Bump pyzx from 0.10.3 to 0.10.5

[dependabot]

Bumps pyzx from 0.10.3 to 0.10.5.

Release notes

Sourced from pyzx's releases.

v0.10.5

Fixed

• string_to_phase no longer returns a constant Poly for a numeric expression that misses its fast parsing path (e.g. the (1)*1/4 form the TikZ importer produces for \frac{\pi}{4}): a parse result without free variables is now collapsed to a Fraction. Such a Poly compared and printed like its numeric value but broke code that branches on the phase type, e.g. to_tensor/to_matrix raised Can't convert diagram with parameters to tensor after a to_tikz/tikz_to_graph round trip (by @​gauthamkanagaraj).
• Correctly handling custom gate identifiers ending in gate (by @​dlyongemallo).
• Circuit.from_qasm routes through the symbolic grammar and accepts parenthesised subexpressions in gate phase arguments, e.g., rz((pi/2 + pi/4)) q[0]; or u3(pi, (a+b)/2, c) q[0]; (by @​dlyongemallo).
• Register identifier validation more closely matches OpenQASM 2 and 3 specs (by @​dlyongemallo).

Added

• Circuit.from_qasm supports parametrised custom gate definitions, e.g., gate phase_kick(theta) q { rz(theta) q; ... } (by @​dlyongemallo).
• The symbolic expression parser (pyzx.symbolic.parse) now accepts division, e.g., theta/2 or (x + y)/2. Divisors must be rational (or complex) constants; division by an integer produces an exact Fraction coefficient. As a side effect, ^ binds tighter than * and /. (by @​dlyongemallo)
• Regression test for unsafe_pivot with boolean pivot phases (by @​dlyongemallo).
• Added support for zooming in D3 drawings (by @​doczenwiry).
• apply_to_boolean_axels opt-in flag to (unsafe_)pauli_push, so boolean parameters are not transformed into non-boolean phase by default (by @​dlyongemallo).

Changed

• Multigraph.edge(s, t) now raises ValueError on ambiguous mixed-type parallel edges or when no matching edge exists, instead of silently returning one; pass et to disambiguate. The et default is now None on BaseGraph.edge (GraphS.edge and GraphIG.edge ignore it). Internal callers were updated to match; as a side effect, PauliWeb.add_edge and PauliWeb.graph_with_errors now propagate this error rather than silently preferring the SIMPLE edge (by @​dlyongemallo).

v0.10.4

Added

• Added the "magic cat" decomposition strategy from https://arxiv.org/pdf/2202.09202 (which previously only existed in Quizx). (by @​mjsutcliffe99).
• settings.strict_phase_types (default True) rejects float phases at set_phase/add_to_phase rather than letting them flow into the graph and crash downstream rewrite rules. Set to False to opt in to automatic conversion to Fraction using settings.float_to_fraction_max_denominator, accepting the resulting precision loss. Fixes crash in full_reduce with non-Clifford spiders (by @​dlyongemallo).

Changelog

Sourced from pyzx's changelog.

[0.10.5] - 2026-08-01

Fixed

• string_to_phase no longer returns a constant Poly for a numeric expression that misses its fast parsing path (e.g. the (1)*1/4 form the TikZ importer produces for \frac{\pi}{4}): a parse result without free variables is now collapsed to a Fraction. Such a Poly compared and printed like its numeric value but broke code that branches on the phase type, e.g. to_tensor/to_matrix raised Can't convert diagram with parameters to tensor after a to_tikz/tikz_to_graph round trip (by @​gauthamkanagaraj).
• Correctly handling custom gate identifiers ending in gate (by @​dlyongemallo).
• Circuit.from_qasm routes through the symbolic grammar and accepts parenthesised subexpressions in gate phase arguments, e.g., rz((pi/2 + pi/4)) q[0]; or u3(pi, (a+b)/2, c) q[0]; (by @​dlyongemallo).
• Register identifier validation more closely matches OpenQASM 2 and 3 specs (by @​dlyongemallo).

Added

• Circuit.from_qasm supports parametrised custom gate definitions, e.g., gate phase_kick(theta) q { rz(theta) q; ... } (by @​dlyongemallo).
• The symbolic expression parser (pyzx.symbolic.parse) now accepts division, e.g., theta/2 or (x + y)/2. Divisors must be rational (or complex) constants; division by an integer produces an exact Fraction coefficient. As a side effect, ^ binds tighter than * and /. (by @​dlyongemallo)
• Regression test for unsafe_pivot with boolean pivot phases (by @​dlyongemallo).
• Added support for zooming in D3 drawings (by @​doczenwiry).
• apply_to_boolean_axels opt-in flag to (unsafe_)pauli_push, so boolean parameters are not transformed into non-boolean phase by default (by @​dlyongemallo).

Changed

• Multigraph.edge(s, t) now raises ValueError on ambiguous mixed-type parallel edges or when no matching edge exists, instead of silently returning one; pass et to disambiguate. The et default is now None on BaseGraph.edge (GraphS.edge and GraphIG.edge ignore it). Internal callers were updated to match; as a side effect, PauliWeb.add_edge and PauliWeb.graph_with_errors now propagate this error rather than silently preferring the SIMPLE edge (by @​dlyongemallo).

[0.10.4] - 2026-07-01

Added

• Added the "magic cat" decomposition strategy from https://arxiv.org/pdf/2202.09202 (which previously only existed in Quizx). (by @​mjsutcliffe99).
• settings.strict_phase_types (default True) rejects float phases at set_phase/add_to_phase rather than letting them flow into the graph and crash downstream rewrite rules. Set to False to opt in to automatic conversion to Fraction using settings.float_to_fraction_max_denominator, accepting the resulting precision loss. Fixes crash in full_reduce with non-Clifford spiders (by @​dlyongemallo).

Commits

• 0f54edb Bump version: 0.10.4 → 0.10.5
• 60923ef Add 0.10.5 section to changelog [skip ci]
• a9ecd26 Fix broken Claude action (hopefully)
• b416daa Added a Claude review workflow that can be called by people with write and tr...
• 3918b2e Merge pull request #488doczenwiry/issue#88
• 28a2ad8 Update FAQ section with some more recent papers and software projects
• 0f79438 Tweak graph docs about multigraphs
• 0282013 Merge branch 'master' into issue#88-zooming
• fc45f64 Merge pull request #486 from dlyongemallo/436-fix_unsafe_pauli_push
• 5ea81ac Clearing all outputs in issue#88.ipynb notebook.
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)