limbo | It currently supports x86 | Emulator library

Your stackblitz doesnt allow ngFor so the setup must be wrong. Here is what I think is the problem:

Your app has a weird data flow, when you subscribe to your observable, you wont receive any data until something calls next on your subject. What you seems to be doing from what I've seen in your stackblitz is:

1. Call to update which is going to to give a value to your Subject
2. You subscribe and its already too late

You can't get the last value from a subject and since you subsribed too late, your console.log from the subscribe is never actually doing anything. You are seeing the console.log related to your update(the one from hello component)

1. You can only change a file input value with another list
   here is how: https://stackoverflow.com/a/52079109/1008999 (also in the example)
2. Using the FileReader is a waste of time, CPU, Encoding & decoding and RAM...
   use URL.createObjectURL instead
3. Don't use canvas.toDataURL... use canvas.toBlob instead
4. Canvas have bad compression, read earlier comment and see the jsfiddle proff...
   If you insist on using canvas to try and squeeze the size down then
   1. First try to see if the image is in a reasonable size first
   2. Compare if the pre existing image file.size is smaller than what the canvas.toBlob provides and choose if you want the old or the new one instead.
   3. If resizing the image isn't enough have a look at this solution that change the quality until a desired file size & image aspect have been meet.

Without any testing, this is how i would have refactor your code too:

Your server isnt making a response after getting the request from the client so the connection is pretty much in limbo for lack of a better word. You need to send a response to client

The consistent difference between existing and non-released films is the attribute data-film-session and data-film-exp. Only add to the list if the entry has one or both of these attributes. This is untested, it may not work, but it's a step in the right direction.

Just create another component named Spinner with the gif image and prop which control the rendering of image.

A query that cannot resolve its residual goals is called a floundering query. Computations can flounder when they are intended to succeed or finitely fail. Residual goals introduce a third state that is neither success nor finitely failed.

Constraint logic programs can fail to invoke certain constraint solvers because variables are insufficiently instantiated or constrained. To avoid floundering in constraint solvers, they often offer labeling as a last resort.

The third state is seen in the top level, when residual goals are listed. And it is implementation dependent how residual goals can be queried. Typical predicates are:

• call_residue_vars(G, L):
  The predicate succeeds whenever the goal G succeeds and unifies L with the newly introduced attributed variables.
• call_residue(G, L):
  The predicate succeeds whenever the goal G succeeds and unifies L with the constraints of the newly introduced attributed variables.

That SWI-Prolog doesn't print the residue variables in the toplevel, is a service by SWI-Prolog, to only show the projected variables. But you can neverthless query the residue:

You need to use checkpointing on the writeStream - it will track offsets for all sources that are used for your operations, and store them in the checkpoint directory, so when you restart application, it will read the offsets for all sources and continue from them. The offset that you specify in readStream is just for case when you don't have checkpoint directory yet - after the first query it will be filled with real offsets, and the value specified in options won't be used (until you remove checkpoint directory).

Read the Structured Streaming documentation to understand how it works.

P.S. The checkpoint that you're using in the last example, is another thing - not for Structured Streaming.

I doubt moving stuff around will improve the performance of the software. But, definitely, having tidy code and a tidy project will improve YOUR performance while working/coding.

Yes, you're safe now, but if your secrets were ever on a remote, you should still rotate them as soon as is convenient.