static Word populateWords(string root)
{
if(words.ContainsKey(root) == false)
{
Word rootWord = new Word();
words.Add(root, rootWord);
rootWord.word = root;
string rootDefinition = definitionsForWord(root);
//rootWord.definition = rootDefinition;
List<string> subDefinitionStrings = allSubDefinitionsFromDefinition(rootDefinition);
rootWord.subDefinitions = new List<SubDefinition>();
rootWord.appearsInDefinitionFor = new List<Word>();
foreach(string subDefinitionString in subDefinitionStrings)
{
SubDefinition subDefinition = new SubDefinition();
subDefinition.subDefinition = subDefinitionString;
List<string> nounStrings = allNounsFromDefinition(subDefinitionString);
rootWord.subDefinitions.Add(subDefinition);
subDefinition.nouns = new List<Word>();
foreach(string nounString in nounStrings)
{
Word noun = populateWords(nounString);
noun.appearsInDefinitionFor.Add(rootWord);
subDefinition.nouns.Add(noun);
}
}
return rootWord;
}
else
{
return words[root];
}
}
//goes through the definition and pulls out all the nouns and their parent subdefinitions
//returns in the order of how rare the noun is, i.e. how little it appears in the
//definitions of other words, (excluding the words that have already been used)
static List<NounDefinitionPair> nounDefinitionPairsOrderedByRarity(Word noun, HashSet<string> usedNouns)
{
//we will get better time complexity if we return a unique list of nouns
HashSet<string> uniqueSubNouns = new HashSet<string>();
List<NounDefinitionPair> pairs = new List<NounDefinitionPair>();
foreach (SubDefinition subDefinition in noun.subDefinitions)
{
foreach (Word subNoun in subDefinition.nouns)
{
if (uniqueSubNouns.Contains(subNoun.word))
{
continue;
}
uniqueSubNouns.Add(subNoun.word);
NounDefinitionPair pair = new NounDefinitionPair();
pair.definition = subDefinition;
pair.noun = subNoun;
pairs.Add(pair);
}
}
pairs = pairs.OrderBy(p => p.noun.numAppearsInDefinitionsForExcludingThese(usedNouns)).ToList();
return pairs;
}