Modify GPIB code to use symbolic constants by DavidCPlatt · Pull Request #601 · pyvisa/pyvisa-py

DavidCPlatt · 2026-06-26T19:32:21Z

Create a file of symbolic constants (IntEnum classes) for the various libgpib interface calls. Use these symbols rather than hard-coded integers. Remove
now-extraneous commentary about what the hard-coded integers mean.

The new file gpib_constants.py is derived from the constants file in gpib_ctypes (which isn't a strict dependency of this project). I've reorganized the values as an individual IntEnum class per API or purpose, with the class name based on the API or purpose, and the member names matching those used in gpib_ctypes. This provides a clean visual match between the ib*() call, class name, and member name each time one of the symbols is used, and should help prevent usage mismatches. In the gpib.py code I modified, I have removed the old commentary which explained the intent of each bare integer as being unnecessary.

This change deserves careful eyeball checking before being merged, to make sure I've gotten the translations between bare-integer and symbol correct in every case. I've reviewed it all two or three times, but you know how much author code reviews are worth! :-)

Closes # (insert issue number if relevant)
Executed black . && isort -c . && flake8 with no errors
The change is fully covered by automated unit tests
Documented in docs/ as appropriate
Added an entry to the CHANGES file

Create a file of symbolic constants (IntEnum classes) for the various libgpib interface calls. Use these symbols rather than hard-coded integers. Remove now-extraneous commentary about what the hard-coded integers mean.

for more information, see https://pre-commit.ci

…nto gpib-enums-clean

codecov · 2026-06-26T19:34:13Z

Codecov Report

❌ Patch coverage is 76.96078% with 47 lines in your changes missing coverage. Please review.
✅ Project coverage is 41.03%. Comparing base (28b0e88) to head (dd28298).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
pyvisa_py/gpib.py	2.08%	47 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #601      +/-   ##
==========================================
+ Coverage   39.18%   41.03%   +1.84%     
==========================================
  Files          27       28       +1     
  Lines        5022     5176     +154     
  Branches      518      517       -1     
==========================================
+ Hits         1968     2124     +156     
+ Misses       3028     3026       -2     
  Partials       26       26

Flag	Coverage Δ
unittests	`41.03% <76.96%> (+1.84%)`	⬆️

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

☔ View full report in Codecov by Harness.
📢 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.

MatthieuDartiailh

Looks fine to me but I will do another pass later.

MatthieuDartiailh · 2026-06-26T19:49:17Z

-                if ifc.ask(2):
-                    ifc.config(2, attribute_state + 96)
+                if ifc.ask(gpib_constants.ask.IbaSAD):
+                    ifc.config(gpib_constants.config.IbcSAD, attribute_state + 96)


I wish my comment about the 96 was more explicit...

Ach. I didn't even consider those NI-specific SAD masks/constants.

I'll add another commit to the PR for these. Shouldn't take long.

Things being as they are, these days, I'm not absolutely certain that saying "96 86'ed" is entirely safe, but what the hell. Done.

Hah! Just realized there was an off-by-one error in the old code - range(96,126) should have been range(96,127). As it was, subaddress 30 wouldn't have been iterated. Fixed now.

Turn 0, 96, and 126 into symbolic constants for the SAD.

for more information, see https://pre-commit.ci

The range() operator which allows iteration through the allowed GPIB sub-addresses (0 through 30, mapped up to 96 through 126) needs to have its stop value set to 127 (LAST_SAD + 1) rather than 126 (LAST_SAD). Otherwise it will look for subaddresses 0 through 29, and not see a device at subaddress 30.

for more information, see https://pre-commit.ci

MatthieuDartiailh

LGTM (minus the changelog formatting)

Co-authored-by: Matthieu Dartiailh <marul@laposte.net>

MatthieuDartiailh · 2026-07-01T08:01:36Z

Can you address the linting/typin failure ? (mostly a bad import I think)

Oops... missed one set of cases where I had initially used GPIBwhatever and now (with imported symbols) need to use gpib_constants.whatever Also, sort the gpib_constants import statement

for more information, see https://pre-commit.ci

Another one I missed: GPIBconfig -> gpib_constants.config

Apparently, in this case, an IntEnum is not sufficiently like an int to be type-matched.

for more information, see https://pre-commit.ci

…nto gpib-enums-clean

DavidCPlatt · 2026-07-02T00:31:46Z

OK, this looks better.

I fixed the two enum-name-change sets (GPIBconfig and GPIBerror) I missed during my original restructuring.

I've moved the import into the proper sorted position. There's one other complaint about an old import, but I don't know if I should touch it now (I didn't, before).

I fixed the type mismatch involving the SAD, although it took a couple of tries for me to realize what was actually going on. Apparently IntEnums don't always behave precisely like integers, from a type-checking perspective.

Please take another look.

MatthieuDartiailh · 2026-07-02T06:35:35Z

Looks good. One more import sorting you can use ruff to fix.

Run ruff on gpib.py, merging two import statements into one.

DavidCPlatt · 2026-07-02T16:22:53Z

Done.

DavidCPlatt and others added 4 commits June 26, 2026 12:24

Modify GPIB code to use symbolic constants

ea1c4e3

Create a file of symbolic constants (IntEnum classes) for the various libgpib interface calls. Use these symbols rather than hard-coded integers. Remove now-extraneous commentary about what the hard-coded integers mean.

[pre-commit.ci] auto fixes from pre-commit.com hooks

b725714

for more information, see https://pre-commit.ci

Document the GPIB symbolic-constants change

c1b2c09

Merge branch 'gpib-enums-clean' of github.com:DavidCPlatt/pyvisa-py i…

68557a2

…nto gpib-enums-clean

MatthieuDartiailh reviewed Jun 26, 2026

View reviewed changes

DavidCPlatt and others added 4 commits June 26, 2026 15:10

Add symbolic constants for secondary device address handling

e7e006e

Turn 0, 96, and 126 into symbolic constants for the SAD.

[pre-commit.ci] auto fixes from pre-commit.com hooks

c7055ae

for more information, see https://pre-commit.ci

[pre-commit.ci] auto fixes from pre-commit.com hooks

dcba3f4

for more information, see https://pre-commit.ci

MatthieuDartiailh reviewed Jun 30, 2026

View reviewed changes

Comment thread CHANGES

Comment thread pyvisa_py/gpib.py

MatthieuDartiailh mentioned this pull request Jun 30, 2026

Add support for NI GPIB-ENET/100 bridge. #587

Open

4 tasks

Update CHANGES per suggestion

024775d

Co-authored-by: Matthieu Dartiailh <marul@laposte.net>

DavidCPlatt and others added 7 commits July 1, 2026 16:10

Fix bad import/use of GPIB error symbols

7916b94

Oops... missed one set of cases where I had initially used GPIBwhatever and now (with imported symbols) need to use gpib_constants.whatever Also, sort the gpib_constants import statement

[pre-commit.ci] auto fixes from pre-commit.com hooks

4feea5d

for more information, see https://pre-commit.ci

Fix GPIBconfig member names

ab05b04

Another one I missed: GPIBconfig -> gpib_constants.config

Coerce "sad" type to int() to avoid lint complaint

6dafd3e

Apparently, in this case, an IntEnum is not sufficiently like an int to be type-matched.

[pre-commit.ci] auto fixes from pre-commit.com hooks

8d90179

for more information, see https://pre-commit.ci

Try the SAD fix a different way...

32f656a

Merge branch 'gpib-enums-clean' of github.com:DavidCPlatt/pyvisa-py i…

9aaccd3

…nto gpib-enums-clean

Fix import using ruff

dd28298

Run ruff on gpib.py, merging two import statements into one.

MatthieuDartiailh merged commit 721c065 into pyvisa:main Jul 2, 2026
18 checks passed

DavidCPlatt deleted the gpib-enums-clean branch July 2, 2026 18:10

Uh oh!

Uh oh!

Conversation

DavidCPlatt commented Jun 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov Bot commented Jun 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

MatthieuDartiailh left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

MatthieuDartiailh Jun 26, 2026

Choose a reason for hiding this comment

Uh oh!

DavidCPlatt Jun 26, 2026

Choose a reason for hiding this comment

Uh oh!

DavidCPlatt Jun 26, 2026

Choose a reason for hiding this comment

Uh oh!

DavidCPlatt Jun 27, 2026

Choose a reason for hiding this comment

Uh oh!

MatthieuDartiailh left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

MatthieuDartiailh commented Jul 1, 2026

Uh oh!

DavidCPlatt commented Jul 2, 2026

Uh oh!

MatthieuDartiailh commented Jul 2, 2026

Uh oh!

DavidCPlatt commented Jul 2, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

DavidCPlatt commented Jun 26, 2026 •

edited

Loading

codecov Bot commented Jun 26, 2026 •

edited

Loading

MatthieuDartiailh left a comment •

edited

Loading