My previous post presented the YQL code required to handle a WebDAV GET request for a “file” in yql storage. To update the file, we need an additional table.

Prerequisites:

• A sherpa record w/ this value in it: {“file1″:”content 1″, “file2″, “content 2″}. The keys and values w/in the JSON can be anything. If you haven’t worked w/ YQL storage before, check out the documentation.
• The code below edited to use your storage record’s select address

Flow:

• The WebDAV client issues a PUT/POST request with “path” set to the key of the value to change, and “contents” set to the updated value.
• The YQL table will extract the stored record, decode it, update the value, encode it, and store it back

Code:

[sourcecode lang=“xml”]

// http://www.json.org/json2.js y.include(‘http://{your domain}/json2.js’);

//credit: http://javascript.crockford.com/remedial.html if (typeof String.prototype.supplant !== ‘function’) { String.prototype.supplant = function (o) { return this.replace(/{([^{}]*)}/g, function (a, b) { var r = o[b]; return typeof r === ‘string’ ? r : a; }); }; }

response.object = function () {

//put queries and results in arrays so we can reuse the var w/o overwriting values var queries = [], results = [];

queries[0] = ‘select * from yql.storage where name="{select}"’.supplant({‘select’:select}), results[0] = y.xmlToJson( y.query( queries[0] ).results );

if (results[0].results.result.value[path]) {

results[0].results.result.value[path] = contents;

queries[1] = “update yql.storage set value=’{json}’ where name=’{update}’” .supplant({ ‘json’ : JSON.stringify(results[0].results.result.value), ‘update’ : update }); results[1] = y.xmlToJson( y.query( queries[1] ).results );

return { “headers” : { “HTTP/1.1 status” : “204”, “Date” : new Date().getTime(), “Location” : “/webdav/” + path, “Content-Length” : contents.length, “Connection” : “close”, “Content-Type” : “text/plain; charset=UTF-8” } } }

results[0].results.result.value[path] = contents;

queries[1] = “update yql.storage set value=’{json}’ where name=’{update}’” .supplant({ ‘json’ : JSON.stringify(results[0].results.result.value), ‘update’ : update }); results[1] = y.xmlToJson( y.query( queries[1] ).results );

return { headers : { “HTTP/1.1 status” : “201”, “Date” : new Date().getTime(), “Location” : ‘/webdav/’ + path, “Content-Length” : contents.length, “Connection” : “close”, “Content-Type” : “text/plain; charset=UTF-8” } }; }(); ]]>

Notes:

• We could use some ruby like this in a rack app to intermediate between YQL and the WebDAV client:

[sourcecode lang=“ruby”] request = Rack::Request.new(env)

contents = ‘’; while part = request.body.read(8192) contents += part end

query = “use ‘{the uri for the YQL table file using the code above}/put.xml’ as table;"+ “insert into table (path, contents) values (’#{file}’, ‘#{contents}’)” host = ‘http://query.yahooapis.com’ path = ‘/v1/public/yql’

response = Net::HTTP.post_form( URI.parse( “#{host}#{path}” ), { ‘q’ => query, ‘debug’ => true } ) doc = REXML::Document.new(response.body) headers = REXML::XPath.first( doc, “///headers” )

you could dump out the headers here

p(headers[0].text)

return an empty success response

[204, {}, ‘’] [/sourcecode]