private static void EnrichWithMissingFiles(BestMatchInfo bestMatchInfo, string outputDirPath)
{
var a = bestMatchInfo.MatchingFiles.First(); // NOTE: there are at least 2 files in the list (see previous note)
for (var i = 1; i < bestMatchInfo.MatchingFiles.Count; i++)
{
var b = bestMatchInfo.MatchingFiles[i];
var nbMissingNumbers = b.Number - a.Number - 1;
var missingNum = a.Number;
while ((nbMissingNumbers--) > 0)
{
missingNum++;
var missingNumStr = bestMatchInfo.RefNumberSeemsPadded ? missingNum.ToPaddedString(bestMatchInfo.RefNumberStr.Length) : missingNum.ToString();
var missingFileName = bestMatchInfo.Prefix + missingNumStr + bestMatchInfo.Suffix;
var missingFilePath = Path.Combine(outputDirPath, missingFileName);
var missingMatchingFile = new MatchingFile(new FileInfo(missingFilePath), missingNumStr, missingNum);
bestMatchInfo.MatchingFiles.Insert(i++, missingMatchingFile);
}
a = b;
}
}
private List <MatchingFile> BuildSearchSummary()
{
// TODO: Deal with accuracy
MatchingFile file = new MatchingFile()
{
accuracy = 0,
fileInfo = new FileInfo(_searchFile),
fileReader = new StreamReader(_searchFile, Encoding.Default),
searchResults = _matches.GroupBy(m => m.Value).ToDictionary(g => g.Key, g => g.Select(m => m.Index))
};
return(new List <MatchingFile> {
file
});
}
public void RowConstructor_CreatesCorrectMatchingFile()
{
const string fileName = "plik.txt";
const string fullPath = "c:\\plik.txt";
const int sizeInBytes = 12345;
var dataTable = new DataTable();
dataTable.Columns.Add("fileName", typeof(string));
dataTable.Columns.Add("fullPath", typeof(string));
dataTable.Columns.Add("sizeInBytes", typeof(long));
dataTable.Rows.Add(fileName, fullPath, sizeInBytes);
var enumerator = dataTable.Rows.GetEnumerator();
enumerator.MoveNext();
var matchingFile = new MatchingFile((DataRow) enumerator.Current, new LinkedList<Match>());
Assert.AreEqual(fileName, matchingFile.FileName);
Assert.AreEqual(fullPath, matchingFile.FullPath);
Assert.AreEqual(sizeInBytes, matchingFile.SizeInBytes);
}
public void RowConstructor_CreatesCorrectMatchingFile()
{
const string fileName = "plik.txt";
const string fullPath = "c:\\plik.txt";
const int sizeInBytes = 12345;
var dataTable = new DataTable();
dataTable.Columns.Add("fileName", typeof(string));
dataTable.Columns.Add("fullPath", typeof(string));
dataTable.Columns.Add("sizeInBytes", typeof(long));
dataTable.Rows.Add(fileName, fullPath, sizeInBytes);
var enumerator = dataTable.Rows.GetEnumerator();
enumerator.MoveNext();
var matchingFile = new MatchingFile((DataRow)enumerator.Current, new LinkedList <Match>());
Assert.AreEqual(fileName, matchingFile.FileName);
Assert.AreEqual(fullPath, matchingFile.FullPath);
Assert.AreEqual(sizeInBytes, matchingFile.SizeInBytes);
}
private static List <MatchingFile> ListMatchingFiles(string prefix, string refNumber, string suffix, IEnumerable <FileInfo> allFiles, out bool refNumberSeemsPadded)
{
var regex = new Regex($"^{Regex.Escape(prefix)}(\\d+){Regex.Escape(suffix)}$");
var matchingFiles = new Dictionary <long, MatchingFile>();
refNumberSeemsPadded = refNumber.StartsWith("0");
foreach (var file in allFiles)
{
var match = regex.Match(file.Name);
if (!match.Success)
{
continue;
}
var fileNumStr = match.Groups[1].Value;
if (!long.TryParse(fileNumStr, out var fileNum))
{
continue;
}
var matchingFile = new MatchingFile(file, fileNumStr, fileNum);
if (!matchingFiles.ContainsKey(fileNum))
{
matchingFiles.Add(fileNum, matchingFile);
}
else
{
// Here, a file with the same number is already found.
// The strategy is here is to prefer a file with a padded number when reference number is padded
// and a non padded number when reference number is not padded
if (refNumberSeemsPadded)
{
if (matchingFile.NumberStr.Length == refNumber.Length)
{
matchingFiles[fileNum] = matchingFile;
}
}
else
{
if (!matchingFile.NumberStr.StartsWith("0"))
{
matchingFiles[fileNum] = matchingFile;
}
}
}
}
var files = matchingFiles.Values.ToList();
files.Sort((m1, m2) =>
{
if (m1.Number == m2.Number)
{
return(0);
}
if (m1.Number < m2.Number)
{
return(-1);
}
return(1);
});
return(files);
}
