ruby on rails - Can I update a single source file on Heroku without recompiling the slug? -

I am working on a rail project which is being hosted in Heroku, I am thinking that the app resumed It is possible to update without a file.

Why I have a bug, but I can not track it down. It works perfectly on my local system, but it seems that stopping midway through processing on Heroki

There are no break points, I am scattering status updates in code. (To be removed later) but adding a line of code to a train app is like a five minute process

Changing the file

Speed ​​file GIT

Commit file

push GIT (all up is fast enough)

Wait for Heroko to drag the arrow, which looks like a gem installed, or At least one gem update.

Change some files to reflect the local URL

Resume the service

Do not run all those things without git Is there a way to push? Probably to add a special parameter to push?

---

Another annoyance is that my GIT now has a bunch of check-ins, which I do not want my cum - to see the workers, (Testing only), and there is no reason to include all these efforts in global source control.

This is a good reason why this is not possible when you push the halook, Slug '(). To provide Scalability to Harka, to provide this slug, only it is read that it can be applied on many dyons which are likely to be distributed in many different physical machines. Each dinosaur runs a different example of your application, while routing mesh ensures that your app requests go to the correct dinase.

Now see if any of these examples was writeable, if you are running 5 dinosaurs then you will have the application running on 5 different instances - if a file is written that this key Type is distributed to your dinosaurs? Yes, Harkak could have considered some kind of shared file system to run applications, but it is complicated. By reading only the file system () this problem has been reduced.

If you have created an app and positioned it in Hiro, but forgot to use the S3 type Iristent Storage, then your The application will upload you files (via such a paperclip like Ruby) only on that dino which it had received and will be lost when the new code is deployed or the application is restored Rarnb is because Dayo receive the latest code from the slug.

If you are debugging against Heroko Dawn, forget that you have a general GIT Arsenal available, git commit --mend . Alternatively, work in a branch and deploy directly on Hiroku ( git pushing your lock to Ltf>: Master , then when you resize the problem) Now need