mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-14 15:14:31 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

move latest to derived data using history

Browse source
This commit is contained in:
Nick Sweeting 2019-03-22 21:38:08 -04:00
parent 69f837bbf6
commit d06775923b
4 changed files with 34 additions and 49 deletions
Download patch file Download diff file Expand all files Collapse all files

30
archivebox/links.py
View file

@ -9,12 +9,6 @@ Link {
title: str,
tags: str,
sources: [str],
latest: {
...,
pdf: 'output.pdf',
wget: 'example.com/1234/index.html',
screenshot: null,
},
history: {
pdf: [
{start_ts, end_ts, duration, cmd, pwd, status, output},
@ -30,7 +24,6 @@ from collections import OrderedDict
from util import (
merge_links,
wget_output_path,
check_link_structure,
check_links_structure,
)
@ -47,30 +40,9 @@ def validate_links(links):
raise SystemExit(1)
for link in links:
link['title'] = unescape(link['title'].strip()) if link['title'].strip() else None
check_link_structure(link)
link['title'] = unescape(link['title']) if link['title'] else None
link['latest'] = link.get('latest') or {}
latest = link['latest']
if not link['latest'].get('wget'):
link['latest']['wget'] = wget_output_path(link)
if not link['latest'].get('pdf'):
link['latest']['pdf'] = None
if not link['latest'].get('screenshot'):
link['latest']['screenshot'] = None
if not link['latest'].get('dom'):
link['latest']['dom'] = None
if not latest.get('favicon'):
latest['favicon'] = None
if not link['latest'].get('title'):
link['latest']['title'] = link['title']
return list(links)