[TC] Sometimes, Internal Viewer (Ctrl-q) fails to search a keyword in a non-unicode (cp949) *.txt

[cpp64]

OS: Windows 11 x64
OS-Locale: Korean
TC: 11.00rc1 x64

[ 1 ]
----- cp949-001.txt
애수
가나다라마바사
아자차카타파하
-----

The cp949-001.txt is displayed well in the Internal Viewer.

But it fails when searching a keyword and copying-and-pasting strings.

When searching "가나다", TC shows a message "Not found, 가나다".

And when copying-and-pasting all the strings from TC into Notepad,
Notepad shows
-----
¾Ö¼ö
°¡³ª´Ù¶ó¸¶¹Ù»ç
¾ÆÀÚÂ÷īŸÆÄÇÏ
-----
.


[ 2 ] without "애수"
----- cp949-002.txt
가나다라마바사
아자차카타파하
-----
is OK in all tests.

[ghisler(Author)]

I will check it, thanks for your report.
Edit:
Can you tell me what is set in the menu "Encoding"? Is it set to
- As configured for current font
or
- ANSI (0)
or
- Korean (949)

I ask because with my current locale, I have to set it to Korean (949), and then both the search and the copying of text work fine.
Could you please try with the other two options?

[cpp64]

2ghisler

I found it.

==============================
OS: Windows 11 x64
OS-Locale: Korean
TC: 11.00rc1 x64

[Font at startup] Autodetect

[Encoding] As configured for current font

[ANSI] Hack,-40,0,D144
--> script: Western (any monospaced font that doesn't have Korean)

[ASCII] Courier New,-96,0,D144
--> script: Western (any monospaced font that doesn't have Korean)

[Variable] Batang,-40,129,D144
--> script: Hangul (any font that has Korean)
==============================

cp949-001.txt with "애수" is opened as ANSI. (Hack) // It seems that "ANSI + Western" is applied, not "ANSI + local code page".

cp949-002.txt without "애수" is opened as ASCII. (Courier New) // This is OK.

------------------------------

[Encoding] ANSI (local code page) (0)
[Encoding] ASCII/DOS (local code page) (1)
[Encoding] cp949
are OK.

However, the displayed font for both English and Korean is "GulimChe".

------------------------------

[Encoding] MacOS (2)
is OK.

"Alphabet + Korean" in TXT:
The displayed font for English is Hack,-40,0,D144
and for Korean is Malgun Gothic (Windows' default font link for Korean).

"Alphabet only" in TXT:
The displayed font for English is MacOS property font.

------------------------------


ps.

Keeping Ctrl-q with Autodetct,

[ . --(down key)--> cp949-001.txt ] cp949-001.txt is opened as ANSI
and
[ cp949-002.txt --(up key)--> cp949-001.txt] cp949-001.txt is opened as ASCII

In some cases, previous encoding with Autodetect seems to affect subsequent encoding.

[ghisler(Author)]

So if I understand you correctly, it misbehaves when using "As configured for current font" when the font does not include Korean encoding?

[cpp64]

So if I understand you correctly, it misbehaves when using "As configured for current font" when the font does not include Korean encoding?

Yes, ANSI + "any monospaced fonts with Font script [Western]" + "As Configured for current font" causes the problem when copying and pasting text from TC to another editor.

// Hangul == Korean
This means that any monospaced fonts without Hangul will always cause the problem.

It seems that the locale of the OS (cp949) does not apply to it.


# Font script: Western
----------------------------------------------------------------------------------------------------
ANSI + Font script [Western] + "As Configured for current font"
----------------------------------------------------------------------------------------------------

Hack (without Korean) that has "Font script, [Western]" causes the problem when the font script is [Western].

D2Coding (eng + kor) that has "Font script, [Western + Hangul]" also causes the problem when the font script is [Western].

This is the same as if "[Encoding] Western Latin1 (1252)" were selected in Lister's Encoding submenu.


# Font script: Hangul (Korean)
----------------------------------------------------------------------------------------------------
ANSI + Font script [Hangul] + "As Configured for current font"
----------------------------------------------------------------------------------------------------

D2Coding (eng + kor) that has "Font script, [Western + Hangul]" is OK when the font script is [Hangul].


ps.

[D2Coding]
https://github.com/naver/d2codingfont

[Hack]
https://github.com/source-foundry/Hack

[ghisler(Author)]

I found the problem: When I create a font with the encoding set to 0, it creates a font with the encoding of current locale/language. But TC was assuming that it would be creating a font with "Western" encoding, because that's what the font choose dialog returns when you choose "Western". So the displayed encoding (Korean) differs from what TC assumes. I will fix it.

[cpp64]

Confirmed fixed in tc-11.00-rc2

[white]

Moderator message from: white » 2023-07-19, 22:53 UTC

Moved to Fixed bugs.