jan/openreplay
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Dev (#3034)

Browse source 
* fix(chalice): changed trend - group by users

* fix(chalice): fixed partial right table issue
This commit is contained in:
Kraiem Taha Yassine 2025-02-13 18:12:21 +01:00 • committed by GitHub
parent ea142b9596
commit b70a641af5
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 39 additions and 37 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
api/chalicelib/core/errors/errors.py
View file

@ -149,11 +149,7 @@ def search(data: schemas.SearchErrorsSchema, project: schemas.ProjectContext, us
params["error_query"] = helper.values_for_operator(value=data.query,
op=schemas.SearchEventOperator.CONTAINS)
main_pg_query = f"""WITH raw_data AS (SELECT DISTINCT session_id
FROM events.errors
{"INNER JOIN public.sessions USING(session_id)" if platform else ""}
WHERE {" AND ".join(pg_sub_query_chart)})
SELECT full_count,
main_pg_query = f"""SELECT full_count,
error_id,
name,
message,
@ -187,11 +183,11 @@ def search(data: schemas.SearchErrorsSchema, project: schemas.ProjectContext, us
FROM (SELECT generated_timestamp AS timestamp,
COUNT(session_id) AS count
FROM generate_series(%(startDate)s, %(endDate)s, %(step_size)s) AS generated_timestamp
LEFT JOIN LATERAL (SELECT *
FROM raw_data
UNION ALL
SELECT NULL AS session_id
WHERE NOT EXISTS(SELECT 1 FROM raw_data)) AS sessions ON (TRUE)
LEFT JOIN LATERAL (SELECT DISTINCT session_id
FROM events.errors
{"INNER JOIN public.sessions USING(session_id)" if platform else ""}
WHERE {" AND ".join(pg_sub_query_chart)}
) AS sessions ON (TRUE)
GROUP BY timestamp
ORDER BY timestamp) AS chart_details) AS chart_details ON (TRUE);"""

2
api/chalicelib/core/errors/errors_ch.py
View file

@ -148,7 +148,7 @@ def search(data: schemas.SearchErrorsSchema, project: schemas.ProjectContext, us
if len(data.events) > errors_condition_count:
subquery_part_args, subquery_part = sessions.search_query_parts_ch(data=data, error_status=data.status,
errors_only=True,
project_id=project_id, user_id=user_id,
project_id=project.project_id, user_id=user_id,
issue=None,
favorite_only=False)
subquery_part = f"INNER JOIN {subquery_part} USING(session_id)"

40
api/chalicelib/core/sessions/sessions_ch.py
View file

@ -34,40 +34,27 @@ def search2_series(data: schemas.SessionsSearchPayloadSchema, project_id: int, d
with ch_client.ClickHouseClient() as cur:
if metric_type == schemas.MetricType.TIMESERIES:
if metric_of == schemas.MetricOfTimeseries.SESSION_COUNT:
query = f"""WITH raw_data AS (SELECT s.session_id AS session_id,
s.datetime AS datetime
{query_part})
SELECT gs.generate_series AS timestamp,
query = f"""SELECT gs.generate_series AS timestamp,
COALESCE(COUNT(DISTINCT processed_sessions.session_id),0) AS count
FROM generate_series(%(startDate)s, %(endDate)s, %(step_size)s) AS gs
LEFT JOIN (SELECT *
FROM raw_data
UNION ALL
SELECT NULL AS session_id,
NULL AS datetime
WHERE NOT EXISTS(SELECT 1 FROM raw_data)) AS processed_sessions ON(TRUE)
LEFT JOIN (SELECT s.session_id AS session_id,
s.datetime AS datetime
{query_part}) AS processed_sessions ON(TRUE)
WHERE processed_sessions.datetime >= toDateTime(timestamp / 1000)
AND processed_sessions.datetime < toDateTime((timestamp + %(step_size)s) / 1000)
GROUP BY timestamp
ORDER BY timestamp;"""
elif metric_of == schemas.MetricOfTimeseries.USER_COUNT:
query = f"""WITH raw_data AS (SELECT s.user_id AS user_id,
s.datetime AS datetime
{query_part}
WHERE isNotNull(s.user_id)
AND s.user_id != '')
SELECT gs.generate_series AS timestamp,
query = f"""SELECT gs.generate_series AS timestamp,
COALESCE(COUNT(DISTINCT processed_sessions.user_id),0) AS count
FROM generate_series(%(startDate)s, %(endDate)s, %(step_size)s) AS gs
LEFT JOIN (SELECT *
FROM raw_data
UNION ALL
SELECT NULL AS user_id,
NULL AS datatime
WHERE NOT EXISTS(SELECT 1 FROM raw_data)) AS processed_sessions ON(TRUE)
WHERE processed_sessions.user_id IS NULL
OR processed_sessions.datetime >= toDateTime(timestamp / 1000)
AND processed_sessions.datetime < toDateTime((timestamp + %(step_size)s) / 1000)
LEFT JOIN (SELECT multiIf(s.user_id IS NOT NULL AND s.user_id != '', s.user_id,
s.user_anonymous_id IS NOT NULL AND s.user_anonymous_id != '',
s.user_anonymous_id, toString(s.user_uuid)) AS user_id,
s.datetime AS datetime
{query_part}) AS processed_sessions ON(TRUE)
WHERE processed_sessions.datetime >= toDateTime(timestamp / 1000)
AND processed_sessions.datetime < toDateTime((timestamp + %(step_size)s) / 1000)
GROUP BY timestamp
ORDER BY timestamp;"""
else:
@ -134,7 +121,8 @@ def search2_series(data: schemas.SessionsSearchPayloadSchema, project_id: int, d
s.pop("main_count")
sessions = {"count": count, "values": helper.list_to_camel_case(sessions)}
return sessions
return helper.complete_missing_steps(rows=sessions, start_timestamp=data.startTimestamp,
end_timestamp=data.endTimestamp, step=step_size, neutral={"count": 0})
def search2_table(data: schemas.SessionsSearchPayloadSchema, project_id: int, density: int,

18
api/chalicelib/utils/helper.py
View file

@ -334,3 +334,21 @@ def cast_session_id_to_string(data):
for key in keys:
data[key] = cast_session_id_to_string(data[key])
return data
from typing import List
def complete_missing_steps(rows: List[dict], start_timestamp: int, end_timestamp: int, step: int, neutral: dict,
time_key: str = "timestamp") -> List[dict]:
result = []
i = 0
for t in range(start_timestamp, end_timestamp, step):
if i >= len(rows) or rows[i][time_key] > t:
neutral[time_key] = t
result.append(neutral.copy())
elif i < len(rows) and rows[i][time_key] == t:
result.append(rows[i])
i += 1
return result