mirror of
https://github.com/Infinidat/infi.clickhouse_orm.git
synced 2024-11-25 18:23:44 +03:00
documentation
This commit is contained in:
parent
5510de597d
commit
3e6677f939
11
README.rst
11
README.rst
|
@ -107,6 +107,9 @@ It is possible to select only a subset of the columns, and the rest will receive
|
||||||
for person in db.select("SELECT first_name FROM my_test_db.person WHERE last_name='Smith'", model_class=Person):
|
for person in db.select("SELECT first_name FROM my_test_db.person WHERE last_name='Smith'", model_class=Person):
|
||||||
print person.first_name
|
print person.first_name
|
||||||
|
|
||||||
|
Ad-Hoc Models
|
||||||
|
*************
|
||||||
|
|
||||||
Specifying a model class is not required. In case you do not provide a model class, an ad-hoc class will
|
Specifying a model class is not required. In case you do not provide a model class, an ad-hoc class will
|
||||||
be defined based on the column names and types returned by the query:
|
be defined based on the column names and types returned by the query:
|
||||||
|
|
||||||
|
@ -115,6 +118,9 @@ be defined based on the column names and types returned by the query:
|
||||||
for row in db.select("SELECT max(height) as max_height FROM my_test_db.person"):
|
for row in db.select("SELECT max(height) as max_height FROM my_test_db.person"):
|
||||||
print row.max_height
|
print row.max_height
|
||||||
|
|
||||||
|
This is a very convenient feature that saves you the need to define a model for each query, while still letting
|
||||||
|
you work with Pythonic column values and an elegant syntax.
|
||||||
|
|
||||||
Counting
|
Counting
|
||||||
--------
|
--------
|
||||||
|
|
||||||
|
@ -139,7 +145,7 @@ StringField String unicode Encoded as UTF-8 when written to C
|
||||||
DateField Date datetime.date Range 1970-01-01 to 2038-01-19
|
DateField Date datetime.date Range 1970-01-01 to 2038-01-19
|
||||||
DateTimeField DateTime datetime.datetime Minimal value is 1970-01-01 00:00:00; Always in UTC
|
DateTimeField DateTime datetime.datetime Minimal value is 1970-01-01 00:00:00; Always in UTC
|
||||||
Int8Field Int8 int Range -128 to 127
|
Int8Field Int8 int Range -128 to 127
|
||||||
Int16Field Int16 int Range -32,768 to 32,767
|
Int16Field Int16 int Range -32768 to 32767
|
||||||
Int32Field Int32 int Range -2147483648 to 2147483647
|
Int32Field Int32 int Range -2147483648 to 2147483647
|
||||||
Int64Field Int64 int/long Range -9223372036854775808 to 9223372036854775807
|
Int64Field Int64 int/long Range -9223372036854775808 to 9223372036854775807
|
||||||
UInt8Field UInt8 int Range 0 to 255
|
UInt8Field UInt8 int Range 0 to 255
|
||||||
|
@ -180,6 +186,9 @@ For a ``SummingMergeTree`` you can optionally specify the summing columns:
|
||||||
engine = engines.SummingMergeTree('EventDate', ('OrderID', 'EventDate', 'BannerID'),
|
engine = engines.SummingMergeTree('EventDate', ('OrderID', 'EventDate', 'BannerID'),
|
||||||
summing_cols=('Shows', 'Clicks', 'Cost'))
|
summing_cols=('Shows', 'Clicks', 'Cost'))
|
||||||
|
|
||||||
|
Data Replication
|
||||||
|
****************
|
||||||
|
|
||||||
Any of the above engines can be converted to a replicated engine (e.g. ``ReplicatedMergeTree``) by adding two parameters, ``replica_table_path`` and ``replica_name``:
|
Any of the above engines can be converted to a replicated engine (e.g. ``ReplicatedMergeTree``) by adding two parameters, ``replica_table_path`` and ``replica_name``:
|
||||||
|
|
||||||
.. code:: python
|
.. code:: python
|
||||||
|
|
Loading…
Reference in New Issue
Block a user