On 2019-03-29, Jürgen Spitzmüller wrote:
> Am Freitag, den 29.03.2019, 10:03 +0000 schrieb Guenter Milde:
>> > So with your solution a significant group of users will stick with
>> > 8bit encodings forever.
>> This is easily solved by changing the default default:
>> diff --git a/src/BufferParams.cpp b/src/BufferParams.cpp
>> index de46966f2e..028889cd66 100644
>> --- a/src/BufferParams.cpp
>> +++ b/src/BufferParams.cpp
>> @@ -425,7 +425,7 @@ BufferParams::BufferParams()
>> fonts_sans_scale[1] = 100;
>> fonts_typewriter_scale[0] = 100;
>> fonts_typewriter_scale[1] = 100;
>> - inputenc = "auto";
>> + inputenc = "utf8";
>> lang_package = "default";
>> graphics_driver = "default";
>> default_output_format = "default";
> This is just even more finger-painting.
> You basically want to defunc auto.
I want to bypass "auto" in order to keep its functionality as-is:
* Currently, the "auto" setting basically¹ means
use a mix of fixed-width 8-bit encodings
For documents with languages using the Latin, Greek, Cyrillic, and
Hebrew script, this allows to encode all supportedt letters even in
cases where utf8 would break something.
¹ For some languages it will be a mix of fixed-width and variable-width
encodings (10 out of the 82 languages with Babel support use
utf8 or legacy variable-width encodings.)
* When changing Encodings in lib/languages, "auto" becomes basically an
alias for "utf8", because "utf8" will be the new default for all²
languages.
² Only "Japanese (platex)" currently requires "utf8-platex".
* With my proposal, the "auto" setting is kept available as-is but not
used for new documents (unless set in a custom template).
To make clear that this is now a legacy setting, it may be renamed.
(We could also rename "Encoding" in lib/languages to something like
"LegacyInputEncoding".)
>> > I have this on my agenda, and I will definitely go on to change
>> > languages for 2.4.0, so be aware that all work you do on templates
>> > might be wasted time.
>> Please don't force this unless
>> a) it does not break any tests, or
>> b) a majority of developers agree that it is OK to break these tests.
> I will. If tests break, we can fix the cause.
The proposed format conversion ("auto" to current buffer default
encoding) will break the export tests for
autotests/export/latex/CJK/en-de-el-ru-zh_CN_language-default.lyx and
autotests/export/latex/en-ar_language-default.lyx. Try compiling
after changing the inputencoding to "Western European (ISO 8859-15)"
manually.
If going ahead with a file format change, please do the changes in a
feature branch, because reverting a file format change is not nice.
> In any case, just painting over the language's default encoding
> everywhere is not a sane solution.
In my opinion, using "utf8" instead of the language's legacy default
encoding for all languages is simple and sane.
Günter
