Scrapyd部署

scrapy爬虫写好后,需要用命令行运行,如果能在网页上操作就比较方便。scrapyd部署就是为了解决这个问题,能够在网页端查看正在执行的任务,也能新建爬虫任务,和终止爬虫任务,功能比较强大。

一、安装

1,安装scrapyd

1
pip install scrapyd

2, 安装 scrapyd-deploy

1
pip install scrapyd-client

windows系统,在c:\python27\Scripts下生成的是scrapyd-deploy,无法直接在命令行里运行scrapd-deploy。
解决办法:
在c:\python27\Scripts下新建一个scrapyd-deploy.bat,文件内容如下:

1
2
@echo off
C:\Python27\python C:\Python27\Scripts\scrapyd-deploy %*

添加环境变量:C:\Python27\Scripts;

二、使用

1,运行scrapyd

首先切换命令行路径到Scrapy项目的根目录下,
要执行以下的命令,需要先在命令行里执行scrapyd,将scrapyd运行起来

1
2
3
4
5
6
7
8
9
10
MacBook-Pro:~ usera$ scrapyd

/usr/local/bin/scrapyd:5: UserWarning: Module _markerlib was already imported from /Library/Python/2.7/site-packages/distribute-0.6.49-py2.7.egg/_markerlib/__init__.pyc, but /System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python is being added to sys.path
from pkg_resources import load_entry_point
2016-09-24 16:00:21+0800 [-] Log opened.
2016-09-24 16:00:21+0800 [-] twistd 15.5.0 (/usr/bin/python 2.7.10) starting up.
2016-09-24 16:00:21+0800 [-] reactor class: twisted.internet.selectreactor.SelectReactor.
2016-09-24 16:00:21+0800 [-] Site starting on 6800
2016-09-24 16:00:21+0800 [-] Starting factory <twisted.web.server.Site instance at 0x102a21518>
2016-09-24 16:00:21+0800 [Launcher] Scrapyd 1.1.0 started: max_proc=16, runner='scrapyd.runner'

2,发布工程到scrapyd

a,配置scrapy.cfg

在scrapy.cfg中,取消#url = http://localhost:6800/前面的“#”,具体如下:,
然后在命令行中切换命令至scrapy工程根目录,运行命令:

1
scrapyd-deploy <target> -p <project>

示例:

1
scrapd-deploy -p MySpider
  • 验证是否发布成功
1
2
3
4
scrapyd-deploy -l

output:
TS http://localhost:6800/

一,开始使用

1,先启动 scrapyd,在命令行中执行:

1
MyMacBook-Pro:MySpiderProject user$ scrapyd

2,创建爬虫任务

1
curl http://localhost:6800/schedule.json -d project=myproject -d spider=spider2
  • bug:
    scrapyd deploy shows 0 spiders by scrapyd-client
    scrapy中有的spider不出现,显示只有0个spiders。
  • 解决
    需要注释掉settings中的
    1
    2
    3
    4
    # LOG_LEVEL = "ERROR"
    # LOG_STDOUT = True
    # LOG_FILE = "/tmp/spider.log"
    # LOG_FORMAT = "%(asctime)s [%(name)s] %(levelname)s: %(message)s"
    When setting LOG_STDOUT=True, scrapyd-deploy will return ‘spiders: 0’. Because the output will be redirected to the file when execute ‘scrapy list’, like this: INFO:stdout:spider-name. Soget_spider_list can not parse it correctly.

3,查看爬虫任务

在网页中输入:http://localhost:6800/

下图为http://localhost:6800/jobs的内容:

4,运行配置

配置文件:C:\Python27\Lib\site-packages\scrapyd-1.1.0-py2.7.egg\scrapyd\default_scrapyd.conf

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
[scrapyd]
eggs_dir = eggs
logs_dir = logs
items_dir = items
jobs_to_keep = 50
dbs_dir = dbs
max_proc = 0
max_proc_per_cpu = 4
finished_to_keep = 100
poll_interval = 5
http_port = 6800
debug = off
runner = scrapyd.runner
application = scrapyd.app.application
launcher = scrapyd.launcher.Launcher

[services]
schedule.json = scrapyd.webservice.Schedule
cancel.json = scrapyd.webservice.Cancel
addversion.json = scrapyd.webservice.AddVersion
listprojects.json = scrapyd.webservice.ListProjects
listversions.json = scrapyd.webservice.ListVersions
listspiders.json = scrapyd.webservice.ListSpiders
delproject.json = scrapyd.webservice.DeleteProject
delversion.json = scrapyd.webservice.DeleteVersion
listjobs.json = scrapyd.webservice.ListJobs

参考

http://www.cnblogs.com/jinhaolin/p/5033733.html
https://scrapyd.readthedocs.io/en/latest/api.html#cancel-json

Author: Binger Chen
Link: http://www.kekefund.com/2016/10/10/scrapyd-deploy/
Copyright Notice: All articles in this blog are licensed under CC BY-NC-SA 4.0 unless stating additionally.