File: test_max_line_size.test_slow

package info (click to toggle)
duckdb 1.5.1-2
  • links: PTS, VCS
  • area: main
  • in suites:
  • size: 299,196 kB
  • sloc: cpp: 865,414; ansic: 57,292; python: 18,871; sql: 12,663; lisp: 11,751; yacc: 7,412; lex: 1,682; sh: 747; makefile: 558
file content (37 lines) | stat: -rw-r--r-- 1,198 bytes parent folder | download | duplicates (4)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
# name: test/sql/copy/csv/test_max_line_size.test_slow
# description: Test lines that exceed the maximum line size
# group: [csv]

statement ok
pragma enable_verification;

# generate CSV file with some long strings some exceeding the maximum line size
statement ok
CREATE TABLE test_base (a INT, b VARCHAR, c INT);

statement ok
INSERT INTO test_base VALUES
    (10, REPEAT('a', 1504857), 20),
    (10, REPEAT('a', 10485760), 20),

statement ok
COPY test_base TO '__TEST_DIR__/test.csv' (HEADER 0)

# value is too big for loading
statement ok
CREATE TABLE test (a INTEGER, b VARCHAR, c INTEGER);

# Error is thrown when line does not fit maximum lines size: note that it may not always throw for exceeding the line
# size by a small margin as the calculation is an estimate based on reads before parsing.
statement error
COPY test FROM '__TEST_DIR__/test.csv';
----
Possible Solution: Change the maximum length size, e.g., max_line_size=10485769

# we can override the max line size
statement ok
COPY test FROM '__TEST_DIR__/test.csv' (max_line_size 10485770);

# also in the read_csv call
statement ok
INSERT INTO test SELECT * FROM read_csv_auto('__TEST_DIR__/test.csv', max_line_size=10485770);