Skip to content

Commit

Permalink
feat: Improve git handler and introduce caching
Browse files Browse the repository at this point in the history
  • Loading branch information
dominik003 committed Aug 5, 2024
1 parent 72a2860 commit 90f51f1
Show file tree
Hide file tree
Showing 11 changed files with 276 additions and 219 deletions.
40 changes: 40 additions & 0 deletions backend/capellacollab/core/cache.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,40 @@
# SPDX-FileCopyrightText: Copyright DB InfraGO AG and contributors
# SPDX-License-Identifier: Apache-2.0


import abc


class Cache(abc.ABC):
@abc.abstractmethod
def get(self, key: str) -> bytes | None:
pass

@abc.abstractmethod
def set(self, key: str, value: bytes) -> None:
pass

@abc.abstractmethod
def delete(self, key: str) -> None:
pass

@abc.abstractmethod
def clear(self) -> None:
pass


class InMemoryCache(Cache):
def __init__(self) -> None:
self.cache: dict[str, bytes] = {}

def get(self, key: str) -> bytes | None:
return self.cache.get(key, None)

def set(self, key: str, value: bytes) -> None:
self.cache[key] = value

def delete(self, key: str) -> None:
self.cache.pop(key)

def clear(self) -> None:
self.cache.clear()
4 changes: 2 additions & 2 deletions backend/capellacollab/projects/toolmodels/diagrams/routes.py
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,7 @@ async def get_diagram_metadata(
) = await handler.get_file_from_repository_or_artifacts_as_json(
"diagram_cache/index.json",
"update_capella_diagram_cache",
"diagram-cache/" + handler.git_model.revision,
"diagram-cache/" + handler.revision,
)
except requests.exceptions.HTTPError:
logger.info("Failed fetching diagram metadata", exc_info=True)
Expand Down Expand Up @@ -83,7 +83,7 @@ async def get_diagram(
_, diagram = await handler.get_file_from_repository_or_artifacts(
f"diagram_cache/{parse.quote(diagram_uuid, safe='')}.svg",
"update_capella_diagram_cache",
"diagram-cache/" + handler.git_model.revision,
"diagram-cache/" + handler.revision,
)
except requests.exceptions.HTTPError:
logger.info("Failed fetching diagram", exc_info=True)
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -71,19 +71,6 @@ def __init__(self, filename: str):
)


class GitInstanceAPIEndpointNotFoundError(core_exceptions.BaseError):
def __init__(self):
super().__init__(
status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
title="Git instance API endpoint not found",
reason=(
"The used Git instance has no API endpoint defined. "
"Please contact your administrator."
),
err_code="GIT_INSTANCE_NO_API_ENDPOINT_DEFINED",
)


class GitPipelineJobNotFoundError(core_exceptions.BaseError):
def __init__(self, job_name: str, revision: str):
super().__init__(
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,6 @@
import base64
import datetime
import io
import json
import typing as t
import zipfile
from urllib import parse
Expand All @@ -18,63 +17,37 @@


class GithubHandler(handler.GitHandler):
async def get_project_id_by_git_url(self) -> str:
@classmethod
async def get_project_id_by_git_url(cls, path: str, *_) -> str:
# Project ID has the format '{owner}/{repo_name}'
return parse.urlparse(self.git_model.path).path[1:]

async def get_last_job_run_id_for_git_model(
self, job_name: str, project_id: str | None = None
) -> tuple[str, str]:
if not project_id:
project_id = await self.get_project_id_by_git_url()
jobs = self.get_last_pipeline_runs(project_id)
return parse.urlparse(path).path[1:]

async def get_last_job_run_id(self, job_name: str) -> tuple[str, str]:
jobs = self.get_last_pipeline_runs()
latest_job = self.__get_latest_successful_job(jobs, job_name)
return (latest_job["id"], latest_job["created_at"])

def get_artifact_from_job_as_json(
self,
project_id: str,
job_id: str,
trusted_path_to_artifact: str,
) -> dict:
return json.loads(
self.get_artifact_from_job(
project_id,
job_id,
trusted_path_to_artifact,
)
)

def get_artifact_from_job_as_content(
self,
project_id: str,
job_id: str,
trusted_path_to_artifact: str,
self, job_id: str, trusted_path_to_artifact: str
) -> bytes:
return self.get_artifact_from_job(
project_id,
job_id,
trusted_path_to_artifact,
job_id, trusted_path_to_artifact
).encode()

def __get_file_from_repository(
self,
project_id: str,
trusted_file_path: str,
revision: str,
headers: dict[str, str] | None = None,
) -> requests.Response:
return requests.get(
f"{self.git_instance.api_url}/repos/{project_id}/contents/{parse.quote(trusted_file_path)}?ref={parse.quote(revision, safe='')}",
f"{self.api_url}/repos/{self.project_id}/contents/{parse.quote(trusted_file_path)}?ref={parse.quote(revision, safe='')}",
timeout=config.requests.timeout,
headers=headers,
)

async def get_file_from_repository(
self,
project_id: str,
trusted_file_path: str,
revision: str | None = None,
def get_file_from_repository(
self, trusted_file_path: str, revision: str | None = None
) -> bytes:
"""
If a repository is public but the permissions are not set correctly, you might be able to download the file without authentication
Expand All @@ -83,15 +56,14 @@ async def get_file_from_repository(
For that purpose first we try to reach it without authentication and only if that fails try to get the file authenticated.
"""
response = self.__get_file_from_repository(
project_id, trusted_file_path, revision or self.git_model.revision
trusted_file_path, revision or self.revision
)

if not response.ok and self.git_model.password:
if not response.ok and self.password:
response = self.__get_file_from_repository(
project_id,
trusted_file_path,
revision=revision or self.git_model.revision,
headers=self.__get_headers(self.git_model.password),
revision=revision or self.revision,
headers=self.__get_headers(self.password),
)

if response.status_code == 404:
Expand All @@ -102,32 +74,26 @@ async def get_file_from_repository(

return base64.b64decode(response.json()["content"])

def get_last_pipeline_runs(
self,
project_id: str,
) -> t.Any:
def get_last_pipeline_runs(self) -> t.Any:
headers = None
if self.git_model.password:
headers = self.__get_headers(self.git_model.password)
if self.password:
headers = self.__get_headers(self.password)
response = requests.get(
f"{self.git_instance.api_url}/repos/{project_id}/actions/runs?branch={parse.quote(self.git_model.revision, safe='')}&per_page=20",
f"{self.api_url}/repos/{self.project_id}/actions/runs?branch={parse.quote(self.revision, safe='')}&per_page=20",
headers=headers,
timeout=config.requests.timeout,
)
response.raise_for_status()
return response.json()["workflow_runs"]

def get_artifact_from_job(
self,
project_id: str,
job_id: str,
trusted_path_to_artifact: str,
self, job_id: str, trusted_path_to_artifact: str
) -> str:
artifact = self.__get_latest_artifact_metadata(project_id, job_id)
artifact = self.__get_latest_artifact_metadata(job_id)
artifact_id = artifact["id"]
artifact_response = requests.get(
f"{self.git_instance.api_url}/repos/{project_id}/actions/artifacts/{artifact_id}/zip",
headers=self.__get_headers(self.git_model.password),
f"{self.api_url}/repos/{self.project_id}/actions/artifacts/{artifact_id}/zip",
headers=self.__get_headers(self.password),
timeout=config.requests.timeout,
)
artifact_response.raise_for_status()
Expand All @@ -137,14 +103,12 @@ def get_artifact_from_job(
)

def get_last_updated_for_file_path(
self, project_id: str, file_path: str, revision: str | None
self, file_path: str, revision: str | None
) -> datetime.datetime | None:
response = requests.get(
f"{self.git_instance.api_url}/repos/{project_id}/commits?path={file_path}&sha={revision or self.git_model.revision}",
f"{self.api_url}/repos/{self.project_id}/commits?path={file_path}&sha={revision or self.revision}",
headers=(
self.__get_headers(self.git_model.password)
if self.git_model.password
else None
self.__get_headers(self.password) if self.password else None
),
timeout=config.requests.timeout,
)
Expand All @@ -169,7 +133,7 @@ def __get_latest_successful_job(self, jobs: list, job_name: str) -> dict:
matched_jobs = [job for job in jobs if job["name"] == job_name]
if not matched_jobs:
raise git_exceptions.GitPipelineJobNotFoundError(
job_name=job_name, revision=self.git_model.revision
job_name=job_name, revision=self.revision
)
matched_jobs.sort(key=lambda job: job["created_at"], reverse=True)
if matched_jobs[0]["conclusion"] == "success":
Expand All @@ -184,10 +148,10 @@ def __get_latest_successful_job(self, jobs: list, job_name: str) -> dict:
job_name, matched_jobs[0]["conclusion"]
)

def __get_latest_artifact_metadata(self, project_id: str, job_id: str):
def __get_latest_artifact_metadata(self, job_id: str):
response = requests.get(
f"{self.git_instance.api_url}/repos/{project_id}/actions/runs/{job_id}/artifacts",
headers=self.__get_headers(self.git_model.password),
f"{self.api_url}/repos/{self.project_id}/actions/runs/{job_id}/artifacts",
headers=self.__get_headers(self.password),
timeout=config.requests.timeout,
)
response.raise_for_status()
Expand Down
Loading

0 comments on commit 90f51f1

Please sign in to comment.