Steps to reproduce:
1. In Portuguese language the thousands separator used is not who is configured in the operating system.
2. In Portuguese should be '111 111,11' (as in French) and not '111.111,11'
I wonder where that is defined. The decimal separator is defined in Menu 'Tools -> Options -> Language'. May be it's "somehow" in a template?
At least I can say that thousands separator setting by Language in Calc dialog 'Format -> Cells -> Numbers -> Language' works, Clicking through the languages will show the different decimal and thousands separators, for example a (thin?) space as thousands separator for Estonian and Esperanto. But for Portuguese dot / comma is used as thousands / decimal separator
<https://en.wikipedia.org/wiki/Decimal_separator> says that that is a correct alternative.
In Germany DIN 1333, DIN 5008 und ISO 31 recommend to use the thin space as thousands separator, but more or less nobody does it.
Is there any public available standard defining the thin space as only allowed separator?
Please read the document from Bureau International des Poid e Measures and approved on the 22nd General Conference:
"," (comma) and "." (dot) should be used for decimal separator and " " (space) should be used for thousand separator.
Portuguese should be "11 111,11" for default.
Such communiques are not decisive for us, LibO will follow the common use in a country.
"... o Censo dos Estados Unidos de 2010, atinge 8.175.133 habitantes ..." seems to tell that the dot as separator is not wrong.
Please feel free to reopen this bug if you can contribute conclusive indication that the thin space separator really is common use
Created attachment 63017 [details]
although not be a bug, it would be good use a space as the thousands separator, it is what is used in handwriting and teach you in school. As you can see, this is possible in excel and google docs (default)
Bug #62128 is very similar to this one. The thousands separator should be customizable and spaces (or thin spaces) should be eligible thousands separators, since practically all the world uses space thousands separators by standard.
Since LibreOffice is being incrementally adopted by nations as default "office" software, and almost every country follows the SI standards (which as cited by our friend bessa, recommend a space as thousands separator and forbid any non-space character as an alternative), solving this bug should be of great importance.
Thin space is a bad choice due to weak font support.
Created attachment 85942 [details]
Python script with all relevant functions to solve this bug implemented.
I know LibreOffice is not coded in Python, but I thought I could help this bug be solved a bit with this contribution. All that is needed is to transcribe this code to C++ and modify accordingly so it works in LibreOffice.
(In reply to comment #7)
> Thin space is a bad choice due to weak font support.
According to the SI standard, a normal space is perfect.
Need independent confirmation from QA - moving to UNCONFIRMED. Thanks
Guys, we are talking about *locale* data, not SI or any other standard. And statements like "practically all the world uses space thousands separators by standard" really don't help, because it might be that it happens to be just the part of the world you know about. This bug is about Portuguese (in Portugal, I guess?) so please don't mix in other observations. I also don't see any Spanish locale mentioned here, so removing ES from the summary again.
At http://www.unicode.org/cldr/charts/26/by_type/numbers.symbols.html#a1ef41eaeb6982d we see that pt_PT (Portuguese in Portugal) uses the space group separator and all other pt locales use the dot separator.
Eike Rathke committed a patch related to this issue.
It has been pushed to "master":
fdo#51008 change [pt-PT] group separator to no-break space
It will be available in 4.4.0.
The patch should be included in the daily builds available at
http://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More
information about daily builds can be found at:
Affected users are encouraged to test the fix and report feedback.