Improve the function to display TemplateOrchestrator as svg

[inmzhang]

As discussed in #70, there can be some improvement ideas on the tqec.display.display_templates_svg function, which includes the following or more:

1. Display the relationships between templates (corner pinning).
2. Find a way to display the "scalable" property of the outer outer edges. For example, outer edges that will scale when scale_to is called are depicted as double-ended arrows whereas those that does not scale are depicted with inward-arrows.

This may also relate to issue #67.

[nelimee]

Better idea about point number 2.

The problem of what I proposed first in 2. is that the arrows of different edges might intersect, which will lead to an hard-to-read representation, which is the opposite of the main goal.
Basically, any visualisation should be obvious at first glance if possible.
In this regard, here are two replacement proposals for 2. :

1. Make "scalable" and "non-scalable" outer edges a different colour.
2. Make "scalable" outer edges dotted and "non-scalable" outer edges plain.

Other ideas are of course welcome.

[inmzhang]

And we should also note that the current implementation relies on the assumption that all the plaquette is of the same square shape. According to #34, this function should also change quite a lot once the assumption is lifted.

[inmzhang]

And we should also note that the current implementation relies on the assumption that all the plaquette is of the same square shape. According to #34, this function should also change quite a lot once the assumption is lifted.

As discussed in #34 and #80, the plaquettes should all be rectangular. This constraint implies that the display_templates_svg function currently works as expected without any modifications, or with only slight adjustments, once issue #80 is merged.

In cases where increment_x is not equal to increment_y, and if there is a need to display the plaquettes¹ as rectangles, some minor code modifications will be required to ensure that displayed as intended.

Footnotes

1. Strictly speaking, the inner grid is not the boundary of the plaquettes in the sense that the shape of a plaquette is useless information. It just shows the tiling which is determined by the increment property of the templates. ↩

[inmzhang]

This issue is related to #107 and #108, we can rename display_templates_svg as display_template_svg and accept a template: Template as the argument. The implementation needs to be changed to accout for both AtomicTemplate and composed TemplateOrchestrator.

[nelimee]

@smburdick might be nice for UF hack? (I know, this is a little bit late, I am sorry)

[giangiac]

@nelimee I was thinking of starting by improving the ASCII representation of the circuit as generated by ComposedTemplate.display_template().
Shall I open a new issue or do you think I can use this?

[nelimee]

@nelimee I was thinking of starting by improving the ASCII representation of the circuit as generated by ComposedTemplate.display_template().
Shall I open a new issue or do you think I can use this?

I think a new issue is the best, this one is specifically about the SVG representation.

[giangiac]

I think a new issue is the best, this one is specifically about the SVG representation.

Opened new issue #234

[KabirDubey]

I'm unfamiliar with this part of the backend but will take on this issue for learning purposes. I'll review the 1/3 meeting, the revised task flow, and every part of the repo linked in this issue.

Please let me know if you'd advise differently and if there have been any updates to this issue I should be aware of.

[nelimee]

I'm unfamiliar with this part of the backend but will take on this issue for learning purposes. I'll review the 1/3 meeting, the revised task flow, and every part of the repo linked in this issue.

Please let me know if you'd advise differently and if there have been any updates to this issue I should be aware of.

There have been no change in this issue, you should be able to safely go on :)

To summarise what would be nice-to-have features:

• It would be nice to be able to distinguish between scalable and non-scalable edges. Basically, if a template is scalable in a given dimension, the 2 edges in that dimension are considered scalable. The idea for the moment is to either use a different colour or use a different type of line to draw the edge, but you are free to propose any other solution, the deciding factor being clearness.
• It would be nice to have a way of seeing which corners are glued together.

Feel free to ask questions here or to send me a mail, we can schedule a meeting if you think this will help you understanding the Templates =)

[KabirDubey]

Stale after refactoring of Template in #317 and the frontend in #325. Current sentiments about desired visuals are in #357.