mirror of
https://github.com/internetarchive/brozzler.git
synced 2025-06-20 12:54:23 -04:00
add info to display of jobless sites in brozzler-webconsole; fix creation of "least_hops" index on the rethinkdb table "pages"
This commit is contained in:
parent
74b229cfb0
commit
f671cf4f11
4 changed files with 13 additions and 7 deletions
|
@ -1,7 +1,7 @@
|
|||
#!/usr/bin/env python
|
||||
'''
|
||||
brozzler-easy - brozzler-worker, warcprox, and pywb all working together in a
|
||||
single process
|
||||
brozzler-easy - brozzler-worker, warcprox, pywb, and brozzler-webconsole all
|
||||
working together in a single process
|
||||
|
||||
Copyright (C) 2016 Internet Archive
|
||||
|
||||
|
|
|
@ -69,7 +69,7 @@ class RethinkDbFrontier:
|
|||
self.r.table("pages").index_create(
|
||||
"least_hops", [
|
||||
self.r.row["site_id"], self.r.row["brozzle_count"],
|
||||
self.r.row["hops_from_seed"]])
|
||||
self.r.row["hops_from_seed"]]).run()
|
||||
if not "jobs" in tables:
|
||||
self.logger.info(
|
||||
"creating rethinkdb table 'jobs' in database %s",
|
||||
|
|
|
@ -73,11 +73,17 @@
|
|||
<thead>
|
||||
<tr>
|
||||
<th>id</th>
|
||||
<th>status</th>
|
||||
<th>started</th>
|
||||
<th>seed url</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr ng-repeat="site in joblessSites">
|
||||
<td><a href="/sites/{{site.id}}">{{site.id}}</a></td>
|
||||
<td>{{site.status}}</td>
|
||||
<td>{{site.start_time}}</td>
|
||||
<td>{{site.seed}}</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
|
|
2
setup.py
2
setup.py
|
@ -32,7 +32,7 @@ def find_package_data(package):
|
|||
|
||||
setuptools.setup(
|
||||
name='brozzler',
|
||||
version='1.1b6.dev74',
|
||||
version='1.1b6.dev75',
|
||||
description='Distributed web crawling with browsers',
|
||||
url='https://github.com/internetarchive/brozzler',
|
||||
author='Noah Levitt',
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue