Harvard Health Content External Web Service Documentation
Images
<img> tags in document bodies will contain an imageid attribute, which is always an integer number. The images are available through the web service.
Images themselves will not be available through SOAP requests, but must be retrieved with HTTP GET requests of the relevant URL's.
The imageid needs to be substituted into a URL such as the following:
http://www.content.health.harvard.edu/[SiteName]/1.img
where 1 is the imageid.
Because all image URLs have the .img extension, the MIME type cannot be inferred by the extension. The HTTP Content-Type header will correctly indicate the MIME type of the image. The supported types are: image/gif, image/jpeg, image/jpg and image/png. Requests for images are also authenticated using the IP address of the requestor so end-user browsers cannot request images directly. The request must be sent to your server which sends its own request to the Krames StayWell servers.
You can trust that the imageid for an image will always be the same. The image referenced by a given ID will never be updated or changed. Therefore it is safe and encouraged to cache images locally or on proxy servers. NOTE: images might be deleted from the UCR if the documents that reference them are removed from the UCR, but their ID's will never be reused. In addition, any document can be changed at any time to refer to different image IDs.
To successfully retrieve an image, a published document containing the image must be licensed to the SiteName that is being used in the request.