filebot | This is a fork of the official FileBot source code which has since been close sourced

Lazy doesn't mean "figure out later whether T is bound", it means "ensure a binding exists for T at compile time, but only create an instance at runtime after I call get". You will still need to make the binding for T available in all cases, but Dagger won't try to create an instance of it until you explicitly ask for it.

Dagger requires that for all uses of Provider and Lazy, the binding T needs to exist at compile time, even if at runtime you do not call it. This ensures that if you do call get() on the Provider or Lazy instance, it does not fail at runtime for a binding it knew was missing at compile time. (Lazy behaves exactly as Provider does, except Lazy remembers the instance it returns regardless of whether the binding was scoped.)

This means that one of your options is to add a binding for ModelMqlBased that returns null or throws an Exception, which normally would be a terrible idea in Dagger, but it would be sufficient for a case where you know at runtime that the Provides method is never called.

A different way to achieve the flexibility you're looking for is with @BindsOptionalOf. This allows you to inject an Optional or Optional>, which resolves to a present value if the binding exists and an absent placeholder if the binding does not.

Browsers can run a very limited set of language runtimes; essentially only Javascript and things that can be recompiled into Javascript. There's not a generic way to take a running desktop application and republish it into a browser.

Docker here is just an intermediate layer. It's very good for things like HTTP servers that are self-contained except for a network port; okay for command-line applications (if you don't mind running it as root with a command line three times as long, it works fine); and bad for GUI applications (works only on Linux, the command lines are long and arcane, and things are still highly likely to look different). It doesn't provide any magic to cross these layers.

You might be able to find some existing software that can fill some of these gaps; perhaps a browser-based X server, or maybe one of the full-blown VM systems can display a VM's desktop in a browser, or maybe you might be able to use the all-but-dead Java applet system, or maybe there's a Java layer that provides a websocket-based client and lets you run your Java-native GUI application with minor modifications.

I'd encourage you to first make your application work, and solve the "how do I actually make it appear the way I want" problem, and only then bring in Docker if it's appropriate. The layer of separation it provides can make many things more difficult especially while you're actively developing or exploring an unknown space.

Here you go.

Go down to "** and does stuff to it ** RIGHT HERE **".

This is where you will search and replace on the new file.

Use read from a here-string to parse the filename into fields, then use date to reformat.

Updated to handle spaces between elements before the date.

Ok I did it with a mix of elements found in similar posts. This command worked for me : docker run -ti --entrypoint=/bin/bash -p 8000:8000 -v /media/HDD/DL/complete:/opt/pyload/Downloads -v /usr/bin:/usr/bin -v /var/run/docker.sock:/var/run/docker.sock --name pyload thk-pyload

I could access the docker from host and run my other container with it. As simply as it is, the volumes were mounted correctly with my host !

This is easily done by strategically using redirection with CALL to establish a lock file. Only one instance can have write access to the lock file. Attempts by a second instance will fail. The first instance will release the lock as soon as it quits, regardless how it terminates.

I use multiple stages of redirection to hide unwanted error messages, yet the main routine has normal stdout and stderr. I add %* to pass the original arguments to the main. The only "odd" situation in main is %0 is :main instead of the executing script. But %~f0 can be used to get the full path to the executing script.

test.bat

As I thought, I'd missed something trivial. filebot has moved to jdk 10, which is why the error was being shown when I was trying to compile with jdk 9.

So, to anyone facing a similar problem, make sure the target jdk of the source you're building is the same as your development kit version.

You have enabled delay environment variable expansion with setlocal EnableDelayedExpansion somewhere above the ECHO line in batch script namer.

Therefore the exclamation mark is not further interpreted as literal character, but as begin/end of a variable reference expanded delayed.

We would need the batch code of namer batch file to help you solving this issue.

It looks like the line is for checking if the file contains .tmp, i.e. has the file extension .tmp and if this is the case, ignore the file.

I suggest to change in your first batch file