public async Task SetupAndRunIndexer()

in tools/issue-labeler/src/SearchIndexCreator/IssueIndex.cs [26:171]
135 lines of code
1 McCabe index (conditional complexity)

        public async Task SetupAndRunIndexer(SearchIndexClient indexClient, SearchIndexerClient indexerClient, AzureOpenAIClient openAIClient)
        {
            // Create an Index  
            Console.WriteLine("Creating/Updating the index...");
            var index = GetSampleIndex();
            await indexClient.CreateOrUpdateIndexAsync(index);
            Console.WriteLine("Index Created/Updated!");

            // Create a Data Source Connection  
            Console.WriteLine("Creating/Updating the data source connection...");
            var dataSource = new SearchIndexerDataSourceConnection(
                $"{_config["IssueIndexName"]}-blob",
                SearchIndexerDataSourceType.AzureBlob,
                connectionString: _config["BlobConnectionString"], // "Connection string" indicating to use managed identity
                container: new SearchIndexerDataContainer($"{_config["IssueIndexName"]}-blob"))
            {
                DataChangeDetectionPolicy = new HighWaterMarkChangeDetectionPolicy("metadata_storage_last_modified"),
                DataDeletionDetectionPolicy = new NativeBlobSoftDeleteDeletionDetectionPolicy()
            };
            indexerClient.CreateOrUpdateDataSourceConnection(dataSource);
            Console.WriteLine("Data Source Created/Updated!");

            // Create a Skillset specifically for chunking
            // Each issue has its associated comments attached to it. Not sure if this will change in the future making chunking potentially unnecessary.
            Console.WriteLine("Creating/Updating the skillset...");
            var skillset = new SearchIndexerSkillset($"{_config["IssueIndexName"]}-skillset", new List<SearchIndexerSkill>
            {  
                // Add required skills here    
                new SplitSkill(
                    new List<InputFieldMappingEntry>
                    {
                        new InputFieldMappingEntry("text") { Source = "/document/Body" }
                    },
                    new List<OutputFieldMappingEntry>
                    {
                        new OutputFieldMappingEntry("textItems") { TargetName = "pages" }
                    })
                {
                    Context = "/document",
                    TextSplitMode = TextSplitMode.Pages,
                    // 10k because token limits are so high but want to experiment with lower chunking.
                    MaximumPageLength = 1000,
                    PageOverlapLength = 100,
                },
                new AzureOpenAIEmbeddingSkill(
                    new List<InputFieldMappingEntry>
                    {
                        new InputFieldMappingEntry("text") { Source = "/document/pages/*" }
                    },
                    new List<OutputFieldMappingEntry>
                    {
                        new OutputFieldMappingEntry("embedding") { TargetName = "text_vector" }
                    }
                )
                {
                    Context = "/document/pages/*",
                    ResourceUri = new Uri(_config["OpenAIEndpoint"]),
                    ModelName = _config["EmbeddingModelName"],
                    DeploymentName = _config["EmbeddingModelName"]
                }
            })
            {
                IndexProjection = new SearchIndexerIndexProjection(new[]
                {
                    new SearchIndexerIndexProjectionSelector(_config["IssueIndexName"], parentKeyFieldName: "parent_id", sourceContext: "/document/pages/*", mappings: new[]
                    {
                        new InputFieldMappingEntry("text_vector")
                        {
                            Source = "/document/pages/*/text_vector"
                        },
                        new InputFieldMappingEntry("chunk")
                        {
                            Source = "/document/pages/*"
                        },
                        new InputFieldMappingEntry("Id")
                        {
                            Source = "/document/Id"
                        },
                        new InputFieldMappingEntry("Title")
                        {
                            Source = "/document/Title"
                        },
                        new InputFieldMappingEntry("Service")
                        {
                            Source = "/document/Service"
                        },
                        new InputFieldMappingEntry("Category")
                        {
                            Source = "/document/Category"
                        },
                        new InputFieldMappingEntry("Author")
                        {
                            Source = "/document/Author"
                        },
                        new InputFieldMappingEntry("Repository")
                        {
                            Source = "/document/Repository"
                        },
                        new InputFieldMappingEntry("CreatedAt")
                        {
                            Source = "/document/CreatedAt"
                        },
                        new InputFieldMappingEntry("Url")
                        {
                            Source = "/document/Url"
                        },
                        new InputFieldMappingEntry("CodeOwner")
                        {
                            Source = "/document/CodeOwner"
                        },
                        // Metadata is needed for updating the document (or atleast last modified not sure of the rest)
                        new InputFieldMappingEntry("metadata_storage_last_modified")
                        {
                            Source = "/document/metadata_storage_last_modified"
                        }
                    })
                })
                {
                    Parameters = new SearchIndexerIndexProjectionsParameters
                    {
                        ProjectionMode = IndexProjectionMode.SkipIndexingParentDocuments
                    }
                }
            };
            await indexerClient.CreateOrUpdateSkillsetAsync(skillset).ConfigureAwait(false);
            Console.WriteLine("Skillset Created/Updated!");

            // Create an Indexer  
            Console.WriteLine("Creating the indexer and running it...");
            var indexer = new SearchIndexer($"{_config["IssueIndexName"]}-indexer", dataSource.Name, _config["IssueIndexName"])
            {
                Description = "Indexer to chunk documents, generate embeddings, and add to the index",
                Parameters = new IndexingParameters()
                {
                    IndexingParametersConfiguration = new IndexingParametersConfiguration()
                    {
                        DataToExtract = BlobIndexerDataToExtract.ContentAndMetadata,
                        ParsingMode = BlobIndexerParsingMode.Json
                    }
                },
                SkillsetName = skillset.Name,
                Schedule = new IndexingSchedule(TimeSpan.FromDays(1)) // Schedule to run every day
            };
            await indexerClient.CreateOrUpdateIndexerAsync(indexer).ConfigureAwait(false);
            Console.WriteLine("Indexer Created/Updated!");
        }