Archives with undetected character sets

[gingerbeardman]

(I have to zip up the LZH archives to add them to GitHub)

Japanese

• bakaha10.lzh.zip (Archiver.app worked OK)
• hanaf131.lzh.zip (Archiver.app worked OK)
• hatiha10.lzh.zip (Archiver.app worked OK)
• kikyo11.lzh.zip (The Unarchiver.app worked OK)
• Sengoku-Hanafuda_ver.1.09.zip (Keka.app prompts for charset selection)

[aonez]

Thanks @gingerbeardman, the charset detection must be improved.

My results where different though. In my tests:

• bakaha10: TUA asks encoding, Archiver and Keka use bad encoding
• hanaf131: TUA asks encoding, Archiver and Keka use bad encoding
• hatiha10: TUA and Archiver ok, Keka uses bad encoding
• kikyo11: Archiver ok, TUA guesses correct encoding, Keka uses bad encoding
• Sengoku-Hanafuda_ver.1.09: Archiver ok, TUA guesses correct encoding, Keka asks encoding

[gingerbeardman]

Let's go with your test results. I'm not sure why mine were different.

[aonez]

Just realized Keka has LHA extraction support but no format detection. I'm fixing that.

[gingerbeardman]

Just an addition to this, whilst Keka extracts these files from #374 correctly, the incorrect charset is used for some filenames:

• tyq01.exe.zip
  • should be お読みください.txt and 遊び方.htm
  • 
• usahana104.exe.zip
  • should be 花札.cnt and 花札.HLP
  • 

[aonez]

Thanks @gingerbeardman! Already noted that on #374 (comment).

It's incredible how many LHA variations there're. Do you think Japanese people still using it for compression? Or are those old files (like golden game gems) that still compressed in that format?

[gingerbeardman]

LHA/LZH just really took off in Japan, because it was created by a Japanese guy. It's still used widely, in fact Windows 7 Japanese comes with LZH folder support! Not sure about Windows 10 Japanese but I assume it will still have it.

https://en.wikipedia.org/wiki/LHA_(file_format)

[aonez]

@gingerbeardman version 1.1.13 just released with most LZH files properly extracted 😊 As for the EXE ones, I'll try to fix that one in #374. It is fairly more complicated to detect the compression method used in some EXE files.

Thanks for the feedback as always!