30:[] 2e:{"title":"SEERdata","cells":[{"id":0,"value":"md`# SEERdata\n

Exploring access to NCI's Surveillance, Epidemiology, and End Results (SEER) program with shared credentials. The key challange is that of sharing keys between public applications under the same origin (in this case observablehq.com), without exposing them to other users. This public live document walks a user through the 1,2,3 steps of creating an API key that another application, being accessed by the same user, can use to call SEER data along side other public or private application .

\n\n\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":10,"value":"md`\n

Access credentials

In 2019, SEER made public an Application Programming Interface (API) for credentialed programatic interoperability at api.seer.cancer.gov. Accessing that URL directs individual users to create personal accounts and 1) manage API keys.

`","pinned":false,"hidden":false,"mode":"ojs"},{"id":33,"value":"md`\n

This should be an Error

\nPublic-facing computational platforms like Observable Notebooks typically support a functionality where API keys and other access credentials are stored for each user. Naturally, that functionality does is not available in the public-facing application, like this one. Calling them should produce an error, as below:\n

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":6,"value":"Secret(\"Your SEER API Key\")","pinned":true,"hidden":false,"mode":"ojs"},{"id":37,"value":"md`\nThis creates a conundrum. In order to disseminate an analytical procedure, the executable application needs to be made public, but with the inconvenience that the user will need to provide individually - for each application - the access credentials for each data source involved in the analysis.\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":41,"value":"md`\n

Share credentials under the same origin

The goal of this document is to manage your SEER API Key in a manner that can be accessed by all applications under the same origin. The solution demonstrated here starts with 2) providing your SEER API here and then verifying that another application, like 3) CaViz (Cancer Visualization) can call SEER data for analyses that merge it with CDC.

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":59,"value":"md`\n

Checking your credentials

You can the API key provided above, you can also for all applications under the shared origin.

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":87,"value":"md`\n

Methods

\nThe management of API keys across an application ecosystem (multiple *pathnames*) under a shared domain relies on a small open source library custom developed for this purpose, seerapi.js (https://github.com/episphere/seerapi).\n

\n`","pinned":true,"hidden":false,"mode":"ojs"},{"id":92,"value":"//require('http://localhost:8000/seerapi/seerapi.js')\nrequire(\"http://episphere.github.io/seerapi/seerapi.js\")","pinned":false,"hidden":false,"mode":"ojs"}],"files":"$W30","stdlib":"1","titling":"auto"}