source: branches/3073_IA_constraint_splitting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Interval/IntervalConstraintsParser.cs @ 18156

#region License Information

/* HeuristicLab
 * Copyright (C) 2002-2019 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
 *
 * This file is part of HeuristicLab.
 *
 * HeuristicLab is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * HeuristicLab is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
 */

#endregion

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text.RegularExpressions;
using HeuristicLab.Data;

namespace HeuristicLab.Problems.DataAnalysis {
  public static class IntervalConstraintsParser {
    public static IEnumerable<IntervalConstraint> ParseInput(string              inputText, string target,
                                                             IEnumerable<string> variables) {
      if (string.IsNullOrEmpty(inputText)) throw new ArgumentNullException("No input text has been provided.");
      if (string.IsNullOrEmpty(target)) throw new ArgumentNullException("No target variable has been provided.");
      if (variables == null) throw new ArgumentNullException("No variables have been provided.");
      if (!variables.Any()) throw new ArgumentException("Variables are empty.");

      var lines = inputText.Split(new[] {"\r\n", "\r", "\n"}, StringSplitOptions.None);
      foreach (var line in lines) {
        var trimmedLine = line.TrimStart();
        //Check for target-variable constraint
        if (trimmedLine.StartsWith("Target:")) {
          var start            = "Target:".Length;
          var end              = trimmedLine.Length;
          var targetConstraint = trimmedLine.Substring(start, end - start);

          var match =
          Regex.Match(targetConstraint,
                        @"(['](.*)[']|(.*[^\s]))" +
                        @"\s*(\bin\b)" +
                        @"\s*([\[])" +
                        @"\s*(\S*)" + // 6: interval lower bound
                        @"\s*(\.{2})" +
                        @"\s*(\S*)" + // 8: interval upper bound
                        @"\s*([\]])" +
                        @"(" +
                          @"\s*,\s*(\S*)\s*=\s*" + // 11: variable name
                          @"([(])\s*" +
                          @"(\S*)\s*" + // 13: region lower bound
                          @"(\.{2})\s*" +
                          @"(\S*)\s*" + // 15: region upper bound
                          @"([)])" +
                        @")*" +
                        @"\s*(<(\S*)>)?"); // 17, 18

          
          if (match.Success) {
            if (match.Groups.Count < 19) throw new ArgumentException("The target-constraint is not complete.", line);

            var targetVariable = match.Groups[1].Value.Trim();
            if (targetVariable.StartsWith("'") && targetVariable.EndsWith("'"))
              targetVariable = targetVariable.Substring(1, targetVariable.Length - 2);

            if (targetVariable != target)
              throw new
                ArgumentException($"The target variable {targetVariable}  does not match the provided target {target}.",
                                  line);

            var lowerBound         = ParseIntervalBounds(match.Groups[6].Value);
            var upperBound         = ParseIntervalBounds(match.Groups[8].Value);
            var expression         = "Target:" + match.Groups[0].Value;
            var parsedTarget       = match.Groups[1].Value.Trim();
            var variable           = targetVariable;
            var isEnabled          = true;
            var numberOfDerivation = 0;
            var interval           = new Interval(lowerBound, upperBound);
            var weight = 1.0;

            if (match.Groups[18].Success && !string.IsNullOrWhiteSpace(match.Groups[18].Value))
              weight = ParseAndValidateDouble(match.Groups[18].Value);

            if (match.Groups[10].Success)
            {
              IntervalCollection regions = new IntervalCollection();
              // option variables found
              for(int idx = 0; idx < match.Groups[10].Captures.Count; ++idx)
              {
                KeyValuePair<string, Interval> region = ParseRegion(
                  match.Groups[11].Captures[idx].Value,
                  match.Groups[13].Captures[idx].Value,
                  match.Groups[15].Captures[idx].Value);
                if (regions.GetReadonlyDictionary().All(r => r.Key != region.Key))
                  regions.AddInterval(region.Key, region.Value);
                else
                  throw new ArgumentException("A constraint cannot contain multiple regions of the same variable.");
              }
              yield return new IntervalConstraint(expression, variable, parsedTarget, numberOfDerivation, interval, regions, weight, isEnabled);
            }
            else
              yield return new IntervalConstraint(expression, variable, parsedTarget, numberOfDerivation, interval, weight, isEnabled);
          }
          else
            throw new ArgumentException("The inserted target constraint is not valid.", line);

          //Check for derivation
        }
        else if (trimmedLine.StartsWith("d") || trimmedLine.StartsWith("\u2202")) {
          var match = Regex.Match(trimmedLine, 
                                    @"([d∂])" +
                                    @"([²³]?)\s*" +
                                    @"(['](.*)[']|(.*[^\s]))\s*" +
                                    @"(\/)\s*" +
                                    @"([d∂])\s*" +
                                    @"(['](.*)[']|(.*[^\s²³]))\s*" +
                                    @"([²³]?)\s*\bin\b\s*" +
                                    @"([\[])\s*" +
                                    @"(\S*)\s*" +
                                    @"(\.{2})\s*" +
                                    @"(\S*)\s*" +
                                    @"([\]])" +
                                    @"(" +
                                      @"\s*,\s*(\S*)\s*=\s*" + // 18: variable name
                                      @"([(])\s*" +
                                      @"(\S*)\s*" + // 20: region lower bound
                                      @"(\.{2})\s*" +
                                      @"(\S*)\s*" + // 22: region upper bound
                                      @"([)])" +
                                    @")*" +
                                    @"\s*(<(\S*)>)?"); // 24, 25

          if (match.Success) {
            if (match.Groups.Count < 26)
              throw new ArgumentException("The given derivation-constraint is not complete.", line);

            var derivationTarget   = match.Groups[3].Value.Trim();
            var derivationVariable = match.Groups[8].Value.Trim();

            if (match.Groups[3].Value.Trim().StartsWith("'") && match.Groups[3].Value.Trim().EndsWith("'"))
              derivationTarget = derivationTarget.Substring(1, derivationTarget.Length - 2);

            if (match.Groups[8].Value.Trim().StartsWith("'") && match.Groups[8].Value.Trim().EndsWith("'"))
              derivationVariable = derivationVariable.Substring(1, derivationVariable.Length - 2);

            if (derivationTarget != target)
              throw new
                ArgumentException($"The target variable {derivationTarget}  does not match the provided target {target}.",
                                  line);

            if (variables.All(v => v != derivationVariable))
              throw new ArgumentException($"The given variable {derivationVariable} does not exist in the dataset.",
                                          line);

            if (match.Groups[2].Value.Trim() != "" || match.Groups[11].Value.Trim() != "") {
              if (match.Groups[2].Value.Trim() == "" || match.Groups[11].Value.Trim() == "")
                throw new ArgumentException("Number of derivation has to be written on both sides.", line);
              if (match.Groups[2].Value.Trim() != match.Groups[11].Value.Trim())
                throw new ArgumentException("Derivation number is not equal on both sides.", line);
            }

            var lowerBound         = ParseIntervalBounds(match.Groups[13].Value);
            var upperBound         = ParseIntervalBounds(match.Groups[15].Value);
            var expression         = match.Groups[0].Value;
            var parsedTarget       = derivationTarget;
            var isEnabled          = true;
            var variable           = derivationVariable;
            var numberOfDerivation = ParseDerivationCount(match.Groups[2].Value.Trim());
            var interval           = new Interval(lowerBound, upperBound);
            var weight             = 1.0;

            if(match.Groups[25].Success && !string.IsNullOrWhiteSpace(match.Groups[25].Value))
              weight = ParseAndValidateDouble(match.Groups[25].Value);

            if(match.Groups[17].Success)
            {
              IntervalCollection regions = new IntervalCollection();
              // option variables found
              for (int idx = 0; idx < match.Groups[17].Captures.Count; ++idx)
              {
                KeyValuePair<string, Interval> region = ParseRegion(
                  match.Groups[18].Captures[idx].Value,
                  match.Groups[20].Captures[idx].Value,
                  match.Groups[22].Captures[idx].Value);
                if (regions.GetReadonlyDictionary().All(r => r.Key != region.Key))
                  regions.AddInterval(region.Key, region.Value);
                else
                  throw new ArgumentException("A constraint cannot contain multiple regions of the same variable.");
              }
              yield return new IntervalConstraint(expression, variable, parsedTarget, numberOfDerivation, interval, regions, weight, isEnabled);
            } else
              yield return new IntervalConstraint(expression, variable, parsedTarget, numberOfDerivation, interval, weight, isEnabled);
          }
          else
            throw new ArgumentException("The inserted derivation constraint is not valid.", line);

          //Check for comment
        }
        else if (trimmedLine.StartsWith("#") || trimmedLine == "") {
          //If it is a comment just continue without saving anything
        }
        else {
          throw new
            ArgumentException("Error at your constraints definition constraints have to start with (Target: | d | \u2202 | #)",
                              line);
        }
      }
    }

    private static KeyValuePair<string, Interval> ParseRegion(string variable, string lb, string ub)
    {
      var regionLb = ParseIntervalBounds(lb);
      var regionUb = ParseIntervalBounds(ub);
      return new KeyValuePair<string, Interval>(variable, new Interval(regionLb, regionUb));
    }

    private static double ParseIntervalBounds(string input) {
      input = input.ToLower();
      switch (input) {
        case "+inf.":
        case "inf.":
          return double.PositiveInfinity;
        case "-inf.":
          return double.NegativeInfinity;
        default: {
          return ParseAndValidateDouble(input);
        }
      }
    }

    private static double ParseAndValidateDouble(string input) {
      var valid = double.TryParse(input, out var value);
      if (!valid) {
        throw new ArgumentException("Invalid Value (Valid Value Format: \"" + FormatPatterns.GetDoubleFormatPattern() + "\")");
      }

      return value;
    }

    private static int ParseDerivationCount(string input) {
      switch (input) {
        case "":
          return 1;
        case "²":
          return 2;
        case "³":
          return 3;
        default:
          int.TryParse(input, out var value);
          return value;
      }
    }
  }
}