Enum Encoding

Source

#[non_exhaustive]pub enum Encoding {
Show 21 variants    Arabic,
    Celtic,
    CentralEuropean,
    Croatian,
    Cyrillic,
    Devanagari,
    Dingbats,
    Farsi,
    Gaelic,
    Greek,
    Gujarati,
    Gurmukhi,
    Hebrew,
    Icelandic,
    Inuit,
    Keyboard,
    Roman,
    Romanian,
    Symbol,
    Thai,
    Turkish,
}

Expand description

The encodings. The generator writes this list and the tables together, so the two always agree.

Variants (Non-exhaustive)§

This enum is marked as non-exhaustive

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

§

Turkish

Implementations§

Source §

impl Encoding

Source

pub fn id(self) -> &'static str

The permanent identifier, for example roman or central-european.

Each encoding has an identifier. Only two have a name in the standard. Refer to Self::whatwg_name.

Source

pub fn apple_name(self) -> &'static str

Apple’s name, for example Mac OS Roman.

Source

pub fn whatwg_name(self) -> Option<&'static str>

The name in the standard, if the standard has one.

The result is Some("macintosh") for Encoding::Roman and Some("x-mac-cyrillic") for Encoding::Cyrillic. For the other 19 encodings the result is None.

Source

pub fn labels(self) -> &'static [&'static str]

The labels in the standard for this encoding.

The list is empty if the standard does not have this encoding.

Source

pub fn is_directional(self) -> bool

Tells you if the table gives a direction to its mappings.

The result is true for Mac OS Arabic, Farsi, and Hebrew. For these three encodings, the direction is also the cause of the condition in Self::encode_is_lossy. But the two conditions are not the same. Mac OS Keyboard has the second condition and not the first.

Source

pub fn encode_is_lossy(self) -> bool

Tells you if two codes give the same text.

If the result is true, one byte can decode to text that encodes to a different byte. The opposite sequence is always correct. To encode text and then to decode it always gives the first text again.

The result is true for Mac OS Arabic, Farsi, Hebrew, and Keyboard. In the first three encodings, the left-to-right form and the right-to-left form of a character have the same code point. Mac OS Keyboard has one such condition at U+2423 OPEN BOX. Apple’s comment for that mapping says “duplicates mapping for 0x61, hence no round-trip”.

Source

pub fn defines_every_byte(self) -> bool

Tells you if all 256 bytes have a mapping.

If the result is true, Self::decode_strict cannot fail. Mac OS Roman is such an encoding. This is the reason that a resource type code of four bytes always decodes.

Source

pub fn from_id(id: &str) -> Option<Self>

Finds the encoding with this Self::id. The text must agree fully.

Source

pub fn from_label(label: &str) -> Option<Self>

Finds the encoding with this label.

This function obeys “get an encoding” in section 4.2 of the standard. It removes the ASCII space characters at the start and at the end. Then it compares the text. A capital letter and a small letter are equivalent.

Only two encodings have labels. For the other 19 encodings the result is None. Use Self::from_id for them.

assert_eq!(Encoding::from_label("  MACINTOSH\n"), Some(Encoding::Roman));
assert_eq!(Encoding::from_label("x-mac-ukrainian"), Some(Encoding::Cyrillic));
assert_eq!(Encoding::from_label("Mac OS Thai"), None);

Source

pub fn decode(self, bytes: &[u8]) -> String

Decodes bytes. A byte with no mapping becomes U+FFFD.

This is the “replacement” error mode in the standard. If Self::defines_every_byte is true, no byte becomes U+FFFD.

Source

pub fn decode_strict(self, bytes: &[u8]) -> Result<String, DecodeError>

Decodes bytes. The first byte with no mapping gives an error.

This is the “fatal” error mode in the standard.

Source

pub fn encode(self, text: &str) -> Result<Vec<u8>, EncodeError>

Encodes text. The first code point with no byte gives an error.

This is the “fatal” error mode in the standard. Use this function for resource data. A type code of four characters must give four bytes. If the encoder replaces a character, you find the wrong resource.

Source

pub fn encode_html(self, text: &str) -> Vec<u8> ⓘ

Encodes text. A code point with no byte becomes &#NNN;.

This is the “html” error mode in the standard. HTML forms need an encoder that cannot fail. The standard gives a warning about this mode. You cannot see a difference between this result and text that contains the same characters. Use Self::encode for all data that is not form data.