auto merge of #15025 : alexcrichton/rust/rollup, r=alexcrichton

Author: bors

src/compiler-rt

@@ -1269,7 +1269,7 @@ fn make_out_name(config: &Config, testfile: &Path, extension: &str) -> Path {
 
 fn aux_output_dir_name(config: &Config, testfile: &Path) -> Path {
     let mut f = output_base_name(config, testfile);
-    match f.filename().map(|s| Vec::from_slice(s).append(bytes!(".libaux"))) {
+    match f.filename().map(|s| Vec::from_slice(s).append(b".libaux")) {
         Some(v) => f.set_filename(v),
         None => ()
     }
@@ -1490,7 +1490,7 @@ fn append_suffix_to_stem(p: &Path, suffix: &str) -> Path {
         (*p).clone()
     } else {
         let stem = p.filestem().unwrap();
-        p.with_filename(Vec::from_slice(stem).append(bytes!("-")).append(suffix.as_bytes()))
+        p.with_filename(Vec::from_slice(stem).append(b"-").append(suffix.as_bytes()))
     }
 }
 

src/compiletest/runtest.rs

@@ -76,7 +76,7 @@ character.
 ~~~
 use std::str;
 
-let x = bytes!(72u8,"ello ",0xF0,0x90,0x80,"World!");
+let x = b"Hello \xF0\x90\x80World!";
 let y = str::from_utf8_lossy(x);
 ~~~
 

src/doc/complement-cheatsheet.md

@@ -378,6 +378,19 @@ the characters `U+0022` (double-quote) (except when followed by at least as
 many `U+0023` (`#`) characters as were used to start the raw string literal) or
 `U+005C` (`\`) do not have any special meaning.
 
+Examples for byte string literals:
+
+~~~~
+b"foo"; br"foo";                     // foo
+b"\"foo\""; br#""foo""#;             // "foo"
+
+b"foo #\"# bar";
+br##"foo #"# bar"##;                 // foo #"# bar
+
+b"\x52"; b"R"; br"R";                // R
+b"\\x52"; br"\x52";                  // \x52
+~~~~
+
 #### Number literals
 
 ~~~~ {.ebnf .gram}

src/doc/rust.md

@@ -0,0 +1,138 @@
+#!/bin/env python
+#
+# Copyright 2014 The Rust Project Developers. See the COPYRIGHT
+# file at the top-level directory of this distribution and at
+# http://rust-lang.org/COPYRIGHT.
+#
+# Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
+# http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
+# <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
+# option. This file may not be copied, modified, or distributed
+# except according to those terms.
+
+import sys
+import subprocess
+import re
+
+
+def main():
+    if len(sys.argv) <= 1:
+        print('Usage: %s [ --apply ] filename1.rs filename2.rs ...'
+              % sys.argv[0])
+    elif sys.argv[1] == '--apply':
+        for filename in sys.argv[2:]:
+            patch(filename)
+    else:
+        for filename in sys.argv[1:]:
+            diff(filename)
+
+
+def patch(filename):
+    source = read(filename)
+    rewritten = rewrite_bytes_macros(source)
+    if rewritten is not None and rewritten != source:
+        write(filename, rewritten)
+
+
+def diff(filename):
+    rewritten = rewrite_bytes_macros(read(filename))
+    if rewritten is not None:
+        p = subprocess.Popen(['diff', '-u', filename, '-'],
+                             stdin=subprocess.PIPE)
+        p.stdin.write(rewritten)
+        p.stdin.close()
+        p.wait()
+
+
+def read(filename):
+    with open(filename, 'rb') as f:
+        return f.read()
+
+
+def write(filename, content):
+    with open(filename, 'wb') as f:
+        f.write(content)
+
+
+def rewrite_bytes_macros(source):
+    rewritten, num_occurrences = BYTES_MACRO_RE.subn(rewrite_one_macro, source)
+    if num_occurrences > 0:
+        return rewritten
+
+
+BYTES_MACRO_RE = re.compile(br'bytes!\(  (?P<args>  [^)]*  )  \)', re.VERBOSE)
+
+
+def rewrite_one_macro(match):
+    try:
+        bytes = parse_bytes(split_args(match.group('args')))
+        return b'b"' + b''.join(map(escape, bytes)) + b'"'
+    except SkipThisRewrite:
+        print('Skipped: %s' % match.group(0).decode('utf8', 'replace'))
+        return match.group(0)
+
+
+class SkipThisRewrite(Exception):
+    pass
+
+
+def split_args(args):
+    previous = b''
+    for arg in args.split(b','):
+        if previous:
+            arg = previous + b',' + arg
+        if arg.count(b'"') % 2 == 0:
+            yield arg
+            previous = b''
+        else:
+            previous = arg
+    if previous:
+        yield previous
+
+
+def parse_bytes(args):
+    for arg in args:
+        arg = arg.strip()
+        if (arg.startswith(b'"') and arg.endswith(b'"')) or (
+                arg.startswith(b"'") and arg.endswith(b"'")):
+            # Escaped newline means something different in Rust and Python.
+            if b'\\\n' in arg:
+                raise SkipThisRewrite
+            for byte in eval(b'u' + arg).encode('utf8'):
+                yield ord(byte)
+        else:
+            if arg.endswith(b'u8'):
+                arg = arg[:-2]
+            # Assume that all Rust integer literals
+            # are valid Python integer literals
+            value = int(eval(arg))
+            assert value <= 0xFF
+            yield value
+
+
+def escape(byte):
+    c = chr(byte)
+    escaped = {
+        b'\0': br'\0',
+        b'\t': br'\t',
+        b'\n': br'\n',
+        b'\r': br'\r',
+        b'\'': b'\\\'',
+        b'\\': br'\\',
+    }.get(c)
+    if escaped is not None:
+        return escaped
+    elif b' ' <= c <= b'~':
+        return chr(byte)
+    else:
+        return ('\\x%02X' % byte).encode('ascii')
+
+
+if str is not bytes:
+    # Python 3.x
+    ord = lambda x: x
+    chr = lambda x: bytes([x])
+
+
+if __name__ == '__main__':
+    main()