Questions about translating Arc42 into Simplified Chinese

[dannyge]

I am preparing to translate arc42 into Chinese, and during this process, I have some questions:

1. I see that the language codes used previously (such as in gradle.properties) are EN, DE, etc. When adding Simplified Chinese, should I use ZH or zh_CN (If a traditional Chinese version is available in the future, they can use zh_HK / zh_TW )?

2. When converting adoc format documents to PDF format using asciidoctor, Font files that support Chinese must be provided (otherwise, the PDF file will display garbled text after rendering). Can I use Noto Sans Simplified Chinese as the default Chinese font?

[dannyge]

see pr : #224

[node]

It's better to use zh-Hans instead of ZH or zh_CN.

In most cases, software documentation will only be translated into Simplified Chinese and Traditional Chinese. It will not be differentiated into more sub-categories by region, except for software interfaces or scenarios involving payments, laws in different regions, and so on.

[dannyge]

It's better to use zh-Hans instead of ZH or zh_CN.

In most cases, software documentation will only be translated into Simplified Chinese and Traditional Chinese. It will not be differentiated into more sub-categories by region, except for software interfaces or scenarios involving payments, laws in different regions, and so on.

@node I agree with your point.

After reviewing the Asciidoctor documentation, the relevant information for the lang attribute states:

"Allowable Values": "BCP 47 language tag (en)"

"Notes": "Language tag specified on the document element of the output document. Refer to the lang and xml:lang attributes section of the HTML specification for acceptable values of this attribute."

According to the "BCP 47" standard, “Chinese written using the Simplified Chinese script” should be represented as zh-Hans.

PS: AsciiDoctor itself uses zh_CN 😕 ,
For example：https://github.com/asciidoctor/asciidoctor/blob/main/data/locale/attributes-zh_CN.adoc

---

However, Use zh-Hans raises another issue:

In the gradle.properties languages is defined using all uppercase letters (the directory names of each language are also in all uppercase letters).

# Configure languages to be built
# Comma-separated list of language codes
languages=EN,DE,FR

In the build.gradle, (at lines 94 and 118):

attributes asciidocAttributes + [
  'imagesdir': "${projectDir}/images",
  'revnumber': versionInfo.revnumber,
  'revdate': versionInfo.revdate,
  'revremark': versionInfo.revremark,
  'lang': language.toLowerCase() // Set the language code
]

This code converts zh-Hans to zh-hans, which does not comply with the "BCP 47" standard.

Should we modify language.toLowerCase() to only lowercase the part before the -?

[gernotstarke]

thanx for this constructive discussion. Both @Hruschka and myself can neither speak nor write anything Chinese, so we have to trust your proposals.

[gernotstarke]

@node can you find the time to check the PR #224 from @dannyge please?

[node]

@node can you find the time to check the PR #224 from @dannyge please?

Certainly, I will check it and review today.
Meanwhile I notice that copilot point out several issues, and I will check all of them.

[node]

In the build.gradle, (at lines 94 and 118):

attributes asciidocAttributes + [
'imagesdir': "${projectDir}/images",
'revnumber': versionInfo.revnumber,
'revdate': versionInfo.revdate,
'revremark': versionInfo.revremark,
'lang': language.toLowerCase() // Set the language code
]
This code converts zh-Hans to zh-hans, which does not comply with the "BCP 47" standard.

Should we modify language.toLowerCase() to only lowercase the part before the -?

That's worth to try. Let us choose zh-Hans first.

[dannyge]

In the build.gradle, (at lines 94 and 118):

attributes asciidocAttributes + [
'imagesdir': "${projectDir}/images",
'revnumber': versionInfo.revnumber,
'revdate': versionInfo.revdate,
'revremark': versionInfo.revremark,
'lang': language.toLowerCase() // Set the language code
]
This code converts zh-Hans to zh-hans, which does not comply with the "BCP 47" standard.

Should we modify language.toLowerCase() to only lowercase the part before the -?

That's worth to try. Let us choose zh-Hans first.

Okay, I'll make some adjustments in the previous PR later.

[node]

How to visit the Chineser version now? Or we need to generate a new webpage for it?

[gernotstarke]

we (Peter, Ralf and/or myself) need to deploy that Chinese version via the arc42-generator... but that means updating THAT repo so it can handle Chinese character set... (possible early December)

Maybe I can manage a manual deploy during the upcoming week - I just got rid of a few customer assignments...

Sorry that's it's still complicated to get the new version up-and-productive... the legacy of our build system is haunting us...