From f5c4671bfbad96bf346bd7e9a21fc4317b4959df Mon Sep 17 00:00:00 2001 From: Indrajith K L Date: Sat, 3 Dec 2022 17:00:20 +0530 Subject: Adds most of the tools --- v_windows/v/vlib/encoding/base64/base64.v | 308 +++++++++++++++++++++ .../v/vlib/encoding/base64/base64_memory_test.v | 59 ++++ v_windows/v/vlib/encoding/base64/base64_test.v | 150 ++++++++++ 3 files changed, 517 insertions(+) create mode 100644 v_windows/v/vlib/encoding/base64/base64.v create mode 100644 v_windows/v/vlib/encoding/base64/base64_memory_test.v create mode 100644 v_windows/v/vlib/encoding/base64/base64_test.v (limited to 'v_windows/v/vlib/encoding/base64') diff --git a/v_windows/v/vlib/encoding/base64/base64.v b/v_windows/v/vlib/encoding/base64/base64.v new file mode 100644 index 0000000..ad06722 --- /dev/null +++ b/v_windows/v/vlib/encoding/base64/base64.v @@ -0,0 +1,308 @@ +// Copyright (c) 2019-2021 Alexander Medvednikov. All rights reserved. +// Use of this source code is governed by an MIT license +// that can be found in the LICENSE file. +// Based off: https://github.com/golang/go/blob/master/src/encoding/base64/base64.go +// Last commit: https://github.com/golang/go/commit/9a93baf4d7d13d7d5c67388c93960d78abc8e11e +module base64 + +const ( + index = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 62, 63, 62, 62, 63, 52, 53, 54, 55, + 56, 57, 58, 59, 60, 61, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, + 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 0, 0, 0, 0, 63, 0, 26, 27, 28, 29, + 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51]! + ending_table = [0, 2, 1]! + enc_table = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/' +) + +union B64_64_datablock { +mut: + data u64 + data_byte [8]byte +} + +union B64_32_datablock { +mut: + data u32 + data_byte [4]byte +} + +// decode decodes the base64 encoded `string` value passed in `data`. +// Please note: If you need to decode many strings repeatedly, take a look at `decode_in_buffer`. +// Example: assert base64.decode('ViBpbiBiYXNlIDY0') == 'V in base 64' +pub fn decode(data string) []byte { + mut size := i64(data.len) * 3 / 4 + if size <= 0 || data.len % 4 != 0 { + return [] + } + size = (size + 3) & ~0x03 // round to the next multiple of 4 (the decoding loop writes multiples of 4 bytes) + unsafe { + buffer := malloc(int(size)) + n := decode_in_buffer(data, buffer) + return buffer.vbytes(n) + } +} + +// decode_str is the string variant of decode +pub fn decode_str(data string) string { + size := data.len * 3 / 4 + if size <= 0 || data.len % 4 != 0 { + return '' + } + unsafe { + buffer := malloc_noscan(size + 1) + buffer[size] = 0 + return tos(buffer, decode_in_buffer(data, buffer)) + } +} + +// encode encodes the `[]byte` value passed in `data` to base64. +// Please note: base64 encoding returns a `string` that is ~ 4/3 larger than the input. +// Please note: If you need to encode many strings repeatedly, take a look at `encode_in_buffer`. +// Example: assert base64.encode('V in base 64') == 'ViBpbiBiYXNlIDY0' +pub fn encode(data []byte) string { + return alloc_and_encode(data.data, data.len) +} + +// encode_str is the string variant of encode +pub fn encode_str(data string) string { + return alloc_and_encode(data.str, data.len) +} + +// alloc_and_encode is a private function that allocates and encodes data into a string +// Used by encode and encode_str +fn alloc_and_encode(src &byte, len int) string { + size := 4 * ((len + 2) / 3) + if size <= 0 { + return '' + } + unsafe { + buffer := malloc_noscan(size + 1) + buffer[size] = 0 + return tos(buffer, encode_from_buffer(buffer, src, len)) + } +} + +// url_decode returns a decoded URL `string` version of +// the a base64 url encoded `string` passed in `data`. +pub fn url_decode(data string) []byte { + mut result := data.replace_each(['-', '+', '_', '/']) + match result.len % 4 { + // Pad with trailing '='s + 2 { result += '==' } // 2 pad chars + 3 { result += '=' } // 1 pad char + else {} // no padding + } + return decode(result) +} + +// url_decode_str is the string variant of url_decode +pub fn url_decode_str(data string) string { + mut result := data.replace_each(['-', '+', '_', '/']) + match result.len % 4 { + // Pad with trailing '='s + 2 { result += '==' } // 2 pad chars + 3 { result += '=' } // 1 pad char + else {} // no padding + } + return decode_str(result) +} + +// url_encode returns a base64 URL encoded `string` version +// of the value passed in `data`. +pub fn url_encode(data []byte) string { + return encode(data).replace_each(['+', '-', '/', '_', '=', '']) +} + +// url_encode_str is the string variant of url_encode +pub fn url_encode_str(data string) string { + return encode_str(data).replace_each(['+', '-', '/', '_', '=', '']) +} + +// assemble64 assembles 8 base64 digits into 6 bytes. +// Each digit comes from the decode map. +// Please note: Invalid base64 digits are not expected and not handled. +fn assemble64(n1 byte, n2 byte, n3 byte, n4 byte, n5 byte, n6 byte, n7 byte, n8 byte) u64 { + return u64(n1) << 58 | u64(n2) << 52 | u64(n3) << 46 | u64(n4) << 40 | u64(n5) << 34 | u64(n6) << 28 | u64(n7) << 22 | u64(n8) << 16 +} + +// assemble32 assembles 4 base64 digits into 3 bytes. +// Each digit comes from the decode map. +// Please note: Invalid base64 digits are not expected and not handled. +fn assemble32(n1 byte, n2 byte, n3 byte, n4 byte) u32 { + return u32(n1) << 26 | u32(n2) << 20 | u32(n3) << 14 | u32(n4) << 8 +} + +// decode_in_buffer decodes the base64 encoded `string` reference passed in `data` into `buffer`. +// decode_in_buffer returns the size of the decoded data in the buffer. +// Please note: The `buffer` should be large enough (i.e. 3/4 of the data.len, or larger) +// to hold the decoded data. +// Please note: This function does NOT allocate new memory, and is thus suitable for handling very large strings. +pub fn decode_in_buffer(data &string, buffer &byte) int { + return decode_from_buffer(buffer, data.str, data.len) +} + +// decode_from_buffer decodes the base64 encoded ASCII bytes from `data` into `buffer`. +// decode_from_buffer returns the size of the decoded data in the buffer. +// Please note: The `buffer` should be large enough (i.e. 3/4 of the data.len, or larger) +// to hold the decoded data. +// Please note: This function does NOT allocate new memory, and is thus suitable for handling very large strings. +pub fn decode_in_buffer_bytes(data []byte, buffer &byte) int { + return decode_from_buffer(buffer, data.data, data.len) +} + +// decode_from_buffer decodes the base64 encoded ASCII bytes from `src` into `dest`. +// decode_from_buffer returns the size of the decoded data in the buffer. +// Please note: The `dest` buffer should be large enough (i.e. 3/4 of the `src_len`, or larger) +// to hold the decoded data. +// Please note: This function does NOT allocate new memory, and is thus suitable for handling very large strings. +// Please note: This function is for internal base64 decoding +fn decode_from_buffer(dest &byte, src &byte, src_len int) int { + if src_len < 4 { + return 0 + } + + mut padding := 0 + if unsafe { src[src_len - 1] == `=` } { + if unsafe { src[src_len - 2] == `=` } { + padding = 2 + } else { + padding = 1 + } + } + + mut d := unsafe { src } + mut b := unsafe { dest } + + unsafe { + mut n_decoded_bytes := 0 // padding bytes are also counted towards this. + mut si := 0 + + mut datablock_64 := B64_64_datablock{ + data: 0 + } + mut datablock_32 := B64_32_datablock{ + data: 0 + } + + for src_len - si >= 8 { + // Converting 8 bytes of input into 6 bytes of output. Storing these in the upper bytes of an u64. + datablock_64.data = assemble64(byte(base64.index[d[si + 0]]), byte(base64.index[d[si + 1]]), + byte(base64.index[d[si + 2]]), byte(base64.index[d[si + 3]]), byte(base64.index[d[ + si + 4]]), byte(base64.index[d[si + 5]]), byte(base64.index[d[si + 6]]), + byte(base64.index[d[si + 7]])) + + // Reading out the individual bytes from the u64. Watch out with endianess. + $if little_endian { + b[n_decoded_bytes + 0] = datablock_64.data_byte[7] + b[n_decoded_bytes + 1] = datablock_64.data_byte[6] + b[n_decoded_bytes + 2] = datablock_64.data_byte[5] + b[n_decoded_bytes + 3] = datablock_64.data_byte[4] + b[n_decoded_bytes + 4] = datablock_64.data_byte[3] + b[n_decoded_bytes + 5] = datablock_64.data_byte[2] + } $else { + b[n_decoded_bytes + 0] = datablock_64.data_byte[0] + b[n_decoded_bytes + 1] = datablock_64.data_byte[1] + b[n_decoded_bytes + 2] = datablock_64.data_byte[2] + b[n_decoded_bytes + 3] = datablock_64.data_byte[3] + b[n_decoded_bytes + 4] = datablock_64.data_byte[4] + b[n_decoded_bytes + 5] = datablock_64.data_byte[5] + } + + n_decoded_bytes += 6 + si += 8 + } + + for src_len - si >= 4 { + datablock_32.data = assemble32(byte(base64.index[d[si + 0]]), byte(base64.index[d[si + 1]]), + byte(base64.index[d[si + 2]]), byte(base64.index[d[si + 3]])) + $if little_endian { + b[n_decoded_bytes + 0] = datablock_32.data_byte[3] + b[n_decoded_bytes + 1] = datablock_32.data_byte[2] + b[n_decoded_bytes + 2] = datablock_32.data_byte[1] + b[n_decoded_bytes + 3] = datablock_32.data_byte[0] + } $else { + b[n_decoded_bytes + 0] = datablock_32.data_byte[0] + b[n_decoded_bytes + 1] = datablock_32.data_byte[1] + b[n_decoded_bytes + 2] = datablock_32.data_byte[2] + b[n_decoded_bytes + 3] = datablock_32.data_byte[3] + } + + n_decoded_bytes += 3 + si += 4 + } + + return n_decoded_bytes - padding + } +} + +// encode_in_buffer base64 encodes the `[]byte` passed in `data` into `buffer`. +// encode_in_buffer returns the size of the encoded data in the buffer. +// Please note: The buffer should be large enough (i.e. 4/3 of the data.len, or larger) to hold the encoded data. +// Please note: The function does NOT allocate new memory, and is suitable for handling very large strings. +pub fn encode_in_buffer(data []byte, buffer &byte) int { + return encode_from_buffer(buffer, data.data, data.len) +} + +// encode_from_buffer will perform encoding from any type of src buffer +// and write the bytes into `dest`. +// Please note: The `dest` buffer should be large enough (i.e. 4/3 of the src_len, or larger) to hold the encoded data. +// Please note: This function is for internal base64 encoding +fn encode_from_buffer(dest &byte, src &byte, src_len int) int { + if src_len == 0 { + return 0 + } + output_length := 4 * ((src_len + 2) / 3) + + mut d := unsafe { src } + mut b := unsafe { dest } + etable := base64.enc_table.str + + mut di := 0 + mut si := 0 + n := (src_len / 3) * 3 + for si < n { + // Convert 3x 8bit source bytes into 4 bytes + unsafe { + val := u32(d[si + 0]) << 16 | u32(d[si + 1]) << 8 | u32(d[si + 2]) + + b[di + 0] = etable[val >> 18 & 0x3F] + b[di + 1] = etable[val >> 12 & 0x3F] + b[di + 2] = etable[val >> 6 & 0x3F] + b[di + 3] = etable[val & 0x3F] + } + si += 3 + di += 4 + } + + remain := src_len - si + if remain == 0 { + return output_length + } + + // Add the remaining small block and padding + unsafe { + mut val := u32(d[si + 0]) << 16 + if remain == 2 { + val |= u32(d[si + 1]) << 8 + } + + b[di + 0] = etable[val >> 18 & 0x3F] + b[di + 1] = etable[val >> 12 & 0x3F] + + match remain { + 2 { + b[di + 2] = etable[val >> 6 & 0x3F] + b[di + 3] = byte(`=`) + } + 1 { + b[di + 2] = byte(`=`) + b[di + 3] = byte(`=`) + } + else { + panic('base64: This case should never occur.') + } + } + } + return output_length +} diff --git a/v_windows/v/vlib/encoding/base64/base64_memory_test.v b/v_windows/v/vlib/encoding/base64/base64_memory_test.v new file mode 100644 index 0000000..be543af --- /dev/null +++ b/v_windows/v/vlib/encoding/base64/base64_memory_test.v @@ -0,0 +1,59 @@ +import encoding.base64 + +fn test_long_encoding() { + repeats := 1000 + input_size := 3000 + + s_original := []byte{len: input_size, init: `a`} + s_encoded := base64.encode(s_original) + s_encoded_bytes := s_encoded.bytes() + s_decoded := base64.decode(s_encoded) + + assert s_encoded.len > s_original.len + assert s_original == s_decoded + + ebuffer := unsafe { malloc(s_encoded.len) } + dbuffer := unsafe { malloc(s_decoded.len) } + defer { + unsafe { free(ebuffer) } + unsafe { free(dbuffer) } + } + // + encoded_size := base64.encode_in_buffer(s_original, ebuffer) + mut encoded_in_buf := []byte{len: encoded_size} + unsafe { C.memcpy(encoded_in_buf.data, ebuffer, encoded_size) } + assert input_size * 4 / 3 == encoded_size + assert encoded_in_buf[0] == `Y` + assert encoded_in_buf[1] == `W` + assert encoded_in_buf[2] == `F` + assert encoded_in_buf[3] == `h` + + assert encoded_in_buf[encoded_size - 4] == `Y` + assert encoded_in_buf[encoded_size - 3] == `W` + assert encoded_in_buf[encoded_size - 2] == `F` + assert encoded_in_buf[encoded_size - 1] == `h` + + assert encoded_in_buf == s_encoded_bytes + + decoded_size := base64.decode_in_buffer(s_encoded, dbuffer) + assert decoded_size == input_size + mut decoded_in_buf := []byte{len: decoded_size} + unsafe { C.memcpy(decoded_in_buf.data, dbuffer, decoded_size) } + assert decoded_in_buf == s_original + + mut s := 0 + for _ in 0 .. repeats { + resultsize := base64.encode_in_buffer(s_original, ebuffer) + s += resultsize + assert resultsize == s_encoded.len + } + + for _ in 0 .. repeats { + resultsize := base64.decode_in_buffer(s_encoded, dbuffer) + s += resultsize + assert resultsize == s_decoded.len + } + + println('Final s: $s') + // assert s == 39147008 +} diff --git a/v_windows/v/vlib/encoding/base64/base64_test.v b/v_windows/v/vlib/encoding/base64/base64_test.v new file mode 100644 index 0000000..8d08de2 --- /dev/null +++ b/v_windows/v/vlib/encoding/base64/base64_test.v @@ -0,0 +1,150 @@ +import encoding.base64 + +struct TestPair { + decoded string + encoded string +} + +const ( + pairs = [ + // RFC 3548 examples + TestPair{'\x14\xfb\x9c\x03\xd9\x7e', 'FPucA9l+'}, + TestPair{'\x14\xfb\x9c\x03\xd9', 'FPucA9k='}, + TestPair{'\x14\xfb\x9c\x03', 'FPucAw=='}, + // RFC 4648 examples + TestPair{'', ''}, + TestPair{'f', 'Zg=='}, + TestPair{'fo', 'Zm8='}, + TestPair{'foo', 'Zm9v'}, + TestPair{'foob', 'Zm9vYg=='}, + TestPair{'fooba', 'Zm9vYmE='}, + TestPair{'foobar', 'Zm9vYmFy'}, + // Wikipedia examples + TestPair{'sure.', 'c3VyZS4='}, + TestPair{'sure', 'c3VyZQ=='}, + TestPair{'sur', 'c3Vy'}, + TestPair{'su', 'c3U='}, + TestPair{'leasure.', 'bGVhc3VyZS4='}, + TestPair{'easure.', 'ZWFzdXJlLg=='}, + TestPair{'asure.', 'YXN1cmUu'}, + TestPair{'sure.', 'c3VyZS4='}, + ] + + man_pair = TestPair{'Man is distinguished, not only by his reason, but by this singular passion from other animals, which is a lust of the mind, that by a perseverance of delight in the continued and indefatigable generation of knowledge, exceeds the short vehemence of any carnal pleasure.', 'TWFuIGlzIGRpc3Rpbmd1aXNoZWQsIG5vdCBvbmx5IGJ5IGhpcyByZWFzb24sIGJ1dCBieSB0aGlzIHNpbmd1bGFyIHBhc3Npb24gZnJvbSBvdGhlciBhbmltYWxzLCB3aGljaCBpcyBhIGx1c3Qgb2YgdGhlIG1pbmQsIHRoYXQgYnkgYSBwZXJzZXZlcmFuY2Ugb2YgZGVsaWdodCBpbiB0aGUgY29udGludWVkIGFuZCBpbmRlZmF0aWdhYmxlIGdlbmVyYXRpb24gb2Yga25vd2xlZGdlLCBleGNlZWRzIHRoZSBzaG9ydCB2ZWhlbWVuY2Ugb2YgYW55IGNhcm5hbCBwbGVhc3VyZS4='} +) + +fn test_decode() { + assert base64.decode(man_pair.encoded) == man_pair.decoded.bytes() + + // Test for incorrect padding. + assert base64.decode('aGk') == ''.bytes() + assert base64.decode('aGk=') == 'hi'.bytes() + assert base64.decode('aGk==') == ''.bytes() + + for i, p in pairs { + got := base64.decode(p.encoded) + if got != p.decoded.bytes() { + eprintln('pairs[$i]: expected = $p.decoded, got = $got') + assert false + } + } +} + +fn test_decode_str() { + assert base64.decode_str(man_pair.encoded) == man_pair.decoded + + // Test for incorrect padding. + assert base64.decode_str('aGk') == '' + assert base64.decode_str('aGk=') == 'hi' + assert base64.decode_str('aGk==') == '' + + for i, p in pairs { + got := base64.decode_str(p.encoded) + if got != p.decoded { + eprintln('pairs[$i]: expected = $p.decoded, got = $got') + assert false + } + } +} + +fn test_encode() { + assert base64.encode(man_pair.decoded.bytes()) == man_pair.encoded + + for i, p in pairs { + got := base64.encode(p.decoded.bytes()) + if got != p.encoded { + eprintln('pairs[$i]: expected = $p.encoded, got = $got') + assert false + } + } +} + +fn test_encode_str() { + assert base64.encode_str(man_pair.decoded) == man_pair.encoded + + for i, p in pairs { + got := base64.encode_str(p.decoded) + if got != p.encoded { + eprintln('pairs[$i]: expected = $p.encoded, got = $got') + assert false + } + } +} + +fn test_url_encode() { + test := base64.url_encode('Hello Base64Url encoding!'.bytes()) + assert test == 'SGVsbG8gQmFzZTY0VXJsIGVuY29kaW5nIQ' +} + +fn test_url_encode_str() { + test := base64.url_encode_str('Hello Base64Url encoding!') + assert test == 'SGVsbG8gQmFzZTY0VXJsIGVuY29kaW5nIQ' +} + +fn test_url_decode() { + test := base64.url_decode('SGVsbG8gQmFzZTY0VXJsIGVuY29kaW5nIQ') + assert test == 'Hello Base64Url encoding!'.bytes() +} + +fn test_url_decode_str() { + test := base64.url_decode_str('SGVsbG8gQmFzZTY0VXJsIGVuY29kaW5nIQ') + assert test == 'Hello Base64Url encoding!' +} + +fn test_encode_null_byte() { + assert base64.encode([byte(`A`), 0, `C`]) == 'QQBD' +} + +fn test_encode_null_byte_str() { + // While this works, bytestr() does a memcpy + s := [byte(`A`), 0, `C`].bytestr() + assert base64.encode_str(s) == 'QQBD' +} + +fn test_decode_null_byte() { + assert base64.decode('QQBD') == [byte(`A`), 0, `C`] +} + +fn test_decode_null_byte_str() { + // While this works, bytestr() does a memcpy + s := [byte(`A`), 0, `C`].bytestr() + assert base64.decode_str('QQBD') == s +} + +fn test_decode_in_buffer_bytes() { + rfc4648_pairs := [ + TestPair{'foob', 'Zm9vYg=='}, + TestPair{'fooba', 'Zm9vYmE='}, + TestPair{'foobar', 'Zm9vYmFy'}, + ] + mut src_dec_buf := []byte{len: 8} + mut src_enc_buf := []byte{len: 8} + mut out_buf := []byte{len: 8} + + for p in rfc4648_pairs { + src_dec_buf = p.decoded.bytes() + src_enc_buf = p.encoded.bytes() + n := base64.decode_in_buffer_bytes(src_enc_buf, out_buf.data) + assert src_dec_buf == out_buf[..n] + } +} -- cgit v1.2.3