OSCON 2012: ql.io and Node.js
Download as PPTX, PDF0 likes891 views
This document discusses building HTTP APIs using node.js and ql.io. It provides examples of using ql.io to consume HTTP APIs cheaply and reduce the number of requests and bandwidth usage. Ql.io allows defining a data source and includes features like mustache templates and jQuery for cross-domain requests. The document encourages building optimized APIs using ql.io's domain specific language that is SQL-inspired and allows implicit orchestration of requests. It provides links to the ql.io code, documentation, demos and a blog for further information.
![Client
[Really important client]
Why don't you give me an
API optimized for my use
Producers
cases?
Server
[Really important producer]
Thanks. Get a number
and stand in the line!
18](https://image.slidesharecdn.com/2012julyosconqlio2-120720005535-phpapp02/85/OSCON-2012-ql-io-and-Node-js-18-320.jpg)
![Client
[Really important client]
Why don't you give me an
API optimized for my use
Producers
cases?
Who gets to
decide the right
Server
thing?
[Really important producer]
Thanks. Get a number
and stand in the line!
19](https://image.slidesharecdn.com/2012julyosconqlio2-120720005535-phpapp02/85/OSCON-2012-ql-io-and-Node-js-19-320.jpg)
OSCON 2012: ql.io and Node.js
- 1. node.js and ql.io Build Your Own July 20, 2012 for OSCON HTTP APIs 1
- 4. How to consume HTTP APIs cheaply and 4
- 5. curl -X <HTTP method> <URI> 5
- 7. 7
- 8. 8
- 9. 9
- 10. 10
- 11. 11
- 12. 12
- 13. 13
- 14. 14
- 15. Use case: Find things from A For each thing, find details from For each thing, find more deta Merge results 15
- 16. Use case: Find products Find dominant categories of pr Look up category info Merge categories with product 16
- 17. Use case: Get stuff from A If A is down, try from B Annotate stuff with other stuff f Ignore some things from the st Join all 17
- 18. Client [Really important client] Why don't you give me an API optimized for my use Producers cases? Server [Really important producer] Thanks. Get a number and stand in the line! 18
- 19. Client [Really important client] Why don't you give me an API optimized for my use Producers cases? Who gets to decide the right Server thing? [Really important producer] Thanks. Get a number and stand in the line! 19
- 20. Build Your Own Optimized APIs 20
- 21. https://github.com/ql -io/ql.io Platform Engineering 21
- 22. A domain specific language for HTTP client programming 22
- 23. cut down lines of code reduce no of requests save bandwidth 23
- 24. 24
- 26. ql.io Installation Modes Web Server Node.js 26
- 27. curl http://<host>/q?s=query 27
- 28. As a Standalone Server Data (internal ) ql.io Node Host Data (externa l)
- 29. # As a node.js module npm install ql.io-engine
- 30. // Use ql.io from node.js var Engine = require('ql.io-engine'); var engine = new Engine({ //OPTIONS }); var script = '…'; engine.execute(script, function(emitter){ emitter.on("end", function(err, res){ … }); });
- 31. Language: Defining a Data Source Include in Request Include in .ql file
- 33. Example: jQuery for Cross Domain Reqs
- 34. Let’s Do It!
- 35. 1. Interop via HTTP 2. SQL inspired 3. Implicit orchestration 4. Failure modes 35
- 36. code https://github.com/ql-io/ql.io docs/demos http://ql.io https://github.com/jcleblanc/ api-masher blog http://ql-io.github.com 36
Editor's Notes
- #29: If you download the standalone server version of ql.io, it will sit on top of your existing node server (or within your node applications) and be able to query your internal data and/or external data sources
- #32: A create table call (to define a data source) may be included in the HTTP request to the ql.io web source or defined in a .ql file, which is added into the /tables directory of your application.
- #33: First we need some mustache power
- #34: Route files are also stored with the .ql extension but are placed under the routes directory
- #35: create table slideshare on select get from "http://www.slideshare.net/rss/user/{user}" resultset "rss.channel";create table github.users on select get from "http://github.com/api/v2/json/user/show/{user}";user = "jcleblanc";slides = select * from slideshare where user="{user}";twitter = select * from github.users where user="{user}";return { "slides": "{slides}", "twitter": "{twitter}"}