AZEN-SGG/Kavita
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Parsing Enhancements (#126)

Browse source 
* More cases for parsing regex

* Implemented the ability to parse "Special" keywords.

* Commented out some unit tests

* More parsing cases

* Fixed unit tests

* Fixed typo in build script

* Fixed a bug where if there was a series with same name, but different capitalization, we wouldn't process it's infos.

* Tons of regex updates to handle more cases.

* More regex tweaking to handle as many cases as possible.

* Bad merge caused the comic parser to break. Fixed with some better regex.
This commit is contained in:
Joseph Milazzo 2021-03-29 15:15:49 -05:00 • committed by GitHub
parent 3e031ab458
commit d9246b7351
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
4 changed files with 118 additions and 39 deletions
Download patch file Download diff file Expand all files Collapse all files

29
API/Services/Tasks/ScannerService.cs
View file

@ -148,13 +148,6 @@ namespace API.Services.Tasks
{
if (parsedSeries == null) throw new ArgumentNullException(nameof(parsedSeries));
// For all parsedSeries, any infos that contain same series name and IsSpecial is true are combined
// foreach (var series in parsedSeries)
// {
// var seriesName = series.Key;
// if (parsedSeries.ContainsKey(seriesName))
// }
// First, remove any series that are not in parsedSeries list
var foundSeries = parsedSeries.Select(s => Parser.Parser.Normalize(s.Key)).ToList();
var missingSeries = library.Series.Where(existingSeries =>
@ -190,7 +183,7 @@ namespace API.Services.Tasks
existingSeries.NormalizedName = Parser.Parser.Normalize(key);
existingSeries.LocalizedName ??= key;
}
// Now, we only have to deal with series that exist on disk. Let's recalculate the volumes for each series
var librarySeries = library.Series.ToList();
Parallel.ForEach(librarySeries, (series) =>
@ -320,7 +313,25 @@ namespace API.Services.Tasks
private void TrackSeries(ParserInfo info)
{
if (info.Series == string.Empty) return;
// Check if normalized info.Series already exists and if so, update info to use that name instead
var normalizedSeries = Parser.Parser.Normalize(info.Series);
var existingName = _scannedSeries.SingleOrDefault(p => Parser.Parser.Normalize(p.Key) == normalizedSeries)
.Key;
if (!string.IsNullOrEmpty(existingName))
{
_logger.LogInformation("Found duplicate parsed infos, merged {Original} into {Merged}", info.Series, existingName);
info.Series = existingName;
}
// TODO: For all parsedSeries, any infos that contain same series name and IsSpecial is true are combined
// foreach (var series in parsedSeries)
// {
// var seriesName = series.Key;
// if (parsedSeries.ContainsKey(seriesName))
// }
_scannedSeries.AddOrUpdate(info.Series, new List<ParserInfo>() {info}, (_, oldValue) =>
{
oldValue ??= new List<ParserInfo>();