连接到 Node.JS 中的数据库
对于某些特殊情况，例如“require”，阻止执行是有意义的，以使事情变得更简单。
我有一个类似的情况，我需要一种方法来建立数据库连接，但被阻止。 而且，因为它只发生一次 - 当应用程序启动并保存在全局对象中以供以后重用时 - 它不会影响性能。
问题：
- 节点没有“睡眠”方法。
- 事件循环有一些技巧，您必须阻止它，但同时允许处理数据库连接的内容。
实际上我已经通过使用 jasmine-node 中的 waitFor 做到了，但是当我查看它的源代码时 - 它非常复杂并且使用 phantomjs C 扩展。
可悲的是，简单的 while(true){...} 东西不起作用。例如，下面的代码不起作用，我相信是因为它阻止了事件循环并且不允许节点处理它等待的事件（在单线程环境中有点死锁:)）。
waitsFor = (fun, message = "waitsFor timeout!", timeout = 1000) ->
start = new Date().getTime()
while true
if fun()
break
else if (new Date().getTime() - start) > timeout
throw new Error(message)
但是，也许可以通过其他一些简单的方式来完成它，而不需要像 phantomjs 和复杂的 C 扩展这样的额外依赖项？
For some special cases, like 'require' it makes sense to block the execution, to make things simpler.
I have a similar case, I need a way to make DB connection, blocked.
And, because it happens only one time - when the app started and saved in global object for later reuse - it will not affect the performance.
The problems:
- node doesn't have the 'sleep' method.
- there's some trickery with event loop, You has to block it, but at the same time allow to process it the database connection stuff.
Actually I already did it, by using waitFor from jasmine-node, but when I looked at it source - it's very complicated and uses phantomjs C-extensions.
And sadly, simple while(true){...} stuff doesn't works. For example, code below doesn't work, I believe because it blocks the event loop and doesn't allow node to process the event it waits for (sort of deadlock in single-threaded environment :) ).
waitsFor = (fun, message = "waitsFor timeout!", timeout = 1000) ->
start = new Date().getTime()
while true
if fun()
break
else if (new Date().getTime() - start) > timeout
throw new Error(message)
But, maybe it's somehow possible to do it in some other simple way, without extra dependencies like phantomjs and complicated C-extensions?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
您的应用程序应首先尝试异步连接到数据库，然后在连接可用时继续执行其处理逻辑。例如：
Your application should first attempt to connect to the database asynchronously and then proceed with its processing logic when the connection is available. For example: