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
|
.. change::
:tags: performance, schema
:tickets: 4379
Rearchitected the schema reflection API to allow some dialects to make use
of high performing batch queries to reflect the schemas of many tables at
once using much fewer queries. The new performance features are targeted
first at the PostgreSQL and Oracle backends, and may be applied to any
dialect that makes use of SELECT queries against system catalog tables to
reflect tables (currently this omits the MySQL and SQLite dialects which
instead make use of parsing the "CREATE TABLE" statement, however these
dialects do not have a pre-existing performance issue with reflection. MS
SQL Server is still a TODO).
The new API is backwards compatible with the previous system, and should
require no changes to third party dialects to retain compatibility;
third party dialects can also opt into the new system by implementing
batched queries for schema reflection.
Along with this change is an updated reflection API that is fully
:pep:`484` typed, features many new methods and some changes.
.. change::
:tags: bug, schema
For SQLAlchemy-included dialects for SQLite, PostgreSQL, MySQL/MariaDB,
Oracle, and SQL Server, the :meth:`.Inspector.has_table`,
:meth:`.Inspector.has_sequence`, :meth:`.Inspector.has_index`,
:meth:`.Inspector.get_table_names` and
:meth:`.Inspector.get_sequence_names` now all behave consistently in terms
of caching: they all fully cache their result after being called the first
time for a particular :class:`.Inspector` object. Programs that create or
drop tables/sequences while calling upon the same :class:`.Inspector`
object will not receive updated status after the state of the database has
changed. A call to :meth:`.Inspector.clear_cache` or a new
:class:`.Inspector` should be used when DDL changes are to be executed.
Previously, the :meth:`.Inspector.has_table`,
:meth:`.Inspector.has_sequence` methods did not implement caching nor did
the :class:`.Inspector` support caching for these methods, while the
:meth:`.Inspector.get_table_names` and
:meth:`.Inspector.get_sequence_names` methods were, leading to inconsistent
results between the two types of method.
Behavior for third party dialects is dependent on whether or not they
implement the "reflection cache" decorator for the dialect-level
implementation of these methods.
.. change::
:tags: change, schema
:tickets: 4379
Improvements to the :class:`.Inspector` object:
* added a method
:meth:`.Inspector.has_schema` that returns if a schema
is present in the target database
* added a method :meth:`.Inspector.has_index` that returns if a table has
a particular index.
* Inspection methods such as :meth:`.Inspector.get_columns` that work
on a single table at a time should now all consistently
raise :class:`_exc.NoSuchTableError` if a
table or view is not found; this change is specific to individual
dialects, so may not be the case for existing third-party dialects.
* Separated the handling of "views" and "materialized views", as in
real world use cases, these two constructs make use of different DDL
for CREATE and DROP; this includes that there are now separate
:meth:`.Inspector.get_view_names` and
:meth:`.Inspector.get_materialized_view_names` methods.
|
