flowscope-core 0.7.0

Core SQL lineage analysis engine
Documentation 
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
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108

# Additional dialect behavior rules

[value_table_functions]
# Functions that return rows (not scalar values)
common = ["UNNEST", "GENERATE_SERIES", "JSON_TABLE"]
postgres = ["UNNEST", "GENERATE_SERIES", "GENERATE_SUBSCRIPTS", "REGEXP_MATCHES"]
bigquery = ["UNNEST"]
snowflake = ["FLATTEN", "SPLIT_TO_TABLE", "STRTOK_SPLIT_TO_TABLE"]
redshift = ["UNNEST"]
mysql = ["JSON_TABLE"]
mssql = ["OPENJSON", "STRING_SPLIT"]
duckdb = ["UNNEST", "GENERATE_SERIES", "RANGE"]
clickhouse = ["ARRAY_JOIN"]
databricks = ["EXPLODE", "EXPLODE_OUTER", "POSEXPLODE", "POSEXPLODE_OUTER", "INLINE", "INLINE_OUTER"]
hive = ["EXPLODE", "POSEXPLODE", "INLINE", "JSON_TUPLE", "PARSE_URL_TUPLE"]
sqlite = []

[null_ordering]
# Where NULLs appear in ORDER BY by default
# nulls_are_large = NULLS LAST (Postgres, Oracle, Snowflake)
# nulls_are_small = NULLS FIRST (BigQuery, MySQL, Spark)
# nulls_are_last = NULLS LAST always
bigquery = "nulls_are_small"
clickhouse = "nulls_are_last"
databricks = "nulls_are_small"
doris = "nulls_are_small"
drill = "nulls_are_last"
duckdb = "nulls_are_last"
hive = "nulls_are_small"
mysql = "nulls_are_small"
oracle = "nulls_are_large"
postgres = "nulls_are_large"
presto = "nulls_are_last"
redshift = "nulls_are_large"
snowflake = "nulls_are_large"
spark = "nulls_are_small"
sqlite = "nulls_are_small"
starrocks = "nulls_are_small"
teradata = "nulls_are_small"
trino = "nulls_are_last"
mssql = "nulls_are_small"

[unnest]
# Dialects with implicit UNNEST support (no explicit CROSS JOIN needed)
implicit_unnest = ["bigquery", "redshift"]

[date_functions]
# Arguments to skip during lineage analysis (unit/part literals)
# Format: function.dialect = [arg_indices_to_skip]

[date_functions.DATEDIFF]
# NOTE: Postgres does NOT have DATEDIFF - uses AGE(), DATE_PART(), or subtraction
# Spark/Databricks have 2-arg datediff(end, start) returning days - no unit arg
redshift = [0]
snowflake = [0]
presto = [0]
trino = [0]
mssql = [0]
mysql = []
hive = []
bigquery = []
duckdb = []
spark = []
databricks = []

[date_functions.DATE_TRUNC]
postgres = [0]
snowflake = [0]
redshift = [0]
presto = [0]
trino = [0]
bigquery = [1]
spark = [0]
databricks = [0]
duckdb = [0]

[date_functions.DATE_ADD]
postgres = []
mysql = []
hive = []
presto = [0]
trino = [0]
bigquery = []
snowflake = [0]
spark = []
mssql = [0]

[date_functions.EXTRACT]
_default = [0]

[date_functions.DATE_PART]
postgres = [0]
snowflake = [0]
redshift = [0]

[date_functions.TIMESTAMP_ADD]
# BigQuery: TIMESTAMP_ADD(timestamp, INTERVAL n UNIT) - arg 0 is timestamp (keep), arg 1 is interval (skip)
bigquery = [1]
# Presto/Trino: date_add(unit, value, timestamp) - arg 0 is unit (skip)
presto = [0]
trino = [0]
# Snowflake: TIMESTAMPADD(unit, value, timestamp) - arg 0 is unit (skip)
snowflake = [0]

[date_functions.TIMESTAMP_SUB]
# Same pattern as TIMESTAMP_ADD
bigquery = [1]