influxparser 0.0.1 → 0.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 300f243df83f6c65d0f5206e32623be63fe60d3e
-  data.tar.gz: 51a7aa6aec089ed64d1b721431ab2dbfdbc3ad06
+  metadata.gz: 599f27d2ff8251400b2b593b79e0ca7facad4aff
+  data.tar.gz: 84c2792855209669e2bc8e4a41aec43c393be2e1
 SHA512:
-  metadata.gz: 6edc1719f7f50394732b52257adacb96fd12445dd94d3124bb58a8bf919d7d179aca26ba2c179e5c71709144eceaa01da61eeea3eb5c60d5693ea06043f8d63c
-  data.tar.gz: fe002ca912eb0783ca328786744672eb525096ec503dd509cb00d21ceb777f545bccb2925698b7dc0b4e1ae2cd49ef89ec3bab5b29dbf8a8294d5657cd84e38b
+  metadata.gz: a5a8a3e29905ac2f31915e7be9dc2e7dcb987f5375a0a80f871720b8e859386397d70c1e73a2b58982db45946a04d0146a82660d6c95dd3f646a2460c3be4335
+  data.tar.gz: 1f97c90e05e42b503ef03283f72cb22ba176f00564c18e356cd998a5dd587f78f2990aa732da1e1e985c8f3f054263149d306efaa15851641c8862c93952d365

data/lib/influxparser.rb

@@ -1,56 +1,79 @@
-class InfluxParser
-    class << self
-        def parse_point(s)
-            point = {}
-            point['_raw'] = s
-            s = s.strip # trim whitespace
-            
-            parts = s.split(/(?<!\\) /) # split on unescaped spaces for the initial sections
-            return false if parts.length < 2 # at bare minimum there needs to be a measurement and some fields
-
-            mparts = parts[0].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
-            point['values'] = {}
-            vparts = parts[1].split(/(?<!\\),/)
-            vparts.each do |v|
-                value = v.split(/(?<!\\)=/)
-                point['values'][unescape_tag value[0]] = unescape_point value[1]
-            end
-
-            if parts.length == 3 # handle the timestamp
-                point['time'] = parts[2]
-            else
-                # no time left for you, on my way to better things
-                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)
-            if s[-1,1] == '"'
-
-                # it's a quoted string and should be unescaped
-                s = s[1..-2] # take the wrapping quotes off
-                return s.gsub(/\\\\/,',').gsub(/\\"/,',')
-            else
-                # it's a number and if it's trailing an i we need to strip it because we're not handling precision here
-                return s.chomp('i')
-            end
-        end
-    end
-end
+# TODO: the tags hash should be absent when there are no tags
+# TODO: numbers shouldn't be strings
+# TODO: numbers which aren't strings are floats
+# TODO: numbers with a trailing i are integers -- ALERT this is actually broken
+# 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)
+            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 = ''
+            last_key = ''
+            point['values'] = {}
+            vparts = s[measurement_end+1..-1].split(/(?<!\\),/)
+            vparts.each do |v|
+                value = v.split(/(?<!\\)=/)
+                last_value = unescape_point value[1]
+                last_key = unescape_tag value[0]
+                # puts "kv:#{last_key}==#{last_value}"
+                point['values'][last_key] = last_value
+            end
+            
+            # 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.rindex(/ /)
+            if has_space
+                time_stamp = last_value[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[0..has_space-1]
+                    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)
+            if s[-1,1] == '"'
+
+                # it's a quoted string and should be unescaped
+                s = s[1..-2] # take the wrapping quotes off
+                return s.gsub(/\\\\/,'\\').gsub(/\\"/,',')
+            else
+                # it's a number and if it's trailing an i we need to strip it because we're not handling precision here
+                return s.chomp('i')
+            end
+        end
+    end
+end

data/test/test_parse_point_from_docs.rb

@@ -0,0 +1,250 @@
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+$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',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
+    # TODO: validate the parsed booleans are bools and expected
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=t 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=T 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=true 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=True 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=TRUE 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=f 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=F 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=false 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=False 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+    point = InfluxParser.parse_point("weather,location=us-midwest temperature=FALSE 1465839830100400200")
+    assert_not_equal(false,point)   # a straight up parse error will false
+
+  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'])
+
+    # 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
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: influxparser
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
 platform: ruby
 authors:
 - Robert Labrie
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-01-27 00:00:00.000000000 Z
+date: 2020-02-03 00:00:00.000000000 Z
 dependencies: []
 description: Unofficial influx line protocol parser
 email: robert.labrie@gmail.com
@@ -17,6 +17,7 @@ extensions: []
 extra_rdoc_files: []
 files:
 - lib/influxparser.rb
+- test/test_parse_point_from_docs.rb
 homepage: 
 licenses: []
 metadata: {}
@@ -40,4 +41,5 @@ rubygems_version: 2.5.2.1
 signing_key: 
 specification_version: 3
 summary: InfluxDB line protocol parser
-test_files: []
+test_files:
+- test/test_parse_point_from_docs.rb