influxparser 0.0.3

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 8a35192cf489f487d8a42343384e1528d5168047
+  data.tar.gz: fa965f4209d17c97bf4bde7508c5cc9fe07cfcd0
+SHA512:
+  metadata.gz: 870efd15c7c9df408b15a2da2dc11134c6c2bd45ff69bdd213b30a3913fb402b0d992626c42bfee7a944dbbfe4624a922baffe6e0c6d5c2cb995170c9a6c74d5
+  data.tar.gz: 194488b6bcb2ed5ce444c603789c6db5ce8b321f2ed15b70567f19b35afd7a0fb5e15cc78be828bf99d30809f0f23ea845a10d5b9da5d0949fe53db4ac43eae6

data/lib/influxparser.rb

@@ -0,0 +1,88 @@
+# TODO: the tags hash should be absent when there are no tags
+# TODO: optional timestamp parsing
+# TODO: time key shouldn't exist if there is no time
+# TODO: deal with improper line protocol
+class InfluxParser
+    class << self
+        def parse_point(s, options = {})
+            default_options = {:parse_types => true}
+            options = default_options.merge(options)
+
+            point = {}
+            point['_raw'] = s
+            s = s.strip # trim whitespace
+            
+            measurement_end = s.index /(?<!\\) /
+
+
+            mparts = s[0..measurement_end-1].split(/(?<!\\),/) # split on unescaped commas for the measurement name and tags
+            point['measurement'] = unescape_measurement mparts[0]
+            
+            # if any tags were attached to the measurement iterate over them now
+            point['tags'] = {}
+            mparts.drop(1).each do |t|
+                tag = t.split(/(?<!\\)=/)
+                point['tags'][unescape_tag tag[0]] = unescape_tag tag[1]
+            end
+            
+            # now iterate over the values
+            last_value_raw = ''
+            last_key = ''
+            point['values'] = {}
+            vparts = s[measurement_end+1..-1].split(/(?<!\\),/)
+            # puts "vparts:#{vparts}"
+            vparts.each do |v|
+                value = v.split(/(?<!\\)=/)
+                last_value_raw = value[1]
+                last_key = unescape_tag value[0]
+                # puts "last k/v #{last_key}==#{last_value_raw}"
+                point['values'][last_key] = unescape_point(value[1],options)
+            end
+            # puts "-----\n#{point['values'].to_yaml}\n"
+            # check for a timestamp in the last value
+            # TODO: I hate this, but it's late and I just want to move past it for now
+            # TODO: what happens if the last character of the last value is an escaped quote?
+            has_space = last_value_raw.rindex(/ /)
+            if has_space
+                time_stamp = last_value_raw[has_space+1..-1] # take everything from the space to the end
+                if time_stamp.index(/"/)
+                    point['time'] = nil
+                else
+                    # it was a timestamp, strip it from the last value and set the timestamp
+                    point['values'][last_key] = unescape_point(last_value_raw[0..has_space-1],options)
+                    point['time'] = time_stamp
+                end    
+            else
+                point['time'] = nil
+            end
+            
+            point
+        end
+        def unescape_measurement(s)
+            s.gsub(/\\ /,' ').gsub(/\\,/,',')
+        end
+        def unescape_tag(s)
+            t = unescape_measurement s
+            t.gsub(/\\=/,'=')
+        end
+        def unescape_point(s,options)
+            # puts "unescape:#{s}"
+            # s = s.gsub(/\\\\/,'\\').gsub(/\\"/,'""') # handle escaped characters if present
+
+            # it is a string, return it
+            return s[1..-2].gsub(/\\\\/,'\\').gsub(/\\"/,'""') if s[0,1] == '"'
+
+            return s.sub(/(?<=\d)i/,'') if (!options[:parse_types]) # the customer doesn't care about types so just return it, but strip the trailing i from an integer because we care
+
+            # handle the booleans
+            return true if ['t','T','true','True','TRUE'].include?(s)
+            return false if ['f','F','false','False','FALSE'].include?(s)
+
+            # by here we have either an unquoted string or some numeric
+            
+            return s.to_f if s =~ /^(\d|\.)+$/ # floats are only digits and dots
+            return s.chomp('i').to_i if s[0..-2] =~ /^(\d)+$/ # trailing i is an integer remove it
+            return s.gsub(/\\\\/,'\\').gsub(/\\"/,'"')
+        end
+    end
+end

data/test/test_parse_point_from_docs.rb

@@ -0,0 +1,281 @@
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), 'lib'))
+require 'test/unit'
+require 'influxparser'
+class TestParsePointFromDocs < Test::Unit::TestCase  # def setup
+  # end
+
+
+  # def teardown
+  # end
+  
+  def test_two_tags
+    point = InfluxParser.parse_point('weather,location=us-midwest,season=summer temperature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    
+    # measurement
+    assert_equal('weather',point['measurement'])
+
+    # tags
+    assert_equal(2,point['tags'].length)
+
+    # check location
+    assert_equal(true,point['tags'].key?('location'))
+    assert_equal('us-midwest',point['tags']['location'])
+
+    # check season
+    assert_equal(true,point['tags'].key?('season'))
+    assert_equal('summer',point['tags']['season'])
+
+    # value
+    assert_equal(true,point['values'].key?('temperature'))
+    assert_equal(82,point['values']['temperature'])
+    
+    # time
+    assert_equal('1465839830100400200',point['time'])
+
+    end
+
+  def test_no_tags
+    point = InfluxParser.parse_point('weather temperature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    # measurement
+    assert_equal('weather',point['measurement'])
+
+    # no tags
+    assert_equal(true,point.key?('tags'))
+    assert_equal(0,point['tags'].length)
+
+    # value
+    assert_equal(true,point['values'].key?('temperature'))
+    assert_equal(82,point['values']['temperature'])
+    
+    # time
+    assert_equal('1465839830100400200',point['time'])
+    
+  end
+
+  def test_two_values
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature=82,humidity=71 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    # measurement
+    assert_equal('weather',point['measurement'])
+
+    # check location
+    assert_equal(true,point['tags'].key?('location'))
+    assert_equal('us-midwest',point['tags']['location'])
+
+    # values
+    assert_equal(2,point['values'].length)
+
+    assert_equal(true,point['values'].key?('temperature'))
+    assert_equal(82,point['values']['temperature'])
+
+    assert_equal(true,point['values'].key?('humidity'))
+    assert_equal(71,point['values']['humidity'])
+
+    # time
+    assert_equal('1465839830100400200',point['time'])
+
+  end
+  def test_timestamp
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature=82')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_nil(point['time'])
+ end
+
+  def test_float
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(82.0,point['values']['temperature'])
+  end
+  def test_integer
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature=82i 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(82,point['values']['temperature'])
+  end
+
+  def test_string
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature="too warm" 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('too warm',point['values']['temperature'])
+  end
+
+  def test_invalid_time
+    # TODO: this should be throwing a parse error
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature=82 "1465839830100400200"')
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+  end
+
+  def test_invalid_string
+    # TODO: this should be throwing a parse error
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature='too warm' 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+  end
+
+  def test_boolean
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=t 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(true,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=T 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(true,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=true 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(true,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=True 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(true,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=TRUE 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(true,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=f 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(false,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=F 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(false,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=false 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(false,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=False 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(false,point['values']['temperature'])
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=FALSE 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(false,point['values']['temperature'])
+
+  end
+  def test_ridiculous_quotes
+    # from the influx docs: Do not double or single quote measurement names, tag keys, tag values, and field keys. It is valid line protocol but InfluxDB assumes that the quotes are part of the name.
+    point = InfluxParser.parse_point('"weather","location"="us-midwest" "temperature"=82 1465839830100400200')
+
+    assert_not_equal(false,point)   # a straight up parse error will false
+    # measurement
+    assert_equal('"weather"',point['measurement'])
+
+    # check tag
+    assert_equal(true,point['tags'].key?('"location"'))
+    assert_equal('"us-midwest"',point['tags']['"location"'])
+
+    # check values
+
+    assert_equal(true,point['values'].key?('"temperature"'))
+    assert_equal(82,point['values']['"temperature"'])
+
+    # time
+    assert_equal('1465839830100400200',point['time'])
+
+
+
+    # Do the same thing for ridiculous single quotes
+    point = InfluxParser.parse_point("'weather','location'='us-midwest' 'temperature'=82 1465839830100400200")
+
+    assert_not_equal(false,point)   # a straight up parse error will false
+    # measurement
+    assert_equal("'weather'",point['measurement'])
+
+    # check tag
+    assert_equal(true,point['tags'].key?("'location'"))
+    assert_equal("'us-midwest'",point['tags']["'location'"])
+
+    # check values
+
+    assert_equal(true,point['values'].key?("'temperature'"))
+    assert_equal(82,point['values']["'temperature'"])
+
+    # time
+    assert_equal('1465839830100400200',point['time'])
+
+  end
+
+  def test_escaping
+    point = InfluxParser.parse_point('weather,location=us\,midwest temperature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('us,midwest',point['tags']['location'])
+
+    point = InfluxParser.parse_point('weather,location=us-midwest temp\=rature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(82,point['values']['temp=rature'])
+
+    point = InfluxParser.parse_point('weather,location\ place=us-midwest temperature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(true,point['tags'].key?('location place'))
+
+    point = InfluxParser.parse_point('wea\,ther,location=us-midwest temperature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('wea,ther',point['measurement'])
+
+    point = InfluxParser.parse_point('wea\ ther,location=us-midwest temperature=82 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('wea ther',point['measurement'])
+
+    point = InfluxParser.parse_point('weather temperature=toohot\"')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('toohot"',point['values']['temperature'])
+
+    
+    # so many slashes -- note they're extra terrible because I need to escape ruby slashes in the test strings
+
+    # forward slash
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature_str="too hot/cold" 1465839830100400201')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('too hot/cold',point['values']['temperature_str'])
+
+    # one slash
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature_str="too hot\cold" 1465839830100400201')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('too hot\cold',point['values']['temperature_str'])
+
+    # two slashes
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature_str="too hot\\\cold" 1465839830100400201')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('too hot\cold',point['values']['temperature_str'])
+
+    # three slashes
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature_str="too hot\\\\\cold" 1465839830100400201')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal("too hot\\\\cold",point['values']['temperature_str'])
+
+    # four slashes
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature_str="too hot\\\\\\\cold" 1465839830100400201')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal("too hot\\\\cold",point['values']['temperature_str'])
+
+    # five slashes
+    point = InfluxParser.parse_point('weather,location=us-midwest temperature_str="too hot\\\\\\\\\cold" 1465839830100400201')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal("too hot\\\\\\cold",point['values']['temperature_str'])
+
+  end
+  def test_types
+    point = InfluxParser.parse_point('weather float=82,integer=82i,impliedstring=helloworld,explicitstring="hello world" 1465839830100400200')
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal(82.0,point['values']['float'])
+    assert_equal(82,point['values']['integer'])
+    assert_equal('helloworld',point['values']['impliedstring'])
+    assert_equal('hello world',point['values']['explicitstring'])
+
+    # do it again but this time without type parsing
+    point = InfluxParser.parse_point('weather float=82,integer=82i,impliedstring=helloworld,explicitstring="hello world" 1465839830100400200',{:parse_types => false})
+    assert_not_equal(false,point)   # a straight up parse error will false
+    assert_equal('82',point['values']['float'])
+    assert_equal('82',point['values']['integer'])
+    assert_equal('helloworld',point['values']['impliedstring'])
+    assert_equal('hello world',point['values']['explicitstring'])
+
+  end
+end

metadata

@@ -0,0 +1,45 @@
+--- !ruby/object:Gem::Specification
+name: influxparser
+version: !ruby/object:Gem::Version
+  version: 0.0.3
+platform: ruby
+authors:
+- Robert Labrie
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2020-02-03 00:00:00.000000000 Z
+dependencies: []
+description: Unofficial influx line protocol parser
+email: robert.labrie@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/influxparser.rb
+- test/test_parse_point_from_docs.rb
+homepage: 
+licenses: []
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.5.2.1
+signing_key: 
+specification_version: 3
+summary: InfluxDB line protocol parser
+test_files:
+- test/test_parse_point_from_docs.rb