Initial commit (history squashed)
Some checks failed
Build Test Deploy / authority-container (push) Has been cancelled
Build Test Deploy / docs (push) Has been cancelled
Build Test Deploy / deploy (push) Has been cancelled
Build Test Deploy / build-test (push) Has been cancelled
Docs CI / lint-and-preview (push) Has been cancelled

This commit is contained in:
2025-10-07 10:14:21 +03:00
commit b97fc7685a
1132 changed files with 117842 additions and 0 deletions

View File

@@ -0,0 +1,3 @@
using System.Runtime.CompilerServices;
[assembly: InternalsVisibleTo("StellaOps.Feedser.Source.Distro.Debian.Tests")]

View File

@@ -0,0 +1,87 @@
using System;
namespace StellaOps.Feedser.Source.Distro.Debian.Configuration;
public sealed class DebianOptions
{
public const string HttpClientName = "feedser.debian";
/// <summary>
/// Raw advisory list published by the Debian security tracker team.
/// Defaults to the Salsa Git raw endpoint to avoid HTML scraping.
/// </summary>
public Uri ListEndpoint { get; set; } = new("https://salsa.debian.org/security-tracker-team/security-tracker/-/raw/master/data/DSA/list");
/// <summary>
/// Base URI for advisory detail pages. Connector appends {AdvisoryId}.
/// </summary>
public Uri DetailBaseUri { get; set; } = new("https://security-tracker.debian.org/tracker/");
/// <summary>
/// Maximum advisories fetched per run to cap backfill effort.
/// </summary>
public int MaxAdvisoriesPerFetch { get; set; } = 40;
/// <summary>
/// Initial history window pulled on first run.
/// </summary>
public TimeSpan InitialBackfill { get; set; } = TimeSpan.FromDays(30);
/// <summary>
/// Resume overlap to accommodate late edits of existing advisories.
/// </summary>
public TimeSpan ResumeOverlap { get; set; } = TimeSpan.FromDays(2);
/// <summary>
/// Request timeout used for list/detail fetches unless overridden via HTTP client.
/// </summary>
public TimeSpan FetchTimeout { get; set; } = TimeSpan.FromSeconds(45);
/// <summary>
/// Optional pacing delay between detail fetches.
/// </summary>
public TimeSpan RequestDelay { get; set; } = TimeSpan.Zero;
/// <summary>
/// Custom user-agent for Debian tracker courtesy.
/// </summary>
public string UserAgent { get; set; } = "StellaOps.Feedser.Debian/0.1 (+https://stella-ops.org)";
public void Validate()
{
if (ListEndpoint is null || !ListEndpoint.IsAbsoluteUri)
{
throw new InvalidOperationException("Debian list endpoint must be an absolute URI.");
}
if (DetailBaseUri is null || !DetailBaseUri.IsAbsoluteUri)
{
throw new InvalidOperationException("Debian detail base URI must be an absolute URI.");
}
if (MaxAdvisoriesPerFetch <= 0 || MaxAdvisoriesPerFetch > 200)
{
throw new InvalidOperationException("MaxAdvisoriesPerFetch must be between 1 and 200.");
}
if (InitialBackfill < TimeSpan.Zero || InitialBackfill > TimeSpan.FromDays(365))
{
throw new InvalidOperationException("InitialBackfill must be between 0 and 365 days.");
}
if (ResumeOverlap < TimeSpan.Zero || ResumeOverlap > TimeSpan.FromDays(14))
{
throw new InvalidOperationException("ResumeOverlap must be between 0 and 14 days.");
}
if (FetchTimeout <= TimeSpan.Zero || FetchTimeout > TimeSpan.FromMinutes(5))
{
throw new InvalidOperationException("FetchTimeout must be positive and less than five minutes.");
}
if (RequestDelay < TimeSpan.Zero || RequestDelay > TimeSpan.FromSeconds(10))
{
throw new InvalidOperationException("RequestDelay must be between 0 and 10 seconds.");
}
}
}

View File

@@ -0,0 +1,637 @@
using System;
using System.Collections.Generic;
using System.Globalization;
using System.Linq;
using System.Net;
using System.Threading;
using System.Threading.Tasks;
using Microsoft.Extensions.Logging;
using Microsoft.Extensions.Options;
using MongoDB.Bson;
using MongoDB.Bson.IO;
using StellaOps.Feedser.Models;
using StellaOps.Feedser.Source.Common;
using StellaOps.Feedser.Source.Common.Fetch;
using StellaOps.Feedser.Source.Distro.Debian.Configuration;
using StellaOps.Feedser.Source.Distro.Debian.Internal;
using StellaOps.Feedser.Storage.Mongo;
using StellaOps.Feedser.Storage.Mongo.Advisories;
using StellaOps.Feedser.Storage.Mongo.Documents;
using StellaOps.Feedser.Storage.Mongo.Dtos;
using StellaOps.Plugin;
namespace StellaOps.Feedser.Source.Distro.Debian;
public sealed class DebianConnector : IFeedConnector
{
private const string SchemaVersion = "debian.v1";
private readonly SourceFetchService _fetchService;
private readonly RawDocumentStorage _rawDocumentStorage;
private readonly IDocumentStore _documentStore;
private readonly IDtoStore _dtoStore;
private readonly IAdvisoryStore _advisoryStore;
private readonly ISourceStateRepository _stateRepository;
private readonly DebianOptions _options;
private readonly TimeProvider _timeProvider;
private readonly ILogger<DebianConnector> _logger;
private static readonly Action<ILogger, string, int, Exception?> LogMapped =
LoggerMessage.Define<string, int>(
LogLevel.Information,
new EventId(1, "DebianMapped"),
"Debian advisory {AdvisoryId} mapped with {AffectedCount} packages");
public DebianConnector(
SourceFetchService fetchService,
RawDocumentStorage rawDocumentStorage,
IDocumentStore documentStore,
IDtoStore dtoStore,
IAdvisoryStore advisoryStore,
ISourceStateRepository stateRepository,
IOptions<DebianOptions> options,
TimeProvider? timeProvider,
ILogger<DebianConnector> logger)
{
_fetchService = fetchService ?? throw new ArgumentNullException(nameof(fetchService));
_rawDocumentStorage = rawDocumentStorage ?? throw new ArgumentNullException(nameof(rawDocumentStorage));
_documentStore = documentStore ?? throw new ArgumentNullException(nameof(documentStore));
_dtoStore = dtoStore ?? throw new ArgumentNullException(nameof(dtoStore));
_advisoryStore = advisoryStore ?? throw new ArgumentNullException(nameof(advisoryStore));
_stateRepository = stateRepository ?? throw new ArgumentNullException(nameof(stateRepository));
_options = (options ?? throw new ArgumentNullException(nameof(options))).Value ?? throw new ArgumentNullException(nameof(options));
_options.Validate();
_timeProvider = timeProvider ?? TimeProvider.System;
_logger = logger ?? throw new ArgumentNullException(nameof(logger));
}
public string SourceName => DebianConnectorPlugin.SourceName;
public async Task FetchAsync(IServiceProvider services, CancellationToken cancellationToken)
{
ArgumentNullException.ThrowIfNull(services);
var cursor = await GetCursorAsync(cancellationToken).ConfigureAwait(false);
var now = _timeProvider.GetUtcNow();
var pendingDocuments = new HashSet<Guid>(cursor.PendingDocuments);
var pendingMappings = new HashSet<Guid>(cursor.PendingMappings);
var fetchCache = new Dictionary<string, DebianFetchCacheEntry>(cursor.FetchCache, StringComparer.OrdinalIgnoreCase);
var touchedResources = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
var listUri = _options.ListEndpoint;
var listKey = listUri.ToString();
touchedResources.Add(listKey);
var existingList = await _documentStore.FindBySourceAndUriAsync(SourceName, listKey, cancellationToken).ConfigureAwait(false);
cursor.TryGetCache(listKey, out var cachedListEntry);
var listRequest = new SourceFetchRequest(DebianOptions.HttpClientName, SourceName, listUri)
{
Metadata = new Dictionary<string, string>(StringComparer.Ordinal)
{
["type"] = "index"
},
AcceptHeaders = new[] { "text/plain", "text/plain; charset=utf-8" },
TimeoutOverride = _options.FetchTimeout,
ETag = existingList?.Etag ?? cachedListEntry?.ETag,
LastModified = existingList?.LastModified ?? cachedListEntry?.LastModified,
};
SourceFetchResult listResult;
try
{
listResult = await _fetchService.FetchAsync(listRequest, cancellationToken).ConfigureAwait(false);
}
catch (Exception ex)
{
_logger.LogError(ex, "Debian list fetch failed");
await _stateRepository.MarkFailureAsync(SourceName, now, TimeSpan.FromMinutes(5), ex.Message, cancellationToken).ConfigureAwait(false);
throw;
}
var lastPublished = cursor.LastPublished ?? (now - _options.InitialBackfill);
var processedIds = new HashSet<string>(cursor.ProcessedAdvisoryIds, StringComparer.OrdinalIgnoreCase);
var newProcessedIds = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
var maxPublished = cursor.LastPublished ?? DateTimeOffset.MinValue;
var processedUpdated = false;
if (listResult.IsNotModified)
{
if (existingList is not null)
{
fetchCache[listKey] = DebianFetchCacheEntry.FromDocument(existingList);
}
}
else if (listResult.IsSuccess && listResult.Document is not null)
{
fetchCache[listKey] = DebianFetchCacheEntry.FromDocument(listResult.Document);
if (!listResult.Document.GridFsId.HasValue)
{
_logger.LogWarning("Debian list document {DocumentId} missing GridFS payload", listResult.Document.Id);
}
else
{
byte[] bytes;
try
{
bytes = await _rawDocumentStorage.DownloadAsync(listResult.Document.GridFsId.Value, cancellationToken).ConfigureAwait(false);
}
catch (Exception ex)
{
_logger.LogError(ex, "Failed to download Debian list document {DocumentId}", listResult.Document.Id);
throw;
}
var text = System.Text.Encoding.UTF8.GetString(bytes);
var entries = DebianListParser.Parse(text);
if (entries.Count > 0)
{
var windowStart = (cursor.LastPublished ?? (now - _options.InitialBackfill)) - _options.ResumeOverlap;
if (windowStart < DateTimeOffset.UnixEpoch)
{
windowStart = DateTimeOffset.UnixEpoch;
}
ProvenanceDiagnostics.ReportResumeWindow(SourceName, windowStart, _logger);
var candidates = entries
.Where(entry => entry.Published >= windowStart)
.OrderBy(entry => entry.Published)
.ThenBy(entry => entry.AdvisoryId, StringComparer.OrdinalIgnoreCase)
.ToList();
if (candidates.Count == 0)
{
candidates = entries
.OrderByDescending(entry => entry.Published)
.ThenBy(entry => entry.AdvisoryId, StringComparer.OrdinalIgnoreCase)
.Take(_options.MaxAdvisoriesPerFetch)
.OrderBy(entry => entry.Published)
.ThenBy(entry => entry.AdvisoryId, StringComparer.OrdinalIgnoreCase)
.ToList();
}
else if (candidates.Count > _options.MaxAdvisoriesPerFetch)
{
candidates = candidates
.OrderByDescending(entry => entry.Published)
.ThenBy(entry => entry.AdvisoryId, StringComparer.OrdinalIgnoreCase)
.Take(_options.MaxAdvisoriesPerFetch)
.OrderBy(entry => entry.Published)
.ThenBy(entry => entry.AdvisoryId, StringComparer.OrdinalIgnoreCase)
.ToList();
}
foreach (var entry in candidates)
{
cancellationToken.ThrowIfCancellationRequested();
var detailUri = new Uri(_options.DetailBaseUri, entry.AdvisoryId);
var cacheKey = detailUri.ToString();
touchedResources.Add(cacheKey);
cursor.TryGetCache(cacheKey, out var cachedDetail);
if (!fetchCache.TryGetValue(cacheKey, out var cachedInRun))
{
cachedInRun = cachedDetail;
}
var metadata = BuildDetailMetadata(entry);
var existingDetail = await _documentStore.FindBySourceAndUriAsync(SourceName, cacheKey, cancellationToken).ConfigureAwait(false);
var request = new SourceFetchRequest(DebianOptions.HttpClientName, SourceName, detailUri)
{
Metadata = metadata,
AcceptHeaders = new[] { "text/html", "application/xhtml+xml" },
TimeoutOverride = _options.FetchTimeout,
ETag = existingDetail?.Etag ?? cachedInRun?.ETag,
LastModified = existingDetail?.LastModified ?? cachedInRun?.LastModified,
};
SourceFetchResult result;
try
{
result = await _fetchService.FetchAsync(request, cancellationToken).ConfigureAwait(false);
}
catch (Exception ex)
{
_logger.LogError(ex, "Failed to fetch Debian advisory {AdvisoryId}", entry.AdvisoryId);
await _stateRepository.MarkFailureAsync(SourceName, now, TimeSpan.FromMinutes(5), ex.Message, cancellationToken).ConfigureAwait(false);
throw;
}
if (result.IsNotModified)
{
if (existingDetail is not null)
{
fetchCache[cacheKey] = DebianFetchCacheEntry.FromDocument(existingDetail);
if (string.Equals(existingDetail.Status, DocumentStatuses.Mapped, StringComparison.Ordinal))
{
pendingDocuments.Remove(existingDetail.Id);
pendingMappings.Remove(existingDetail.Id);
}
}
continue;
}
if (!result.IsSuccess || result.Document is null)
{
continue;
}
fetchCache[cacheKey] = DebianFetchCacheEntry.FromDocument(result.Document);
pendingDocuments.Add(result.Document.Id);
pendingMappings.Remove(result.Document.Id);
if (_options.RequestDelay > TimeSpan.Zero)
{
try
{
await Task.Delay(_options.RequestDelay, cancellationToken).ConfigureAwait(false);
}
catch (TaskCanceledException)
{
break;
}
}
if (entry.Published > maxPublished)
{
maxPublished = entry.Published;
newProcessedIds.Clear();
processedUpdated = true;
}
if (entry.Published == maxPublished)
{
newProcessedIds.Add(entry.AdvisoryId);
processedUpdated = true;
}
}
}
}
}
if (fetchCache.Count > 0 && touchedResources.Count > 0)
{
var stale = fetchCache.Keys.Where(key => !touchedResources.Contains(key)).ToArray();
foreach (var key in stale)
{
fetchCache.Remove(key);
}
}
if (!processedUpdated && cursor.LastPublished.HasValue)
{
maxPublished = cursor.LastPublished.Value;
newProcessedIds = new HashSet<string>(cursor.ProcessedAdvisoryIds, StringComparer.OrdinalIgnoreCase);
}
var updatedCursor = cursor
.WithPendingDocuments(pendingDocuments)
.WithPendingMappings(pendingMappings)
.WithFetchCache(fetchCache);
if (processedUpdated && maxPublished > DateTimeOffset.MinValue)
{
updatedCursor = updatedCursor.WithProcessed(maxPublished, newProcessedIds);
}
await UpdateCursorAsync(updatedCursor, cancellationToken).ConfigureAwait(false);
}
public async Task ParseAsync(IServiceProvider services, CancellationToken cancellationToken)
{
ArgumentNullException.ThrowIfNull(services);
var cursor = await GetCursorAsync(cancellationToken).ConfigureAwait(false);
if (cursor.PendingDocuments.Count == 0)
{
return;
}
var remaining = cursor.PendingDocuments.ToList();
var pendingMappings = cursor.PendingMappings.ToList();
foreach (var documentId in cursor.PendingDocuments)
{
cancellationToken.ThrowIfCancellationRequested();
var document = await _documentStore.FindAsync(documentId, cancellationToken).ConfigureAwait(false);
if (document is null)
{
remaining.Remove(documentId);
continue;
}
if (!document.GridFsId.HasValue)
{
_logger.LogWarning("Debian document {DocumentId} missing GridFS payload", document.Id);
await _documentStore.UpdateStatusAsync(document.Id, DocumentStatuses.Failed, cancellationToken).ConfigureAwait(false);
remaining.Remove(documentId);
continue;
}
var metadata = ExtractMetadata(document);
if (metadata is null)
{
_logger.LogWarning("Debian document {DocumentId} missing required metadata", document.Id);
await _documentStore.UpdateStatusAsync(document.Id, DocumentStatuses.Failed, cancellationToken).ConfigureAwait(false);
remaining.Remove(documentId);
continue;
}
byte[] bytes;
try
{
bytes = await _rawDocumentStorage.DownloadAsync(document.GridFsId.Value, cancellationToken).ConfigureAwait(false);
}
catch (Exception ex)
{
_logger.LogError(ex, "Failed to download Debian document {DocumentId}", document.Id);
throw;
}
var html = System.Text.Encoding.UTF8.GetString(bytes);
DebianAdvisoryDto dto;
try
{
dto = DebianHtmlParser.Parse(html, metadata);
}
catch (Exception ex)
{
_logger.LogWarning(ex, "Failed to parse Debian advisory {AdvisoryId}", metadata.AdvisoryId);
await _documentStore.UpdateStatusAsync(document.Id, DocumentStatuses.Failed, cancellationToken).ConfigureAwait(false);
remaining.Remove(document.Id);
continue;
}
var payload = ToBson(dto);
var dtoRecord = new DtoRecord(Guid.NewGuid(), document.Id, SourceName, SchemaVersion, payload, _timeProvider.GetUtcNow());
await _dtoStore.UpsertAsync(dtoRecord, cancellationToken).ConfigureAwait(false);
await _documentStore.UpdateStatusAsync(document.Id, DocumentStatuses.PendingMap, cancellationToken).ConfigureAwait(false);
remaining.Remove(document.Id);
if (!pendingMappings.Contains(document.Id))
{
pendingMappings.Add(document.Id);
}
}
var updatedCursor = cursor
.WithPendingDocuments(remaining)
.WithPendingMappings(pendingMappings);
await UpdateCursorAsync(updatedCursor, cancellationToken).ConfigureAwait(false);
}
public async Task MapAsync(IServiceProvider services, CancellationToken cancellationToken)
{
ArgumentNullException.ThrowIfNull(services);
var cursor = await GetCursorAsync(cancellationToken).ConfigureAwait(false);
if (cursor.PendingMappings.Count == 0)
{
return;
}
var pendingMappings = cursor.PendingMappings.ToList();
foreach (var documentId in cursor.PendingMappings)
{
cancellationToken.ThrowIfCancellationRequested();
var dtoRecord = await _dtoStore.FindByDocumentIdAsync(documentId, cancellationToken).ConfigureAwait(false);
var document = await _documentStore.FindAsync(documentId, cancellationToken).ConfigureAwait(false);
if (dtoRecord is null || document is null)
{
pendingMappings.Remove(documentId);
continue;
}
DebianAdvisoryDto dto;
try
{
dto = FromBson(dtoRecord.Payload);
}
catch (Exception ex)
{
_logger.LogError(ex, "Failed to deserialize Debian DTO for document {DocumentId}", documentId);
await _documentStore.UpdateStatusAsync(documentId, DocumentStatuses.Failed, cancellationToken).ConfigureAwait(false);
pendingMappings.Remove(documentId);
continue;
}
var advisory = DebianMapper.Map(dto, document, _timeProvider.GetUtcNow());
await _advisoryStore.UpsertAsync(advisory, cancellationToken).ConfigureAwait(false);
await _documentStore.UpdateStatusAsync(documentId, DocumentStatuses.Mapped, cancellationToken).ConfigureAwait(false);
pendingMappings.Remove(documentId);
LogMapped(_logger, dto.AdvisoryId, advisory.AffectedPackages.Length, null);
}
var updatedCursor = cursor.WithPendingMappings(pendingMappings);
await UpdateCursorAsync(updatedCursor, cancellationToken).ConfigureAwait(false);
}
private async Task<DebianCursor> GetCursorAsync(CancellationToken cancellationToken)
{
var state = await _stateRepository.TryGetAsync(SourceName, cancellationToken).ConfigureAwait(false);
return state is null ? DebianCursor.Empty : DebianCursor.FromBson(state.Cursor);
}
private async Task UpdateCursorAsync(DebianCursor cursor, CancellationToken cancellationToken)
{
var document = cursor.ToBsonDocument();
await _stateRepository.UpdateCursorAsync(SourceName, document, _timeProvider.GetUtcNow(), cancellationToken).ConfigureAwait(false);
}
private static Dictionary<string, string> BuildDetailMetadata(DebianListEntry entry)
{
var metadata = new Dictionary<string, string>(StringComparer.Ordinal)
{
["debian.id"] = entry.AdvisoryId,
["debian.published"] = entry.Published.ToString("O", CultureInfo.InvariantCulture),
["debian.title"] = entry.Title,
["debian.package"] = entry.SourcePackage
};
if (entry.CveIds.Count > 0)
{
metadata["debian.cves"] = string.Join(' ', entry.CveIds);
}
return metadata;
}
private static DebianDetailMetadata? ExtractMetadata(DocumentRecord document)
{
if (document.Metadata is null)
{
return null;
}
if (!document.Metadata.TryGetValue("debian.id", out var id) || string.IsNullOrWhiteSpace(id))
{
return null;
}
if (!document.Metadata.TryGetValue("debian.published", out var publishedRaw)
|| !DateTimeOffset.TryParse(publishedRaw, CultureInfo.InvariantCulture, DateTimeStyles.AssumeUniversal, out var published))
{
published = document.FetchedAt;
}
var title = document.Metadata.TryGetValue("debian.title", out var t) ? t : id;
var package = document.Metadata.TryGetValue("debian.package", out var pkg) && !string.IsNullOrWhiteSpace(pkg)
? pkg
: id;
IReadOnlyList<string> cveList = Array.Empty<string>();
if (document.Metadata.TryGetValue("debian.cves", out var cvesRaw) && !string.IsNullOrWhiteSpace(cvesRaw))
{
cveList = cvesRaw
.Split(' ', StringSplitOptions.TrimEntries | StringSplitOptions.RemoveEmptyEntries)
.Where(static s => !string.IsNullOrWhiteSpace(s))
.Select(static s => s!)
.Distinct(StringComparer.OrdinalIgnoreCase)
.ToArray();
}
return new DebianDetailMetadata(
id.Trim(),
new Uri(document.Uri, UriKind.Absolute),
published.ToUniversalTime(),
title,
package,
cveList);
}
private static BsonDocument ToBson(DebianAdvisoryDto dto)
{
var packages = new BsonArray();
foreach (var package in dto.Packages)
{
var packageDoc = new BsonDocument
{
["package"] = package.Package,
["release"] = package.Release,
["status"] = package.Status,
};
if (!string.IsNullOrWhiteSpace(package.IntroducedVersion))
{
packageDoc["introduced"] = package.IntroducedVersion;
}
if (!string.IsNullOrWhiteSpace(package.FixedVersion))
{
packageDoc["fixed"] = package.FixedVersion;
}
if (!string.IsNullOrWhiteSpace(package.LastAffectedVersion))
{
packageDoc["last"] = package.LastAffectedVersion;
}
if (package.Published.HasValue)
{
packageDoc["published"] = package.Published.Value.UtcDateTime;
}
packages.Add(packageDoc);
}
var references = new BsonArray(dto.References.Select(reference =>
{
var doc = new BsonDocument
{
["url"] = reference.Url
};
if (!string.IsNullOrWhiteSpace(reference.Kind))
{
doc["kind"] = reference.Kind;
}
if (!string.IsNullOrWhiteSpace(reference.Title))
{
doc["title"] = reference.Title;
}
return doc;
}));
return new BsonDocument
{
["advisoryId"] = dto.AdvisoryId,
["sourcePackage"] = dto.SourcePackage,
["title"] = dto.Title,
["description"] = dto.Description ?? string.Empty,
["cves"] = new BsonArray(dto.CveIds),
["packages"] = packages,
["references"] = references,
};
}
private static DebianAdvisoryDto FromBson(BsonDocument document)
{
var advisoryId = document.GetValue("advisoryId", "").AsString;
var sourcePackage = document.GetValue("sourcePackage", advisoryId).AsString;
var title = document.GetValue("title", advisoryId).AsString;
var description = document.TryGetValue("description", out var desc) ? desc.AsString : null;
var cves = document.TryGetValue("cves", out var cveArray) && cveArray is BsonArray cvesBson
? cvesBson.OfType<BsonValue>()
.Select(static value => value.ToString())
.Where(static s => !string.IsNullOrWhiteSpace(s))
.Select(static s => s!)
.ToArray()
: Array.Empty<string>();
var packages = new List<DebianPackageStateDto>();
if (document.TryGetValue("packages", out var packageArray) && packageArray is BsonArray packagesBson)
{
foreach (var element in packagesBson.OfType<BsonDocument>())
{
packages.Add(new DebianPackageStateDto(
element.GetValue("package", sourcePackage).AsString,
element.GetValue("release", string.Empty).AsString,
element.GetValue("status", "unknown").AsString,
element.TryGetValue("introduced", out var introducedValue) ? introducedValue.AsString : null,
element.TryGetValue("fixed", out var fixedValue) ? fixedValue.AsString : null,
element.TryGetValue("last", out var lastValue) ? lastValue.AsString : null,
element.TryGetValue("published", out var publishedValue)
? publishedValue.BsonType switch
{
BsonType.DateTime => DateTime.SpecifyKind(publishedValue.ToUniversalTime(), DateTimeKind.Utc),
BsonType.String when DateTimeOffset.TryParse(publishedValue.AsString, out var parsed) => parsed.ToUniversalTime(),
_ => (DateTimeOffset?)null,
}
: null));
}
}
var references = new List<DebianReferenceDto>();
if (document.TryGetValue("references", out var referenceArray) && referenceArray is BsonArray refBson)
{
foreach (var element in refBson.OfType<BsonDocument>())
{
references.Add(new DebianReferenceDto(
element.GetValue("url", "").AsString,
element.TryGetValue("kind", out var kind) ? kind.AsString : null,
element.TryGetValue("title", out var titleValue) ? titleValue.AsString : null));
}
}
return new DebianAdvisoryDto(
advisoryId,
sourcePackage,
title,
description,
cves,
packages,
references);
}
}

View File

@@ -0,0 +1,22 @@
using System;
using System.Threading;
using System;
using Microsoft.Extensions.DependencyInjection;
using StellaOps.Plugin;
namespace StellaOps.Feedser.Source.Distro.Debian;
public sealed class DebianConnectorPlugin : IConnectorPlugin
{
public const string SourceName = "distro-debian";
public string Name => SourceName;
public bool IsAvailable(IServiceProvider services) => services is not null;
public IFeedConnector Create(IServiceProvider services)
{
ArgumentNullException.ThrowIfNull(services);
return ActivatorUtilities.CreateInstance<DebianConnector>(services);
}
}

View File

@@ -0,0 +1,53 @@
using System;
using Microsoft.Extensions.Configuration;
using Microsoft.Extensions.DependencyInjection;
using StellaOps.DependencyInjection;
using StellaOps.Feedser.Core.Jobs;
using StellaOps.Feedser.Source.Distro.Debian.Configuration;
namespace StellaOps.Feedser.Source.Distro.Debian;
public sealed class DebianDependencyInjectionRoutine : IDependencyInjectionRoutine
{
private const string ConfigurationSection = "feedser:sources:debian";
private const string FetchSchedule = "*/30 * * * *";
private const string ParseSchedule = "7,37 * * * *";
private const string MapSchedule = "12,42 * * * *";
private static readonly TimeSpan FetchTimeout = TimeSpan.FromMinutes(6);
private static readonly TimeSpan ParseTimeout = TimeSpan.FromMinutes(10);
private static readonly TimeSpan MapTimeout = TimeSpan.FromMinutes(10);
private static readonly TimeSpan LeaseDuration = TimeSpan.FromMinutes(5);
public IServiceCollection Register(IServiceCollection services, IConfiguration configuration)
{
ArgumentNullException.ThrowIfNull(services);
ArgumentNullException.ThrowIfNull(configuration);
services.AddDebianConnector(options =>
{
configuration.GetSection(ConfigurationSection).Bind(options);
options.Validate();
});
var scheduler = new JobSchedulerBuilder(services);
scheduler
.AddJob<DebianFetchJob>(
DebianJobKinds.Fetch,
cronExpression: FetchSchedule,
timeout: FetchTimeout,
leaseDuration: LeaseDuration)
.AddJob<DebianParseJob>(
DebianJobKinds.Parse,
cronExpression: ParseSchedule,
timeout: ParseTimeout,
leaseDuration: LeaseDuration)
.AddJob<DebianMapJob>(
DebianJobKinds.Map,
cronExpression: MapSchedule,
timeout: MapTimeout,
leaseDuration: LeaseDuration);
return services;
}
}

View File

@@ -0,0 +1,37 @@
using System;
using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.Options;
using StellaOps.Feedser.Source.Common.Http;
using StellaOps.Feedser.Source.Distro.Debian.Configuration;
namespace StellaOps.Feedser.Source.Distro.Debian;
public static class DebianServiceCollectionExtensions
{
public static IServiceCollection AddDebianConnector(this IServiceCollection services, Action<DebianOptions> configure)
{
ArgumentNullException.ThrowIfNull(services);
ArgumentNullException.ThrowIfNull(configure);
services.AddOptions<DebianOptions>()
.Configure(configure)
.PostConfigure(static options => options.Validate());
services.AddSourceHttpClient(DebianOptions.HttpClientName, (sp, httpOptions) =>
{
var options = sp.GetRequiredService<IOptions<DebianOptions>>().Value;
httpOptions.BaseAddress = options.DetailBaseUri.GetLeftPart(UriPartial.Authority) is { Length: > 0 } authority
? new Uri(authority, UriKind.Absolute)
: new Uri("https://security-tracker.debian.org/", UriKind.Absolute);
httpOptions.Timeout = options.FetchTimeout;
httpOptions.UserAgent = options.UserAgent;
httpOptions.AllowedHosts.Clear();
httpOptions.AllowedHosts.Add(options.DetailBaseUri.Host);
httpOptions.AllowedHosts.Add(options.ListEndpoint.Host);
httpOptions.DefaultRequestHeaders["Accept"] = "text/html,application/xhtml+xml,text/plain;q=0.9,application/json;q=0.8";
});
services.AddTransient<DebianConnector>();
return services;
}
}

View File

@@ -0,0 +1,27 @@
using System;
using System.Collections.Generic;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal sealed record DebianAdvisoryDto(
string AdvisoryId,
string SourcePackage,
string? Title,
string? Description,
IReadOnlyList<string> CveIds,
IReadOnlyList<DebianPackageStateDto> Packages,
IReadOnlyList<DebianReferenceDto> References);
internal sealed record DebianPackageStateDto(
string Package,
string Release,
string Status,
string? IntroducedVersion,
string? FixedVersion,
string? LastAffectedVersion,
DateTimeOffset? Published);
internal sealed record DebianReferenceDto(
string Url,
string? Kind,
string? Title);

View File

@@ -0,0 +1,177 @@
using System;
using System.Collections.Generic;
using System.Linq;
using MongoDB.Bson;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal sealed record DebianCursor(
DateTimeOffset? LastPublished,
IReadOnlyCollection<string> ProcessedAdvisoryIds,
IReadOnlyCollection<Guid> PendingDocuments,
IReadOnlyCollection<Guid> PendingMappings,
IReadOnlyDictionary<string, DebianFetchCacheEntry> FetchCache)
{
private static readonly IReadOnlyCollection<string> EmptyIds = Array.Empty<string>();
private static readonly IReadOnlyCollection<Guid> EmptyGuidList = Array.Empty<Guid>();
private static readonly IReadOnlyDictionary<string, DebianFetchCacheEntry> EmptyCache =
new Dictionary<string, DebianFetchCacheEntry>(StringComparer.OrdinalIgnoreCase);
public static DebianCursor Empty { get; } = new(null, EmptyIds, EmptyGuidList, EmptyGuidList, EmptyCache);
public static DebianCursor FromBson(BsonDocument? document)
{
if (document is null || document.ElementCount == 0)
{
return Empty;
}
DateTimeOffset? lastPublished = null;
if (document.TryGetValue("lastPublished", out var lastValue))
{
lastPublished = lastValue.BsonType switch
{
BsonType.String when DateTimeOffset.TryParse(lastValue.AsString, out var parsed) => parsed.ToUniversalTime(),
BsonType.DateTime => DateTime.SpecifyKind(lastValue.ToUniversalTime(), DateTimeKind.Utc),
_ => null,
};
}
var processed = ReadStringArray(document, "processedIds");
var pendingDocuments = ReadGuidArray(document, "pendingDocuments");
var pendingMappings = ReadGuidArray(document, "pendingMappings");
var cache = ReadCache(document);
return new DebianCursor(lastPublished, processed, pendingDocuments, pendingMappings, cache);
}
public BsonDocument ToBsonDocument()
{
var document = new BsonDocument
{
["pendingDocuments"] = new BsonArray(PendingDocuments.Select(static id => id.ToString())),
["pendingMappings"] = new BsonArray(PendingMappings.Select(static id => id.ToString())),
};
if (LastPublished.HasValue)
{
document["lastPublished"] = LastPublished.Value.UtcDateTime;
}
if (ProcessedAdvisoryIds.Count > 0)
{
document["processedIds"] = new BsonArray(ProcessedAdvisoryIds);
}
if (FetchCache.Count > 0)
{
var cacheDoc = new BsonDocument();
foreach (var (key, entry) in FetchCache)
{
cacheDoc[key] = entry.ToBsonDocument();
}
document["fetchCache"] = cacheDoc;
}
return document;
}
public DebianCursor WithPendingDocuments(IEnumerable<Guid> ids)
=> this with { PendingDocuments = ids?.Distinct().ToArray() ?? EmptyGuidList };
public DebianCursor WithPendingMappings(IEnumerable<Guid> ids)
=> this with { PendingMappings = ids?.Distinct().ToArray() ?? EmptyGuidList };
public DebianCursor WithProcessed(DateTimeOffset published, IEnumerable<string> ids)
=> this with
{
LastPublished = published.ToUniversalTime(),
ProcessedAdvisoryIds = ids?.Where(static id => !string.IsNullOrWhiteSpace(id))
.Select(static id => id.Trim())
.Distinct(StringComparer.OrdinalIgnoreCase)
.ToArray() ?? EmptyIds
};
public DebianCursor WithFetchCache(IDictionary<string, DebianFetchCacheEntry>? cache)
{
if (cache is null || cache.Count == 0)
{
return this with { FetchCache = EmptyCache };
}
return this with { FetchCache = new Dictionary<string, DebianFetchCacheEntry>(cache, StringComparer.OrdinalIgnoreCase) };
}
public bool TryGetCache(string key, out DebianFetchCacheEntry entry)
{
if (FetchCache.Count == 0)
{
entry = DebianFetchCacheEntry.Empty;
return false;
}
return FetchCache.TryGetValue(key, out entry!);
}
private static IReadOnlyCollection<string> ReadStringArray(BsonDocument document, string field)
{
if (!document.TryGetValue(field, out var value) || value is not BsonArray array)
{
return EmptyIds;
}
var list = new List<string>(array.Count);
foreach (var element in array)
{
if (element.BsonType == BsonType.String)
{
var str = element.AsString.Trim();
if (!string.IsNullOrEmpty(str))
{
list.Add(str);
}
}
}
return list;
}
private static IReadOnlyCollection<Guid> ReadGuidArray(BsonDocument document, string field)
{
if (!document.TryGetValue(field, out var value) || value is not BsonArray array)
{
return EmptyGuidList;
}
var list = new List<Guid>(array.Count);
foreach (var element in array)
{
if (Guid.TryParse(element.ToString(), out var guid))
{
list.Add(guid);
}
}
return list;
}
private static IReadOnlyDictionary<string, DebianFetchCacheEntry> ReadCache(BsonDocument document)
{
if (!document.TryGetValue("fetchCache", out var value) || value is not BsonDocument cacheDocument || cacheDocument.ElementCount == 0)
{
return EmptyCache;
}
var cache = new Dictionary<string, DebianFetchCacheEntry>(StringComparer.OrdinalIgnoreCase);
foreach (var element in cacheDocument.Elements)
{
if (element.Value is BsonDocument entry)
{
cache[element.Name] = DebianFetchCacheEntry.FromBson(entry);
}
}
return cache;
}
}

View File

@@ -0,0 +1,12 @@
using System;
using System.Collections.Generic;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal sealed record DebianDetailMetadata(
string AdvisoryId,
Uri DetailUri,
DateTimeOffset Published,
string Title,
string SourcePackage,
IReadOnlyList<string> CveIds);

View File

@@ -0,0 +1,76 @@
using System;
using MongoDB.Bson;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal sealed record DebianFetchCacheEntry(string? ETag, DateTimeOffset? LastModified)
{
public static DebianFetchCacheEntry Empty { get; } = new(null, null);
public static DebianFetchCacheEntry FromDocument(StellaOps.Feedser.Storage.Mongo.Documents.DocumentRecord document)
=> new(document.Etag, document.LastModified);
public static DebianFetchCacheEntry FromBson(BsonDocument document)
{
if (document is null || document.ElementCount == 0)
{
return Empty;
}
string? etag = null;
DateTimeOffset? lastModified = null;
if (document.TryGetValue("etag", out var etagValue) && etagValue.BsonType == BsonType.String)
{
etag = etagValue.AsString;
}
if (document.TryGetValue("lastModified", out var modifiedValue))
{
lastModified = modifiedValue.BsonType switch
{
BsonType.String when DateTimeOffset.TryParse(modifiedValue.AsString, out var parsed) => parsed.ToUniversalTime(),
BsonType.DateTime => DateTime.SpecifyKind(modifiedValue.ToUniversalTime(), DateTimeKind.Utc),
_ => null,
};
}
return new DebianFetchCacheEntry(etag, lastModified);
}
public BsonDocument ToBsonDocument()
{
var document = new BsonDocument();
if (!string.IsNullOrWhiteSpace(ETag))
{
document["etag"] = ETag;
}
if (LastModified.HasValue)
{
document["lastModified"] = LastModified.Value.UtcDateTime;
}
return document;
}
public bool Matches(StellaOps.Feedser.Storage.Mongo.Documents.DocumentRecord document)
{
if (document is null)
{
return false;
}
if (!string.Equals(document.Etag, ETag, StringComparison.Ordinal))
{
return false;
}
if (LastModified.HasValue && document.LastModified.HasValue)
{
return LastModified.Value.UtcDateTime == document.LastModified.Value.UtcDateTime;
}
return !LastModified.HasValue && !document.LastModified.HasValue;
}
}

View File

@@ -0,0 +1,326 @@
using System;
using System.Collections.Generic;
using System.Globalization;
using System.Linq;
using AngleSharp.Html.Dom;
using AngleSharp.Html.Parser;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal static class DebianHtmlParser
{
public static DebianAdvisoryDto Parse(string html, DebianDetailMetadata metadata)
{
ArgumentException.ThrowIfNullOrEmpty(html);
ArgumentNullException.ThrowIfNull(metadata);
var parser = new HtmlParser();
var document = parser.ParseDocument(html);
var description = ExtractDescription(document) ?? metadata.Title;
var references = ExtractReferences(document, metadata);
var packages = ExtractPackages(document, metadata.SourcePackage, metadata.Published);
return new DebianAdvisoryDto(
metadata.AdvisoryId,
metadata.SourcePackage,
metadata.Title,
description,
metadata.CveIds,
packages,
references);
}
private static string? ExtractDescription(IHtmlDocument document)
{
foreach (var table in document.QuerySelectorAll("table"))
{
if (table is not IHtmlTableElement tableElement)
{
continue;
}
foreach (var row in tableElement.Rows)
{
if (row.Cells.Length < 2)
{
continue;
}
var header = row.Cells[0].TextContent?.Trim();
if (string.Equals(header, "Description", StringComparison.OrdinalIgnoreCase))
{
return NormalizeWhitespace(row.Cells[1].TextContent);
}
}
// Only the first table contains the metadata rows we need.
break;
}
return null;
}
private static IReadOnlyList<DebianReferenceDto> ExtractReferences(IHtmlDocument document, DebianDetailMetadata metadata)
{
var references = new List<DebianReferenceDto>();
var seen = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
// Add canonical Debian advisory page.
var canonical = new Uri($"https://www.debian.org/security/{metadata.AdvisoryId.ToLowerInvariant()}");
references.Add(new DebianReferenceDto(canonical.ToString(), "advisory", metadata.Title));
seen.Add(canonical.ToString());
foreach (var link in document.QuerySelectorAll("a"))
{
var href = link.GetAttribute("href");
if (string.IsNullOrWhiteSpace(href))
{
continue;
}
string resolved;
if (Uri.TryCreate(href, UriKind.Absolute, out var absolute))
{
resolved = absolute.ToString();
}
else if (Uri.TryCreate(metadata.DetailUri, href, out var relative))
{
resolved = relative.ToString();
}
else
{
continue;
}
if (!seen.Add(resolved))
{
continue;
}
var text = NormalizeWhitespace(link.TextContent);
string? kind = null;
if (text.StartsWith("CVE-", StringComparison.OrdinalIgnoreCase))
{
kind = "cve";
}
else if (resolved.Contains("debian.org/security", StringComparison.OrdinalIgnoreCase))
{
kind = "advisory";
}
references.Add(new DebianReferenceDto(resolved, kind, text));
}
return references;
}
private static IReadOnlyList<DebianPackageStateDto> ExtractPackages(IHtmlDocument document, string defaultPackage, DateTimeOffset published)
{
var table = FindPackagesTable(document);
if (table is null)
{
return Array.Empty<DebianPackageStateDto>();
}
var accumulators = new Dictionary<string, PackageAccumulator>(StringComparer.OrdinalIgnoreCase);
string currentPackage = defaultPackage;
foreach (var body in table.Bodies)
{
foreach (var row in body.Rows)
{
if (row.Cells.Length < 4)
{
continue;
}
var packageCell = NormalizeWhitespace(row.Cells[0].TextContent);
if (!string.IsNullOrWhiteSpace(packageCell))
{
currentPackage = ExtractPackageName(packageCell);
}
if (string.IsNullOrWhiteSpace(currentPackage))
{
continue;
}
var releaseRaw = NormalizeWhitespace(row.Cells[1].TextContent);
var versionRaw = NormalizeWhitespace(row.Cells[2].TextContent);
var statusRaw = NormalizeWhitespace(row.Cells[3].TextContent);
if (string.IsNullOrWhiteSpace(releaseRaw))
{
continue;
}
var release = NormalizeRelease(releaseRaw);
var key = $"{currentPackage}|{release}";
if (!accumulators.TryGetValue(key, out var accumulator))
{
accumulator = new PackageAccumulator(currentPackage, release, published);
accumulators[key] = accumulator;
}
accumulator.Apply(statusRaw, versionRaw);
}
}
return accumulators.Values
.Where(static acc => acc.ShouldEmit)
.Select(static acc => acc.ToDto())
.OrderBy(static dto => dto.Release, StringComparer.OrdinalIgnoreCase)
.ThenBy(static dto => dto.Package, StringComparer.OrdinalIgnoreCase)
.ToArray();
}
private static IHtmlTableElement? FindPackagesTable(IHtmlDocument document)
{
foreach (var table in document.QuerySelectorAll("table"))
{
if (table is not IHtmlTableElement tableElement)
{
continue;
}
var header = tableElement.Rows.FirstOrDefault();
if (header is null || header.Cells.Length < 4)
{
continue;
}
var firstHeader = NormalizeWhitespace(header.Cells[0].TextContent);
var secondHeader = NormalizeWhitespace(header.Cells[1].TextContent);
var thirdHeader = NormalizeWhitespace(header.Cells[2].TextContent);
if (string.Equals(firstHeader, "Source Package", StringComparison.OrdinalIgnoreCase)
&& string.Equals(secondHeader, "Release", StringComparison.OrdinalIgnoreCase)
&& string.Equals(thirdHeader, "Version", StringComparison.OrdinalIgnoreCase))
{
return tableElement;
}
}
return null;
}
private static string NormalizeRelease(string release)
{
var trimmed = release.Trim();
var parenthesisIndex = trimmed.IndexOf('(');
if (parenthesisIndex > 0)
{
trimmed = trimmed[..parenthesisIndex].Trim();
}
return trimmed;
}
private static string ExtractPackageName(string value)
{
var trimmed = value.Split(' ', StringSplitOptions.TrimEntries | StringSplitOptions.RemoveEmptyEntries).FirstOrDefault();
if (string.IsNullOrWhiteSpace(trimmed))
{
return value.Trim();
}
if (trimmed.EndsWith(")", StringComparison.Ordinal) && trimmed.Contains('('))
{
trimmed = trimmed[..trimmed.IndexOf('(')];
}
return trimmed.Trim();
}
private static string NormalizeWhitespace(string value)
=> string.IsNullOrWhiteSpace(value)
? string.Empty
: string.Join(' ', value.Split((char[]?)null, StringSplitOptions.RemoveEmptyEntries | StringSplitOptions.TrimEntries));
private sealed class PackageAccumulator
{
private readonly DateTimeOffset _published;
public PackageAccumulator(string package, string release, DateTimeOffset published)
{
Package = package;
Release = release;
_published = published;
Status = "unknown";
}
public string Package { get; }
public string Release { get; }
public string Status { get; private set; }
public string? IntroducedVersion { get; private set; }
public string? FixedVersion { get; private set; }
public string? LastAffectedVersion { get; private set; }
public bool ShouldEmit =>
!string.Equals(Status, "not_affected", StringComparison.OrdinalIgnoreCase)
|| IntroducedVersion is not null
|| FixedVersion is not null;
public void Apply(string statusRaw, string versionRaw)
{
var status = statusRaw.ToLowerInvariant();
var version = string.IsNullOrWhiteSpace(versionRaw) ? null : versionRaw.Trim();
if (status.Contains("fixed", StringComparison.OrdinalIgnoreCase))
{
FixedVersion = version;
if (!string.Equals(Status, "open", StringComparison.OrdinalIgnoreCase))
{
Status = "resolved";
}
return;
}
if (status.Contains("vulnerable", StringComparison.OrdinalIgnoreCase)
|| status.Contains("open", StringComparison.OrdinalIgnoreCase))
{
IntroducedVersion ??= version;
if (!string.Equals(Status, "resolved", StringComparison.OrdinalIgnoreCase))
{
Status = "open";
}
LastAffectedVersion = null;
return;
}
if (status.Contains("not affected", StringComparison.OrdinalIgnoreCase)
|| status.Contains("not vulnerable", StringComparison.OrdinalIgnoreCase))
{
Status = "not_affected";
IntroducedVersion = null;
FixedVersion = null;
LastAffectedVersion = null;
return;
}
if (status.Contains("end-of-life", StringComparison.OrdinalIgnoreCase) || status.Contains("end of life", StringComparison.OrdinalIgnoreCase))
{
Status = "end_of_life";
return;
}
Status = statusRaw;
}
public DebianPackageStateDto ToDto()
=> new(
Package: Package,
Release: Release,
Status: Status,
IntroducedVersion: IntroducedVersion,
FixedVersion: FixedVersion,
LastAffectedVersion: LastAffectedVersion,
Published: _published);
}
}

View File

@@ -0,0 +1,11 @@
using System;
using System.Collections.Generic;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal sealed record DebianListEntry(
string AdvisoryId,
DateTimeOffset Published,
string Title,
string SourcePackage,
IReadOnlyList<string> CveIds);

View File

@@ -0,0 +1,107 @@
using System;
using System.Collections.Generic;
using System.Globalization;
using System.Text.RegularExpressions;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal static class DebianListParser
{
private static readonly Regex HeaderRegex = new("^\\[(?<date>[^\\]]+)\\]\\s+(?<id>DSA-\\d{4,}-\\d+)\\s+(?<title>.+)$", RegexOptions.Compiled);
private static readonly Regex CveRegex = new("CVE-\\d{4}-\\d{3,7}", RegexOptions.IgnoreCase | RegexOptions.Compiled);
public static IReadOnlyList<DebianListEntry> Parse(string? content)
{
if (string.IsNullOrWhiteSpace(content))
{
return Array.Empty<DebianListEntry>();
}
var entries = new List<DebianListEntry>();
var currentCves = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
DateTimeOffset currentDate = default;
string? currentId = null;
string? currentTitle = null;
string? currentPackage = null;
foreach (var rawLine in content.Split('\n'))
{
var line = rawLine.TrimEnd('\r');
if (string.IsNullOrWhiteSpace(line))
{
continue;
}
if (line[0] == '[')
{
if (currentId is not null && currentTitle is not null && currentPackage is not null)
{
entries.Add(new DebianListEntry(
currentId,
currentDate,
currentTitle,
currentPackage,
currentCves.Count == 0 ? Array.Empty<string>() : new List<string>(currentCves)));
}
currentCves.Clear();
currentId = null;
currentTitle = null;
currentPackage = null;
var match = HeaderRegex.Match(line);
if (!match.Success)
{
continue;
}
if (!DateTimeOffset.TryParseExact(
match.Groups["date"].Value,
new[] { "dd MMM yyyy", "d MMM yyyy" },
CultureInfo.InvariantCulture,
DateTimeStyles.AssumeUniversal | DateTimeStyles.AdjustToUniversal,
out currentDate))
{
continue;
}
currentId = match.Groups["id"].Value.Trim();
currentTitle = match.Groups["title"].Value.Trim();
var separatorIndex = currentTitle.IndexOf(" - ", StringComparison.Ordinal);
currentPackage = separatorIndex > 0
? currentTitle[..separatorIndex].Trim()
: currentTitle.Split(' ', StringSplitOptions.RemoveEmptyEntries | StringSplitOptions.TrimEntries).FirstOrDefault();
if (string.IsNullOrWhiteSpace(currentPackage))
{
currentPackage = currentId;
}
continue;
}
if (line[0] == '{')
{
foreach (Match match in CveRegex.Matches(line))
{
if (match.Success && !string.IsNullOrWhiteSpace(match.Value))
{
currentCves.Add(match.Value.ToUpperInvariant());
}
}
}
}
if (currentId is not null && currentTitle is not null && currentPackage is not null)
{
entries.Add(new DebianListEntry(
currentId,
currentDate,
currentTitle,
currentPackage,
currentCves.Count == 0 ? Array.Empty<string>() : new List<string>(currentCves)));
}
return entries;
}
}

View File

@@ -0,0 +1,266 @@
using System;
using System.Collections.Generic;
using System.Linq;
using StellaOps.Feedser.Models;
using StellaOps.Feedser.Normalization.Distro;
using StellaOps.Feedser.Source.Common;
using StellaOps.Feedser.Storage.Mongo.Documents;
namespace StellaOps.Feedser.Source.Distro.Debian.Internal;
internal static class DebianMapper
{
public static Advisory Map(
DebianAdvisoryDto dto,
DocumentRecord document,
DateTimeOffset recordedAt)
{
ArgumentNullException.ThrowIfNull(dto);
ArgumentNullException.ThrowIfNull(document);
var aliases = BuildAliases(dto);
var references = BuildReferences(dto, recordedAt);
var affectedPackages = BuildAffectedPackages(dto, recordedAt);
var fetchProvenance = new AdvisoryProvenance(
DebianConnectorPlugin.SourceName,
"document",
document.Uri,
document.FetchedAt.ToUniversalTime());
var mappingProvenance = new AdvisoryProvenance(
DebianConnectorPlugin.SourceName,
"mapping",
dto.AdvisoryId,
recordedAt);
return new Advisory(
advisoryKey: dto.AdvisoryId,
title: dto.Title ?? dto.AdvisoryId,
summary: dto.Description,
language: "en",
published: dto.Packages.Select(p => p.Published).Where(p => p.HasValue).Select(p => p!.Value).Cast<DateTimeOffset?>().DefaultIfEmpty(null).Min(),
modified: recordedAt,
severity: null,
exploitKnown: false,
aliases: aliases,
references: references,
affectedPackages: affectedPackages,
cvssMetrics: Array.Empty<CvssMetric>(),
provenance: new[] { fetchProvenance, mappingProvenance });
}
private static string[] BuildAliases(DebianAdvisoryDto dto)
{
var aliases = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
if (!string.IsNullOrWhiteSpace(dto.AdvisoryId))
{
aliases.Add(dto.AdvisoryId.Trim());
}
foreach (var cve in dto.CveIds ?? Array.Empty<string>())
{
if (!string.IsNullOrWhiteSpace(cve))
{
aliases.Add(cve.Trim());
}
}
return aliases.OrderBy(a => a, StringComparer.OrdinalIgnoreCase).ToArray();
}
private static AdvisoryReference[] BuildReferences(DebianAdvisoryDto dto, DateTimeOffset recordedAt)
{
if (dto.References is null || dto.References.Count == 0)
{
return Array.Empty<AdvisoryReference>();
}
var references = new List<AdvisoryReference>();
foreach (var reference in dto.References)
{
if (string.IsNullOrWhiteSpace(reference.Url))
{
continue;
}
try
{
var provenance = new AdvisoryProvenance(
DebianConnectorPlugin.SourceName,
"reference",
reference.Url,
recordedAt);
references.Add(new AdvisoryReference(
reference.Url,
NormalizeReferenceKind(reference.Kind),
reference.Kind,
reference.Title,
provenance));
}
catch (ArgumentException)
{
// Ignore malformed URLs while keeping the rest of the advisory intact.
}
}
return references.Count == 0
? Array.Empty<AdvisoryReference>()
: references
.OrderBy(r => r.Url, StringComparer.OrdinalIgnoreCase)
.ToArray();
}
private static string? NormalizeReferenceKind(string? value)
{
if (string.IsNullOrWhiteSpace(value))
{
return null;
}
return value.Trim().ToLowerInvariant() switch
{
"advisory" or "dsa" => "advisory",
"cve" => "cve",
"patch" => "patch",
_ => null,
};
}
private static AdvisoryProvenance BuildPackageProvenance(DebianPackageStateDto package, DateTimeOffset recordedAt)
=> new(DebianConnectorPlugin.SourceName, "affected", $"{package.Package}:{package.Release}", recordedAt);
private static IReadOnlyList<AffectedPackage> BuildAffectedPackages(DebianAdvisoryDto dto, DateTimeOffset recordedAt)
{
if (dto.Packages is null || dto.Packages.Count == 0)
{
return Array.Empty<AffectedPackage>();
}
var packages = new List<AffectedPackage>(dto.Packages.Count);
foreach (var package in dto.Packages)
{
if (string.IsNullOrWhiteSpace(package.Package))
{
continue;
}
var provenance = new[] { BuildPackageProvenance(package, recordedAt) };
var ranges = BuildVersionRanges(package, recordedAt);
packages.Add(new AffectedPackage(
AffectedPackageTypes.Deb,
identifier: package.Package.Trim(),
platform: package.Release,
versionRanges: ranges,
statuses: Array.Empty<AffectedPackageStatus>(),
provenance: provenance));
}
return packages;
}
private static IReadOnlyList<AffectedVersionRange> BuildVersionRanges(DebianPackageStateDto package, DateTimeOffset recordedAt)
{
var provenance = new AdvisoryProvenance(
DebianConnectorPlugin.SourceName,
"range",
$"{package.Package}:{package.Release}",
recordedAt);
var introduced = package.IntroducedVersion;
var fixedVersion = package.FixedVersion;
var lastAffected = package.LastAffectedVersion;
if (string.IsNullOrWhiteSpace(introduced) && string.IsNullOrWhiteSpace(fixedVersion) && string.IsNullOrWhiteSpace(lastAffected))
{
return Array.Empty<AffectedVersionRange>();
}
var extensions = new Dictionary<string, string>(StringComparer.Ordinal)
{
["debian.release"] = package.Release,
["debian.status"] = package.Status
};
AddExtension(extensions, "debian.introduced", introduced);
AddExtension(extensions, "debian.fixed", fixedVersion);
AddExtension(extensions, "debian.lastAffected", lastAffected);
var primitives = BuildEvrPrimitives(introduced, fixedVersion, lastAffected);
return new[]
{
new AffectedVersionRange(
rangeKind: "evr",
introducedVersion: introduced,
fixedVersion: fixedVersion,
lastAffectedVersion: lastAffected,
rangeExpression: BuildRangeExpression(introduced, fixedVersion, lastAffected),
provenance: provenance,
primitives: primitives is null && extensions.Count == 0
? null
: new RangePrimitives(
SemVer: null,
Nevra: null,
Evr: primitives,
VendorExtensions: extensions.Count == 0 ? null : extensions))
};
}
private static EvrPrimitive? BuildEvrPrimitives(string? introduced, string? fixedVersion, string? lastAffected)
{
var introducedComponent = ParseEvr(introduced);
var fixedComponent = ParseEvr(fixedVersion);
var lastAffectedComponent = ParseEvr(lastAffected);
if (introducedComponent is null && fixedComponent is null && lastAffectedComponent is null)
{
return null;
}
return new EvrPrimitive(introducedComponent, fixedComponent, lastAffectedComponent);
}
private static EvrComponent? ParseEvr(string? value)
{
if (!DebianEvr.TryParse(value, out var evr) || evr is null)
{
return null;
}
return new EvrComponent(
evr.Epoch,
evr.Version,
evr.Revision.Length == 0 ? null : evr.Revision);
}
private static string? BuildRangeExpression(string? introduced, string? fixedVersion, string? lastAffected)
{
var parts = new List<string>();
if (!string.IsNullOrWhiteSpace(introduced))
{
parts.Add($"introduced:{introduced.Trim()}");
}
if (!string.IsNullOrWhiteSpace(fixedVersion))
{
parts.Add($"fixed:{fixedVersion.Trim()}");
}
if (!string.IsNullOrWhiteSpace(lastAffected))
{
parts.Add($"last:{lastAffected.Trim()}");
}
return parts.Count == 0 ? null : string.Join(" ", parts);
}
private static void AddExtension(IDictionary<string, string> extensions, string key, string? value)
{
if (!string.IsNullOrWhiteSpace(value))
{
extensions[key] = value.Trim();
}
}
}

View File

@@ -0,0 +1,46 @@
using System;
using System.Threading;
using System.Threading.Tasks;
using StellaOps.Feedser.Core.Jobs;
namespace StellaOps.Feedser.Source.Distro.Debian;
internal static class DebianJobKinds
{
public const string Fetch = "source:debian:fetch";
public const string Parse = "source:debian:parse";
public const string Map = "source:debian:map";
}
internal sealed class DebianFetchJob : IJob
{
private readonly DebianConnector _connector;
public DebianFetchJob(DebianConnector connector)
=> _connector = connector ?? throw new ArgumentNullException(nameof(connector));
public Task ExecuteAsync(JobExecutionContext context, CancellationToken cancellationToken)
=> _connector.FetchAsync(context.Services, cancellationToken);
}
internal sealed class DebianParseJob : IJob
{
private readonly DebianConnector _connector;
public DebianParseJob(DebianConnector connector)
=> _connector = connector ?? throw new ArgumentNullException(nameof(connector));
public Task ExecuteAsync(JobExecutionContext context, CancellationToken cancellationToken)
=> _connector.ParseAsync(context.Services, cancellationToken);
}
internal sealed class DebianMapJob : IJob
{
private readonly DebianConnector _connector;
public DebianMapJob(DebianConnector connector)
=> _connector = connector ?? throw new ArgumentNullException(nameof(connector));
public Task ExecuteAsync(JobExecutionContext context, CancellationToken cancellationToken)
=> _connector.MapAsync(context.Services, cancellationToken);
}

View File

@@ -0,0 +1,17 @@
<Project Sdk="Microsoft.NET.Sdk">
<PropertyGroup>
<TargetFramework>net10.0</TargetFramework>
<ImplicitUsings>enable</ImplicitUsings>
<Nullable>enable</Nullable>
</PropertyGroup>
<ItemGroup>
<ProjectReference Include="../StellaOps.Plugin/StellaOps.Plugin.csproj" />
<ProjectReference Include="../StellaOps.Feedser.Source.Common/StellaOps.Feedser.Source.Common.csproj" />
<ProjectReference Include="../StellaOps.Feedser.Models/StellaOps.Feedser.Models.csproj" />
<ProjectReference Include="../StellaOps.Feedser.Normalization/StellaOps.Feedser.Normalization.csproj" />
<ProjectReference Include="../StellaOps.Feedser.Storage.Mongo/StellaOps.Feedser.Storage.Mongo.csproj" />
</ItemGroup>
</Project>