Nonreducing

[andrewgsavage]

• Closes # (insert issue number)
• Executed pre-commit run --all-files with no errors
• The change is fully covered by automated unit tests
• Documented in docs/ as appropriate
• Added an entry to the CHANGES file

[codspeed-hq]

Merging this PR will not alter performance

✅ 448 untouched benchmarks

---

_{Comparing andrewgsavage:nonreducing (af3cd64) with master (86f37c0)}

[andrewgsavage]

I think support for this can be added in a few functions, eg

mm_per_mm = ureg.NonReducingUnitsContainer([ureg.mm, 1/ureg.mm])
Quantity(1, mm_per_mm)
Quantity(1,"").to(mm_per_mm)

I am unsure on using this in more places as it would change behaviour where units currently cancel out.

[andrewgsavage]

I am unsure on using this in more places as it would change behaviour where units currently cancel out.

Could set behaviour with a registry setting like auto_reduce_dimensions

[mutricyl]

For information I have au use case that made me open #2256
I often use transfer function in my industry. You have to compute the response of the system when excited and divide it by the amplitude of the exciting process. To be more specific vessel motion (meter) divided by wave amplitude (meter).
As I always normalize results with wave amplitude (m), I use a custom formatter to force displaying dimensionless to m/m

[andrewgsavage]

For information I have au use case that made me open #2256 I often use transfer function in my industry. You have to compute the response of the system when excited and divide it by the amplitude of the exciting process. To be more specific vessel motion (meter) divided by wave amplitude (meter). As I always normalize results with wave amplitude (m), I use a custom formatter to force displaying dimensionless to m/m

would you be willing to look through this PR and review it?

[andrewgsavage]

For information I have au use case that made me open #2256 I often use transfer function in my industry. You have to compute the response of the system when excited and divide it by the amplitude of the exciting process. To be more specific vessel motion (meter) divided by wave amplitude (meter). As I always normalize results with wave amplitude (m), I use a custom formatter to force displaying dimensionless to m/m

would you be willing to look through this PR and review it?

poke @mutricyl

[mutricyl]

Hey Andrew, sorry will not be available top have a look before march. Vacation time :-/ Le dim. 15 févr. 2026, 23:57, andrewgsavage ***@***.***> a écrit :

…

*andrewgsavage* left a comment (hgrecco/pint#1993) <#1993 (comment)> For information I have au use case that made me open #2256 <#2256> I often use transfer function in my industry. You have to compute the response of the system when excited and divide it by the amplitude of the exciting process. To be more specific vessel motion (meter) divided by wave amplitude (meter). As I always normalize results with wave amplitude (m), I use a custom formatter to force displaying dimensionless to m/m would you be willing to look through this PR and review it? poke @mutricyl <https://github.com/mutricyl> — Reply to this email directly, view it on GitHub <#1993 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A4JRUQC6L2VCNMND777HXC34MD2XFAVCNFSM6AAAAACUP66NKOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTSMBVGMZTMNRYGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[mutricyl]

Andrew,
I have played a little bit with the NonReducing feature but I came accross the following issue:

ureg_t = UnitRegistry(auto_reduce_units=True)
ureg_f = UnitRegistry(auto_reduce_units=False)

ureg_t('1.0 kN.m') # fine: <Quantity(1.0, 'kilonewton * meter')>
ureg_f('1.0 kN.m') # NOK: 
# pint\util.py", line 628, in __copy__
#     out._one = self._one
#               ^^^^^^^^^
# AttributeError: 'NonReducingUnitsContainer' object has no attribute '_one'
ureg_f.Unit('kN.m') # OK; <Unit('kilonewton * meter')>
ureg_f(kN) # OK: <Quantity(1, 'kilonewton')>
ureg_f('m/s/m') # NOK: same AttributeError as above in pint.util.UnitsContainer.__copy__

Parsing of a new Quantity with auto_reduce_units set to false sometimes fails when it come to copy the UnitContainer. The Error is not present when auto_reduce_units is set to tue or when Unit method is used.