fix: Avoid implicit join when using join with unnest

[snapiri]

When using JOIN with UNNEST statements, and then creating a SELECT statement based on it, the UNNESTed table will appear twice in the FROM clause, causing an implicit join of the table with itself

Sample of problematic code:

from sqlalchemy import ARRAY, Column, create_engine, func, Integer, MetaData, Table, column, select
db_metadata = MetaData()
engine = create_engine("bigquery://", future=True)
column_defs1 = [Column("id", Integer, nullable=True)]
column_defs2 = [Column("ids", ARRAY(Integer), nullable=True), Column("dummy", Integer, nullable=True)]
table1 = Table("table1", db_metadata, *column_defs1)
table2 = Table("table2", db_metadata, *column_defs2)
unnested_col_name = "unnested_ids"
unnested_ids = func.unnest(table2.c.ids).alias(unnested_col_name)
unnested_id_col = column(unnested_col_name)
q = select(table1.c.id, table2.c.dummy).select_from(unnested_ids.join(table1, table1.c.id == unnested_id_col))
# This is where things go bad
q = select("*").select_from(q.subquery())
compiled = q.compile(engine)
print(str(compiled))

Generates the following output:

SELECT *
FROM (SELECT `table1`.`id` AS `id`, `table2`.`dummy` AS `dummy`
FROM `table2`, `table2` `table2_1`, unnest(`table2_1`.`ids`) AS `unnested_ids` JOIN `table1` ON `table1`.`id` = `unnested_ids`) AS `anon_1`

Co-Authored-By: @yuval-bavli

Fixes#368

[snapiri]

@tswast / @prash-mi Would appreciate your review if possible.

[snapiri]

@chalmerlowe / @Linchin,
kind reminder - could you please review?

[Linchin]

Hi @snapiri, cover test is also failing

nox > coverage report --show-missing --fail-under=100
Name Stmts Miss Branch BrPart Cover Missing
------------------------------------------------------------------------------------
sqlalchemy_bigquery/__init__.py 12 0 0 0 100%
sqlalchemy_bigquery/_helpers.py 43 0 14 0 100%
sqlalchemy_bigquery/_struct.py 47 0 16 0 100%
sqlalchemy_bigquery/_types.py 53 0 18 0 100%
sqlalchemy_bigquery/base.py 497 0 152 2 99% 275->281, 276->281
sqlalchemy_bigquery/geography.py 51 0 14 0 100%
sqlalchemy_bigquery/parse_url.py 125 0 56 0 100%
sqlalchemy_bigquery/version.py 1 0 0 0 100%
tests/unit/__init__.py 0 0 0 0 100%
tests/unit/conftest.py 49 0 4 0 100%
tests/unit/fauxdbi.py 214 0 70 0 100%
tests/unit/test__struct.py 33 0 0 0 100%
tests/unit/test_catalog_functions.py 77 0 6 0 100%
tests/unit/test_comments.py 28 0 0 0 100%
tests/unit/test_compiler.py 68 0 0 0 100%
tests/unit/test_compliance.py 61 0 8 0 100%
tests/unit/test_dialect_types.py 7 0 2 0 100%
tests/unit/test_engine.py 30 0 0 0 100%
tests/unit/test_geography.py 58 0 0 0 100%
tests/unit/test_helpers.py 94 0 0 0 100%
tests/unit/test_like_reescape.py 17 0 0 0 100%
tests/unit/test_parse_url.py 90 0 10 0 100%
tests/unit/test_select.py 157 0 22 0 100%
tests/unit/test_sqlalchemy_bigquery.py 76 0 4 0 100%
tests/unit/test_version.py 6 0 0 0 100%
tests/unit/test_view.py 7 0 0 0 100%
------------------------------------------------------------------------------------
TOTAL 1901 0 396 2 99%
Coverage failure: total of 99 is less than fail-under=100

[snapiri]

Hi @Linchin,

Thanks for the feedback.
I fixed cover and added more tests.
Would appreciate any feedback

[snapiri]

@tswast / @Linchin / @chalmerlowe
Anyone care to review?

[snapiri]

@tswast - Not sure I know how to read the Kokoro results - I see that the system-3.8 tests failed with the following error:

google.api_core.exceptions.ServiceUnavailable: 503 GET https://bigquery.googleapis.com/bigquery/v2/projects/precise-truck-742/jobs/72b3d73c-939c-4d7f-ab50-06f073ecac58?location=US&prettyPrint=false: The service is currently unavailable.

Also I see that there were 6 runs where the first failed and the 2-6 succeeded.

Anything I should change?

[tswast]

@snapiri That looks like a backend issue, not anything related to this code. I'll trigger a re-run.