Writing X12 EDI with StediOctober 16, 2022

This tutorial covers end-to-end architecture and code for writing X12 EDI documents with Stedi. This implementation demonstrates one possible way to interact with Stedi's APIs to achieve a typical EDI write workload; your implementation may include some or all of these products depending on your particular systems and requirements.

Get the files: See the matching GitHub repository to get all the files for this tutorial
Watch a video walkthrough: See a video walkthrough for this tutorial

The main orchestration point of this demo is a Stedi function called write-outbound-edi, which is written in TypeScript. For this workload, the function is the entry point and expected to be called along with a JSON payload that represents the source data that needs to be converted into the X12 EDI document.

As the illustration below shows, the write-outbound-edi function performs several steps:

  1. Accepts a JSON payload of the source data.

  2. Calls Stash to generate a control number for the EDI document.

  3. Passes the incoming source JSON to Mappings using a predefined Mapping. The Mapping converts the source data to a predefined Guide (which is a JSON Schema).

  4. Combines the mapped result, control number & guide id before calling the EDI Translate API.

  5. The EDI Translate API retrieves the guide, validates that the input conforms to the guide schema, and generates the X12 EDI document.

  6. The function finally saves the EDI string as a file in a Bucket.

  7. A trading partner can retrieve the EDI from Buckets using SFTP.

Software architecture diagram of Stedi products to write EDI

Resource directories

Each subdirectory within the resources directory contains templates that can be used to create an EDI document for a specific transaction set from sample JSON input. Each template directory includes:

  • guide.json: a Guide used to generate the EDI document
  • map.json: a Mapping that converts the sample JSON input to the schema of the Guide
  • input.json: the sample JSON input to the workflow
    Note, each input.json file MUST include a transactionSet property whose value matches the name of the directory in which it exists. For example:
    "transactionSet": "X12-850"
    The transactionSet attribute is used by the write-outbound-edi function to identify the appropriate guide and mapping to use.


  1. Node.js (npm version must be 7.x or greater)

  2. Clone the Github repo and install the necessary dependencies:

    git clone https://github.com/Stedi-Demos/write-edi-demo.git
    cd write-edi-demo
    npm ci
  3. This project uses dotenv to manage the environmental variables required. You must create a .env file in the root directory of this repo and add two environment variables:

    • STEDI_API_KEY: Your Stedi API Key - used to deploy the function and internally to interact with product APIs. If you don't already have one, you can generate an API Key here.
    • ENABLED_TRANSACTION_SETS: a comma separated list of transaction sets for which you would like to be able to generate EDI documents. The values in the list MUST match available subdirectory names under the resources directory. The names are case-sensitive. Note: you can always come back and add or remove entries from this list. After doing so, you'll just need to re-run the create-guides, create-mappings, and deploy steps described below in order to apply the changes.

example .env file:


The subsequent setup scripts will use the ENABLED_TRANSACTION_SETS environment variable to determine which resources to deploy to your account.

  1. Create the EDI Guides by running:

    npm run create-guides

    For each guide that is created, an environment variable entry will automatically be added to the .env file. The output of the script will include a list of the environment variables that have been added:

    Updated .env file with 2 guide entries:
  2. Create the Mappings by running:

    npm run create-mappings

    For each mapping that is created, an environment variable entry will automatically be added to the .env file. The output of the script will include a list of the environment variables that have been added:

    Updated .env file with 2 mapping entries:
  3. Create the Stash Keyspace to store control numbers:

    npm run create-keyspace

    The Stash Keyspace is used to generate (and increment) control numbers to use when generating the EDI documents.

  4. Configure the Buckets (one for SFTP access and one for tracking function executions):

    npm run configure-buckets

    For each bucket, an environment variable entry will automatically be added to the .env file. The output of the script will include a list of the environment variables that have been added:

    Updated .env file with 2 bucket entries:
  5. [Optional] Provision an SFTP user by visiting the SFTP Web UI. This is optional, and is something you can always come back and do later. Creating an SFTP user will allow you to log in to retrieve the generated EDI documents via SFTP. When you do create a user, be sure to record the password (it will not be shown again).

Setup & Deployment

The tutorial repo includes a basic deployment script to bundle and deploy the write-outbound-edi function to Stedi. To deploy you must complete the following steps:

  1. Confirm that your .env file contains the necessary environment variables:

    • For each transaction set in the ENABLED_TRANSACTION_SETS list:
      • a <TXN_SET>_GUIDE_ID variable
      • a <TXN_SET>_MAPPING_ID variable

    It should look something like the following:


    Other than the Stedi API key, you shouldn’t store secrets in environment variables. If your function needs access to passwords, API keys, and the like, store them in Stash and access Stash from your function. The Stedi API key is the only exception, because you need it to access other Stedi products, including Stash.

  2. To deploy the function:

    npm run deploy

    This should produce the following output:

    > stedi-write-edi-demo@1.1.0 deploy
    > ts-node ./src/setup/deploy.ts
    Deploying write-outbound-edi
    Done write-outbound-edi
    Deploy completed at: 9/14/2022, 08:48:43 PM

Invoking the function

Once deployed, you may access the Function Web UI and perform the following steps.

  1. In the Functions List view you should see a function labelled write-outbound-edi, click on it's name to view its details.

  2. Clicking on the Edit environment variables link will allow you to see the variables that were populated during the deploy step. The values will include the values from your .env file, as well as values from the .resource_ids files for the transaction sets that you included in the ENABLED_TRANSACTION_SETS list.

  3. Click the Edit execution payload link, paste the contents of src/resources/X12-850/input.json into the payload modal, and click save. You can choose an input.json for any one of the transaction sets included in the ENABLED_TRANSACTION_SETS list, but we'll show X12-850 in the examples below.

  4. Hit the Execute button, if successful the Output should look similar to the following (this example output was created using the input for the X12-850 transaction set):

      "statusCode": 200,
      "bucketName": "4c22f54a-9ecf-41c8-b404-6a1f20674953-sftp",
      "key": "trading_partners/ANOTHERMERCH/outbound/000000001.edi",
      "body": "ISA*00*          *00*          *ZZ*AMERCHANT      *14*ANOTHERMERCH   *220915*0218*U*00501*000000001*0*T*>~GS*OW*WRITEDEMO*072271711TMS*20220915*021828*000000001*X*005010~ST*850*000000001~BEG*00*DS*365465413**20220830~REF*CO*ACME-4567~REF*ZZ*Thank you for your business~PER*OC*Marvin Acme*TE*973-555-1212*EM*marvin@acme.com~TD5****ZZ*FHD~N1*ST*Wile E Coyote*92*123~N3*111 Canyon Court~N4*Phoenix*AZ*85001*US~PO1*item-1*0008*EA*400**VC*VND1234567*SK*ACM/8900-400~PID*F****400 pound anvil~PO1*item-2*0004*EA*125**VC*VND000111222*SK*ACM/1100-001~PID*F****Detonator~CTT*2~AMT*TT*3700~SE*16*000000001~GE*1*000000001~IEA*1*000000001~"
  5. You can view the file using the Buckets Web View. As shown above, the output of the function includes the bucketName and key (path within the bucket) of where the generated EDI was saved.

  6. You may also connect using your preferred SFTP client and the user credentials provisioned earlier to retrieve the file.

  7. You can also view the other resources that were created during setup in the UIs for the associated products:

    • Guides Web UI: a guide for each transaction set that you enabled
    • Mappings Web UI: a mapping for each transaction set that you enabled
    • Stash Web UI: a keyspace for tracking and incrementing control numbers for generated X12 EDI files

Function execution tracking

The write-outbound-edi function uses the bucket referenced by the EXECUTIONS_BUCKET_NAME environment variable to track details about each invocation of the function.

  1. At the beginning of each invocation, an executionId is generated. The executionId is a hash of the function name and the input payload. This allows subsequent invocations of the function with the same payload to be processed as retries.

  2. The JSON input to the function is written to the executions bucket in the following location:

  3. If any failures are encountered during the execution of the function, the details of the failure are written to the following location:

  4. Upon successful invocation of the function (either on the initial invocation, or on a subsequent retry with the same input payload), the input.json as well as the failure.json from previous failed invocations (if present) are deleted. Therefore, any items present in the executions bucket represent in-progress executions, or previously failed invocations (with details about the failure).