DynamoRevit for Design Automation

This repo contains supporting files, scripts and documentation that can help streamline onboarding and running Dynamo graphs in Design automation for Revit.

Overview of the standard design automation process

• Create an AppBundle, a collection of dlls that represents a revit cloud addin.
• Publish an Activity, metadata that describes some of the characteristics of the job that will run your AppBundle, like the inputs and outputs.
• Upload files to Cloud Storage, the files that will be referenced by the Activity.
• Submit a WorkItem, a job.
• Download the Results.

Revit add-in on Design Automation Walkthrough

Prerequisites

• Go to APS Developer portal.
• Create an app.
• Grab the Client ID and Client Secret.
• Clone this repo to your local system.
• To use postman, download and install it from here.
• There are 2 files you need to import in the postman app, a collection and an environment file.
• Set the current environment to the loaded DA4DynamoRevit
• Set the work directory in postman to the location where you cloned this repo locally

Tasks

For general instructions see the Walkthrough.

Below there are instructions on how to use postman to run design automation work items. Switch to the Execute a Dynamo Revit Add-in collection and go through the Tasks one by one. Some of the tasks also contain Delete commands which can be helpful if you want a clean slate.

• Task 1 - Convert Revit Add-in.

  • The addin is already created and you find it at DynamoRevitDA.zip.
  • If you need to use python nodes please also see pythonDependencies.zip.
  • If you need to run dynamo packages, you can zip up the packages in a single zipped with the name packages.zip. See this example packages.zip.

• Task 2 – Obtain an Access Token.

  • Run the Get an Access Token request in the Task 2 - Obtain an Access Token section of the postman collection.

• Task 3 – Create a Nickname.

  • Run the Create Nickname request in the Task 3 - Create a Nickname section of the postman collection.

• Task 4 – Upload the AppBundle.

  • Register the AppBundle by running Register the AppBundle request.
  • Upload the AppBundle by running the Upload the AppBundle request.
  • Create an alias for the AppBundle by running the Create an Alias for the AppBundle request.

• Task 5 – Publish an Activity.

  • Create a new Activity by running the Create a New Activity request.
  • Create an alias for the Activity by running the Create an Alias to the Activity request. Notes:

  The activity must use specific parameters to be compatible with the dynamo AppBundle.

  • parameters - the DynamoAddin requires the following parameters (but not limited to):

    • An .rvt file (get parameter) on which to operate (will be automatically opened as the current revit model). You can use DeleteWalls.rvt as an example.
    • A run.json file (get parameter). This must respect the Player specifications for a GraphRunRequest. You can use run.json as an example.
    • A pythonDependencies.zip (get parameter). This zip can be downloaded from pythonDependencies.zip. It contains the Python.Included.dll and Python.ScyPy.dll normally included with Dynamo out of the box. These must be uploaded separately from Dynamo because they are too large to bundle up with the DynamoAddin (DesignAutomation has a size limit on the AppBundles you can upload set at 100 MB).
    • A result.json (put parameter) can be included. This file will contain the graph run output (as produced by Dynamo Player).
    • A result.rvt (put parameter) can be included in order to download it. This file will contain the provided revit model after the dynamo run.

    Here is an example of an activity:

    curl -X POST 'https://developer.api.autodesk.com/da/us-east/v3/activities' \
      -H 'Content-Type: application/json' \
      -H 'Authorization: Bearer <ACCESS_TOKEN>' \
      -d '{
            "id": "DeleteWallsActivity13",
            "commandLine": [ "$(engine.path)\\\\revitcoreconsole.exe /i \"$(args[rvtFile].path)\" /al \"$(appbundles[DynamoDATest].path)\"" ],
            "parameters": {
             "rvtFile": {
                "zip": false,
                "ondemand": false,
                "verb": "get",
                "description": "Input Revit model",
                "required": true,
                "localName": "$(rvtFile)"
              },
             "runRequest": {
                "zip": false,
                "ondemand": false,
                "verb": "get",
                "description": "Input Revit model",
                "required": false,
                "localName": "run.json"
              },
              "pythonLibs": {
                "zip": true,
                "ondemand": false,
                "verb": "get",
                "description": "Python libs",
                "required": false,
                "localName": "pythonDependencies"
              },
              "dynResult": {
                "zip": false,
                "ondemand": false,
                "verb": "put",
                "description": "Results",
                "required": false,
                "localName": "result.json"
              },
              "rvtResult": {
                "zip": false,
                "ondemand": false,
                "verb": "put",
                "description": "Results",
                "required": false,
                "localName": "result.rvt"
              }
            },
            "engine": "Autodesk.Revit+2026",
            "appbundles": [ "DynamoDATest.DynamoDATest+test" ],
            "description": "Deletes walls from Revit file."
          }'

• Task 6 – Prepare Cloud Storage.

  • If not already created, run the Create a Bucket request to create a bucket for all storage items.
  • The get parameters described at the previous step must be uploaded as individual files. Run all the requests in the Task 6 - Prepare cloud storage folder.
    • The .rvt file - run the tasks inside the RvtFile subfolder.
    • The run.json file - run the tasks inside the RvtFile subfolder.
    • The run.dyn file - run the tasks inside the RvtFile subfolder.
    • The pythonpythonDependencies.zip - run the tasks inside the RvtFile subfolder.
    • The packages.zip - run the tasks inside the packages.zip subfolder.

• Task 7 - Submit a WorkItem.

  • Submit a workitem by running the Create a WorkItem request.
  • Check the work status by running the Check Status of a WorkItem request.

  Here is an example of an activity:

    curl -X POST 'https://developer.api.autodesk.com/da/us-east/v3/workitems' \
      -H 'Content-Type: application/json' \
      -H 'Authorization: Bearer <ACCESS_TOKEN>' \
      -d '{
            "activityId": "DynamoDATest.DeleteWallsActivity7+test",
            "arguments": {
              "rvtFile": {
                "url": "urn:adsk.objects:os.object:dynamodatestbucket/DeleteWalls.rvt",
                  "verb": "get",
                  "headers": {
                    "Authorization": "Bearer <ACCESS_TOKEN>"
                  }
              },
              "runRequest": {
                "url": "urn:adsk.objects:os.object:dynamodatestbucket/run.json",
                  "verb": "get",
                  "headers": {
                    "Authorization": "Bearer <ACCESS_TOKEN>"
                  }
              },
              "pythonLibs": {
                "url": "urn:adsk.objects:os.object:dynamodatestbucket/da_files/pythonDependencies.zip",
                  "verb": "get",
                  "headers": {
                    "Authorization": "Bearer <ACCESS_TOKEN>"
                  }
              },
              "dynResult": {
                  "url": `urn:adsk.objects:os.object:dynamodatestbucket/result.json`,
                  "verb": "put",
                  "headers": {
                      "Authorization": `Bearer <ACCESS_TOKEN>`
                  }
              },
              "rvtResult": {
                  "url": `urn:adsk.objects:os.object:dynamodatestbucket/result.rvt`,
                  "verb": "put",
                  "headers": {
                      "Authorization": `Bearer <ACCESS_TOKEN>`
                  }
              }
            }
          }'

• Task 8 - Download the Results

  • Run the Get S3 Download URL for Resulting RVT file and Download Resulting RVT File tasks to download the revit file.
  • Run the Get S3 Download URL for result.json and Download Result.json to download the dynamo output. The result will look like this

Useful scripts

You can perform above tasks using scripts in this repository.

Important

Copy .env.example and save it as .env and fill in the CLIENT_ID and CLIENT_SECRET and update other fields as needed.

How to run bash scripts

Important

• Install jq.
• Prepare environment variables.
  • You can use direnv to load values from .env as environment variables.
  • Alternatively, you can copy use an .envrc file. .envrc.example and save it as .envrc and source it source .envrc to load the environment variables.

You use these bash scripts to perform all of the tasks listed above. e.g ./scripts/get_appbundles.sh.

• Task 1 - Convert Revit Add-in

  • Download addin - ./scripts/download_artifacts.sh

• Task 2 – Obtain an Access Token

  • Get access token - ./scripts/get_access_token.sh

• Task 3 – Create a Nickname

  • Create nickname - ./scripts/create_nickname.sh <nickname>

• Task 4 – Upload the AppBundle

  • Register the AppBundle - ./scripts/register_appbundle.sh <app_id> => output <appbundle_json>
  • Upload the AppBundle. Use DynamoRevitDA.zip. - ./scripts/upload_appbundle.sh <appbundle_json>.
  • Create an alias for the AppBundle. - ./scripts/create_appbuldle_alias.sh <app_id> <app_alias>
  • Create a new version of the AppBundle. - ./scripts/update_appbundle.sh <app_id>
  • Point the alias to the new version of the AppBundle. - ./scripts/update_appbundle_alias.sh <app_id> <app_alias> <app_version>

• Task 5 – Publish an Activity

  • Create activity - ./scripts/create_activity.sh <nickname> <app_id> <app_alias> <activity_id> => output activity.json
  • Create activity alias - ./scripts/create_activity_alias.sh <activity_id> <activity_alias>
  • Download report. Report will look like this - ./scripts/download_report.sh

• Task 6 – Prepare Cloud Storage

  • Create bucket - ./scripts/create_bucket.sh <bucket_key> => output bucket.json
  • Create signed upload url - ./scripts/create_signed_upload_url.sh <bucket_key> <object_key> => output <object_key>_upload.json
  • Upload file to bucket - ./scripts/upload_to_bucket.sh <upload_json> <file>
  • Complete upload - ./scripts/complete_upload.sh bucket_key> <object_key> <upload_json>

  The get parameters described at the previous step must be uploaded as individual files:

  • The .rvt file
  • The run.json file
  • The run.dyn file
  • The pythonpythonDependencies.zip

• Task 7 - Submit a WorkItem

  • Create workitem - ./scripts/create_workitem.sh <nickname> <activity_id> <activity_alias> <bucket_key> => output workitem_<timestamp>.json
  • Get workitem status - ./scripts/get_workitem_status.sh <workitem_id> => output workitem_<workitem_id>.json
  • Get report - ./scripts/get_report.sh <workitem_status_json> => output report.txt

• Task 8 - Download the Results

  • Create signed download url - ./scripts/create_signed_download_url.sh <bucket_key> <object_key> => output <object_key>_download.json
  • Download result - ./scripts/download_result.sh <download_json> <file>

  Result will look like this

Run DA with Revit cloud models

Only the postman collections have a working example The only 2 sections that are specific for cloud models are:

1. Task 5 - Create a Revit Cloud Model Activity Here we create an activity without requiring a revit file input to be supplied
   • Removed the /i \"$(args[rvtFile].path)\" part from the "Commandline"
   • Removed the rvtFile parameter
   • Added a new setup paramter to specify the location of the cloud model
2. Task 7 - Create a Revit Cloud Model WorkItem Here we adjust the Workitem for the Activity described above
   • Added a new argument

     "setup": {
         "url": "data:application/json, {'SaveRevitFile': true, 'CloudModel': {'Region': 'US', 'ProjectGuid': '3e0d80bb-a0fd-4930-84f8-72ff40ef12d2', 'ModelGuid': 'ff4e6d0b-5c64-41ce-9292-617e1c61296c'}}"
     }
     

   • Added a new argument

     "adsk3LeggedToken": "dummy_value"
     

     This is necesary for DA to be able to open cloud shared models. The adsk3LeggedToken can be generated by following these steps https://aps.autodesk.com/en/docs/design-automation/v3/developers_guide/revit_specific/revit-cloud-model-integration/#authentication-requirements

Run DA with javascript: design_automation.js

Important

Install nodejs. Run npm install

You can also use this js script for all the update tasks once the resources are created. e.g node design_automation.js all.

• Run node design_automation.js all to set up all tasks (from 2 to 8)
• You can also run individual tasks, by specifying one or more of the following arguments:
  • nick- Creates a nickname
  • bundle- Creates an app bundle
  • activity - Creates an activity
  • rvt - Uploads the revit file
  • python - Uploads the pythonDependencies.zip
  • run - Uploads the run.json request
  • work - Creates and submits a WorkItem request
  • result - Polls for results (ands saves the results locally log.txt, result.json and result.rvt)

Useful tools

Design Automation Tools portal.

Tip

You can also see your AppBundles, Activities and WorkItems in Design Automation Tools portal.

Autodesk Platform Services vscode extension.

Tip

You can browse your APS resources in VSCode with Autodesk Platform Services extension.