datasift 3.5.2 → 3.6.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b02269afad134b1065df764ff2a4c4400a96d55c
-  data.tar.gz: b7f6cede19b4a62ee7a1eeaa3784fa01bcba3c1a
+  metadata.gz: ced2e15492b4c926ee5329ed799ff6c0edc4ef1f
+  data.tar.gz: 97a2ee4cee298c822517913341db932bcae1aec2
 SHA512:
-  metadata.gz: 0ecb7edc5b1e12329c021b156b245cc1eaba269116618f5daad8c5486c854ee368e28325623c84089da199a9ed497fa69ba91718711946d7285f5b1b0f855366
-  data.tar.gz: 1aaaf7b124c91ab0be40d32c21a16feab784291c50b7feda1dcba3d0e5649ee03ac3f1aa233c836acd59a6c1ebd503f449e77030d60dc4c7d9fd1a0f1220d840
+  metadata.gz: 0d16d5dd9ed6c4a0a4a0b9e55734859fe7ea17c15be46cd953ce3a81728337e5f31ec5980fb440b70996f4f5648476dea2fb050e8ae35143f61c6e4a62014e0b
+  data.tar.gz: a3617a40236f7dec6fe1d7412f2f95af5075d313c4278c3b57b26e6ea516b8d76039c3e67709b817733528a9c98e636d18342dd5d1a384355a89368730f8e23f

data/CHANGELOG.md

@@ -1,14 +1,16 @@
 CHANGELOG
 ================================
-## v.3.5.2 (2016-11-08)
-### Fixes
-* Uses correct timestamp params in PYLON Sample API calls. `start_time` -> `start` and `end_time` -> `end`
-
-## v.3.5.1 (2016-10-06)
+## v.4.0 (Under development)
+### Planned Features
+* Better thought out API design in a more traditional Gem style
+* More of an SDK style library; for example, you should perform actions (start/stop/analyze) on a PYLON recording object
+* Designed to make the most of DataSift's latest API version and features
+* Designed for Ruby 2.3+. Use features like keyword parameters across the board
+
+## v.3.6.0 (2016-02-19)
 ### Added
-* Explicit support for 500 and 502 responses from the API
-### Fixes
-* Resolves [#86](https://github.com/datasift/datasift-ruby/issues/86); not handling 502 responses well
+* Support for the [/pylon/update](dev.datasift.com/pylon/docs/api/pylon-api-endpoints/pylonupdate) API endpoint
+* Support for [API v1.3](http://dev.datasift.com/docs/api/api-changelog)
 
 ## v.3.5.0 (2015-11-13)
 ### Added

data/VERSION

@@ -1 +1 @@
-3.5.2
+3.6.0

data/examples/auth.rb

@@ -5,7 +5,7 @@ class DataSiftExample
     @config = {
       username: 'DATASIFT_USERNAME',
       api_key: 'DATASIFT_API_KEY',
-      api_version: 'v1.2'
+      api_version: 'v1.3'
     }
     @params = {
       output_type: 's3',

data/examples/{pylon_eg.rb → pylon/pylon_api_v1.2_eg.rb}

@@ -1,4 +1,8 @@
-require './auth'
+##
+# This script runs through all PYLON API endpoints using v1.2 of the API
+##
+
+require './../auth'
 class AnalysisApi < DataSiftExample
   def initialize
     super
@@ -20,15 +24,18 @@ class AnalysisApi < DataSiftExample
       token = @datasift.account_identity_token.create(
         identity_id,
         'facebook',
-        'YOUR_TOKEN'
+        '125595667777713|5aef9cfdb31d8be64b87204c3bca820f'
       )
       puts token[:data].to_json
 
       puts "\nNow make PYLON API calls using the Identity's API key"
-      @config.merge!(api_key: identity[:data][:api_key])
+      @config.merge!(
+        api_key: identity[:data][:api_key],
+        api_version: 'v1.2'
+      )
       @datasift = DataSift::Client.new(@config)
 
-      csdl = 'return { fb.content contains "data" }'
+      csdl = "return { fb.content any \"data, #{Time.now}\" }"
 
       puts "Check this CSDL is valid: #{csdl}"
       puts "Valid? #{@datasift.pylon.valid?(csdl)}"
@@ -75,7 +82,7 @@ class AnalysisApi < DataSiftExample
           target: 'fb.author.age'
         }
       }
-      filter = 'fb.content contains "starbucks"'
+      filter = ''
       puts @datasift.pylon.analyze(
         hash,
         params,

data/examples/pylon/pylon_api_v1.3_eg.rb

@@ -0,0 +1,201 @@
+##
+# This script runs through all PYLON API endpoints using v1.3 of the API
+##
+
+require './../auth'
+class AnalysisApi < DataSiftExample
+  def initialize
+    super
+    run_analysis
+  end
+
+  def run_analysis
+    begin
+      puts "Create a new identity to make PYLON API calls"
+      identity = @datasift.account_identity.create(
+        "RUBY_LIB_#{Time.now.to_i}",
+        "active",
+        false
+      )
+      identity_id = identity[:data][:id]
+      puts identity[:data].to_json
+
+      puts "\nCreate a Token for our Identity"
+      token = @datasift.account_identity_token.create(
+        identity_id,
+        'facebook',
+        '125595667777713|5aef9cfdb31d8be64b87204c3bca820f'
+      )
+      puts token[:data].to_json
+
+      puts "\nNow make PYLON API calls using the Identity's API key"
+      @pylon_config = @config.dup
+      @pylon_config.merge!(
+        api_key: identity[:data][:api_key],
+        api_version: 'v1.3'
+      )
+      @datasift = DataSift::Client.new(@pylon_config)
+
+      csdl = "return { fb.all.content any \"data, #{Time.now}\" }"
+
+      puts "Check this CSDL is valid: #{csdl}"
+      puts "Valid? #{@datasift.pylon.valid?(csdl)}"
+
+      puts "\nCompile my CSDL"
+      compiled = @datasift.pylon.compile csdl
+      hash = compiled[:data][:hash]
+      puts "Hash: #{hash}"
+
+      puts "\nStart recording with hash #{hash}"
+      recording = @datasift.pylon.start(
+        hash,
+        'Facebook Pylon Test Recording'
+      )
+      puts recording[:data].to_json
+
+      puts "\nSleep for 10 seconds to record a little data"
+      sleep(10)
+
+      puts "\nGet details of our running recording by ID"
+      puts @datasift.pylon.get('', recording[:data][:id])[:data].to_json
+
+      puts "\nYou can also list running recordings"
+      puts @datasift.pylon.list[:data].to_json
+
+      puts "\nFrequency distribution analysis on fb.author.country"
+      params = {
+        analysis_type: 'freqDist',
+        parameters: {
+          threshold: 3,
+          target: 'fb.author.country'
+        }
+      }
+      puts @datasift.pylon.analyze(
+        '',
+        params,
+        '',
+        nil,
+        nil,
+        recording[:data][:id]
+      )[:data].to_json
+
+      puts "\nFrequency distribution analysis on fb.author.age with filter"
+      params = {
+        analysis_type: 'freqDist',
+        parameters: {
+          threshold: 1,
+          target: 'fb.author.age'
+        }
+      }
+      filter = 'fb.parent.content any "facebook"'
+      puts @datasift.pylon.analyze(
+        '',
+        params,
+        filter,
+        nil,
+        nil,
+        recording[:data][:id]
+      )[:data].to_json
+
+      puts "\nTime series analysis"
+      params = {
+        analysis_type: 'timeSeries',
+        parameters: {
+          interval: 'hour',
+          span: 12
+        }
+      }
+      filter = ''
+      start_time = Time.now.to_i - (60 * 60 * 24 * 7) # 7 days ago
+      end_time = Time.now.to_i
+      puts @datasift.pylon.analyze(
+        '',
+        params,
+        filter,
+        start_time,
+        end_time,
+        recording[:data][:id]
+      )[:data].to_json
+
+      puts "\nFrequency Distribution with nested queries. Find the top three " \
+        "age groups for each gender by country"
+      filter = ''
+      params = {
+        analysis_type: 'freqDist',
+        parameters: {
+          threshold: 4,
+          target: 'fb.author.country'
+        },
+        child: {
+          analysis_type: 'freqDist',
+          parameters: {
+            threshold: 2,
+            target: 'fb.author.gender'
+          },
+          child: {
+            analysis_type: 'freqDist',
+            parameters: {
+              threshold: 3,
+              target: 'fb.author.age'
+            }
+          }
+        }
+      }
+      start_time = Time.now.to_i - (60 * 60 * 24 * 7)
+      end_time = Time.now.to_i
+      puts @datasift.pylon.analyze(
+        '',
+        params,
+        filter,
+        start_time,
+        end_time,
+        recording[:data][:id]
+      )[:data].to_json
+
+      puts "\nTags analysis"
+      puts @datasift.pylon.tags('',recording[:data][:id])[:data].to_json
+
+      puts "\nGet Public Posts"
+      puts @datasift.pylon.sample(
+        '',
+        10,
+        Time.now.to_i - (60 * 60), # from 1hr ago
+        Time.now.to_i, # to 'now'
+        'fb.content contains_any "your, filter, terms"',
+        recording[:data][:id]
+      )[:data].to_json
+
+      puts "\nv1.3+ of the API allows you to update the name or hash of recordings;"
+      puts "\nBefore update:"
+      puts @datasift.pylon.get(recording[:data][:id])[:data].to_json
+
+      new_hash = @datasift.pylon.compile("fb.content any \"data, #{Time.now}\"")[:data][:hash]
+
+      puts "\nAfter update:"
+      puts @datasift.pylon.update(
+        recording[:data][:id],
+        new_hash,
+        "Updated at #{Time.now}"
+      )[:data].to_json
+
+      puts "\nStop recording filter with the recording ID #{recording[:data][:id]}"
+      puts @datasift.pylon.stop('', recording[:data][:id])[:data].to_json
+      sleep(3)
+      puts "\nYou can also restart a stopped recording by recording ID #{recording[:data][:id]}"
+      puts @datasift.pylon.restart(recording[:data][:id])[:data].to_json
+
+      # Cleanup.
+      # Stop the recording again to clean up
+      sleep(3)
+      @datasift.pylon.stop('', recording[:data][:id])[:data].to_json
+      # Disable the identity created for this example
+      @datasift = DataSift::Client.new(@config)
+      @datasift.account_identity.update(identity_id, '', 'disabled')
+
+      rescue DataSiftError => dse
+        puts dse.inspect
+    end
+  end
+end
+
+AnalysisApi.new

data/lib/datasift.rb

@@ -230,14 +230,8 @@ module DataSift
       begin
         code = e.http_code
         body = e.http_body
-        error = nil
         if code && body
-          begin
-            error = MultiJson.load(body)
-          rescue MultiJson::ParseError
-            # In cases where we receive 502 responses, Nginx may send HTML rather than JSON
-            error = body
-          end
+          error = MultiJson.load(body)
           response_on_error = {
             :data => nil,
             :datasift => {
@@ -309,10 +303,6 @@ module DataSift
       raise UnprocessableEntityError.new(code, body, response)
     when 429
       raise TooManyRequestsError.new(code, body, response)
-    when 500
-      raise InternalServerError.new(code, body, response)
-    when 502
-      raise BadGatewayError.new(code, body, response)
     when 503
       raise ServiceUnavailableError.new(code, body, response)
     when 504

data/lib/errors.rb

@@ -68,14 +68,6 @@ end
 class TooManyRequestsError < DataSiftError
 end
 
-# Standard error returned when receiving a 500 response from the API
-class InternalServerError < DataSiftError
-end
-
-# Standard error returned when receiving a 502 response from the API
-class BadGatewayError < DataSiftError
-end
-
 # Standard error returned when receiving a 503 response from the API
 class ServiceUnavailableError < DataSiftError
 end

data/lib/pylon.rb

@@ -30,10 +30,13 @@ module DataSift
     # Perform /pylon/get API call to query status of your PYLON recordings
     #
     # @param hash [String] Hash you with the get the status for
+    # @param id [String] The ID of the PYLON recording to get
     # @return [Object] API reponse object
-    def get(hash)
-      fail BadParametersError, 'hash is required' if hash.empty?
-      params = { hash: hash }
+    def get(hash = '', id = '')
+      fail BadParametersError, 'hash or id is required' if hash.empty? && id.empty?
+      params = {}
+      params.merge!(hash: hash) unless hash.empty?
+      params.merge!(id: id) unless id.empty?
 
       DataSift.request(:GET, 'pylon/get', @config, params)
     end
@@ -56,15 +59,46 @@ module DataSift
       DataSift.request(:GET, 'pylon/get', @config, params)
     end
 
+    # Perform /pylon/update API call to update a given PYLON Recording
+    #
+    # @param id [String] The ID of the PYLON recording to update
+    # @param hash [String] The CSDL filter hash this recording should be subscribed to
+    # @param name [String] Update the name of your recording
+    # @return [Object] API reponse object
+    def update(id, hash = '', name = '')
+      params = {id: id}
+      params.merge!(hash: hash) unless hash.empty?
+      params.merge!(name: name) unless name.empty?
+
+      DataSift.request(:PUT, 'pylon/update', @config, params)
+    end
+
     # Start recording a PYLON filter by making an /pylon/start API call
     #
     # @param hash [String] CSDL you wish to begin (or resume) recording
     # @param name [String] Give your recording a name. Required when starting a
+    # @param id [String] ID of the recording you wish to start
+    #   new recording
+    # @return [Object] API reponse object
+    def start(hash = '', name = '', id = '')
+      fail BadParametersError, 'hash or id is required' if hash.empty? && id.empty?
+      params = {}
+      params.merge!(hash: hash) unless hash.empty?
+      params.merge!(name: name) unless name.empty?
+      params.merge!(id: id) unless id.empty?
+
+      DataSift.request(:PUT, 'pylon/start', @config, params)
+    end
+
+    # Restart an existing PYLON recording by making an /pylon/start API call with a recording ID
+    #
+    # @param id [String] CSDL you wish to begin (or resume) recording
+    # @param name [String] Give your recording a name. Required when starting a
     #   new recording
     # @return [Object] API reponse object
-    def start(hash = '', name = '')
-      fail BadParametersError, 'hash is required' if hash.empty?
-      params = { hash: hash }
+    def restart(id, name = '')
+      fail BadParametersError, 'id is required' if id.empty?
+      params = { id: id }
       params.merge!(name: name) unless name.empty?
 
       DataSift.request(:PUT, 'pylon/start', @config, params)
@@ -73,10 +107,13 @@ module DataSift
     # Stop an active PYLON recording by making an /pylon/stop API call
     #
     # @param hash [String] CSDL you wish to stop recording
+    # @param id [String] ID of the recording you wish to stop
     # @return [Object] API reponse object
-    def stop(hash)
-      fail BadParametersError, 'hash is required' if hash.empty?
-      params = { hash: hash }
+    def stop(hash = '', id = '')
+      fail BadParametersError, 'hash or id is required' if hash.empty? && id.empty?
+      params = {}
+      params.merge!(hash: hash) unless hash.empty?
+      params.merge!(id: id) unless id.empty?
 
       DataSift.request(:PUT, 'pylon/stop', @config, params)
     end
@@ -92,14 +129,14 @@ module DataSift
     # @param filter [String] Optional PYLON CSDL for a query filter
     # @param start_time [Integer] Optional start timestamp for filtering by date
     # @param end_time [Integer] Optional end timestamp for filtering by date
+    # @param id [String] ID of the recording you wish to analyze
     # @return [Object] API reponse object
-    def analyze(hash = '', parameters = '', filter = '', start_time = nil, end_time = nil)
-      fail BadParametersError, 'hash is required' if hash.empty?
+    def analyze(hash = '', parameters = '', filter = '', start_time = nil, end_time = nil, id = '')
+      fail BadParametersError, 'hash or id is required' if hash.empty? && id.empty?
       fail BadParametersError, 'parameters is required' if parameters.empty?
-      params = {
-        hash: hash,
-        parameters: parameters
-      }
+      params = { parameters: parameters }
+      params.merge!(hash: hash) unless hash.empty?
+      params.merge!(id: id) unless id.empty?
       params.merge!(filter: filter) unless filter.empty?
       params.merge!(start: start_time) unless start_time.nil?
       params.merge!(end: end_time) unless end_time.nil?
@@ -111,10 +148,13 @@ module DataSift
     #   recording
     #
     # @param hash [String] Hash of the recording you wish to query
+    # @param id [String] ID of the recording you wish to query
     # @return [Object] API reponse object
-    def tags(hash)
-      fail BadParametersError, 'hash is required' if hash.empty?
-      params = { hash: hash }
+    def tags(hash = '', id = '')
+      fail BadParametersError, 'hash or id is required' if hash.empty? && id.empty?
+      params = {}
+      params.merge!(hash: hash) unless hash.empty?
+      params.merge!(id: id) unless id.empty?
 
       DataSift.request(:GET, 'pylon/tags', @config, params)
     end
@@ -126,13 +166,16 @@ module DataSift
     # @param start_time [Integer] Optional start timestamp for filtering by date
     # @param end_time [Integer] Optional end timestamp for filtering by date
     # @param filter [String] Optional PYLON CSDL for a query filter
+    # @param id [String] ID of the recording you wish to sample
     # @return [Object] API reponse object
-    def sample(hash = '', count = nil, start_time = nil, end_time = nil, filter = '')
-      fail BadParametersError, 'hash is required' if hash.empty?
-      params = { hash: hash }
+    def sample(hash = '', count = nil, start_time = nil, end_time = nil, filter = '', id = '')
+      fail BadParametersError, 'hash or id is required' if hash.empty? && id.empty?
+      params = {}
+      params.merge!(hash: hash) unless hash.empty?
+      params.merge!(id: id) unless id.empty?
       params.merge!(count: count) unless count.nil?
-      params.merge!(start: start_time) unless start_time.nil?
-      params.merge!(end: end_time) unless end_time.nil?
+      params.merge!(start_time: start_time) unless start_time.nil?
+      params.merge!(end_time: end_time) unless end_time.nil?
 
       if filter.empty?
         DataSift.request(:GET, 'pylon/sample', @config, params)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: datasift
 version: !ruby/object:Gem::Version
-  version: 3.5.2
+  version: 3.6.0
 platform: ruby
 authors:
 - DataSift
@@ -10,7 +10,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-11-08 00:00:00.000000000 Z
+date: 2016-03-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rest-client
@@ -110,7 +110,8 @@ files:
 - examples/managed_source_ig_eg.rb
 - examples/pull.rb
 - examples/push_eg.rb
-- examples/pylon_eg.rb
+- examples/pylon/pylon_api_v1.2_eg.rb
+- examples/pylon/pylon_api_v1.3_eg.rb
 - lib/account.rb
 - lib/account_identity.rb
 - lib/account_identity_limit.rb
@@ -202,8 +203,9 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: 1.3.5
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.6.3
+rubygems_version: 2.4.8
 signing_key: 
 specification_version: 4
 summary: DataSift is a simple wrapper for the DataSift API.
 test_files: []
+has_rdoc: