TSA/chapter-organizer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Enhance event occurrence parsing to skip unmatched high school section headers

Browse Source 
This commit introduces a new property to track skipped high school section headers in the EventOccurrenceParseResult and EventOccurrenceParserResult classes. The EventOccurrenceParser has been updated to gracefully skip HS section headers that do not match any event definitions, improving the parsing logic. Additionally, the LocationParsingConfiguration has been removed from the EventOccurrenceParser, simplifying its constructor. Unit tests have been updated to reflect these changes and ensure correct behavior during parsing.
This commit is contained in:
poprhythm
2026-01-09 00:14:19 -05:00
parent f916cfad6b
commit 19e5ef0675
10 changed files with 279 additions and 372 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Core/Parsers/EventOccurrenceParser.cs
+65 -22
View File
@@ -13,19 +13,18 @@ public class EventOccurrenceParserResult
{
public IDictionary<EventDefinition, List<Entities.EventOccurrence>> Occurrences { get; set; } = new Dictionary<EventDefinition, List<Entities.EventOccurrence>>();
public List<ParsingIssue> Issues { get; set; } = new();
public List<string> SkippedHSSectionHeaders { get; set; } = new();
}
public class EventOccurrenceParser
{
private FileSystemInfo _txtFile;
private ICollection<EventDefinition> _events;
private LocationParsingConfiguration? _locationConfig;
public EventOccurrenceParser(FileSystemInfo txtFile, ICollection<EventDefinition> events, LocationParsingConfiguration? locationConfig = null)
public EventOccurrenceParser(FileSystemInfo txtFile, ICollection<EventDefinition> events)
{
_events = events;
_txtFile = txtFile;
_locationConfig = locationConfig;
}
public EventOccurrenceParserResult Parse()
@@ -34,6 +33,8 @@ public class EventOccurrenceParser
var occurrences = result.Occurrences;
var issues = result.Issues;
EventDefinition? currentEventDefinition = null;
bool inContinuationMode = false;
bool inHSSection = false;
var lines = File.ReadLines(_txtFile.FullName);
foreach (var (line, index) in lines.Select((line, index) => (line, index + 1)))
@@ -44,11 +45,19 @@ public class EventOccurrenceParser
// Skip empty lines
if (EventOccurrenceParsers.LineClassifier.IsEmptyLine(normalizedLine))
{
// Empty lines break continuation mode
inContinuationMode = false;
continue;
}
// Skip comment lines (starting with "#") - use grammar parser
if (EventOccurrenceParsers.LineClassifier.IsCommentLine(normalizedLine))
{
// Comment lines break continuation mode
inContinuationMode = false;
continue;
}
// Try to parse occurrence line using grammar parser
var occurrenceLine = EventOccurrenceGrammar.TryParseOccurrenceLine(normalizedLine);
@@ -61,10 +70,23 @@ public class EventOccurrenceParser
{
var (eventNamePart, schoolLevel) = sectionHeader.Value;
// Section headers break continuation mode
inContinuationMode = false;
// Use fuzzy matching to find the best matching event definition
var evt = EventOccurrenceParsers.SectionHeaderMatcher.MatchEventDefinition(eventNamePart, _events);
if (evt == null)
{
// Check if this is an HS event - if so, skip gracefully
if (schoolLevel.Equals("HS", StringComparison.OrdinalIgnoreCase))
{
result.SkippedHSSectionHeaders.Add(normalizedLine);
currentEventDefinition = null; // Skip subsequent occurrences
inHSSection = true; // Mark that we're in an HS section
continue; // No issue created
}
// For non-HS unmatched headers, create issue as before
var bestRatio = EventOccurrenceParsers.SectionHeaderMatcher.GetBestMatchRatio(eventNamePart, _events);
issues.Add(new ParsingIssue
{
@@ -76,12 +98,16 @@ public class EventOccurrenceParser
continue;
}
currentEventDefinition = evt;
inHSSection = false; // Reset HS section flag for MS events
continue;
}
// Check for General Schedule/Session using grammar parser
if (EventOccurrenceParsers.SectionHeaderMatcher.IsGeneralSchedule(normalizedLine))
{
// General schedule breaks continuation mode
inContinuationMode = false;
inHSSection = false; // Reset HS section flag
currentEventDefinition = EventDefinition.GeneralSchedule;
continue;
}
@@ -89,9 +115,22 @@ public class EventOccurrenceParser
// Also check for simple "MS" or "HS" in line (backward compatibility)
if (EventOccurrenceParsers.SectionHeaderMatcher.HasSchoolLevel(normalizedLine))
{
// Section headers break continuation mode
inContinuationMode = false;
var evt = EventOccurrenceParsers.SectionHeaderMatcher.MatchEventDefinition(normalizedLine, _events);
if (evt == null)
{
// Check if this is an HS event - if so, skip gracefully
if (normalizedLine.Contains("HS", StringComparison.OrdinalIgnoreCase))
{
result.SkippedHSSectionHeaders.Add(normalizedLine);
currentEventDefinition = null; // Skip subsequent occurrences
inHSSection = true; // Mark that we're in an HS section
continue; // No issue created
}
// For non-HS unmatched headers, create issue as before
var bestRatio = EventOccurrenceParsers.SectionHeaderMatcher.GetBestMatchRatio(normalizedLine, _events);
issues.Add(new ParsingIssue
{
@@ -103,11 +142,18 @@ public class EventOccurrenceParser
continue;
}
currentEventDefinition = evt;
inHSSection = false; // Reset HS section flag for MS events
continue;
}
// Skip continuation lines (lines that look like they're continuing from previous line)
if (EventOccurrenceParsers.LineClassifier.IsContinuationLine(normalizedLine))
// Check if line starts with "*" to enter continuation mode
if (normalizedLine.TrimStart().StartsWith("*", StringComparison.Ordinal))
{
inContinuationMode = true;
}
// Skip continuation lines (in continuation mode OR line starts with "*" or is parenthetical)
if (inContinuationMode || EventOccurrenceParsers.LineClassifier.IsContinuationLine(normalizedLine))
{
continue;
}
@@ -127,6 +173,15 @@ public class EventOccurrenceParser
continue;
}
// Occurrence lines break continuation mode
inContinuationMode = false;
// Skip occurrences under HS sections (they won't match any event definition)
if (inHSSection)
{
continue;
}
var (occurrenceName, month, dayOfMonthStr, timeAndLocation) = occurrenceLine.Value;
// Remove weekday suffix from occurrence name if present
@@ -151,23 +206,8 @@ public class EventOccurrenceParser
// timeAndLocation is already normalized (hyphens normalized) since normalizedLine was sanitized
// Parse time and location using configurable patterns
EventOccurrenceParsers.TimeLocationParser.Parse(timeAndLocation, _locationConfig, out string time, out string location, out bool locationParseSuccess);
// Track location parsing failure if patterns are configured but none matched
if (!locationParseSuccess && !string.IsNullOrWhiteSpace(location))
{
if (_locationConfig != null && _locationConfig.LocationPatterns.Any())
{
issues.Add(new ParsingIssue
{
LineNumber = index,
LineContent = normalizedLine,
IssueType = ParsingIssueType.LocationParseFailure,
Message = $"Location '{location}' does not match any configured pattern"
});
}
}
// Parse time and location - extract time using regex, then use everything after time as location
EventOccurrenceParsers.TimeLocationParser.Parse(timeAndLocation, out string time, out string location);
// Parse date
DateOnly? startDate = null;
@@ -222,6 +262,9 @@ public class EventOccurrenceParser
if (!occurrences.ContainsKey(eventDefinition))
occurrences.Add(eventDefinition, []);
occurrences[eventDefinition].Add(eventOccurrence);
// Reset HS section flag when we successfully parse an occurrence (means we're in a valid section)
inHSSection = false;
}
return result;