editor tracking on the layer
flc.manager.update_definition({"editorTrackingInfo":{"enableEditorTracking":"false","enableOwnershipAccessControl":"false","allowOthersToUpdate":"true","allowOthersToDelete":"true","allowOthersToQuery":"true","allowAnonymousToUpdate":"true","allowAnonymousToDelete":"true"}})

This will create a standard feature service in your organization whose item id you can pass. Note that this will turn off some of the security considerations in your LKL data.

If you wish, you can also:

Create a blank point layer (Content > Create > Feature Layer > Build a layer > Points) and then add your requisite fields to that layer. Ensure the field_name value is the same, not just the alias. Ensure you create an extra field named "created_user" if you take this approach

Location Tracking must be enabled for your organization to use this script. You must be either at least a track viewer in order to use this script.

In addition, the user may want to use Python to perform the join itself. While we recommend you use the "Export results of the Join features analysis tool as a hosted feature layer view" option while using AGOL (info about that can be found here), Enterprise does not support this feature as of 10.8. While joining is not supported in the script out of the box, you can easily modify it and perform a left join using the Pandas library.

Let's take an example where the we're joining LKL data to an external layer with a 1:1 relationship. This feature service will store the "Status" of each worker. The join will be on the "Creator" field in each service.

1. Add extra fields into your layer cloned from above to support the fields from the joined layer
2. After the applyEdits call has been made in the script (line 101, mirror_layer.edit_features), re-query the layer and store it as a sdf
3. Get external layer and convert to an SDF
4. Get most recent surveys submitted by each user
5. Use pandas to perform a left join on the two SDFs
6. For each user in the joined data, find its corresponding feature and append attributes. Then re-post the data.

Our code looks something like

# convert LKL to sdf
lkl_sdf = mirror_layer.query('1=1', as_df=True)
status_layer_item = gis.content.get('status_layer_id')
# get SDF of your extenral layer
status_sdf = status_layer_item.layers[0].query(where='1=1', as_df=True)
# perform merge
overlap_rows = pandas.merge(left = lkl_sdf, right = status_sdf, how='left', left_on='created_user', right_on='Creator')
updated_features = []
features = mirror_layer.query('1=1').features
for feature in features:
    # return row with matching created_user
    merged_row = overlap_rows.loc[overlap_rows['created_user'] == feature.attributes['created_user']]
    # try to update a field "Status", if fails assign not updated
    try:
        feature.attributes['Status'] = merged_row['Status'].values[0]
    except IndexError:
        feature.attributes['Status'] = "Not Reported"
    # repeat this for every field you want to update
    updated_features.append(feature)
# post features with joined data over    
updates = mirror_layer.edit_features(updates=updated_features)

Supports Python 3.6+

---

Other than the authentication arguments (username, password, org) the script uses the following parameters:

• -item_id - Item id in your portal of the feature service you want to mirror data into. See above for how to create this service. Required
• -lkl-layer-url <tracks_layer_url> - The URL to the LKL layer from a Track View / Location Tracking Service you want to utilize. This URL should end in /1 (since we're targeting the second layer in the feature collection). Required
• -log-file <logFile> - The log file to use for logging messages. Optional

Example Usage 1 - Mirror LKL data from the lkl layer url into the layer with the listed item id

python mirror_lkl_layer.py -u username -p password -org https://arcgis.com -item-id a05eee7b1cs5461db0e1ef1c1c4abe18 -lkl-layer-url https://locationservices9.arcgis.com/US6xjA1Nc8bW1aoA/arcgis/rest/services/f1087713d8934d5b8218dda736c26af4_Track_View/FeatureServer/1

What it does

1. First the script uses the provided credentials to authenticate with AGOL to get the required token
2. Gets the LKL layer, which is provided to the script
3. Gets the layer that you are mirroring data into
4. Checks whether the LKL is already in the mirrored layer or is new
5. Posts updates with both new and existing features using the edit_features functionality in the Python API