'metadata_storage_name' always null: Missing or empty value '/document/blob_name'

Using Azure's AI Search Service to chunk and read textfiles, the target field values are always NULL.

I tried using a simplified example, just for the fields metadata_blob_name / title and metadata_blob_path / source.

Warning

This warning appears after running the indexer:

Message: Could not generate projection from input '/document'. Check the 'source' or 'sourceContext' property of your projection in your skillset. =$(/document) ?map { "source": $(/document/blob_path), "title": $(/document/blob_name) }

Details: Missing or empty value '/document/blob_name'.

Skillset

{
  "@odata.etag": "<REDACTED>",
  "name": "<REDACTED>",
  "description": "Text extract",
  "skills": [
    {
      "@odata.type": "#Microsoft.Skills.Util.DocumentExtractionSkill",
      "name": "extract-text",
      "context": "/document",
      "parsingMode": "text",
      "dataToExtract": "contentAndMetadata",
      "inputs": [
        {
          "name": "file_data",
          "source": "/document/file_data",
          "inputs": []
        }
      ],
      "outputs": [
        {
          "name": "content",
          "targetName": "raw_content"
        },
        {
          "name": "metadata_storage_name",
          "targetName": "blob_name"
        },
        {
          "name": "metadata_storage_path",
          "targetName": "blob_path"
        }
      ],
      "configuration": {}
    }
  ],
  "indexProjections": {
    "selectors": [
      {
        "targetIndexName": "<redacted>",
        "parentKeyFieldName": "parent_id",
        "sourceContext": "/document",
        "mappings": [
          {
            "name": "title",
            "source": "/document/blob_name",
            "inputs": []
          },
          {
            "name": "source",
            "source": "/document/blob_path",
            "inputs": []
          }
        ]
      }
    ],
    "parameters": {
      "projectionMode": "skipIndexingParentDocuments"
    }
  }
}

Index

{
  "@odata.etag": "<REDACTED>",
  "name": "<REDACTED>",
  "fields": [
    {
      "name": "id",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": true,
      "analyzer": "keyword",
      "synonymMaps": []
    },
    {
      "name": "title",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": true,
      "facetable": false,
      "key": false,
      "analyzer": "standard.lucene",
      "synonymMaps": []
    },
    {
      "name": "source",
      "type": "Edm.String",
      "searchable": false,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": true,
      "facetable": false,
      "key": false,
      "synonymMaps": []
    },
    {
      "name": "parent_id",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "stored": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "analyzer": "keyword",
      "synonymMaps": []
    }
  ],
  "similarity": {
    "@odata.type": "#Microsoft.Azure.Search.BM25Similarity"
  },
  "vectorSearch": {
    "algorithms": [
      {
        "name": "<redacted>",
        "kind": "hnsw",
        "hnswParameters": {
          "metric": "cosine",
          "m": 4,
          "efConstruction": 400,
          "efSearch": 500
        }
      }
    ],
    "profiles": [
      {
        "name": "<REDACTED>",
        "algorithm": "<REDACTED>"
      }
    ]
  }
}

Indexer

{
  "@odata.context": "<redacted>/$metadata#indexers/$entity",
  "@odata.etag": "<redacted>",
  "name": "<redacted>",
  "dataSourceName": "<redacted>",
  "skillsetName": "<redacted>",
  "targetIndexName": "<redacted>",
  "parameters": {
    "batchSize": 50,
    "maxFailedItems": 0,
    "maxFailedItemsPerBatch": 0,
    "configuration": {
      "dataToExtract": "contentAndMetadata",
      "parsingMode": "text",
      "allowSkillsetToReadFileData": true,
      "indexedFileNameExtensions": ".md",
      "excludedFileNameExtensions": "",
      "imageAction": "none",
      "pdfTextRotationAlgorithm": "none"
    }
  },
  "fieldMappings": [
    {
      "sourceFieldName": "metadata_storage_path",
      "targetFieldName": "id",
      "mappingFunction": {
        "name": "base64Encode",
        "parameters": null
      }
    }
  ],
  "outputFieldMappings": [],
  "cache": {
    "id": "<redacted>",
    "enableReprocessing": true,
    "storageConnectionString": "<redacted>",
    "identity": null
  }
}

Example file 01_keywords.md

# ContentTaxonomy/Keywords
# Variant 1 (v2025-10-06)
# 
# keyword- and alias lists

## topic1
- Keywords: lorem, ipsum, dolor, sit, amet

## topic2
- Keywords: lorem, ipsum, dolor, sit, amet

Search Explorer (NULL value result)

JSON-Search-Query:

{
  "search": "*",
  "select": "*",
  "count": true
}

RESULT:

{
  "@odata.context": "<redacted>/$metadata#docs(*)",
  "@odata.count": 5,
  "value": [
    {
      "@search.score": 1,
      "id": "<redacted (valid base64)>",
      "title": null,
      "source": null,
      "parent_id": "<redacted (valid base64)>"
    },
    // ...

Question: Why are the fields title and source always NULL ?

What I tried:

• resetting the Indexer cache
• using "targetName": "/document/blob_name" in the skillset (outputs)
• The run is processing 5 documents