--- title: "|| (concatenation)" slug: "dev-database-reference-sql-functions-string-concat-operator" updated: 2026-10-30T18:18:56Z published: 2026-10-30T18:18:56Z canonical: "docs.vastdata.com/dev-database-reference-sql-functions-string-concat-operator" --- > ## 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. # || (concatenation) *Applies to VAST Cluster 5.5.* *[Functions](/develop/docs/dev-database-reference-sql-functions-index) › [Text basics](/develop/docs/dev-database-reference-sql-functions-string-text-basics) › Concatenation.* The `||` operator joins two strings, two lists or two `BLOB` values. Other values are converted to text first. If either side is NULL, the result is NULL; to skip NULLs instead, use [concat](/develop/docs/dev-database-reference-sql-functions-string-concat). ## Syntax ``` a || b ``` | Argument | Type | Description | | --- | --- | --- | | `a`, `b` | any | The values to join. Two lists give a list, two `BLOB` values give a `BLOB`; otherwise each side is converted to text. A list can't be joined with a value that isn't a list. | **Returns:** `VARCHAR`; a list for two lists; a `BLOB` for two `BLOB` values. **Aliases:** none. The function form with NULL skipping is [concat](/develop/docs/dev-database-reference-sql-functions-string-concat). ## 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_concat_operator`. 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 four files, each with a directory, a name and an extension. File 3 has no extension, and the last row has no values: ``` 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_concat_operator"' 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, "") 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}.files" run(f"CREATE SCHEMA IF NOT EXISTS {SCHEMA}") run(f"CREATE TABLE IF NOT EXISTS {TABLE} (id BIGINT, dir VARCHAR, name VARCHAR, ext VARCHAR)") run(f"INSERT INTO {TABLE} VALUES " "(1, '/data/logs', 'app', 'log'), " "(2, '/data/raw', 'events', 'json'), " "(3, '/data', 'README', NULL), " "(4, NULL, NULL, NULL)") ``` ``` [] [] [(4,)] ``` ## What it does `||` joins its two sides, with nothing between them. Chain it to join more: `a || ' ' || b`. What it returns depends on the types of the two sides: | Sides | Example | Result | | --- | --- | --- | | Two strings | `'Ada' \|\| ' Lovelace'` | `'Ada Lovelace'` | | A string and a number, boolean or date | `'a' \|\| 1`, `'a' \|\| true`, `DATE '2026-10-16' \|\| '-x'` | `'a1'`, `'atrue'`, `'2026-10-16-x'` | | Two numbers | `1 \|\| 2` | `'12'`, a `VARCHAR` | | Two lists | `[1] \|\| [2]` | `[1, 2]`, a list | | Two `BLOB` values | `'\xAA'::BLOB \|\| '\xBB'::BLOB` | the two bytes `AA BB`, a `BLOB` | | A `BLOB` and a string | `'\xAA'::BLOB \|\| '-x'` | `'\xAA-x'`, a `VARCHAR` holding the `BLOB`'s text form | | A `BLOB` and a number | `'\xAA'::BLOB \|\| 1` | `'\xAA1'`, a `VARCHAR` | | A NULL on either side | `'a' \|\| NULL`, `[1] \|\| NULL` | `NULL`, also for a list | The NULL rule applies to the whole chain: `'a' || 'b' || NULL` is `NULL`, so one missing value empties the whole result. This is the main difference from [concat](/develop/docs/dev-database-reference-sql-functions-string-concat), which skips NULLs. Arithmetic binds tighter than `||`, and `||` binds tighter than a comparison: `1 + 2 || 'x'` is `'3x'`, and `'a' || 'b' = 'ab'` is `true`. ## NULLs and edge cases | Input | Result | | --- | --- | | `'a' \|\| NULL` | `NULL` | | `NULL \|\| 'a'` | `NULL` | | `'a' \|\| 'b' \|\| NULL` | `NULL` | | `'' \|\| '-x'` (empty string) | `'-x'` | | `1 \|\| 2` | `'12'` | | `1.5 \|\| '-x'` | `'1.5-x'` | | `'a' \|\| 1` | `'a1'` | | `'a' \|\| true` | `'atrue'` | | `DATE '2026-10-16' \|\| '-x'` | `'2026-10-16-x'` | | `[1] \|\| [2]` | `[1, 2]` | | `'\xAA'::BLOB \|\| '\xBB'::BLOB` | `b'\xaa\xbb'` in Python: a `BLOB` | | `'\xAA'::BLOB \|\| '-x'` | `'\xAA-x'`: a `VARCHAR` (Python shows the backslash doubled) | | A string and a list, such as `'a' \|\| [1]` | Error: `Cannot concatenate types VARCHAR and INTEGER[] - an explicit cast is required` | | A list and a string, such as `['a', 'b'] \|\| '-x'` | Error: `Cannot concatenate types VARCHAR[] and VARCHAR - an explicit cast is required` | When a side is the bare literal `NULL`, the result is a NULL with no string type: `typeof('a' || NULL)` is `"NULL"`, and an Arrow client sees a 32-bit integer column. A NULL from a `VARCHAR` column, as in [Examples](/develop/docs/dev-database-reference-sql-functions-string-concat-operator#examples), gives a `VARCHAR` column. The same cases, as the cluster returns them. They use the table and helpers from [Set up the examples](/develop/docs/dev-database-reference-sql-functions-string-concat-operator#set-up-the-examples): ``` run("SELECT 'a' || NULL, NULL || 'a', 'a' || 'b' || NULL, '' || '-x'") run("SELECT 1 || 2, 1.5 || '-x', 'a' || 1, 'a' || true, DATE '2026-10-16' || '-x'") run("SELECT [1] || [2], '\\xAA'::BLOB || '\\xBB'::BLOB, '\\xAA'::BLOB || '-x'") run("SELECT 'a' || [1]") run("SELECT ['a', 'b'] || '-x'") ``` ``` [(None, None, None, '-x')] [('12', '1.5-x', 'a1', 'atrue', '2026-10-16-x')] [([1, 2], b'\xaa\xbb', '\\xAA-x')] InternalError: Binder Error: Cannot concatenate types VARCHAR and INTEGER[] - an explicit cast is required InternalError: Binder Error: Cannot concatenate types VARCHAR[] and VARCHAR - an explicit cast is required ``` ## Examples The full path of each file. File 3 has no extension, so its whole path is `None`: ``` run(f"SELECT id, dir || '/' || name || '.' || ext FROM {TABLE} ORDER BY id LIMIT 10") ``` ``` [(1, '/data/logs/app.log'), (2, '/data/raw/events.json'), (3, None), (4, None)] ``` To keep the path when the extension is missing, use the NULL rule on the optional part only: `'.' || ext` is NULL when `ext` is, and `coalesce` replaces it with an empty string: ``` run(f"SELECT id, dir || '/' || name || coalesce('.' || ext, '') FROM {TABLE} ORDER BY id LIMIT 10") ``` ``` [(1, '/data/logs/app.log'), (2, '/data/raw/events.json'), (3, '/data/README'), (4, None)] ``` Find a file by its path without the extension: ``` run(f"SELECT id FROM {TABLE} WHERE dir || '/' || name = '/data/logs/app' ORDER BY id LIMIT 10") ``` ``` [(1,)] ``` ## 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`). VAST DataBase can evaluate a comparison of a column with a constant, but not an expression built from a column, so `||` on a column is always computed by the Query Engine: - **In the `SELECT` list,** the scan reads the columns and the Query Engine joins them for each row it returns. - **In `WHERE`,** the scan sends the columns for every row, and a `FILTER` step in the Query Engine joins them and keeps the rows that match. No rows are skipped in storage. The plan of the path example shows the `FILTER` step above the scan: ``` plan(f"SELECT id FROM {TABLE} WHERE dir || '/' || name = '/data/logs/app'") ``` ``` ┌───────────────────────────┐ │ PROJECTION │ │ ──────────────────── │ │ #2 │ │ │ └─────────────┬─────────────┘ ┌─────────────┴─────────────┐ │ FILTER │ │ ──────────────────── │ │ (((dir || '/') || name) = │ │ '/data/logs/app') │ │ │ └─────────────┬─────────────┘ ┌─────────────┴─────────────┐ │ VAST_SCAN │ │ ──────────────────── │ │ Function: VAST_SCAN │ │ │ │ Projections: │ │ dir │ │ name │ │ id │ │ │ └───────────────────────────┘ ``` ## Make it faster - **Compare the parts, not the joined string.** `dir = '/data/logs' AND name = 'app'` finds the same file, but each part is a comparison of a raw column with a constant, which is pushed down to VAST DataBase, so rows that don't match never leave storage. The plan has no `FILTER` step: ``` run(f"SELECT id FROM {TABLE} WHERE dir = '/data/logs' AND name = 'app' ORDER BY id LIMIT 10") plan(f"SELECT id FROM {TABLE} WHERE dir = '/data/logs' AND name = 'app'") ``` ``` [(1,)] ┌───────────────────────────┐ │ VAST_SCAN │ │ ──────────────────── │ │ Function: VAST_SCAN │ │ Projections: id │ │ │ └───────────────────────────┘ ``` - **Store the joined value if you filter on it often.** Compute it when you write the rows and keep it in its own column: `path = '/data/logs/app.log'` is pushed down, `dir || '/' || name || '.' || ext = '...'` isn't. - **Combine it with a pushed-down condition.** In `id > 1 AND dir || '/' || name = '...'`, `id > 1` is still evaluated in VAST DataBase, so `||` runs only on the rows that pass it. ## Limitations and errors - **A list joins only with a list.** `['a', 'b'] || '-x'` and `'a' || [1]` fail with `Cannot concatenate types ... - an explicit cast is required`. To put a list's text form into a string, use [concat_ws](/develop/docs/dev-database-reference-sql-functions-string-concat-ws), which converts lists to text. - **One NULL makes the result NULL.** Use [concat](/develop/docs/dev-database-reference-sql-functions-string-concat) or [concat_ws](/develop/docs/dev-database-reference-sql-functions-string-concat-ws) to skip NULLs, or `coalesce` on the optional parts. - **`BLOB` and string give a string.** Only two `BLOB` values give a `BLOB`. With a string or a number on the other side, the `BLOB` is converted to its text form. ``` run("SELECT ['a', 'b'] || '-x'") run("SELECT concat_ws('', ['a', 'b'], '-x')") ``` ``` InternalError: Binder Error: Cannot concatenate types VARCHAR[] and VARCHAR - an explicit cast is required [('[a, b]-x',)] ``` ## Similar functions | Function | NULL handling | Result for `'a'`, `NULL`, `'b'` (`\|\|` gives `NULL`) | Result for `[1, 2]` and `[3]` (`\|\|` gives `[1, 2, 3]`) | | --- | --- | --- | --- | | [concat](/develop/docs/dev-database-reference-sql-functions-string-concat) | skips NULLs | `concat('a', NULL, 'b')` is `'ab'` | `concat([1, 2], [3])` is `[1, 2, 3]` | | [concat_ws](/develop/docs/dev-database-reference-sql-functions-string-concat-ws) | skips NULLs and the separator next to them | `concat_ws('-', 'a', NULL, 'b')` is `'a-b'` | `concat_ws(',', [1, 2], [3])` is `'[1, 2],[3]'`, a string | | [list_concat](/develop/docs/dev-database-reference-sql-functions-lists-list-concat) | treats a NULL list as empty | (lists only) | `[1, 2, 3]` | | [BLOB concatenation](/develop/docs/dev-database-reference-sql-functions-binary-hash-uuid) | | (`BLOB` values) | (`BLOB` values) | ``` run("SELECT 'a' || NULL || 'b', concat('a', NULL, 'b'), concat_ws('-', 'a', NULL, 'b')") run("SELECT [1, 2] || [3], concat([1, 2], [3]), concat_ws(',', [1, 2], [3]), list_concat([1, 2], [3])") ``` ``` [(None, 'ab', 'a-b')] [([1, 2, 3], [1, 2, 3], '[1, 2],[3]', [1, 2, 3])] ``` Clean up: ``` run(f"DROP TABLE IF EXISTS {TABLE}") run(f"DROP SCHEMA IF EXISTS {SCHEMA}") conn.close() ``` ``` [] [] ``` ## See also - [Text basics](/develop/docs/dev-database-reference-sql-functions-string-text-basics): all the text functions. - [concat](/develop/docs/dev-database-reference-sql-functions-string-concat) and [concat_ws](/develop/docs/dev-database-reference-sql-functions-string-concat-ws): join with NULLs skipped. - [Concatenation of lists](/develop/docs/dev-database-reference-sql-functions-operators-and-predicates#concatenation) and [list_concat](/develop/docs/dev-database-reference-sql-functions-lists-list-concat). - [Comparing and concatenating BLOB values](/develop/docs/dev-database-reference-sql-functions-binary-hash-uuid). - [Where filters run](/develop/docs/dev-database-reference-sql-select#where-filters-run) and [pushdowns](/develop/docs/dev-database-reference-pushdowns). - [EXPLAIN](/develop/docs/dev-database-reference-sql-explain): how to read a plan.