mca blog

the last SSDS 'sprint' (kinda like a beta release only 'sprintier') had a major update - support for BLOB data. users can now read/write/delete binary items using the REST interface for SSDS. very nice. this adds a whole new level to the power and flexibility of SSDS

despite a few drawbacks (limited control of the binary item's metadata and some URL-encoding bugs), i've been very keen on using SSDS to handle binary data. mostly because it allowes|forces me to level-up on my HTTP programming foo to handle binary data requests and responses. to this point, i've been taking shortcuts and only working with text-based data. this covers the better portion of most web requests, but leaves out very important stuff. not just images, but also PDF and the whole range of download-able binary formats (.ZIP, .DOC, etc.). i avoided this work for two reasons: 1) i was lazy, and 2) i was dumb. well, avoid no more.

it took some work tonight, but i finally got binary data support rolled into my HttpClient class and into the Entity class within the SSDS-Proxy library. this also meant updating my internal Caching class to make sure it could handle binary data. i still have some rough edges to sand out (mostly dealing with server-side content-negotiation and caching the proper media type), but it should be clean and zippy by the end of the week.

i first plan on updating the SSDS Provisioning Client. that will mean users will be able to upload, update, and delete binary data via the Web app. and once that's done, i've got some ideas on how to leverage binary data support in SSDS in general. if time allows, i'll have another demo app by the end of the month. i'm rubbing my hands together already[g].