public DBTextSummary(TextSummaryAndStructures textAndStruct)
{
_id = ObjectId.GenerateNewId();
id = textAndStruct.summary.id;
meta = textAndStruct.summary.meta;
idType = textAndStruct.summary.idType;
wordsCount = textAndStruct.summary.wordsCount;
uniqueWordsCount = textAndStruct.summary.uniqueWordsCount;
summaryDurationSec = textAndStruct.summary.summaryDurationSec;
createdAt = textAndStruct.summary.createdAt;
lengthsStructure = textAndStruct.lengthsStructure;
frequencyStructure = textAndStruct.frequencyStructure;
mostFrequentWord = textAndStruct.summary.mostFrequentWord;
leastFrequentWord = textAndStruct.summary.leastFrequentWord;
longestWord = textAndStruct.summary.longestWord;
shortestWord = textAndStruct.summary.shortestWord;
}
/// <summary>
/// This function will fetch the content of document if it is a link that was provided.
/// It will also ensure that the text is pre-processed.
///
/// The text will then be submitted to the inventory algorithm which will return a data structure
/// summarizing the result of the inventory.
/// This data structure is augmented iwith meta data and returned to the caller
/// </summary>
/// <param name="textSrc">The Text to be processed or a http link to a document that must be processed</param>
/// <returns>Awaitable object that eventually resolves into a Data Structure that contains the summary of the inventory algorithm</returns>
public async Task <TextSummaryAndStructures> Handle(string textSrc)
{
string sanitizedText = "";
string textID = GetHashString(textSrc); // hash is used as id in the db to cache the result of the processing of the input text
string textMeta = "";
string pattern = @"^(((https?:\/\/)|(www\.))[^\s]+)$";
Match m = Regex.Match(textSrc, pattern, RegexOptions.IgnoreCase);
// Get and sanitize text
if (m.Success)
{
sanitizedText = await textPreprocessor.FetchCorpus(textSrc, true);
textMeta = textSrc;
}
else
{
sanitizedText = textPreprocessor.Sanitize(textSrc);
textMeta = textSrc.Substring(0, textSrc.Length < 50 ? textSrc.Length : 50);
}
// Prepare for inventory algorithm
IDictionary <string, int> freqs;
IDictionary <int, LinkedList <string> > lengths;
IInventoryItem mostFrequentToken;
IInventoryItem longestToken;
IInventoryItem leastFrequentToken;
IInventoryItem shortestToken;
int textLength = 0;
// Perform inventory on text
double durationMs = textInventorier.Process(sanitizedText, out freqs, out lengths, out mostFrequentToken, out longestToken, out leastFrequentToken, out shortestToken, out textLength);
// Serialize data structure for db
string strFreqs;
string strLengths;
textInventorier.Serialize(freqs, lengths, out strFreqs, out strLengths);
// Pack summary object
TextSummary textSummary = new TextSummary(
textID,
textMeta,
textLength,
freqs.Count,
mostFrequentToken,
longestToken,
leastFrequentToken,
shortestToken,
durationMs / 1000,
new List <IQueryResult>()
//results
);
// Pack wrapper for summary object
TextSummaryAndStructures res = new TextSummaryAndStructures();
res.frequencyStructure = strFreqs;
res.lengthsStructure = strLengths;
res.summary = textSummary;
// return
return(res);
}
