China DOS Union

Strange, after generation, I compared it.
http://www.cn-dos.net/forum/attachment.php?aid=950&checkid=de74b&download=1
with this one for differences

After generating the GBK HZK, I compared it with the HZK16.GBK provided here and found differences. So I uploaded it to Wuyou (here I can only upload 100K) for you. Download it and see what's going on? To prevent everyone from using problematic Chinese character libraries, I added a password. The password for the compressed package is: benshi. Download address: floor 259 http://bbs.wuyou.com/viewthread.php?tid=100870&pid=1008822&page=26&extra=page%3D1

I checked the HZK16.GBK that comes with UCWIN and found that it is not calculated according to the formula "(ch1-0x81)*190+(ch2-0x40)-(ch2/128)"! That is the order of Chinese characters for generating the GBK codebook, not the arrangement order in the font. The GBK font is stored in partitions! In order to be compatible with the HZK16 of GB2312, it first stores areas 1 to 94 in the original order, and then stores the extended parts in 3 other areas. For details, see the following list:

　　GBK character set range
　Equivalent to
"GB area code" Partition　　　High || Low　　
—————————————————————————————
Areas 1~94　　●GB Chinese characters: A1~FE || A1~FE
—　—　—　—　—　—　—　—　—　—　—　—　—　—　—
Areas 95~126　　●Extended Chinese characters: 81~A0 || A1~FE
Areas 127~159 　●Extended Chinese characters: 81~A0 || 40~A0
Areas 159+~254+　●Extended Chinese characters: A1~FE || 40~A0
—————————————————————————————
It seems that this is the order when I observed. However, you can check it again. Only after confirming it is correct can you start adjusting your program. I'm sleepy, going to sleep~~~

Expert user, expert thread!

Originally posted by 本是 at 01:48 AM:
I checked the HZK16.GBK that comes with UCWIN and found that it is not according to the formula

That is the order of Chinese characters for generating the GBK codebook, not the arrangement order in the font.
GBK characters ...

Headache, too lazy to study, it's so complicated that reading will be troublesome later! Just use
“(ch1-0x81)*190+(ch2-0x40)-(ch2/128)” to locate it,

Give the font a new name: GBK order font, and the previous one is the compatible font.

You try to use the above code to read the GBK font library I generated to see if it succeeded?

Originally posted by download at 2007-3-25 12:52 AM:
Not open source, bundles N rogue plugins, and charges a registration fee of 50 yuan per machine ^_^

Hmm, applause, the future Bill Gates of China

Originally posted by Ben Shi at 01:48 AM:
I checked the HZK16.GBK that comes with UCWIN and found that it is not calculated according to the formula
"(ch1-0x81)*190+(ch2-0x40)-(ch2/128)"!
That is the order of Chinese characters in generating the GBK codebook, not the arrangement order in the font library.
GBK characters...

The previous compatible area seems to have 94/96 Chinese characters per area, not 190. It's really troublesome to arrange like this

Originally posted by qingyangecho at 01:36 PM:

Well, applause, the future Bill Gates of China

Bill is not something to envy, overworked too much,
The current situation looks much older than the actual age
It's more practical to continue being my virgin...

Originally posted by download at 2007-3-26 02:27:
Can you try using the above code to read the GBK font I generated and see if it works?

I have browsed it with the font display program, and it is consistent with the order in GBK.TXT
(formula "(ch1-0x81)*190+(ch2-0x40)-(ch2/128)").
However, the font pattern order of each Chinese character is in the format that can be loaded into VGA text fonts,
that is, a single-byte type with 16 bytes on the left and right, not the double-byte type of ordinary Chinese fonts!!!

Therefore, the font library generated in this way is not a universal font library!
Suggestion: Adjust to a double-byte format for a single Chinese character, and the entire font library to a universal Chinese font library format.

It seems there is hope for using it in BCDW?

Really don't understand, passing by!

Originally posted by Ben Shi at 2007-3-26 05:22:
Suggestion: Adjust to... The entire character set is in the universal Chinese character set format.

Universal HZK16.GBK library
The calculation method of the record number in units of 32 bytes is as follows
(Assuming that the high byte of the Chinese character is in ch1 and the low byte is in ch2):
if ch1>A0h then
if ch2>A0h then '1-94:A1~FE || A1~FE
rec=(ch1-A1h)*94+(ch2-A1h)
else '159-254:A1~FE || 40~A0
rec=94*94+31*94+31*96+(ch1-A1h)*96+(ch2-40h)-(ch2/128)
endif
else
if ch2>A0h then '95-126:81~A0 || A1~FE
rec=94*94+(ch1-81h)*94+(ch2-A1h)
else '127-159:81~A0 || 40~A0
rec=94*94+31*94+(ch1-81h)*96+(ch2-40h)-(ch2/128)
endif
endif

[ Last edited by Ben Shi on 2007-3-26 at 09:03 AM ]

The part in red should be ch2\128

The formula is incorrect, and there are duplicate codes.