severity 347173 minor thanks > > Please patch debian sources until upstream integrates it. I have sent a patch and bug in upstream at http://sourceware.org/bugzilla/show_bug.cgi?id=2125 Currently I have contacted Mr. George Prutean (a known Romanian linguist) that, hopefully will put me in contact with somebody at the Academy and we will clarify this issue. > Hi Eddy, > > could you please have a look at #119528? > I do not know what to do with this bugreport. Note: I have answered regarding all the changes that I could check Coding iso-8859-16 - is not that widely used in Romania and applications that support it are really rare. As etch's (secondary) target is usage of UTF-8 and currently most people use iso-8859-2 or utf-8 I propose that we switch to utf-8 (support is still scarce - note the mc issues, mailers issues and so on) or stick momentanely with -2 (but will have to switch later to UTF-8). abday: This is wired, all days are reduced to one letter (preceded by one space) except Tuesday and Wednesday. These are two letters wide as the Romanian equivalents start both with letter m. So the abbreviations are " D", " L", "Ma", "Mi", " J", " V", " S" instead of "Du", "Lu", "Ma", "Mi", "Jo", "Vi", "Sâ". To be honest the mostly used abbreviations are " D", " L", " M", " M", " J", " V", " S" (yes, with the confusion embedded ;-) and I think Ionel was trying to address this issue and make the abbreviations not weird to Romanians (Jo and Sâ are the most weird), but my honest oppinion is that he two letter abbreviation is better. [I tested at home on a gentoo system the localization of the date with Ionel's ro_RO file] Note: the << >> signs were produced by the test.sh script, see sources. Attached are the results of the tests. What I can say is that: - as one can easily see the default ro_RO locale is broken (not recognised) - the ro_RO.ISO-8859-16 is not recognised (I feel that I am doing something wrong here) - Saturday has the pre-92 orthography (î instead of â) - Tuesday is written with the correct code for t, - t comma below (U021B); There is, however, the fear that using 21B might induce compatibility problems (Windows has no real support for t comma) - there is no transliteration available for ţ (21B), ş(219) (and their upper case counter parts) to the incorrect-but-looking-right-highly-used-preently t cedilla, s cedilla - so if 21B is not available, then one will see nothing (afair it was not transliteration to t either) - it is correct to use lower case characters for a month name and day name (ianuarie instead of Ianuarie and luni instead of Luni) - I am still uncertain if the abbreviated month should be capitalised, but I recall that it shouldn't be (so, ian is correct instead of Ian) - the date format is weird: the result of LC_ALL=ro_RO.UTF-8 LANG=ro_RO.UTF-8 date is "joi, 12 ianuarie 2006, 08h 44'26''", but nobody uses this format for displaying time (although is correct for geographical locations :) . Instead locale's time representation << 08:44:26 >> should be used. - AFAIK, the correct abmon for Noiembrie is "nov" (due to the latin origin oof the month name - I don't know why the abmon for ianuarie, iunie, iulie are not jan jun jul, but as I can see this exception is consistent as all are starting with "i") - I haven't tested the number groupping and the decimal separator but they should be "." for groups' separation and "," for decimal simbol (how can I test this?) - the current Romanian alphabet has 31 letters (the 26 letters in the English one + ăâîşţ) and they are inserted as follows : ă, â after a, ş after s, ţ after t. Yes, Q W and Y are a part of the alphabet, contrary to Ionel's statement. - abday as Ionel implemented has several flaws: It is inconsistent, can't be alligned properly on the left, it is rarely used except in calendars, where there is a disambiguity even if one letter abbreviations are used. The ISO-8859-2 encoding has t cedilla and s cedilla (which are incorrect, but widely used in Romania) while -16 has indeed t comma and s comma, but I think the one true solution is to use UTF-8 by default in Romanian. It is corect that t cedilla belongs to no nation and it is incorrect, but many fonts have a representation for t sedilla code, but most of them are represented as t comma and support is present. The Y for yes is welcome, as many applications are half translated or not at all translated in Romanian. This is a good safety net. The two letter abday Lu Ma Mi Jo Vi Sâ Du has none of the afore mentioned problems. I am working on a better solution and waiting for the input for the Academy, then I will send another mail. -- Regards, EddyP ============================================= "Imagination is more important than knowledge" A.Einstein
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = Locale : ro_RO Locale's month: Abrev. month: Jan Long month: January Abrev. month: Feb Long month: February Abrev. month: Mar Long month: March Abrev. month: Apr Long month: April Abrev. month: May Long month: May Abrev. month: Jun Long month: June Abrev. month: Jul Long month: July Abrev. month: Aug Long month: August Abrev. month: Sep Long month: September Abrev. month: Oct Long month: October Abrev. month: Nov Long month: November Abrev. month: Dec Long month: December Locale's day: Abrev. day: Sun Long day: Sunday Abrev. day: Mon Long day: Monday Abrev. day: Tue Long day: Tuesday Abrev. day: Wed Long day: Wednesday Abrev. day: Thu Long day: Thursday Abrev. day: Fri Long day: Friday Abrev. day: Sat Long day: Saturday Locale's date and time: << Thu Jan 12 08:44:26 2006 >> Locale's short date representation: << 01/12/06 >> Locale's time representation: << 08:44:26 >> the result of LC_ALL=ro_RO LANG=ro_RO date Thu Jan 12 08:44:26 EET 2006 = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = Locale : ro_RO.ISO-8859-16 Locale's month: Abrev. month: Jan Long month: January Abrev. month: Feb Long month: February Abrev. month: Mar Long month: March Abrev. month: Apr Long month: April Abrev. month: May Long month: May Abrev. month: Jun Long month: June Abrev. month: Jul Long month: July Abrev. month: Aug Long month: August Abrev. month: Sep Long month: September Abrev. month: Oct Long month: October Abrev. month: Nov Long month: November Abrev. month: Dec Long month: December Locale's day: Abrev. day: Sun Long day: Sunday Abrev. day: Mon Long day: Monday Abrev. day: Tue Long day: Tuesday Abrev. day: Wed Long day: Wednesday Abrev. day: Thu Long day: Thursday Abrev. day: Fri Long day: Friday Abrev. day: Sat Long day: Saturday Locale's date and time: << Thu Jan 12 08:44:26 2006 >> Locale's short date representation: << 01/12/06 >> Locale's time representation: << 08:44:26 >> the result of LC_ALL=ro_RO.ISO-8859-16 LANG=ro_RO.ISO-8859-16 date Thu Jan 12 08:44:26 EET 2006 = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = Locale : ro_RO.UTF-8 Locale's month: Abrev. month: ian Long month: ianuarie Abrev. month: feb Long month: februarie Abrev. month: mar Long month: martie Abrev. month: apr Long month: aprilie Abrev. month: mai Long month: mai Abrev. month: iun Long month: iunie Abrev. month: iul Long month: iulie Abrev. month: aug Long month: august Abrev. month: sep Long month: septembrie Abrev. month: oct Long month: octombrie Abrev. month: noi Long month: noiembrie Abrev. month: dec Long month: decembrie Locale's day: Abrev. day: D Long day: duminicÄ? Abrev. day: L Long day: luni Abrev. day: Ma Long day: marÈ?i Abrev. day: Mi Long day: miercuri Abrev. day: J Long day: joi Abrev. day: V Long day: vineri Abrev. day: S Long day: sîmbÄ?tÄ? Locale's date and time: << joi 12 ianuarie 2006 08:44:26 >> Locale's short date representation: << 2006-01-12 >> Locale's time representation: << 08:44:26 >> the result of LC_ALL=ro_RO.UTF-8 LANG=ro_RO.UTF-8 date joi, 12 ianuarie 2006, 08h 44'26''
Attachment:
test.sh
Description: Bourne shell script