feat: Capture pytket's output permutation explicitly in the hugr connectivity by aborgna-q · Pull Request #1075 · Quantinuum/tket2

aborgna-q · 2025-08-26T11:23:14Z

Now the permutation gets reflected in the wire structure.

A (circ := Circuit(2).SWAP(0,1)).replace_SWAPs() now gets encoded directly as

graph LR
    subgraph 0 ["(0) Module"]
        direction LR
        subgraph 1 ["(1) [**FuncDefn: #quot;#quot;**]"]
            direction LR
            style 1 stroke:#832561,stroke-width:3px
            2["(2) Input"]
            3["(3) Output"]
            2--"0:1<br>qubit"-->3
            2--"1:0<br>qubit"-->3
        end
    end

Before we only stored the permutation in the circuit's metadata

codecov · 2025-08-26T11:27:21Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 79.09%. Comparing base (5cb6293) to head (cef2a80).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1075      +/-   ##
==========================================
+ Coverage   79.01%   79.09%   +0.07%     
==========================================
  Files         115      115              
  Lines       13416    13390      -26     
  Branches    12634    12608      -26     
==========================================
- Hits        10601    10591      -10     
+ Misses       2158     2144      -14     
+ Partials      657      655       -2

Flag	Coverage Δ
python	`92.83% <ø> (ø)`
rust	`78.24% <100.00%> (+0.08%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

…cture

lmondada

I think there is an error in your docs, but otherwise I like it. This makes more sense to me than the metadata trick you had before.

One question: does this represent a breaking change in serialisation? I.e. are there circuits that users may have stored for which the ordering of the outputs will not be parsed properly?

lmondada · 2025-08-26T13:15:52Z

tket/src/serialize/pytket/decoder/wires.rs

+    /// This originates from pytket's `implicit_permutation` field.
+    ///
+    /// For a circuit with three qubit registers [q0, q1, q2] and an implicit
+    /// permutation {q0 -> q1, q1 -> q2, q2 -> q0}, `output_to_input` will be {1


I don't think output_to_input refers to anything anymore. Update?

lmondada · 2025-08-26T13:19:58Z

tket/src/serialize/pytket/decoder/wires.rs

    /// A map from pytket register hashes to the latest up-to-date [`TrackedQubit`] referencing it.
    ///
-    /// The map keys are kept in the order they were defined in the circuit.
+    /// The map keys are kept in the order we expect to see them at the circuits output.


I'm confused. In fn known_pytket_qubits you reorder your outputs using output_qubit_permutation, so that would indicate that they are not ordered by output order?

Furthermore, this is an IndexMap, so I'm not even sure what the order would be.

same comment 4 lines below.

Ah, I forgot to rollback this comment after trying a change here.

IndexMap is a HashMap that keeps the insertion order of its keys, so we can traverse it / find elements by index, etc.

lmondada · 2025-08-26T13:25:17Z

tket/src/serialize/pytket.rs

 pub const METADATA_Q_REGISTERS: &str = "TKET1.qubit_registers";
 /// The reordered qubit registers in the output, if an implicit permutation was applied.
+#[deprecated(
+    note = "Output registers now always match the input register order.",


I don't find this message very clear. Do you mean something like

Suggested change

note = "Output registers now always match the input register order.",

note = "Qubit permutations are now represented explicitly in DFG output ordering .",

lmondada · 2025-08-26T13:27:26Z

tket/src/extension.rs

+#[deprecated(
+    note = "TKET.tk1op payload is stored as a op parameter, not a metadata field.",
+    since = "0.14.1"
+)]


I don't understand how this deprecation relates to the rest of the PR, but that might not be important.

Sorry, it's a drive-by added while deprecating the other constants.

aborgna-q · 2025-08-26T13:49:20Z

One question: does this represent a breaking change in serialisation? I.e. are there circuits that users may have stored for which the ordering of the outputs will not be parsed properly?

Technically yes. Stored hugrs extracted from pytket circuits with implicit_permutations won't preserve the reordering if re-encoded into pytket.

The HUGR -> pytket is not currently used externally though, so I don't think we have to worry about the break.

So we can make a release with #1075. BEGIN_COMMIT_OVERRIDE feat: Capture pytket's output permutation explicitly in the hugr connectivity (#1075) END_COMMIT_OVERRIDE

So we can make a release with #1075. BEGIN_COMMIT_OVERRIDE feat: Capture pytket's output permutation explicitly in the hugr connectivity (#1075) END_COMMIT_OVERRIDE ...#1076 didn't actually touch the python files -.- I swear I'm not farming github stats.

🤖 I have created a release *beep* *boop* --- ## [0.12.5](tket-py-v0.12.4...tket-py-v0.12.5) (2025-08-26) ### Features * Capture pytket's output permutation explicitly in the hugr connectivity ([#1075](#1075)) ([5fc9614](5fc9614)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). --------- Co-authored-by: Agustín Borgna <agustin.borgna@quantinuum.com>

## 🤖 New release * `tket`: 0.14.0 -> 0.15.0 (✓ API compatible changes) * `tket-qsystem`: 0.20.1 -> 0.21.0 (✓ API compatible changes) <details><summary>Changelog</summary> ## `tket` <blockquote> ## [0.15.0](tket-v0.14.0...tket-v0.15.0) - 2025-09-15 ### Bug Fixes - [**breaking**] Fix rotation -> float param type conversion ([#1061](#1061)) - Pytket barrier operations not being decoded ([#1069](#1069)) - Always load parameter expressions as half turns in the decoder ([#1083](#1083)) - Move attribute to come after all the cases. ([#1112](#1112)) ### New Features - Capture pytket's output permutation explicitly in the hugr connectivity ([#1075](#1075)) - Add ResourceScope ([#1052](#1052)) - [**breaking**] Remove unnecessary Arc from PytketDecoder method ([#1114](#1114)) - [**breaking**] Remove deprecated definitions ([#1113](#1113)) </blockquote> ## `tket-qsystem` <blockquote> ## [0.21.0](tket-qsystem-v0.20.1...tket-qsystem-v0.21.0) - 2025-09-15 ### Bug Fixes - [**breaking**] Fix rotation -> float param type conversion ([#1061](#1061)) - Pytket barrier operations not being decoded ([#1069](#1069)) - *(qystem)* fix angle bug in CZ decomposition ([#1080](#1080)) - Always load parameter expressions as half turns in the decoder ([#1083](#1083)) ### New Features - Add a `borrow_array` type replacement pass ([#975](#975)) - Add gpu module ([#1090](#1090)) - [**breaking**] Remove unnecessary Arc from PytketDecoder method ([#1114](#1114)) - [**breaking**] Remove deprecated definitions ([#1113](#1113)) ### Refactor - [**breaking**] Factor out wasm extension code into compute module ([#1089](#1089)) </blockquote> </details> --- This PR was generated with [release-plz](https://github.com/release-plz/release-plz/). --------- Co-authored-by: Agustín Borgna <121866228+aborgna-q@users.noreply.github.com> Co-authored-by: Alec Edgington <54802828+cqc-alec@users.noreply.github.com>

aborgna-q requested a review from a team as a code owner August 26, 2025 11:23

aborgna-q requested a review from lmondada August 26, 2025 11:23

feat: Capture pytket's output permutation explicitly in the hugr stru…

b1891f4

…cture

aborgna-q force-pushed the ab/explicit-output-permutation branch from e9fc8b9 to b1891f4 Compare August 26, 2025 12:14

aborgna-q changed the title ~~feat: Capture pytket's output permutation explicitly in the hugr structure~~ feat: Capture pytket's output permutation explicitly in the hugr connectivity Aug 26, 2025

lmondada approved these changes Aug 26, 2025

View reviewed changes

doc fixes

cef2a80

aborgna-q added this pull request to the merge queue Aug 26, 2025

aborgna-q removed this pull request from the merge queue due to a manual request Aug 26, 2025

aborgna-q enabled auto-merge August 26, 2025 13:50

aborgna-q added this pull request to the merge queue Aug 26, 2025

Merged via the queue into main with commit be48088 Aug 26, 2025
20 checks passed

aborgna-q deleted the ab/explicit-output-permutation branch August 26, 2025 13:59

hugrbot mentioned this pull request Aug 26, 2025

chore: release #1064

Closed

aborgna-q mentioned this pull request Aug 26, 2025

chore: Dummy change to trigger python release #1076

Merged

aborgna-q mentioned this pull request Aug 26, 2025

chore: Dummy change to trigger python release #1077

Merged

hugrbot mentioned this pull request Aug 26, 2025

chore(py): release tket-py 0.12.5 #1078

Merged

hugrbot mentioned this pull request Sep 15, 2025

chore: release #1115

Merged

hugrbot mentioned this pull request Sep 16, 2025

chore: release #1119

Merged

hugrbot mentioned this pull request Jan 16, 2026

chore(py): release tket-py 0.12.16 #1365

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: Capture pytket's output permutation explicitly in the hugr connectivity#1075

feat: Capture pytket's output permutation explicitly in the hugr connectivity#1075
aborgna-q merged 2 commits intomainfrom
ab/explicit-output-permutation

aborgna-q commented Aug 26, 2025 •

edited

Loading

Uh oh!

codecov bot commented Aug 26, 2025 •

edited

Loading

Uh oh!

lmondada left a comment

Uh oh!

lmondada Aug 26, 2025

Uh oh!

lmondada Aug 26, 2025

Uh oh!

lmondada Aug 26, 2025

Uh oh!

aborgna-q Aug 26, 2025

Uh oh!

lmondada Aug 26, 2025

Uh oh!

lmondada Aug 26, 2025

Uh oh!

aborgna-q Aug 26, 2025

Uh oh!

aborgna-q commented Aug 26, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

	note = "Output registers now always match the input register order.",
	note = "Qubit permutations are now represented explicitly in DFG output ordering .",

Conversation

aborgna-q commented Aug 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov bot commented Aug 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

lmondada left a comment

Choose a reason for hiding this comment

Uh oh!

lmondada Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

lmondada Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

lmondada Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

aborgna-q Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

lmondada Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

lmondada Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

aborgna-q Aug 26, 2025

Choose a reason for hiding this comment

Uh oh!

aborgna-q commented Aug 26, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

aborgna-q commented Aug 26, 2025 •

edited

Loading

codecov bot commented Aug 26, 2025 •

edited

Loading