DevComp/zip-rs-wasm
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Initial file name/comment encoding support

Browse source
This commit is contained in:
Mathijs van de Nes 2014-09-14 17:25:40 +02:00
parent d65cb2d8f5
commit 723e8b0b12
8 changed files with 582 additions and 30 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
src/bin/extract.rs
View file

@ -10,8 +10,8 @@ fn main()
for file in zipcontainer.files()
{
println!("{}", file.file_name_string());
let comment = file.file_comment_string();
println!("{}", file.file_name);
let comment = &file.file_comment;
if comment.len() > 0 { println!(" File comment: {}", comment); }
if file.uncompressed_size == 0 { continue }

4
src/bin/write_sample.rs
View file

@ -18,10 +18,10 @@ fn doit(filename: &str) -> std::io::IoResult<()>
let mut zip = zip::ZipWriter::new(file);
try!(zip.start_file(b"test/hello_world.txt", zip::types::Stored));
try!(zip.start_file("test/hello_world.txt", zip::types::Stored));
try!(zip.write(b"Hello, World!\n"));
try!(zip.start_file(b"test/lorem_ipsum.txt", zip::types::Deflated));
try!(zip.start_file("test/lorem_ipsum.txt", zip::types::Deflated));
try!(zip.write(LOREM_IPSUM));
zip.finalize()

554
src/cp437.rs Normal file
View file

@ -0,0 +1,554 @@
/// Conversion module to convert a String to IBM codepage 437
pub fn to_string(input: &[u8]) -> String
{
input.iter().map(|c| to_char(*c)).collect()
}
pub fn from_string(input: &str) -> Vec<u8>
{
input.chars().map(|c| from_char(c)).collect()
}
fn to_char(input: u8) -> char
{
let output = match input
{
0x00 => 0x0000,
0x01 => 0x0001,
0x02 => 0x0002,
0x03 => 0x0003,
0x04 => 0x0004,
0x05 => 0x0005,
0x06 => 0x0006,
0x07 => 0x0007,
0x08 => 0x0008,
0x09 => 0x0009,
0x0a => 0x000a,
0x0b => 0x000b,
0x0c => 0x000c,
0x0d => 0x000d,
0x0e => 0x000e,
0x0f => 0x000f,
0x10 => 0x0010,
0x11 => 0x0011,
0x12 => 0x0012,
0x13 => 0x0013,
0x14 => 0x0014,
0x15 => 0x0015,
0x16 => 0x0016,
0x17 => 0x0017,
0x18 => 0x0018,
0x19 => 0x0019,
0x1a => 0x001a,
0x1b => 0x001b,
0x1c => 0x001c,
0x1d => 0x001d,
0x1e => 0x001e,
0x1f => 0x001f,
0x20 => 0x0020,
0x21 => 0x0021,
0x22 => 0x0022,
0x23 => 0x0023,
0x24 => 0x0024,
0x25 => 0x0025,
0x26 => 0x0026,
0x27 => 0x0027,
0x28 => 0x0028,
0x29 => 0x0029,
0x2a => 0x002a,
0x2b => 0x002b,
0x2c => 0x002c,
0x2d => 0x002d,
0x2e => 0x002e,
0x2f => 0x002f,
0x30 => 0x0030,
0x31 => 0x0031,
0x32 => 0x0032,
0x33 => 0x0033,
0x34 => 0x0034,
0x35 => 0x0035,
0x36 => 0x0036,
0x37 => 0x0037,
0x38 => 0x0038,
0x39 => 0x0039,
0x3a => 0x003a,
0x3b => 0x003b,
0x3c => 0x003c,
0x3d => 0x003d,
0x3e => 0x003e,
0x3f => 0x003f,
0x40 => 0x0040,
0x41 => 0x0041,
0x42 => 0x0042,
0x43 => 0x0043,
0x44 => 0x0044,
0x45 => 0x0045,
0x46 => 0x0046,
0x47 => 0x0047,
0x48 => 0x0048,
0x49 => 0x0049,
0x4a => 0x004a,
0x4b => 0x004b,
0x4c => 0x004c,
0x4d => 0x004d,
0x4e => 0x004e,
0x4f => 0x004f,
0x50 => 0x0050,
0x51 => 0x0051,
0x52 => 0x0052,
0x53 => 0x0053,
0x54 => 0x0054,
0x55 => 0x0055,
0x56 => 0x0056,
0x57 => 0x0057,
0x58 => 0x0058,
0x59 => 0x0059,
0x5a => 0x005a,
0x5b => 0x005b,
0x5c => 0x005c,
0x5d => 0x005d,
0x5e => 0x005e,
0x5f => 0x005f,
0x60 => 0x0060,
0x61 => 0x0061,
0x62 => 0x0062,
0x63 => 0x0063,
0x64 => 0x0064,
0x65 => 0x0065,
0x66 => 0x0066,
0x67 => 0x0067,
0x68 => 0x0068,
0x69 => 0x0069,
0x6a => 0x006a,
0x6b => 0x006b,
0x6c => 0x006c,
0x6d => 0x006d,
0x6e => 0x006e,
0x6f => 0x006f,
0x70 => 0x0070,
0x71 => 0x0071,
0x72 => 0x0072,
0x73 => 0x0073,
0x74 => 0x0074,
0x75 => 0x0075,
0x76 => 0x0076,
0x77 => 0x0077,
0x78 => 0x0078,
0x79 => 0x0079,
0x7a => 0x007a,
0x7b => 0x007b,
0x7c => 0x007c,
0x7d => 0x007d,
0x7e => 0x007e,
0x7f => 0x007f,
0x80 => 0x00c7,
0x81 => 0x00fc,
0x82 => 0x00e9,
0x83 => 0x00e2,
0x84 => 0x00e4,
0x85 => 0x00e0,
0x86 => 0x00e5,
0x87 => 0x00e7,
0x88 => 0x00ea,
0x89 => 0x00eb,
0x8a => 0x00e8,
0x8b => 0x00ef,
0x8c => 0x00ee,
0x8d => 0x00ec,
0x8e => 0x00c4,
0x8f => 0x00c5,
0x90 => 0x00c9,
0x91 => 0x00e6,
0x92 => 0x00c6,
0x93 => 0x00f4,
0x94 => 0x00f6,
0x95 => 0x00f2,
0x96 => 0x00fb,
0x97 => 0x00f9,
0x98 => 0x00ff,
0x99 => 0x00d6,
0x9a => 0x00dc,
0x9b => 0x00a2,
0x9c => 0x00a3,
0x9d => 0x00a5,
0x9e => 0x20a7,
0x9f => 0x0192,
0xa0 => 0x00e1,
0xa1 => 0x00ed,
0xa2 => 0x00f3,
0xa3 => 0x00fa,
0xa4 => 0x00f1,
0xa5 => 0x00d1,
0xa6 => 0x00aa,
0xa7 => 0x00ba,
0xa8 => 0x00bf,
0xa9 => 0x2310,
0xaa => 0x00ac,
0xab => 0x00bd,
0xac => 0x00bc,
0xad => 0x00a1,
0xae => 0x00ab,
0xaf => 0x00bb,
0xb0 => 0x2591,
0xb1 => 0x2592,
0xb2 => 0x2593,
0xb3 => 0x2502,
0xb4 => 0x2524,
0xb5 => 0x2561,
0xb6 => 0x2562,
0xb7 => 0x2556,
0xb8 => 0x2555,
0xb9 => 0x2563,
0xba => 0x2551,
0xbb => 0x2557,
0xbc => 0x255d,
0xbd => 0x255c,
0xbe => 0x255b,
0xbf => 0x2510,
0xc0 => 0x2514,
0xc1 => 0x2534,
0xc2 => 0x252c,
0xc3 => 0x251c,
0xc4 => 0x2500,
0xc5 => 0x253c,
0xc6 => 0x255e,
0xc7 => 0x255f,
0xc8 => 0x255a,
0xc9 => 0x2554,
0xca => 0x2569,
0xcb => 0x2566,
0xcc => 0x2560,
0xcd => 0x2550,
0xce => 0x256c,
0xcf => 0x2567,
0xd0 => 0x2568,
0xd1 => 0x2564,
0xd2 => 0x2565,
0xd3 => 0x2559,
0xd4 => 0x2558,
0xd5 => 0x2552,
0xd6 => 0x2553,
0xd7 => 0x256b,
0xd8 => 0x256a,
0xd9 => 0x2518,
0xda => 0x250c,
0xdb => 0x2588,
0xdc => 0x2584,
0xdd => 0x258c,
0xde => 0x2590,
0xdf => 0x2580,
0xe0 => 0x03b1,
0xe1 => 0x00df,
0xe2 => 0x0393,
0xe3 => 0x03c0,
0xe4 => 0x03a3,
0xe5 => 0x03c3,
0xe6 => 0x00b5,
0xe7 => 0x03c4,
0xe8 => 0x03a6,
0xe9 => 0x0398,
0xea => 0x03a9,
0xeb => 0x03b4,
0xec => 0x221e,
0xed => 0x03c6,
0xee => 0x03b5,
0xef => 0x2229,
0xf0 => 0x2261,
0xf1 => 0x00b1,
0xf2 => 0x2265,
0xf3 => 0x2264,
0xf4 => 0x2320,
0xf5 => 0x2321,
0xf6 => 0x00f7,
0xf7 => 0x2248,
0xf8 => 0x00b0,
0xf9 => 0x2219,
0xfa => 0x00b7,
0xfb => 0x221a,
0xfc => 0x207f,
0xfd => 0x00b2,
0xfe => 0x25a0,
0xff => 0x00a0,
_ => 0xfffd,
};
::std::char::from_u32(output).unwrap()
}
fn from_char(input: char) -> u8
{
let output = match input as u32
{
0x0000 => 0x00,
0x0001 => 0x01,
0x0002 => 0x02,
0x0003 => 0x03,
0x0004 => 0x04,
0x0005 => 0x05,
0x0006 => 0x06,
0x0007 => 0x07,
0x0008 => 0x08,
0x0009 => 0x09,
0x000a => 0x0a,
0x000b => 0x0b,
0x000c => 0x0c,
0x000d => 0x0d,
0x000e => 0x0e,
0x000f => 0x0f,
0x0010 => 0x10,
0x0011 => 0x11,
0x0012 => 0x12,
0x0013 => 0x13,
0x0014 => 0x14,
0x0015 => 0x15,
0x0016 => 0x16,
0x0017 => 0x17,
0x0018 => 0x18,
0x0019 => 0x19,
0x001a => 0x1a,
0x001b => 0x1b,
0x001c => 0x1c,
0x001d => 0x1d,
0x001e => 0x1e,
0x001f => 0x1f,
0x0020 => 0x20,
0x0021 => 0x21,
0x0022 => 0x22,
0x0023 => 0x23,
0x0024 => 0x24,
0x0025 => 0x25,
0x0026 => 0x26,
0x0027 => 0x27,
0x0028 => 0x28,
0x0029 => 0x29,
0x002a => 0x2a,
0x002b => 0x2b,
0x002c => 0x2c,
0x002d => 0x2d,
0x002e => 0x2e,
0x002f => 0x2f,
0x0030 => 0x30,
0x0031 => 0x31,
0x0032 => 0x32,
0x0033 => 0x33,
0x0034 => 0x34,
0x0035 => 0x35,
0x0036 => 0x36,
0x0037 => 0x37,
0x0038 => 0x38,
0x0039 => 0x39,
0x003a => 0x3a,
0x003b => 0x3b,
0x003c => 0x3c,
0x003d => 0x3d,
0x003e => 0x3e,
0x003f => 0x3f,
0x0040 => 0x40,
0x0041 => 0x41,
0x0042 => 0x42,
0x0043 => 0x43,
0x0044 => 0x44,
0x0045 => 0x45,
0x0046 => 0x46,
0x0047 => 0x47,
0x0048 => 0x48,
0x0049 => 0x49,
0x004a => 0x4a,
0x004b => 0x4b,
0x004c => 0x4c,
0x004d => 0x4d,
0x004e => 0x4e,
0x004f => 0x4f,
0x0050 => 0x50,
0x0051 => 0x51,
0x0052 => 0x52,
0x0053 => 0x53,
0x0054 => 0x54,
0x0055 => 0x55,
0x0056 => 0x56,
0x0057 => 0x57,
0x0058 => 0x58,
0x0059 => 0x59,
0x005a => 0x5a,
0x005b => 0x5b,
0x005c => 0x5c,
0x005d => 0x5d,
0x005e => 0x5e,
0x005f => 0x5f,
0x0060 => 0x60,
0x0061 => 0x61,
0x0062 => 0x62,
0x0063 => 0x63,
0x0064 => 0x64,
0x0065 => 0x65,
0x0066 => 0x66,
0x0067 => 0x67,
0x0068 => 0x68,
0x0069 => 0x69,
0x006a => 0x6a,
0x006b => 0x6b,
0x006c => 0x6c,
0x006d => 0x6d,
0x006e => 0x6e,
0x006f => 0x6f,
0x0070 => 0x70,
0x0071 => 0x71,
0x0072 => 0x72,
0x0073 => 0x73,
0x0074 => 0x74,
0x0075 => 0x75,
0x0076 => 0x76,
0x0077 => 0x77,
0x0078 => 0x78,
0x0079 => 0x79,
0x007a => 0x7a,
0x007b => 0x7b,
0x007c => 0x7c,
0x007d => 0x7d,
0x007e => 0x7e,
0x007f => 0x7f,
0x00c7 => 0x80,
0x00fc => 0x81,
0x00e9 => 0x82,
0x00e2 => 0x83,
0x00e4 => 0x84,
0x00e0 => 0x85,
0x00e5 => 0x86,
0x00e7 => 0x87,
0x00ea => 0x88,
0x00eb => 0x89,
0x00e8 => 0x8a,
0x00ef => 0x8b,
0x00ee => 0x8c,
0x00ec => 0x8d,
0x00c4 => 0x8e,
0x00c5 => 0x8f,
0x00c9 => 0x90,
0x00e6 => 0x91,
0x00c6 => 0x92,
0x00f4 => 0x93,
0x00f6 => 0x94,
0x00f2 => 0x95,
0x00fb => 0x96,
0x00f9 => 0x97,
0x00ff => 0x98,
0x00d6 => 0x99,
0x00dc => 0x9a,
0x00a2 => 0x9b,
0x00a3 => 0x9c,
0x00a5 => 0x9d,
0x20a7 => 0x9e,
0x0192 => 0x9f,
0x00e1 => 0xa0,
0x00ed => 0xa1,
0x00f3 => 0xa2,
0x00fa => 0xa3,
0x00f1 => 0xa4,
0x00d1 => 0xa5,
0x00aa => 0xa6,
0x00ba => 0xa7,
0x00bf => 0xa8,
0x2310 => 0xa9,
0x00ac => 0xaa,
0x00bd => 0xab,
0x00bc => 0xac,
0x00a1 => 0xad,
0x00ab => 0xae,
0x00bb => 0xaf,
0x2591 => 0xb0,
0x2592 => 0xb1,
0x2593 => 0xb2,
0x2502 => 0xb3,
0x2524 => 0xb4,
0x2561 => 0xb5,
0x2562 => 0xb6,
0x2556 => 0xb7,
0x2555 => 0xb8,
0x2563 => 0xb9,
0x2551 => 0xba,
0x2557 => 0xbb,
0x255d => 0xbc,
0x255c => 0xbd,
0x255b => 0xbe,
0x2510 => 0xbf,
0x2514 => 0xc0,
0x2534 => 0xc1,
0x252c => 0xc2,
0x251c => 0xc3,
0x2500 => 0xc4,
0x253c => 0xc5,
0x255e => 0xc6,
0x255f => 0xc7,
0x255a => 0xc8,
0x2554 => 0xc9,
0x2569 => 0xca,
0x2566 => 0xcb,
0x2560 => 0xcc,
0x2550 => 0xcd,
0x256c => 0xce,
0x2567 => 0xcf,
0x2568 => 0xd0,
0x2564 => 0xd1,
0x2565 => 0xd2,
0x2559 => 0xd3,
0x2558 => 0xd4,
0x2552 => 0xd5,
0x2553 => 0xd6,
0x256b => 0xd7,
0x256a => 0xd8,
0x2518 => 0xd9,
0x250c => 0xda,
0x2588 => 0xdb,
0x2584 => 0xdc,
0x258c => 0xdd,
0x2590 => 0xde,
0x2580 => 0xdf,
0x03b1 => 0xe0,
0x00df => 0xe1,
0x0393 => 0xe2,
0x03c0 => 0xe3,
0x03a3 => 0xe4,
0x03c3 => 0xe5,
0x00b5 => 0xe6,
0x03c4 => 0xe7,
0x03a6 => 0xe8,
0x0398 => 0xe9,
0x03a9 => 0xea,
0x03b4 => 0xeb,
0x221e => 0xec,
0x03c6 => 0xed,
0x03b5 => 0xee,
0x2229 => 0xef,
0x2261 => 0xf0,
0x00b1 => 0xf1,
0x2265 => 0xf2,
0x2264 => 0xf3,
0x2320 => 0xf4,
0x2321 => 0xf5,
0x00f7 => 0xf6,
0x2248 => 0xf7,
0x00b0 => 0xf8,
0x2219 => 0xf9,
0x00b7 => 0xfa,
0x221a => 0xfb,
0x207f => 0xfc,
0x00b2 => 0xfd,
0x25a0 => 0xfe,
0x00a0 => 0xff,
_ => 0x3f, // For safety, us the question mark '?'
};
output
}
#[cfg(test)]
mod test
{
#[test]
fn to_char_valid()
{
for i in ::std::iter::range_inclusive(0x00, 0xFF)
{
super::to_char(i);
}
}
}

1
src/lib.rs
View file

@ -17,3 +17,4 @@ mod crc32;
mod reader;
pub mod types;
mod writer;
mod cp437;

2
src/reader.rs
View file

@ -21,7 +21,7 @@ use flate2::FlateReader;
///
/// for file in zip.files()
/// {
/// println!("Filename: {}", file.file_name_string());
/// println!("Filename: {}", file.file_name);
/// let mut file_reader = try!(zip.read_file(file));
/// let first_byte = try!(file_reader.read_byte());
/// println!("{}", first_byte);

20
src/spec.rs
View file

@ -25,6 +25,7 @@ pub fn central_header_to_zip_file<R: Reader+Seek>(reader: &mut R) -> IoResult<Zi
try!(reader.read_le_u16());
let flags = try!(reader.read_le_u16());
let encrypted = flags & 1 == 1;
let is_utf8 = flags & (1 << 11) != 0;
let compression_method = try!(reader.read_le_u16());
let last_mod_time = try!(reader.read_le_u16());
let last_mod_date = try!(reader.read_le_u16());
@ -38,9 +39,20 @@ pub fn central_header_to_zip_file<R: Reader+Seek>(reader: &mut R) -> IoResult<Zi
try!(reader.read_le_u16());
try!(reader.read_le_u32());
let offset = try!(reader.read_le_u32()) as i64;
let file_name = try!(reader.read_exact(file_name_length));
let file_name_raw = try!(reader.read_exact(file_name_length));
try!(reader.read_exact(extra_field_length));
let file_comment = try!(reader.read_exact(file_comment_length));
let file_comment_raw = try!(reader.read_exact(file_comment_length));
let file_name = match is_utf8
{
true => String::from_utf8_lossy(file_name_raw.as_slice()).into_string(),
false => ::cp437::to_string(file_name_raw.as_slice()),
};
let file_comment = match is_utf8
{
true => String::from_utf8_lossy(file_comment_raw.as_slice()).into_string(),
false => ::cp437::to_string(file_comment_raw.as_slice()),
};
// Remember end of central header
let return_position = try!(reader.tell()) as i64;
@ -97,7 +109,7 @@ pub fn write_local_file_header<T: Writer>(writer: &mut T, file: &ZipFile) -> IoR
try!(writer.write_le_u32(file.uncompressed_size as u32));
try!(writer.write_le_u16(file.file_name.len() as u16));
try!(writer.write_le_u16(0));
try!(writer.write(file.file_name.as_slice()));
try!(writer.write(::cp437::from_string(file.file_name.as_slice()).as_slice()));
Ok(())
}
@ -122,7 +134,7 @@ pub fn write_central_directory_header<T: Writer>(writer: &mut T, file: &ZipFile)
try!(writer.write_le_u16(0));
try!(writer.write_le_u32(0));
try!(writer.write_le_u32(file.header_start as u32));
try!(writer.write(file.file_name.as_slice()));
try!(writer.write(::cp437::from_string(file.file_name.as_slice()).as_slice()));
Ok(())
}

19
src/types.rs
View file

@ -58,26 +58,11 @@ pub struct ZipFile
/// Size of the file when extracted
pub uncompressed_size: u64,
/// Name of the file
pub file_name: Vec<u8>,
pub file_name: String,
/// File comment
pub file_comment: Vec<u8>,
pub file_comment: String,
/// Specifies where the local header of the file starts
pub header_start: u64,
/// Specifies where the compressed data of the file starts
pub data_start: u64,
}
impl ZipFile
{
/// Lossy UTF-8 interpretation of the file name
pub fn file_name_string(&self) -> String
{
String::from_utf8_lossy(self.file_name.as_slice()).into_string()
}
/// Lossy UTF-8 interpretation of the file comment
pub fn file_comment_string(&self) -> String
{
String::from_utf8_lossy(self.file_comment.as_slice()).into_string()
}
}

8
src/writer.rs
View file

@ -28,7 +28,7 @@ enum GenericZipWriter<W>
/// let w = std::io::BufWriter::new(&mut buf);
/// let mut zip = zip::ZipWriter::new(w);
///
/// try!(zip.start_file(b"hello_world.txt", zip::types::Stored));
/// try!(zip.start_file("hello_world.txt", zip::types::Stored));
/// try!(zip.write(b"Hello, World!"));
///
/// // Optionally finish the zip. (this is also done on drop)
@ -99,7 +99,7 @@ impl<W: Writer+Seek> ZipWriter<W>
}
/// Start a new file for with the requested compression method.
pub fn start_file(&mut self, name: &[u8], compression: types::CompressionMethod) -> IoResult<()>
pub fn start_file(&mut self, name: &str, compression: types::CompressionMethod) -> IoResult<()>
{
try!(self.finish_file());
@ -120,8 +120,8 @@ impl<W: Writer+Seek> ZipWriter<W>
crc32: 0,
compressed_size: 0,
uncompressed_size: 0,
file_name: Vec::from_slice(name),
file_comment: Vec::new(),
file_name: String::from_str(name),
file_comment: String::new(),
header_start: header_start,
data_start: self.stats.start,
});