---
title: "json_serialize_sql"
slug: "dev-database-reference-sql-functions-json-json-serialize-sql"
updated: 2026-10-30T18:19:06Z
published: 2026-10-30T18:19:06Z
canonical: "docs.vastdata.com/dev-database-reference-sql-functions-json-json-serialize-sql"
---

> ## Documentation Index
> Fetch the complete documentation index at: https://docs.vastdata.com/llms.txt
> Use this file to discover all available pages before exploring further.

# json_serialize_sql

*Applies to VAST Cluster 5.5.*

*[Functions](/develop/docs/dev-database-reference-sql-functions-index) › [Parsing and JSON functions](/develop/docs/dev-database-reference-sql-functions-json) › SQL as JSON.* Parses the text of one or more `SELECT` statements and returns their syntax tree as a JSON document. Use it to inspect queries in a program, for example to find the tables a saved query reads. [json_deserialize_sql](/develop/docs/dev-database-reference-sql-functions-json-json-deserialize-sql) turns the tree back into SQL text.

## Syntax

```
json_serialize_sql(sql)
json_serialize_sql(sql, skip_null := true, skip_empty := true, format := true)
```

| Argument | Type | Description |
| --- | --- | --- |
| `sql` | `VARCHAR` | The text of one or more `SELECT` statements, separated by `;`. |
| `skip_null` | `BOOLEAN`, named | Leave out members whose value is `null`. |
| `skip_empty` | `BOOLEAN`, named | Leave out empty lists and objects. |
| `format` | `BOOLEAN`, named | Write the JSON over several lines, indented. |

The options are named: pass them as `name := value`, in any combination.

**Returns:** `JSON`. The driver returns it as text (a Python `str`).

## Set up the examples

The examples connect with the ADBC driver, as in [VAST Query Engine (ADBC)](/develop/docs/dev-database-integrations-adbc), and work in a schema called `docs_sql_fn_json_json_serialize_sql`. Set `QE_ENDPOINT`, `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY` and `VASTDB_BUCKET` ([SQL reference](/develop/docs/dev-database-reference-sql-index#example-connect-and-use-a-transaction)), and run the blocks in order in one Python session. Below each block is the output you should see.

This block connects, defines the `run()` and `plan()` helpers, and creates a table of saved queries: two `SELECT` statements, two statements in one text, an `INSERT`, and a NULL:

```
import os
import re

import adbc_driver_manager
import adbc_driver_manager.dbapi as dbapi
import adbc_driver_vastdb

BUCKET = os.environ["VASTDB_BUCKET"]
SCHEMA = f'"{BUCKET}/docs_sql_fn_json_json_serialize_sql"'

conn = dbapi.connect(
    driver=adbc_driver_vastdb.get_driver_path(),
    db_kwargs={
        "vast.db.endpoint": os.environ["QE_ENDPOINT"],
        "vast.db.access_key": os.environ["AWS_ACCESS_KEY_ID"],
        "vast.db.secret_key": os.environ["AWS_SECRET_ACCESS_KEY"],
    },
    autocommit=True,
)

def server_message(e):
    """The server's message in a driver error, without the request ID or the echoed statement."""
    text = re.sub(r"\[req=0x[0-9a-f]+\]: ", "", str(e))
    found = (re.search(r'message: "(.*?)", request_id', text, re.S)
             or re.search(r"description: (.*)\)", text, re.S))
    message = found.group(1) if found else text.strip().splitlines()[-1]
    message = re.sub(r"^Plan query: '.*?' failed with: ", "", message, flags=re.S)
    message = re.split(r"\\n|\n", message)[0].replace('\\"', '"')
    return message.replace(BUCKET, "<bucket>")

def run(sql):
    """Run one statement. Print the rows it returns, or the error."""
    try:
        with conn.cursor() as cur:
            cur.execute(sql)
            rows = cur.fetchall() if cur.description else []
    except (adbc_driver_manager.Error, OSError) as e:
        print(f"{type(e).__name__}: {server_message(e)}")
    else:
        print(rows)

def plan(sql):
    """Print the plan without its ~N Rows lines: the table's row count, which can still be 0 just after an insert."""
    with conn.cursor() as cur:
        cur.execute("EXPLAIN " + sql)
        steps = cur.fetchall()[0][1]
    print("\n".join(line for line in steps.splitlines() if "Rows" not in line))

TABLE = f"{SCHEMA}.saved_queries"
run(f"CREATE SCHEMA IF NOT EXISTS {SCHEMA}")
run(f"CREATE TABLE IF NOT EXISTS {TABLE} (id BIGINT, q VARCHAR)")
run(f"INSERT INTO {TABLE} VALUES "
    "(1, 'SELECT name FROM people WHERE id = 1'), "
    "(2, 'select a + 1 from t where b = ''x'' order by 1 limit 2'), "
    "(3, 'SELECT 1; SELECT 2'), "
    "(4, 'INSERT INTO x VALUES (1)'), "
    "(5, NULL)")
```

```
[]
[]
[(5,)]
```

## What it does

`json_serialize_sql` parses the text without running it, and returns an object:

- **For `SELECT` statements,** `&quot;error&quot;: false` and a `statements` array with one syntax tree per statement. Each tree is under `node`, and its `type` is `SELECT_NODE`. A `WITH` query is a `SELECT` statement too.
- **For anything else,** such as an `INSERT` or text that isn't SQL, `&quot;error&quot;: true`, with `error_type` (for example `parser`) and `error_message`. The query doesn't fail.

The tree of the shortest query, with the null and empty members left out:

```
run("SELECT json_serialize_sql('SELECT 1', skip_null := true, skip_empty := true)")
```

```
[('{"error":false,"statements":[{"node":{"type":"SELECT_NODE","select_list":[{"class":"CONSTANT","type":"VALUE_CONSTANT","query_location":7,"value":{"type":{"id":"INTEGER"},"is_null":false,"value":1}}],"from_table":{"type":"EMPTY","query_location":18446744073709551615},"aggregate_handling":"STANDARD_HANDLING"}}]}',)]
```

Read single fields from the tree by path, with `-&gt;&gt;` or [json_extract_string](/develop/docs/dev-database-reference-sql-functions-json-json-extract-string):

```
run("SELECT json_serialize_sql('SELECT 1 AS x') ->> '$.error', "
    "json_serialize_sql('SELECT 1 AS x') ->> '$.statements[0].node.type'")
run("SELECT json_array_length(json_serialize_sql('SELECT 1; SELECT 2'), '$.statements')")
```

```
[('false', 'SELECT_NODE')]
[(2,)]
```

## NULLs and edge cases

| Input | Result |
| --- | --- |
| `NULL` | `NULL`, with or without options |
| `&apos;&apos;` (empty text) | `&quot;error&quot;: false` |
| `&apos;WITH a AS (SELECT 1 AS x) SELECT x FROM a&apos;` | `&quot;error&quot;: false` |
| `&apos;INSERT INTO x VALUES (1)&apos;` | `&quot;error&quot;: true` |
| `&apos;not sql&apos;` | `&quot;error&quot;: true`, `error_type` `parser`, `error_message` `syntax error at or near &quot;not&quot;` |
| `skip_null := true` | a shorter document |
| `format := true` | a longer document, over several lines |

```
run("SELECT json_serialize_sql(NULL), json_serialize_sql(NULL, skip_null := true)")
run("SELECT json_serialize_sql('') ->> '$.error', "
    "json_serialize_sql('WITH a AS (SELECT 1 AS x) SELECT x FROM a') ->> '$.error', "
    "json_serialize_sql('INSERT INTO x VALUES (1)') ->> '$.error'")
run("SELECT json_serialize_sql('not sql') ->> '$.error', json_serialize_sql('not sql') ->> '$.error_type', "
    "json_serialize_sql('not sql') ->> '$.error_message'")
run("SELECT length(json_serialize_sql('SELECT 1')) > length(json_serialize_sql('SELECT 1', skip_null := true)), "
    "length(json_serialize_sql('SELECT 1', format := true)) > length(json_serialize_sql('SELECT 1'))")
```

```
[(None, None)]
[('false', 'false', 'true')]
[('true', 'parser', 'syntax error at or near "not"')]
[(True, True)]
```

## Examples

Check each saved query, and count its statements. The `INSERT` gives `true` for the error, and the NULL row gives NULL:

```
run(f"SELECT id, json_serialize_sql(q) ->> '$.error', json_array_length(json_serialize_sql(q), '$.statements') "
    f"FROM {TABLE} ORDER BY id LIMIT 10")
```

```
[(1, 'false', 1), (2, 'false', 1), (3, 'false', 2), (4, 'true', None), (5, None, None)]
```

Find the table that each single-statement query reads:

```
run(f"SELECT id, json_serialize_sql(q) ->> '$.statements[0].node.from_table.table_name' "
    f"FROM {TABLE} WHERE id < 3 ORDER BY id LIMIT 10")
```

```
[(1, 'people'), (2, 't')]
```

## How it runs in VAST

The Query Engine runs on the cluster's CNodes, and a query on a table starts with a scan in VAST DataBase (`VAST_SCAN`). `json_serialize_sql` parses the text of each row in the Query Engine; the statements aren't run. In the `SELECT` list, the scan reads the column and a projection step parses it for each row it returns:

```
plan(f"SELECT json_serialize_sql(q) FROM {TABLE}")
```

```
┌───────────────────────────┐
│         PROJECTION        │
│    ────────────────────   │
│   json_serialize_sql(q)   │
│                           │
└─────────────┬─────────────┘
┌─────────────┴─────────────┐
│         VAST_SCAN         │
│    ────────────────────   │
│    Function: VAST_SCAN    │
│       Projections: q      │
│                           │
└───────────────────────────┘
```

## Make it faster

- **Read only the fields you need.** A whole tree is large even for a short query; extract the fields by path, as in the examples, rather than returning the document.
- **Select the rows with a condition on a column.** A comparison such as `id &lt; 3` is evaluated in VAST DataBase, so only those rows are parsed.
- **Use `skip_null` and `skip_empty`** when you do return the tree: they leave out the members that carry no information.

## Limitations and errors

- **`SELECT` statements only.** Other statements give `&quot;error&quot;: true` instead of a tree.
- **Don't return the whole error object.** Besides `error`, `error_type` and `error_message`, it holds long diagnostic fields, such as a server stack trace. Read the three fields by path.
- **The options are named.** A positional option, such as `json_serialize_sql(&apos;SELECT 1&apos;, 1)`, fails with `No function matches the given name and argument types &apos;json_serialize_sql(STRING_LITERAL, INTEGER_LITERAL)&apos;`. A misspelt name fails with `json_serialize_sql: Unknown argument &apos;skip_nulls&apos;`.
- **Text only.** A number or a boolean fails with `No function matches the given name and argument types`.
- **A tree written with `skip_null` and `skip_empty` can't be turned back into SQL** by [json_deserialize_sql](/develop/docs/dev-database-reference-sql-functions-json-json-deserialize-sql). Serialize without them if you need the SQL again.

```
run("SELECT json_serialize_sql('SELECT 1', 1)")
run("SELECT json_serialize_sql('SELECT 1', skip_nulls := true)")
run("SELECT json_serialize_sql(123)")
```

```
InternalError: Binder Error: No function matches the given name and argument types 'json_serialize_sql(STRING_LITERAL, INTEGER_LITERAL)'. You might need to add explicit type casts.
InternalError: Binder Error: json_serialize_sql: Unknown argument 'skip_nulls'
InternalError: Binder Error: No function matches the given name and argument types 'json_serialize_sql(INTEGER_LITERAL)'. You might need to add explicit type casts.
```

## Similar functions

| Function | What it does | Result for `&apos;SELECT 1 AS x&apos;` |
| --- | --- | --- |
| `json_serialize_sql` | SQL text to a JSON syntax tree | a JSON document with `&quot;error&quot;: false` |
| [json_deserialize_sql](/develop/docs/dev-database-reference-sql-functions-json-json-deserialize-sql) | a JSON syntax tree back to SQL text | `json_deserialize_sql(json_serialize_sql(&apos;SELECT 1 AS x&apos;))` gives `SELECT 1 AS x` |
| [EXPLAIN](/develop/docs/dev-database-reference-sql-explain) | the plan the Query Engine runs | (a statement, not a function) |

```
run("SELECT json_deserialize_sql(json_serialize_sql('SELECT 1 AS x'))")
```

```
[('SELECT 1 AS x',)]
```

Clean up:

```
run(f"DROP TABLE IF EXISTS {TABLE}")
run(f"DROP SCHEMA IF EXISTS {SCHEMA}")
conn.close()
```

```
[]
[]
```

## See also

- [Parsing and JSON functions](/develop/docs/dev-database-reference-sql-functions-json): all the JSON functions.
- [json_deserialize_sql](/develop/docs/dev-database-reference-sql-functions-json-json-deserialize-sql) and [json_extract_string](/develop/docs/dev-database-reference-sql-functions-json-json-extract-string).
- [EXPLAIN](/develop/docs/dev-database-reference-sql-explain): how to read a plan.