quickwit-oss/quickwit · error

the snippet field `{}` must be of type `Str`, got `{}`

Error message

the snippet field `{}` must be of type `Str`, got `{}`

What it means

Snippet fields must be of type Str (text). If the requested snippet field exists in the schema but has a non-Str field type (u64, bytes, json, etc.), validation rejects it with this error naming the actual type. Snippet generation only supports text content.

Source

Thrown at quickwit/quickwit-search/src/root.rs:339

fn validate_requested_snippet_fields(
    schema: &Schema,
    snippet_fields: &[String],
) -> anyhow::Result<()> {
    for field_name in snippet_fields {
        let field_entry = schema
            .get_field(field_name)
            .map(|field| schema.get_field_entry(field))?;
        match field_entry.field_type() {
            FieldType::Str(text_options) => {
                if !text_options.is_stored() {
                    return Err(anyhow::anyhow!(
                        "the snippet field `{}` must be stored",
                        field_name
                    ));
                }
            }
            other => {
                return Err(anyhow::anyhow!(
                    "the snippet field `{}` must be of type `Str`, got `{}`",
                    field_name,
                    other.value_type().name()
                ));
            }
        }
    }
    Ok(())
}

fn simplify_search_request_for_scroll_api(req: &SearchRequest) -> crate::Result<SearchRequest> {
    if req.search_after.is_some() {
        return Err(SearchError::InvalidArgument(
            "search_after cannot be used in a scroll context".to_string(),
        ));
    }

    // We do not mutate

View on GitHub (pinned to a39730c5cd)

Solutions

1. Remove the non-text field from snippet_fields; snippets are only supported for text fields.
2. Only pass fields programmatically filtered to schema field type Str.
3. If snippets on the field are truly needed, change the field type to text in the index config (requires reindexing).

Example fix

// before
{"search": ["status"], "snippet_fields": ["status"]} // status is u64
// after
{"search": ["status"], "snippet_fields": ["message"]} // message is text

fn only_text_fields(schema: &Schema, fields: &[String]) -> Vec<String> {
    fields.iter().filter(|f| {
        schema.get_field(f).map(|fd| matches!(schema.get_field_entry(fd).field_type(), FieldType::Str(_))).unwrap_or(false)
    }).cloned().collect()
}

fn is_str_field(schema: &Schema, name: &str) -> bool {
    schema.get_field(name).map(|f| matches!(schema.get_field_entry(f).field_type(), FieldType::Str(_))).unwrap_or(false)
}

match search(req.clone()).await {
    Err(e) if e.to_string().contains("must be of type `Str`") => {
        let mut req = req; req.snippet_fields = vec![]; search(req).await
    }
    r => r,
}

When it happens

Trigger: Calling search with snippet_fields containing a field whose schema type is not text (e.g. u64, datetime, json, bytes).

Common situations: Client UI sends the same field list for both retrieval and snippets; schema field type changed (e.g. text to json) while request templates still request snippets on it.

Understand the failure class

Background: Type mismatch errors: IllegalArgumentException, TypeError and type guards across 150 open-source libraries — this error's family across 150 libraries.

Related errors

• the snippet field `{}` must be stored
• unknown tokenizer `{}` for field `{}`
• invalid named document. there are more than 1 value associat
• the `{key}` value has to be a json object
• `record`, `tokenizer`, and `fieldnorms` parameters are allow

AI-assisted analysis of quickwit-oss/quickwit@a39730c5cd (2026-09-08). Data as JSON: /api/errors/679baa8c8940c2ed. Report an issue: GitHub.