Thank you for maintaining this important plugin. Is Your Feature Re

Hey <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url=

<a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/us

Hey <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url=

FR: Ignore hyphens in "Space between Chinese Japanese or Korean and English or numbers" about obsidian-linter HOT 11 CLOSED

RyotaUshio commented on September 21, 2024

FR: Ignore hyphens in "Space between Chinese Japanese or Korean and English or numbers"

from obsidian-linter.

Comments (11)

pjkaufman commented on September 21, 2024

Hey @RyotaUshio , I think I am having trouble following why hyphens would be usend instead of their fullwidth equivalent. Could you elaborate on this?

from obsidian-linter.

RyotaUshio commented on September 21, 2024

@pjkaufman Thank you for your attention. To be honest, I'm not sure about the exact reason.
But I'm pretty sure that at least 99% of us Japanese use half-width hyphens - instead of the full-width − in this situation.

One reason I've come up with is that the full-width hyphen is much harder or more tedious to type than the half-width hyphen.
This is due to our keyboard layout and the nature of Japanese IMEs.
In my quick test, the full-width one required 7 keystrokes while the half-width one required only one.

from obsidian-linter.

pjkaufman commented on September 21, 2024

Hey @RyotaUshio , do you think it would make sense to expose the punctuation/symbol list for English letter symbols to check for around the CJK characters?

For example, right now that logic lives here:

obsidian-linter/src/rules/space-between-chinese-japanese-or-korean-and-english-or-numbers.ts

Lines 26 to 42 in 416f37f

 const head = /(\p{sc=Han}|\p{sc=Katakana}|\p{sc=Hiragana}|\p{sc=Hangul})( *)(\[[^[]*\]\(.*\)|`[^`]*`|\w+|[-+'"([¥$]|\*[^*])/gmu; 

 const tail = /(\[[^[]*\]\(.*\)|`[^`]*`|\w+|[-+;:'"°%$)\]]|[^*]\*)( *)(\p{sc=Han}|\p{sc=Katakana}|\p{sc=Hiragana}|\p{sc=Hangul})/gmu; 

 // inline math, inline code, markdown links, and wiki links are an exception in that even though they are to be ignored we want to keep a space around these types when surrounded by CJK characters 

 const regexEscapedIgnoreExceptionPlaceHolders = `${IgnoreTypes.link.placeholder}|${IgnoreTypes.inlineMath.placeholder}|${IgnoreTypes.inlineCode.placeholder}|${IgnoreTypes.wikiLink.placeholder}`.replaceAll('{', '\\{').replaceAll('}', '\\}'); 

 const ignoreExceptionsHead = new RegExp(`(\\p{sc=Han}|\\p{sc=Katakana}|\\p{sc=Hiragana}|\\p{sc=Hangul})( *)(${regexEscapedIgnoreExceptionPlaceHolders})`, 'gmu'); 

 const ignoreExceptionsTail = new RegExp(`(${regexEscapedIgnoreExceptionPlaceHolders})( *)(\\p{sc=Han}|\\p{sc=Katakana}|\\p{sc=Hiragana}|\\p{sc=Hangul})`, 'gmu'); 

 const addSpaceAroundChineseJapaneseKoreanAndEnglish = function(text: string): string { 

 return text.replace(head, '$1 $3').replace(tail, '$1 $3'); 

 }; 

 let newText = ignoreListOfTypes([IgnoreTypes.italics, IgnoreTypes.bold], text, addSpaceAroundChineseJapaneseKoreanAndEnglish); 

 newText = newText.replace(ignoreExceptionsHead, '$1 $3').replace(ignoreExceptionsTail, '$1 $3'); 

 newText = updateItalicsText(newText, addSpaceAroundChineseJapaneseKoreanAndEnglish); 

 newText = updateBoldText(newText, addSpaceAroundChineseJapaneseKoreanAndEnglish);

Essentially there are two sets of punctuation that are looked at:

Punctuation/symbols before CJK that are considered to be English: -+'"([¥$]
Punctuation/symbols after CJK that are considered to be English: -+;:'"°%$)]

Note: I think * is also checked for, but I think I would keep that hidden since it is a part of a markdown syntax check which could get hairy to try to allow the user to customize.

I can expose these two values as a list of characters to consider to be English when they are found in their respective locations. This would also fix #1036 .

How does that sound? Any suggestions around this?

from obsidian-linter.

pjkaufman commented on September 21, 2024

A dev note: I will need to handle an empty list to make sure that we do not end up with an empty capture group since I think that is invalid in regex. So I need to handle that to make sure someone does not hose the linter's ability to run by doing that.

from obsidian-linter.

RyotaUshio commented on September 21, 2024

By "expose" do you mean adding a setting to customize the list?

from obsidian-linter.

pjkaufman commented on September 21, 2024

Yes. Sorry. I forgot to say that it would be a user setting that would be a text box with all of those characters in it that could be edited. It is not a list in the way it would be displayed in the UI.

from obsidian-linter.

pjkaufman commented on September 21, 2024

I think I have the code ready for this kind of change, but I need to test in Obsidian and with a couple of edge cases to make sure it does not die. Once I do that, I will go ahead and see about making the change. But if you want to give feedback on what I have written above, feel free to do so.

from obsidian-linter.

RyotaUshio commented on September 21, 2024

Thanks for elaborating, it sounds great! I really appreciate your quick work 🙏

from obsidian-linter.

pjkaufman commented on September 21, 2024

No problem. Today I am trying to knock out some issues and requests before I do the next release and start focusing back on better handling of mass linting operations.

from obsidian-linter.

pjkaufman commented on September 21, 2024

I have merged the change in question. It should be on master and go out in the next release. It looks to be working, but there may be some issues with it. Feel free to let us know if there are any issues.

from obsidian-linter.

RyotaUshio commented on September 21, 2024

Thank you so much!! From my quick test, it seems to be working perfectly. I'll let you know if I noticed anything.

from obsidian-linter.

FR: Ignore hyphens in "Space between Chinese Japanese or Korean and English or numbers" about obsidian-linter HOT 11 CLOSED

Comments (11)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	const head = /(\p{sc=Han}\|\p{sc=Katakana}\|\p{sc=Hiragana}\|\p{sc=Hangul})( )(\[[^[]\]\(.\)\|`[^`]`\|\w+\|[-+'"([¥$]\|\[^])/gmu;
	const tail = /(\[[^[]\]\(.\)\|`[^`]`\|\w+\|[-+;:'"°%$)\]]\|[^]\)( )(\p{sc=Han}\|\p{sc=Katakana}\|\p{sc=Hiragana}\|\p{sc=Hangul})/gmu;
	// inline math, inline code, markdown links, and wiki links are an exception in that even though they are to be ignored we want to keep a space around these types when surrounded by CJK characters
	const regexEscapedIgnoreExceptionPlaceHolders = `${IgnoreTypes.link.placeholder}\|${IgnoreTypes.inlineMath.placeholder}\|${IgnoreTypes.inlineCode.placeholder}\|${IgnoreTypes.wikiLink.placeholder}`.replaceAll('{', '\\{').replaceAll('}', '\\}');
	const ignoreExceptionsHead = new RegExp(`(\\p{sc=Han}\|\\p{sc=Katakana}\|\\p{sc=Hiragana}\|\\p{sc=Hangul})( *)(${regexEscapedIgnoreExceptionPlaceHolders})`, 'gmu');
	const ignoreExceptionsTail = new RegExp(`(${regexEscapedIgnoreExceptionPlaceHolders})( *)(\\p{sc=Han}\|\\p{sc=Katakana}\|\\p{sc=Hiragana}\|\\p{sc=Hangul})`, 'gmu');
	const addSpaceAroundChineseJapaneseKoreanAndEnglish = function(text: string): string {
	return text.replace(head, '$1 $3').replace(tail, '$1 $3');
	};

	let newText = ignoreListOfTypes([IgnoreTypes.italics, IgnoreTypes.bold], text, addSpaceAroundChineseJapaneseKoreanAndEnglish);

	newText = newText.replace(ignoreExceptionsHead, '$1 $3').replace(ignoreExceptionsTail, '$1 $3');

	newText = updateItalicsText(newText, addSpaceAroundChineseJapaneseKoreanAndEnglish);

	newText = updateBoldText(newText, addSpaceAroundChineseJapaneseKoreanAndEnglish);