Skip to content

Include Unicode Normalization as part of the resource comparison #5765

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Fernando Henrique Inocêncio Borba Ferreira (fhibf) wants to merge 1 commit into
base: main
Choose a base branch
from
+61 −10
Open

Include Unicode Normalization as part of the resource comparison #5765

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
44 changes: 43 additions & 1 deletion src/Microsoft.Health.Fhir.CosmosDb.UnitTests/Features/Storage/CosmosFhirDataStoreTests.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,8 +10,8 @@
using System.Net.Http;
using System.Reflection;
using System.Runtime.CompilerServices;
using System.Text;
using System.Threading;
using System.Threading.Tasks;
using Hl7.Fhir.Model;
using Hl7.Fhir.Serialization;
using Microsoft.Azure.Cosmos;
Expand Down Expand Up @@ -88,6 +88,48 @@ public CosmosFhirDataStoreTests()
requestContextAccessor);
}

[Fact]
public void GivenCanonicallyEquivalentJapaneseText_WhenComparingResourcesWithoutVersionMetadata_ThenTheDataMatches()
{
const string existingData = "{\"resourceType\":\"Patient\",\"id\":\"id1\",\"meta\":{\"versionId\":\"1\",\"lastUpdated\":\"2026-08-26T21:00:00+00:00\"},\"name\":[{\"text\":\"\u304C\"}]}";
const string incomingData = "{\"resourceType\":\"Patient\",\"id\":\"id1\",\"meta\":{\"versionId\":\"2\",\"lastUpdated\":\"2026-08-26T21:01:00+00:00\"},\"name\":[{\"text\":\"\u304B\u3099\"}]}";

var lastModified = new DateTimeOffset(2026, 8, 26, 21, 0, 0, TimeSpan.Zero);
var request = new ResourceRequest(HttpMethod.Put, "http://fhir");

var existingResource = new FhirCosmosResourceWrapper(
"id1",
"1",
"Patient",
new RawResource(existingData, FhirResourceFormat.Json, isMetaSet: true),
request,
lastModified,
deleted: false,
history: false,
searchIndices: null,
compartmentIndices: null,
lastModifiedClaims: null);

var incomingResource = new FhirCosmosResourceWrapper(
"id1",
"2",
"Patient",
new RawResource(incomingData, FhirResourceFormat.Json, isMetaSet: true),
request,
lastModified.AddMinutes(1),
deleted: false,
history: false,
searchIndices: null,
compartmentIndices: null,
lastModifiedClaims: null);

var existingResourceData = CosmosFhirDataStore.RemoveVersionIdAndLastUpdatedFromMeta(existingResource);
var incomingResourceData = CosmosFhirDataStore.RemoveVersionIdAndLastUpdatedFromMeta(incomingResource);

Assert.Equal(existingResourceData, incomingResourceData);
Assert.True(existingResourceData.IsNormalized(NormalizationForm.FormC));
}

[Fact]
public async Task GivenAQuery_WhenASinglePageReturnsRequestedCount_ASingleQueryIsPerformced()
{
Expand Down
27 changes: 18 additions & 9 deletions src/Microsoft.Health.Fhir.CosmosDb/Features/Storage/CosmosFhirDataStore.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,22 +12,20 @@
using System.Linq;
using System.Net;
using System.Text;
using System.Text.RegularExpressions;
using System.Threading;
using System.Threading.Tasks;
using EnsureThat;
using Hl7.Fhir.Utility;
using Microsoft.Azure.Cosmos;
using Microsoft.Azure.Cosmos.Scripts;
using Microsoft.Extensions.Logging;
using Microsoft.Extensions.Options;
using Microsoft.Health.Abstractions.Exceptions;
using Microsoft.Health.Core;
using Microsoft.Health.Core.Features.Context;
using Microsoft.Health.Extensions.DependencyInjection;
using Microsoft.Health.Fhir.Core.Configs;
using Microsoft.Health.Fhir.Core.Exceptions;
using Microsoft.Health.Fhir.Core.Extensions;
using Microsoft.Health.Fhir.Core.Features;
using Microsoft.Health.Fhir.Core.Features.Conformance;
using Microsoft.Health.Fhir.Core.Features.Context;
using Microsoft.Health.Fhir.Core.Features.Definition;
Expand Down Expand Up @@ -432,10 +430,13 @@ await retryPolicy.ExecuteAsync(
// If not a delete then check if its an update with no data change
if (!cosmosWrapper.IsDeleted)
{
// check if the new resource data is same as existing resource data
if (string.Equals(RemoveVersionIdAndLastUpdatedFromMeta(existingItemResource), RemoveVersionIdAndLastUpdatedFromMeta(cosmosWrapper), StringComparison.Ordinal))
string existingData = RemoveVersionIdAndLastUpdatedFromMeta(existingItemResource);
string newData = RemoveVersionIdAndLastUpdatedFromMeta(cosmosWrapper);

// Check if the new resource data is same as existing resource data
if (string.Equals(existingData, newData, StringComparison.Ordinal))
{
// Do not store the duplicate data, for a update with no impact - returning existingItemResource as no updates
// Do not store the duplicate data, for a update with no impact - returning existingItemResource as no updates.
return new UpsertOutcome(existingItemResource, SaveOutcomeType.Updated);
}
}
Expand Down Expand Up @@ -876,11 +877,19 @@ private static string RemoveTrailingZerosFromMillisecondsForAGivenDate(DateTimeO
return formattedDate.Replace(milliseconds, trimmedMilliseconds, StringComparison.Ordinal);
}

private static string RemoveVersionIdAndLastUpdatedFromMeta(FhirCosmosResourceWrapper resourceWrapper)
internal static string RemoveVersionIdAndLastUpdatedFromMeta(FhirCosmosResourceWrapper resourceWrapper)
{
var versionToReplace = resourceWrapper.RawResource.IsMetaSet ? resourceWrapper.Version : "1";
var rawResource = resourceWrapper.RawResource.Data.Replace($"\"versionId\":\"{versionToReplace}\"", string.Empty, StringComparison.Ordinal);
return rawResource.Replace($"\"lastUpdated\":\"{RemoveTrailingZerosFromMillisecondsForAGivenDate(resourceWrapper.LastModified)}\"", string.Empty, StringComparison.Ordinal);
var rawResource = resourceWrapper.RawResource.Data
.Replace($"\"versionId\":\"{versionToReplace}\"", string.Empty, StringComparison.Ordinal)
.Replace($"\"lastUpdated\":\"{RemoveTrailingZerosFromMillisecondsForAGivenDate(resourceWrapper.LastModified)}\"", string.Empty, StringComparison.Ordinal);

// StringComparison.Ordinal performs a byte-by-byte comparison without any Unicode normalization. This is problematic for Japanese characters because: Japanese characters
// can be represented in multiple equivalent Unicode forms: NFD (Canonical Decomposition) or NFC (Canonical Composition).
// If a client sends Japanese text in one normalization form(e.g., NFD) but a previous version stored it in another form(e.g., NFC), the Ordinal comparison will fail even though
// the characters look identical.
// English/ASCII characters don't have normalization variants, so this only affects languages like Japanese, Korean, Vietnamese, etc.
return rawResource.Normalize(NormalizationForm.FormC);
}

public async Task BuildAsync(ICapabilityStatementBuilder builder, CancellationToken cancellationToken)
Expand Down
Loading