Bug 77834 - FORMATTING: Option to keep numeric precision (significant figures) as entered
Summary: FORMATTING: Option to keep numeric precision (significant figures) as entered
Status: NEW
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Calc (show other bugs)
(earliest affected) release
Hardware: All All
: low enhancement
Assignee: Not Assigned
Whiteboard: BSA
Depends on:
Blocks: Number-Format
  Show dependency treegraph
Reported: 2014-04-23 20:32 UTC by DF
Modified: 2020-11-07 22:50 UTC (History)
6 users (show)

See Also:
Crash report or crash signature:
Regression By:


Note You need to log in before you can comment on or make changes to this bug.
Description DF 2014-04-23 20:32:15 UTC
For entering numeric data with decimal places, should have an option to keep set the number of decimal places as per the precision input by the user; not automatically remove any trailing zeros.

For example if one enters 0.10, Libre office will truncate to 0.1 and the user has to manually change the number of decimal places. It would be useful if Libre office could keep the number of decimal places based on the number of trailing zeros entered.

Another related enhancement would be to have the option to display a specified number of significant figures instead of number of decimal places. And carry it forward in calculations. For example one may input a list of data (ex 120, 34.567, 0.890) and want to display all of them with the same number of significant figures (say 3). This would require changing the number of decimal places for each cell which is impractical for large datasets. This could be extended to calculations where adding or multiplying figures would output a value with the correct number of significant figures (ex 0.10 * 10. = 1.0)

These enhancements would be incredibly useful for scientific data where significant figures are important and need to be tracked.
Operating System: All
Version: release
Comment 1 m.a.riosv 2014-04-25 23:01:38 UTC
Hi DF,
If you want an enhancement please change the importance from normal to enhancement.

Seems there are several request of enhancement, please do one report bug for every request.
Comment 2 DF 2014-04-25 23:36:03 UTC
Split significant figures part as bug# 77951.
Comment 3 Jean-Baptiste Faure 2014-06-22 20:30:57 UTC
Please could you describe a use-case where it is important to keep the trailing zeros ?

Set status to NEEDINFO. Please set it back to UNCONFIRMED once you provided the requested informations. Thank you for your understanding.

Best regards. JBF
Comment 4 DF 2014-06-23 17:34:42 UTC
With most scientific data significant figures are very important. The number of trailing zeros are used as placeholders denoting the precision of the number. https://en.wikipedia.org/wiki/Significant_figures privides a more indepth explanation.
Another example would be currency where one might include two trailing zeros after the decimal place even for whole values eg $1.00.

This feature request is simply for steamlining the procedure for specifying the precision. Currently one must manually right-click -> format cell -> change decimal places. It could be set automatically based on the number of decimal places entered by the user.
Example if I type or paste in 1.00, it should remain as 1.00, not automatically be set to 1.
Comment 5 Jean-Baptiste Faure 2014-06-28 18:34:16 UTC
Many users want to type the fewest possible characters, for example type 1 and get 1.00 by formatting the cells with 2 decimal digits.

Best regards. JBF
Comment 6 DF 2014-07-01 18:07:10 UTC
Yes, that is also useful. Ideally it would be easy to switch between the two.
Comment 7 Joel Madero 2014-07-16 20:47:50 UTC
Perfectly reasonable request (ie. do not alter user inputed numerical value). From what I can see, there is no way to currently do this. For instance:

If you want A1 = 2.10 and B1 = 2.100

There is no setting that can apply to both A1 and B1 that would allow this.

One option is to add an auto correct option that ignores decimal replacements.

Low - currently it seems like a very corner case for when this would be needed. You could always be MORE precise than you need to by making a format something like #.0000000000000000000000000000000000000000 and that should cover everything. (Obviously exagerating there, you could make the 0 to the largest number of trailing zeroes in your data set and then everything else would just be "more precise" than really needed but that should be fine)
Comment 8 kie000 2015-08-31 13:10:54 UTC
Just to add my voice, I'd say this would be a very nice feature to have and it should be an option, it is something I googled for and expected to find.
Comment 9 Joel Madero 2015-08-31 17:50:28 UTC
Please don't change the top section - it's fine to say you're interested in it too but "affects me too" is not a reason to increase the importance. Thanks
Comment 10 DF 2015-08-31 21:39:00 UTC
I'd agree adding greater precision is usually better than less, but in the case of at least scientific research the correct number of significant figures is quite important. The other examples such as currency I mentioned earlier, or maybe entering a product number which contains a decimal place and has lots of trailing zeros.

Ya, I think just being able to disable the auto-correct option for removing trailing zeros would be ideal. Infact the only reason I can think of for that auto-correct feature is strictly aesthetics. And how many people enter trailing zeros that they want auto-removed? In my experience it's be more of a burden than a benefit. So I'd almost suspect disabling that auto-correct feature by default would be more useful for most people.

Although this isn't a super critical enhancement, I think this would be too difficult to implement and for those who it would benefit it would make this software more streamlined to use and thus stand out that much more. So I'd suggest boosting the importance of it.
Comment 11 b. 2020-11-01 09:47:38 UTC
supporting the request, 

'7,5' (decimal) entered is not! the same as '7,5000' (decimal) entered! 

7,5 is a placeholder for the range from 7,450000000000000 to 7,549999999999999 if we stay with 16 digit precision, 

7,5000 is a placeholder for the range 7,499950000000000 to 7,500049999999999, 

it carries some precision information in it which is thrown away when truncating the trailing zeros, 

additional: user sometimes want a formatted display of columns, without first having to manipulate the cell format, 

and additional: from such a display respecting the input values you can spot some typing errors which is more difficult with display / values messed up by truncating zeroes and / or overriding cell format, 

shouldn't be too difficult to implement, an option to automatically change cell format acc. input, and store trailing zeroes in content.xml ...
Comment 12 Jean-Baptiste Faure 2020-11-07 20:47:35 UTC
(In reply to b. from comment #11)
> [...]
> shouldn't be too difficult to implement [...]

Hmmm, not the best way to convince a developer to work on this enhancement. If you know that, why do you don't do that yourself?

Rule #1: never tell a developer that a function is easy to implement.

Best regards. JBF
Comment 13 b. 2020-11-07 22:50:22 UTC


thanks for the hint, but ... a little late, my time machine is in repair, 

'trying myself' ... besides not being a c-programmer i tried to get in touch with LO code ... looks like it will need some more tries ... shortcomings were: 
- no 'pig picture' what's handeled where, how things should work, 
- debugger problems to catch 'threads', 
- too few comments in code to get 'the idea', 
- no info which other functions might be affected by changes, 
- 'c' inherently cryptic, 

thought to leave it to those who learned theese things earlier in live and be myself a bug-spotter, 

if there would be something like a mentoring service answering questions like 'where is cell format info stored?', 'which module handeles input processing' and plenty! similar i could give it a try ...