Yellow science stuck inside sorter (overtaking problem)

Issue description

Taking advantage of the convergence of production times, I made a contraption where first assembly should make a white and yellow science, while next assembly red ones.
Unfortunately, exit sorter from assembly got permanently stuck, most likely because white science filled up sorter before nitrogen was provided to manufacture yellow science.
Removing white science from belt didn't help (see https://streamable.com/c2ga1v ).
Destroying-rebuilding sorter helped for a short period of time, then system locked up again.

Expected behavior

Sorter that is connected to output of assembly should allow to transport goods that can be transported, while leaving others inside. This property should be transitional to allow to connect many sorters to receive specific goods.

Reproduction steps

Using "Truck Import - Off" can block transport of Liquid Nitrogen Tank to help making system stuck.

Game version

v0.4.12c

Attachments

stuckScience.save

Files not imported: stuckScience.save

This is working as intended. You're producing 48 Lab Equipment 1 a minute and only using 24 in the same assembly for Lab Equipment 2. It's always going to back up. The only way to resolve this would be to have a second assembler handling the overflow of Lab Equipment 1.

You're producing 48 Lab Equipment 1 a minute and only using 24 in the same assembly for Lab Equipment 2

You seem to not understand what is shown in the picture. Assembler cannot produce both items at the same time, because engine of this game do not allow such operation.
It is expected to produce White until saturated then Yellow. In average, 2 * 8 White (20 seconds) and 2 * 4 Yellow (40 seconds) per minute..

The only way to resolve this would be to have a second assembler

You seem to not know this game deeply. Let me list my ways of handling this issue:

1. Leave assembler unconnected at all, trucks will do job of non-blocking items movement. That's how did it work at start. But using pickups to move White from exit of assembler to its own entry seems silly.
2. To have Red Lab Equipment assembler connected directly to exit of this mixed assembler - contrary to Sorter, assembler do not have any problem with cherry-picking Yellow. Unfortunately, this still requires using pickups to move White Lab Equipment from exit of assembler to its own entry, and this still seems silly. This was my middle solution.
3. Use Storage as "non broken sorter", see picture. This is my current solution. Somewhat Storage, contrary to broken Sorter, also do not have problem with cherry picking White. This workaround forces use of pickups to receive Yellow.

However, each of workarounds above is a workaround. And if there is a workaround, there has to be defect that is a cause. And a defect is that sorter - contrary to Assembler and Storage - have an implementation that blocks itself.

All you have to do is to prioritize production of golden lab equipment and it will work because golden lab equipment will always have the highest output priority. At this point this machine is choking because it is overproducing white lab equipment. I understand the use case to have sorter solving this automatically by having dedicated queues for filtered and unfiltered products and let them pass that way. But not sure how feasible that is in other scenarios. So let's keep it in backlog for now.