Maraiah/maraiah/text.rs

//! Text conversion utilities.

/// Formats a binary size string for any given number.
///
/// # Examples
///
/// ```
/// use maraiah::text::to_binsize;
///
/// assert_eq!(to_binsize(5000), "5kB".to_string());
/// ```
pub fn to_binsize(n: u64) -> String
{
   const NAMES: [&str; 4] = ["kB", "MB", "GB", "TB"];

   // empty size
   if n == 0 {
      return String::from("empty");
   }

   // terabytes, gigabytes, megabytes, kilobytes
   for i in (1..=4).rev() {
      let pow = 1000_u64.pow(i as u32);
      if n >= pow {
         return format!("{:1}{}", n / pow, NAMES[i - 1]);
      }
   }

   // or, just bytes
   format!("{} {}", n, if n == 1 {"byte"} else {"bytes"})
}

/// Encodes or decodes a string in the terminal encryption format.
pub fn fuck_string(s: &[u8]) -> Vec<u8>
{
   let mut v = s.to_vec();
   let l     = s.len();
   let mut p = 0;

   for _ in 0..l / 4 {
      p += 2;
      v[p] ^= 0xfe;
      v[p + 1] ^= 0xed;
      p += 2;
   }

   for _ in 0..l % 4 {
      v[p] ^= 0xfe;
      p += 1;
   }

   v
}

/// Reads a Pascal-style byte string with bounds checking.
///
/// # Examples
///
/// ```
/// use maraiah::text::pascal_str;
///
/// assert_eq!(pascal_str(b"\x0bhello world"), b"hello world"[..].into());
/// assert_eq!(pascal_str(b"\x0chello world"), None);
/// assert_eq!(pascal_str(&[]), None);
/// ```
pub fn pascal_str(b: &[u8]) -> Option<&[u8]>
{
   let s = usize::from(*b.get(0)?);
   b.get(1..=s)
}

/// Converts input from Mac Roman to a Unicode string.
///
/// # Examples
///
/// ```
/// use maraiah::text::mac_roman_conv;
///
/// assert_eq!(mac_roman_conv(b"p\x8cth"), "påth");
/// assert_eq!(mac_roman_conv(b"I\xd5ve"), "I’ve");
/// ```
pub fn mac_roman_conv(s: &[u8]) -> String
{
   let mut v = String::with_capacity(s.len());

   for &c in s.iter() {
      let c = match c {
         0x80..=0xFF => TR[usize::from(c) & 0x7F],
         b'\r'       => '\n',
         c           => char::from(c),
      };

      v.push(c);
   }

   v
}

/// Converts a C-style string from Mac Roman to Unicode.
///
/// # Examples
///
/// ```
/// use maraiah::text::mac_roman_cstr;
///
/// assert_eq!(mac_roman_cstr(b"I\xd5ve awaken\0ed"), "I’ve awaken");
/// assert_eq!(mac_roman_cstr(b"I\xd5ve awaken\0"), "I’ve awaken");
/// assert_eq!(mac_roman_cstr(b"I\xd5ve awaken"), "I’ve awaken");
/// ```
#[inline]
pub fn mac_roman_cstr(s: &[u8]) -> String
{
   if let Some(n) = memchr::memchr(0, s) {
      mac_roman_conv(&s[..n])
   } else {
      mac_roman_conv(s)
   }
}

/// Converts input from a Unicode string to Mac Roman.
///
/// # Examples
///
/// ```
/// use maraiah::text::to_mac_roman;
///
/// assert_eq!(to_mac_roman("påth"), b"p\x8cth".to_vec());
/// assert_eq!(to_mac_roman("I’ve\n"), b"I\xd5ve\r".to_vec());
/// ```
pub fn to_mac_roman(s: &str) -> Vec<u8>
{
   let mut v = Vec::with_capacity(s.len());

   for c in s.chars() {
      let c = match c {
         '\n' => b'\r',
         '\0'..='\x7f' => c as u8,
         c => {
            if let Some(c) = TR.iter().position(|&o| o == c) {
               c as u8 + 0x80
            } else {
               0x7f
            }
         }
      };

      v.push(c);
   }

   v
}

/// Pads the output with zeroes.
///
/// # Examples
///
/// ```
/// use maraiah::text::{pad_zero, to_mac_roman};
///
/// let s = to_mac_roman("påth");
///
/// assert_eq!(pad_zero(s.clone(), 8), b"p\x8cth\0\0\0\0".to_vec());
/// assert_eq!(pad_zero(s.clone(), 6), b"p\x8cth\0\0".to_vec());
/// assert_eq!(pad_zero(s.clone(), 4), b"p\x8cth".to_vec());
/// assert_eq!(pad_zero(s.clone(), 2), b"p\x8cth".to_vec());
/// assert_eq!(pad_zero(s.clone(), 0), b"p\x8cth".to_vec());
/// ```
pub fn pad_zero(mut v: Vec<u8>, n: usize) -> Vec<u8>
{
   for _ in v.len()..n {
      v.push(0);
   }

   v
}

const TR: [char; 128] =
   ['\u{00c4}', '\u{00c5}', '\u{00c7}', '\u{00c9}', '\u{00d1}', '\u{00d6}',
    '\u{00dc}', '\u{00e1}', '\u{00e0}', '\u{00e2}', '\u{00e4}', '\u{00e3}',
    '\u{00e5}', '\u{00e7}', '\u{00e9}', '\u{00e8}', '\u{00ea}', '\u{00eb}',
    '\u{00ed}', '\u{00ec}', '\u{00ee}', '\u{00ef}', '\u{00f1}', '\u{00f3}',
    '\u{00f2}', '\u{00f4}', '\u{00f6}', '\u{00f5}', '\u{00fa}', '\u{00f9}',
    '\u{00fb}', '\u{00fc}', '\u{2020}', '\u{00b0}', '\u{00a2}', '\u{00a3}',
    '\u{00a7}', '\u{2022}', '\u{00b6}', '\u{00df}', '\u{00ae}', '\u{00a9}',
    '\u{2122}', '\u{00b4}', '\u{00a8}', '\u{2260}', '\u{00c6}', '\u{00d8}',
    '\u{221e}', '\u{00b1}', '\u{2264}', '\u{2265}', '\u{00a5}', '\u{00b5}',
    '\u{2202}', '\u{2211}', '\u{220f}', '\u{03c0}', '\u{222b}', '\u{00aa}',
    '\u{00ba}', '\u{03a9}', '\u{00e6}', '\u{00f8}', '\u{00bf}', '\u{00a1}',
    '\u{00ac}', '\u{221a}', '\u{0192}', '\u{2248}', '\u{2206}', '\u{00ab}',
    '\u{00bb}', '\u{2026}', '\u{00a0}', '\u{00c0}', '\u{00c3}', '\u{00d5}',
    '\u{0152}', '\u{0153}', '\u{2013}', '\u{2014}', '\u{201c}', '\u{201d}',
    '\u{2018}', '\u{2019}', '\u{00f7}', '\u{25ca}', '\u{00ff}', '\u{0178}',
    '\u{2044}', '\u{20ac}', '\u{2039}', '\u{203a}', '\u{fb01}', '\u{fb02}',
    '\u{2021}', '\u{00b7}', '\u{201a}', '\u{201e}', '\u{2030}', '\u{00c2}',
    '\u{00ca}', '\u{00c1}', '\u{00cb}', '\u{00c8}', '\u{00cd}', '\u{00ce}',
    '\u{00cf}', '\u{00cc}', '\u{00d3}', '\u{00d4}', '\u{f8ff}', '\u{00d2}',
    '\u{00da}', '\u{00db}', '\u{00d9}', '\u{0131}', '\u{02c6}', '\u{02dc}',
    '\u{00af}', '\u{02d8}', '\u{02d9}', '\u{02da}', '\u{00b8}', '\u{02dd}',
    '\u{02db}', '\u{02c7}'];

#[test]
fn to_binsize_integrals()
{
   assert_eq!(to_binsize(0), "empty");
   assert_eq!(to_binsize(1), "1 byte");
   assert_eq!(to_binsize(2), "2 bytes");
   assert_eq!(to_binsize(999), "999 bytes");
   assert_eq!(to_binsize(1000), "1kB");
   assert_eq!(to_binsize(1000 * 7), "7kB");
   assert_eq!(to_binsize(1000 * 1000), "1MB");
   assert_eq!(to_binsize(1000 * 1000 * 7), "7MB");
   assert_eq!(to_binsize(1000 * 1000 * 1000), "1GB");
   assert_eq!(to_binsize(1000 * 1000 * 1000 * 7), "7GB");
   assert_eq!(to_binsize(1000 * 1000 * 1000 * 1000), "1TB");
   assert_eq!(to_binsize(1000 * 1000 * 1000 * 1000 * 7), "7TB");
}

// EOF