Start implementing the parser

src/text_to_sql_search.cr

@@ -6,13 +6,23 @@ module TextToSqlSearch
 		@operators_allowed = ["<", ">", "=", ":"]
 		@negation_words    = ["!", "no", "not"]
 		@ignored_words     = ["with", "than"]
+		@split_regex       = /"(.*?)"|\s+|([()><:=+!-])/
+		@strip             = true
+		@default_joiner    = "AND"
+		@default_negative  = false
+		@first_element     = "1"
 
 		def initialize; end
 
 		YAML.mapping(
-			operators_allowed:   {type: Array(String)},
-			negation_words:      {type: Array(String)},
-			ignored_words:       {type: Array(String)},
+			operators_allowed: {type: Array(String)},
+			negation_words:    {type: Array(String)},
+			ignored_words:     {type: Array(String)},
+			split_regex:       {type: Regex},
+			strip:             {type: Bool},
+			default_joiner:    {type: String},
+			default_negative:  {type: Bool},
+			first_element:     {type: String},
 		)
 	end
 
@@ -33,5 +43,32 @@ module TextToSqlSearch
 			end
 		end
 
+	def parse( input)
+		# These happen one time to prepare the input line for parsing
+		input= input.strip if config.strip
+		tokens= input.split( config.split_regex).
+		tokens.reject! {|t| t.blank? }
+
+		# These initialize one-time and accumulate values as parsing advances
+		terms= config.first_element # Accumulated SQL WHERE terms
+		values= [] of String        # Their corresponding values (replacements for "?"s)
+		count= tokens.size          # Total number of tokens we have for parsing
+		i= 0                        # Current position in tokens array
+
+		# These are re-set to these same defaults after each key=value is fully processed and a new one begins
+		joiner= config.default_joiner      # Default joiner to put in SQL between (key=value) pairs
+		negative= config.default_negative  # Are we in positive match or negative by default?
+		content_before= ""                 # Small accumulator for pass-thru characters like opening parentheses
+
+		while i< count
+			token= tokens[i].strip if config.strip
+
+			if config.ignored_words.includes? token
+				i+= 1
+				next
+			end
+
+		end
+	end
 	end
 end