Is there some mechanism to decode HTML charset

[cybexr]

Symptom : xidel cannot treat charset correctly.
Example: xidel.exe --html 5.htm -e "//div[@Class='qxName']/a" --stdin-encoding=oem >5o.htm --output-encoding=input

This 5.htm has , but the outpupt is incorrect.

After I convert the 5.htm file to UTF8 encoded, the output is fine. It seems xidel always treat input file as UTF8 , am I missed something? or xidel can only work like this .

[benibela]

What are the encodings of the files?

With -output-encoding=input 5o.htm should have the same encoding as 5.htm

--stdin-encoding=oem is ignored here, it is only used when you do - <5.htm

[cybexr]

the 5.htm file is GBK encoded, which is DBCS. it's quite different from european-language( all are SBCS) https://en.wikipedia.org/wiki/GBK_(character_encoding)
https://en.wikipedia.org/wiki/DBCS

actually the 5.htm is download from a web-site http-response with header:Content-Type: text/html; charset=gb2312 .

[cybexr]

tryed again, just with option : -output-encoding=input result is correct. both on windows&centos.
maybe I'm confused with debugging so much http response yesterday.
Thank you benibela !