chore(release): 5.0.0 — security boundary + deferred breaking changes - #101
Merged
Conversation
The SPD3303X/-E command set has no protection subsystem (QS0503X-E01B p.36), so ovp_level/ocp_level never armed anything on real hardware. Flip has_ovp/has_ocp to False for these models per the deprecation promised in the 4.1.0 changelog, and drop the now-dead FutureWarning blocks -- the existing has_ovp/has_ocp guards raise NotImplementedError for both getter and setter instead. Guard the same calls in the psu_advanced_features example so its real-world-scenario demo keeps running against the SPD3303X mock.
Unmatched PSU/AWG/DAQ queries now raise TimeoutError by default, matching real hardware, instead of silently returning "". Pass strict=False to restore the pre-5.0.0 lenient behavior. BREAKING CHANGE: code relying on unmatched PSU/AWG/DAQ mock queries returning "" must now pass MockConnection(..., strict=False) explicitly.
MockConnection.read_raw() no longer falls through to the legacy C<n>:WF? DAT2/DESC block on a modern-dialect (SDS800X HD/SDS5000X) instance. That back-compat shim (Task 18) was documented as answering "until v5.0.0"; the modern programming guide has zero occurrences of "WF?" and the driver's capture path already uses :WAVeform:PREamble?/:WAVeform:DATA? exclusively. A modern instance that receives an unrecognized/legacy waveform read now raises TimeoutError, matching real hardware. Legacy-dialect behavior is unchanged -- build_waveform_response still serves C<n>:WF? on legacy scopes.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Cuts v5.0.0 (MAJOR). Bundles the merged web-gateway security boundary (PR #100) with the three breaking changes the 4.1.0 changelog deferred to v5.0.0, and bumps the version.
Breaking changes
From the security boundary (PR #100, already on main):
/api/*request and the WebSocket stream;/docsand/redocremoved (schema at/api/openapi.json, token required); reference files move to a JSON metadata format (scpi-web references migrate).Landed on this branch — the three 4.1.0-deferred breaks, now due:
ovp_level/ocp_levelraiseNotImplementedError(the SPD3303X has no protection subsystem;has_ovp/has_ocpare nowFalse). They emitted aFutureWarningin 4.1.0.MockConnectionstrict-by-default — an unmatched PSU/AWG/DAQ query raisesTimeoutErrorlike real hardware; passstrict=Falsefor the old lenient"".WF?shim removed — a modern-dialect mock no longer answers the legacyC<n>:WF?read; legacy scopes unaffected.How it was built
Each break landed subagent-driven with review. Notably, the strict-default flip broke only 1 test — the mock is already faithful enough (from the wire-forms work) that PSU/AWG/DAQ tests don't hit unmatched queries. The OVP/OCP change surfaced an example script that would have broken the smoke test; it was guarded.
Verification
pyproject.toml,scpi_control/__init__.py(__version__, feeds provenance + GUI About),CHANGELOG.md([Unreleased]→[5.0.0] - 2026-07-24).After merge
Release cut by creating the GitHub Release through
gh(firespublish.yml→ PyPI as the maintainer, not the bot; the tag also triggers the desktop builds). PyPI publish is irreversible.Note:
docs/about/changelog.md(a separate docs-site mirror) has been stale since 3.3.0 — pre-existing drift, not addressed here.