summaryrefslogtreecommitdiffstats
path: root/third_party/rust/encoding_rs/doc/Big5.txt
diff options
context:
space:
mode:
Diffstat (limited to 'third_party/rust/encoding_rs/doc/Big5.txt')
-rw-r--r--third_party/rust/encoding_rs/doc/Big5.txt16
1 files changed, 16 insertions, 0 deletions
diff --git a/third_party/rust/encoding_rs/doc/Big5.txt b/third_party/rust/encoding_rs/doc/Big5.txt
new file mode 100644
index 0000000000..61e8fd5801
--- /dev/null
+++ b/third_party/rust/encoding_rs/doc/Big5.txt
@@ -0,0 +1,16 @@
+/// This is Big5 with HKSCS with mappings to more recent Unicode assignments
+/// instead of the Private Use Area code points that have been used historically.
+/// It is believed to be able to decode existing Web content in a way that makes
+/// sense.
+///
+/// To avoid form submissions generating data that Web servers don't understand,
+/// the encoder doesn't use the HKSCS byte sequences that precede the unextended
+/// Big5 in the lexical order.
+///
+/// [Index visualization](https://encoding.spec.whatwg.org/big5.html),
+/// [Visualization of BMP coverage](https://encoding.spec.whatwg.org/big5-bmp.html)
+///
+/// This encoding is designed to be suited for decoding the Windows code page 950
+/// and its HKSCS patched "951" variant such that the text makes sense, given
+/// assignments that Unicode has made after those encodings used Private Use
+/// Area characters.