Drawing a ControlledQubitUnitary

[antalszava]

Context:
We consider drawing a ControlledQubitUnitary using the circuit drawer, which raises an error due to having no special rule for drawing this operation.

Description of the Change:
Adds a rule for rendering the matrix and the operation.

Benefits:
Draws ControlledQubitUnitary instances.

Possible Drawbacks:
N/A

Related GitHub Issues:
Closes #1171

[github-actions]

Hello. You may have forgotten to update the changelog!
Please edit .github/CHANGELOG.md with:

• A one-to-two sentence description of the change. You may include a small working example for new features.
• A link back to this PR.
• Your name (or GitHub username) in the contributors section.

[codecov]

Codecov Report

Merging #1174 (d92c6bc) into master (d0d4f81) will increase coverage by 3.31%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1174      +/-   ##
==========================================
+ Coverage   94.81%   98.12%   +3.31%     
==========================================
  Files         144      144              
  Lines       10836    10847      +11     
==========================================
+ Hits        10274    10644     +370     
+ Misses        562      203     -359     

Impacted Files	Coverage Δ
...ennylane/circuit_drawer/representation_resolver.py	99.33% <100.00%> (+0.04%)	⬆️
pennylane/ops/qubit.py	98.30% <100.00%> (+<0.01%)	⬆️
pennylane/math/tensorbox.py	96.02% <0.00%> (+1.32%)	⬆️
pennylane/beta/devices/default_tensor.py	95.22% <0.00%> (+1.70%)	⬆️
pennylane/devices/default_qubit_tf.py	90.66% <0.00%> (+2.66%)	⬆️
pennylane/math/numpy_box.py	98.57% <0.00%> (+2.85%)	⬆️
pennylane/beta/devices/default_tensor_tf.py	87.50% <0.00%> (+3.12%)	⬆️
pennylane/qnode.py	98.67% <0.00%> (+3.96%)	⬆️
pennylane/interfaces/torch.py	100.00% <0.00%> (+4.81%)	⬆️
pennylane/interfaces/tf.py	97.14% <0.00%> (+5.71%)	⬆️
... and 10 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d0d4f81...d92c6bc. Read the comment docs.

[josh146]

Nice :)

[josh146]

If these attributes are required by the circuit drawer, we should probably document it somewhere.

Are there any other operations that are missing these attributes?

[antalszava]

Are there any other operations that are missing these attributes?

They seem to be specific to ControlledQubitUnitary (at least with the current structure of PennyLane). Storing these two are required because:

• control_wires is merged into self.wires with the target wires at a later stage;
• the unitary is expanded to the entire subsystem using identities on the control qubits and only the expanded unitary is stored in self.data.

Therefore, introducing self.control_wires and self.U provides a convenient way to gather both the control wires and the unitary.

We could actually consider separating control and target wires natively for other more defined operations, as it can be convenient to have the separation be available easily.

If these attributes are required by the circuit drawer, we should probably document it somewhere.

Indeed, at the moment they are well used by the circuit drawer, but it can easily be that other features will also use them. Any thoughts on where they could be documented? For attributes of operations, we don't seem to have any specific docstrings/docs.

[josh146]

Any thoughts on where they could be documented?

This is a good question which I don't think has a good answer 🤔

[DSGuala]

Great!