Implement GOST 16876-71:1983 Russian (`gost-rus-cyrl-latn-16876-71-1983`)

[ronaldtse]

I was able to find GOST 1983 here: 7th_UNCSGN_econf.91_3_Add1.html.zip

• it is the OCR'ed from this file: https://unstats.un.org/unsd/geoinfo/UNGEGN/docs/7th-uncsgn-docs/econf/7th_UNCSGN_econf.91_3_Add1.pdf

We should Implement GOST 16876-71:1983 Russian (gost-rus-cyrl-latn-16876-71-1983)

Originally posted by @ronaldtse in https://github.com/interscript/interscript/issues/76#issuecomment-635052191

[CAMOBAP]

@ronaldtse I have found original GOST document here http://vsegost.com/Catalog/45/45002.shtml

I have noted that for some letters it allow to select different results: for example: Щ -> Šč or Ŝ

If I will compare attached screenshots with http://vsegost.com/Catalog/45/45002.shtml it looks inconsistent, because something's it select the first variant (without brackets) or second (within brackets)

@ronaldtse what is your opinion should we follow https://unstats.un.org/unsd/geoinfo/UNGEGN/docs/7th-uncsgn-docs/econf/7th_UNCSGN_econf.91_3_Add1.pdf or http://vsegost.com/Catalog/45/45002.shtml ?

If you ask me if we need real GOST 16876-71 we should check http://vsegost.com/Catalog/45/45002.shtml

P.S. How do we really know that GOST 1983 and GOST 16876-71 refer to the same document?

[ronaldtse]

P.S. How do we really know that GOST 1983 and GOST 16876-71 refer to the same document?

It was stated in one of the UNGEGN documents.

I think GOST 16876-71 provides two systems, the first has no choice, the second allows choice of the rules.

Have you found any discrepancy between https://unstats.un.org/unsd/geoinfo/UNGEGN/docs/7th-uncsgn-docs/econf/7th_UNCSGN_econf.91_3_Add1.pdf and http://vsegost.com/Catalog/45/45002.shtml ?

[CAMOBAP]

Have you found any discrepancy between

Formally no

But to give you a more complete picture about GOST 16876-71:

• It defines two tables: one with diacritics, second without
• Even for the one with diacritics, it allows multiple choice
• This is why I was a bit confused by UNGEGN doc (the doc from top comment), it selects first variant for all letters except Щ. For this letter, it selects the second variant: Šč instead of first one: Ŝ

Let me confirm again that formally this isn't an issue

[CAMOBAP]

@ronaldtse I have found relation GOST 1983 and GOST 16876-71

16 April 1982 - the year when this GOST was officially introduced 13 May 1983 - contains an updated table and according to it it 100% equal to the one which attached it top comment

Official document https://rosreestr.ru/upload/documenty/doc_169.doc

13 May 1983 edition eliminate pluralism of translation, so staring from it there is exists only one way to 'convert' characters

[CAMOBAP]

PR is ready for review https://github.com/interscript/interscript/pull/296

[CAMOBAP]

Done

[ronaldtse]

Thanks @CAMOBAP !