Snuba SDK for generating SnQL queries.
Project description
See here for the full documentation.
Status
Examples
Snuba SDK is a tool that allows requests to Snuba to be built programatically. A Request consists of a Query, the dataset the Query is targeting, the AppID of the Request, and any flags for the Request. A Query object is a code representation of a SnQL or MQL query, and has a number of attributes corresponding to different parts of the query.
Requests and Queries can be created directly:
request = Request(
dataset = "discover",
app_id = "myappid",
tenant_ids = {"referrer": "my_referrer", "organization_id": 1234}
query = Query(
match=Entity("events"),
select=[
Column("title"),
Function("uniq", [Column("event_id")], "uniq_events"),
],
groupby=[Column("title")],
where=[
Condition(Column("timestamp"), Op.GT, datetime.datetime(2021, 1, 1)),
Condition(Column("project_id"), Op.IN, Function("tuple", [1, 2, 3])),
],
limit=Limit(10),
offset=Offset(0),
granularity=Granularity(3600),
),
flags = Flags(debug=True)
)Queries can also be built incrementally:
query = (
Query("discover", Entity("events"))
.set_select(
[Column("title"), Function("uniq", [Column("event_id")], "uniq_events")]
)
.set_groupby([Column("title")])
.set_where(
[
Condition(Column("timestamp"), Op.GT, datetime.datetime.(2021, 1, 1)),
Condition(Column("project_id"), Op.IN, Function("tuple", [1, 2, 3])),
]
)
.set_limit(10)
.set_offset(0)
.set_granularity(3600)
)Once the request is built, it can be translated into a Snuba request that can be sent to Snuba.
# Outputs a formatted Snuba request
request.serialize()It can also be printed in a more human readable format.
# Outputs a formatted Snuba request
print(request.print())This outputs:
{
"dataset": "discover",
"app_id": "myappid",
"query": "MATCH (events) SELECT title, uniq(event_id) AS uniq_events BY title WHERE timestamp > toDateTime('2021-01-01T00:00:00.000000') AND project_id IN tuple(1, 2, 3) LIMIT 10 OFFSET 0 GRANULARITY 3600",
"debug": true
}If an expression in the query is invalid (e.g. Column(1)) then an InvalidExpressionError exception will be thrown. If there is a problem with a query, it will throw an InvalidQueryError exception when .validate() or .translate() is called. If there is a problem with the Request or the Flags, an InvalidRequestError or InvalidFlagError will be thrown respectively.
MQL Examples
MQL queries can be built in a similar way to SnQL queries. However they use a MetricsQuery object instead of a Query object. The query argument of a MetricsQuery is either a Timeseries or Formula, which is a mathemtical formula of Timeseries.
The other arguments to the MetricsQuery are meta data about how to run the query, e.g. start/end timestamps, the granularity, limits etc.
MetricsQuery(
query=Formula(
ArithmeticOperator.DIVIDE.value,
[
Timeseries(
metric=Metric(
public_name="transaction.duration",
),
aggregate="sum",
),
1000,
],
),
start=NOW,
end=NOW + timedelta(days=14),
rollup=Rollup(interval=3600, totals=None, granularity=3600),
scope=MetricsScope(
org_ids=[1], project_ids=[11], use_case_id="transactions"
),
limit=Limit(100),
offset=Offset(5),
)Contributing to the SDK
Please refer to CONTRIBUTING.rst.
License
Licensed under FSL-1.0-Apache-2.0, see LICENSE.md.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
File details
Details for the file snuba_sdk-2.0.31-py2.py3-none-any.whl.
File metadata
- Download URL: snuba_sdk-2.0.31-py2.py3-none-any.whl
- Upload date:
- Size: 46.9 kB
- Tags: Python 2, Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | a3a9f20f4661669860b67658594cfab0b7f71493045f854544410a2fba03c41e |
|
| MD5 | 03d14c8bc10ed22e111263abf319037c |
|
| BLAKE2b-256 | ed9a3b4b31482cdb6d85d19da66c411cd743b95ab6d8b166c3e53d7f06007a4c |
