python脚本运行的问题

我现在执行一个python脚本。例如 print 123。 然后运行。
在运行的过程中,我将 print 123 修改成 print 3333.
怎么样让这个python自动运行。 求解。

阅读 4.5k
2 个回答

比较合适和常见的处理方法,是另外做一个daemon,定时去轮询文件,比如检查文件的修改时间,md5值等,一旦发现文件被修改,就重新用subprocess模块调用下这个文件,大多数python做热部署的代码都是这么搞的。
举个例子
来自https://github.com/jimjose/pyreload/blob/master/pyreload.py
flask里面也有类似的file reloader机制 可以参考 http://stackoverflow.com/questions/16344756/auto-reloading-python-flask-app-upon-code-changes

"""
A file monitor and module reloader.
Major part of the code is taken directly from the original "paste.reloader" module. I just created this stripped down version for one of my projects and wanted to share if anyone else is interested.
**all credits goes to original authors @ Paste **
http://pythonpaste.org/modules/reloader.html
Use this like:
..code-block:: Python
    import pyreload
    pyreload.install(verbose=True, poll_interval=1)
Now this will watch for any changes in the files and will reload the corresponding module automatically, so there is no need to restart the python server.
"""

import os
import imp
import sys
import time
import threading

def install(verbose=True, poll_interval=1):
    """
    Install the reloading monitor.
    """
    mon = Monitor(verbose=verbose,poll_interval=poll_interval)
    t = threading.Thread(target=mon.periodic_reload)
    t.setDaemon(True)
    t.start()

class Monitor(object):
    def __init__(self, verbose, poll_interval):
        self.verbose = verbose;
        self.module_mtimes = {}
        self.poll_interval = poll_interval

    def periodic_reload(self):
        while True:
            self.check_reload()
            time.sleep(self.poll_interval)

    def check_reload(self):
        filenames = []
        for module in sys.modules.values():
            try:
                filename = module.__file__
            except (AttributeError, ImportError), exc:
                continue

            try:
                stat = os.stat(filename)
                if stat:
                    mtime = stat.st_mtime
                else:
                    mtime = 0
            except (OSError, IOError):
                continue

            if filename.endswith('.pyc') and os.path.exists(filename[:-1]):
                mtime = max(os.stat(filename[:-1]).st_mtime, mtime)
            elif filename.endswith('$py.class') and \
                    os.path.exists(filename[:-9] + '.py'):
                mtime = max(os.stat(filename[:-9] + '.py').st_mtime, mtime)
            if not self.module_mtimes.has_key(filename):
                self.module_mtimes[filename] = mtime
            elif self.module_mtimes[filename] < mtime:
                if self.verbose:
                    print >> sys.stderr, (
                        "pyreload: %s changed; reloading..." % filename)
                try:
                    imp.reload( module )
                    if self.verbose:
                        print >> sys.stderr, ("pyreload: reloaded: %s" % module.__name__ )
                except Exception as e:
                    if self.verbose:
                        print >> sys.stderr, ("pyreload: reloaded: %s" % e.message )
                    pass

                del self.module_mtimes[filename]
新手上路,请多包涵

tornado也有这样地机制,具体可以看tornado源码,使用上比如:

    config_evn = settings.get("config").split(".")[-1].lower()
    if settings.get("debug") or config_evn == "debug":
        # 检测py文件的变动, 自动reload新的代码, 无需重启服务器.
        # 仅用于开发和测试阶段, 否则每次修改都重启Tornado服务.
        instance = tornado.ioloop.IOLoop.instance()
        tornado.autoreload.start(instance)
        instance.start()
    else:
        tornado.ioloop.IOLoop.instance().start()
撰写回答
你尚未登录,登录后可以
  • 和开发者交流问题的细节
  • 关注并接收问题和回答的更新提醒
  • 参与内容的编辑和改进,让解决方法与时俱进
推荐问题