tea/browsertrix
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

frontend: configure running with a fixed 'replay.json', auth headers passed via separate config (#899)

Browse Source 
wabac.js will reload the replay.json on 403 with new token (will be in next version of wabac.js)
presign urls: make presign timeout configurable (in minutes), defaults to 60 mins
dockerfile: fix configuring RWP_BASE_URL
This commit is contained in:
Ilya Kreymer 2023-06-08 11:26:26 -07:00 committed by GitHub
parent d286555396
commit 4428184aea
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 16 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
backend/btrixcloud/crawls.py
View File

@ -220,7 +220,9 @@ class CrawlOps:
self.crawl_configs.set_crawl_ops(self)
self.presign_duration = int(os.environ.get("PRESIGN_DURATION_SECONDS", 3600))
self.presign_duration_seconds = (
int(os.environ.get("PRESIGN_DURATION_MINUTES", 60)) * 60
)
async def init_index(self):
"""init index for crawls db collection"""
@ -463,7 +465,7 @@ class CrawlOps:
print("no files")
return
delta = timedelta(seconds=self.presign_duration)
delta = timedelta(seconds=self.presign_duration_seconds)
updates = []
out_files = []
@ -475,7 +477,7 @@ class CrawlOps:
if not presigned_url or now >= file_.expireAt:
exp = now + delta
presigned_url = await get_presigned_url(
org, file_, self.crawl_manager, self.presign_duration
org, file_, self.crawl_manager, self.presign_duration_seconds
)
updates.append(
(

2
chart/templates/configmap.yaml
View File

@ -62,6 +62,8 @@ data:
RERUN_LAST_MIGRATION: "{{ .Values.rerun_last_migration }}"
PRESIGN_DURATION_MINUTES: "{{ .Values.storage_presign_duration_minutes | default 60 }}"
---
apiVersion: v1

7
frontend/Dockerfile
View File

@ -1,7 +1,4 @@
# syntax=docker/dockerfile:1.4
# central place to configure the production replayweb.page loading prefix
ARG RWP_BASE_URL=https://cdn.jsdelivr.net/npm/replaywebpage/
FROM docker.io/library/node:16 as build_deps
WORKDIR /app
@ -30,10 +27,12 @@ COPY --link src ./src/
ARG GIT_COMMIT_HASH
ARG GIT_BRANCH_NAME
ARG VERSION
ARG RWP_BASE_URL=https://cdn.jsdelivr.net/npm/replaywebpage/
ENV GIT_COMMIT_HASH=${GIT_COMMIT_HASH} \
GIT_BRANCH_NAME=${GIT_BRANCH_NAME} \
VERSION=${VERSION}
VERSION=${VERSION} \
RWP_BASE_URL=${RWP_BASE_URL}
# Prevent Docker caching node_modules
RUN yarn build && \

10
frontend/src/pages/org/crawl-detail.ts
View File

@ -571,11 +571,12 @@ export class CrawlDetail extends LiteElement {
}
private renderReplay() {
const bearer = this.authState?.headers?.Authorization?.split(" ", 2)[1];
//const replaySource = `/api/orgs/${this.crawl?.oid}/crawls/${this.crawlId}/replay.json?auth_bearer=${bearer}`;
const replaySource = `/api/orgs/${this.crawl?.oid}/crawls/${this.crawlId}/replay.json`;
// for now, just use the first file until multi-wacz support is fully implemented
const replaySource = `/api/orgs/${this.crawl?.oid}/crawls/${this.crawlId}/replay.json?auth_bearer=${bearer}`;
//const replaySource = this.crawl?.resources?.[0]?.path;
const headers = this.authState?.headers;
const config = JSON.stringify({headers});
const canReplay = replaySource && this.hasFiles;
@ -587,6 +588,7 @@ export class CrawlDetail extends LiteElement {
<replay-web-page
source="${replaySource}"
coll="${ifDefined(this.crawl?.id)}"
config="${config}"
replayBase="/replay/"
noSandbox="true"
noCache="true"