dgraph-io/dgraph · error

while processing '%s', brackets are not closed properly

Error message

while processing '%s', brackets are not closed properly

What it means

GetFullTextTokens tokenizes one value with the FullTextTokenizer for the given language. Like GetTokens, it requires exactly one argument in funcArgs; more or fewer is an API misuse and returns this error.

Source

Thrown at chunker/json_parser.go:233

		nq.ObjectValue = &api.Value{Val: &api.Value_IntVal{IntVal: v}}

	case string:
		// Default value is considered as S P * deletion.
		if v == "" && op == DeleteNquads {
			nq.ObjectValue = &api.Value{Val: &api.Value_DefaultVal{DefaultVal: x.Star}}
			return nil
		}

		if vf, err := types.ParseVFloat(v); err == nil && len(vf) != 0 {
			nq.ObjectValue = &api.Value{Val: &api.Value_Vfloat32Val{Vfloat32Val: types.FloatArrayAsBytes(vf)}}
			return nil
		}

		// Handle the uid function in upsert block
		s := stripSpaces(v)
		if strings.HasPrefix(s, "uid(") || strings.HasPrefix(s, "val(") {
			if !strings.HasSuffix(s, ")") {
				return fmt.Errorf("while processing '%s', brackets are not closed properly", s)
			}
			nq.ObjectId = s
			return nil
		}

		// In RDF, we assume everything is default (types.DefaultID), but in JSON we assume string
		// (StringID). But this value will be checked against the schema so we don't overshadow a
		// password value (types.PasswordID) - Issue#2623
		nq.ObjectValue = &api.Value{Val: &api.Value_StrVal{StrVal: v}}

	case float64:
		if v == 0 && op == DeleteNquads {
			nq.ObjectValue = &api.Value{Val: &api.Value_DefaultVal{DefaultVal: x.Star}}
			return nil
		}
		nq.ObjectValue = &api.Value{Val: &api.Value_DoubleVal{DoubleVal: v}}

	case bool:

View on GitHub (pinned to 759e242be6)

Solutions

1. Join multiple values into one string and pass a single-element slice: tok.GetFullTextTokens([]string{text}, lang).
2. Validate len(funcArgs)==1 before calling.
3. Fix the query parser to forward only one value argument for full-text functions.

Example fix

// before
toks, err := tok.GetFullTextTokens([]string{"alpha", "beta"}, "en")
// after
toks, err := tok.GetFullTextTokens([]string{"alpha beta"}, "en")

if len(funcArgs) != 1 {
    return fmt.Errorf("GetFullTextTokens expects exactly 1 argument, got %d", len(funcArgs))
}

tokens, err := tok.GetFullTextTokens(funcArgs, lang)
if err != nil && strings.Contains(err.Error(), "Function requires 1 arguments") {
    return nil, fmt.Errorf("full-text functions accept a single text value; got %d", len(funcArgs))
}

When it happens

Trigger: Calling tok.GetFullTextTokens([]string{}, "en") or tok.GetFullTextTokens([]string{"a", "b"}, "en") — any slice whose length is not 1. Reached via full-text query handling and tests (getStringTokens, uidsForMatch paths).

Common situations: Full-text search functions (alloftext/anyoftext) receiving multiple values from parsed query args; splitting text into multiple args instead of one combined string.

Related errors

• illegal rune found "%c", expecting {
• Malformed JSON
• UID must be present and non-zero while deleting edges
• facet value can only be string/number/bool
• facets format should be of type map for scalarlist predicate

AI-assisted analysis of dgraph-io/dgraph@759e242be6 (2026-09-01). Data as JSON: /api/errors/dba2f14326009562. Report an issue: GitHub.