---

yaml
---
r: 144252
b: refs/heads/try2
c: 4e3dbf9
h: refs/heads/master
v: v3

Author: bors

[refs]

@@ -5,7 +5,7 @@ refs/heads/snap-stage3: 78a7676898d9f80ab540c6df5d4c9ce35bb50463
 refs/heads/try: 519addf6277dbafccbb4159db4b710c37eaa2ec5
 refs/tags/release-0.1: 1f5c5126e96c79d22cb7862f75304136e204f105
 refs/heads/ndm: f3868061cd7988080c30d6d5bf352a5a5fe2460b
-refs/heads/try2: ffb6404c5ade3af113738c3cb72fd853e82a379e
+refs/heads/try2: 4e3dbf959a5124481d1e3ec9b0b30f48ac6dd4f0
 refs/heads/dist-snap: ba4081a5a8573875fed17545846f6f6902c8ba8d
 refs/tags/release-0.2: c870d2dffb391e14efb05aa27898f1f6333a9596
 refs/tags/release-0.3: b5f0d0f648d9a6153664837026ba1be43d3e2503

branches/try2/src/etc/unicode.py

@@ -26,11 +26,15 @@ def fetch(f):
 def load_unicode_data(f):
     fetch(f)
     gencats = {}
+    combines = []
     canon_decomp = {}
     compat_decomp = {}
     curr_cat = ""
+    curr_combine = ""
     c_lo = 0
     c_hi = 0
+    com_lo = 0
+    com_hi = 0
     for line in fileinput.input(f):
         fields = line.split(";")
         if len(fields) != 15:
@@ -69,7 +73,21 @@ def load_unicode_data(f):
             c_lo = code
             c_hi = code
 
-    return (canon_decomp, compat_decomp, gencats)
+        if curr_combine == "":
+            curr_combine = combine
+            com_lo = code
+            com_hi = code
+
+        if curr_combine == combine:
+            com_hi = code
+        else:
+            if curr_combine != "0":
+                combines.append((com_lo, com_hi, curr_combine))
+            curr_combine = combine
+            com_lo = code
+            com_hi = code
+
+    return (canon_decomp, compat_decomp, gencats, combines)
 
 
 def load_derived_core_properties(f):
@@ -178,50 +196,149 @@ def emit_property_module_old(f, mod, tbl):
         f.write("    }\n\n")
     f.write("}\n")
 
-def emit_decomp_module(f, canon, compat):
+def format_table_content(f, content, indent):
+    line = " "*indent
+    first = True
+    for chunk in content.split(","):
+        if len(line) + len(chunk) < 98:
+            if first:
+                line += chunk
+            else:
+                line += ", " + chunk
+            first = False
+        else:
+            f.write(line + ",\n")
+            line = " "*indent + chunk
+    f.write(line)
+
+def emit_decomp_module(f, canon, compat, combine):
     canon_keys = canon.keys()
     canon_keys.sort()
 
     compat_keys = compat.keys()
     compat_keys.sort()
-    f.write("mod decompose {\n\n");
-    f.write("    export canonical, compatibility;\n\n")
-    f.write("    fn canonical(c: char, i: block(char)) "
-            + "{ d(c, i, false); }\n\n")
-    f.write("    fn compatibility(c: char, i: block(char)) "
-            +"{ d(c, i, true); }\n\n")
-    f.write("    fn d(c: char, i: block(char), k: bool) {\n")
+    f.write("pub mod decompose {\n");
+    f.write("    use option::Option;\n");
+    f.write("    use option::{Some, None};\n");
+    f.write("    use vec::ImmutableVector;\n");
+    f.write("""
+    fn bsearch_table(c: char, r: &'static [(char, &'static [char])]) -> Option<&'static [char]> {
+        use cmp::{Equal, Less, Greater};
+        match r.bsearch(|&(val, _)| {
+            if c == val { Equal }
+            else if val < c { Less }
+            else { Greater }
+        }) {
+            Some(idx) => {
+                let (_, result) = r[idx];
+                Some(result)
+            }
+            None => None
+        }
+    }\n
+""")
 
-    f.write("        if c <= '\\x7f' { i(c); ret; }\n")
+    f.write("""
+    fn bsearch_range_value_table(c: char, r: &'static [(char, char, u8)]) -> u8 {
+        use cmp::{Equal, Less, Greater};
+        match r.bsearch(|&(lo, hi, _)| {
+            if lo <= c && c <= hi { Equal }
+            else if hi < c { Less }
+            else { Greater }
+        }) {
+            Some(idx) => {
+                let (_, _, result) = r[idx];
+                result
+            }
+            None => 0
+        }
+    }\n\n
+""")
 
-    # First check the canonical decompositions
-    f.write("        // Canonical decomposition\n")
-    f.write("        alt c {\n")
+    f.write("    // Canonical decompositions\n")
+    f.write("    static canonical_table : &'static [(char, &'static [char])] = &[\n")
+    data = ""
+    first = True
     for char in canon_keys:
-        f.write("          %s {\n" % escape_char(char))
+        if not first:
+            data += ","
+        first = False
+        data += "(%s,&[" % escape_char(char)
+        first2 = True
         for d in canon[char]:
-            f.write("            d(%s, i, k);\n"
-                    % escape_char(d))
-        f.write("          }\n")
+            if not first2:
+                data += ","
+            first2 = False
+            data += escape_char(d)
+        data += "])"
+    format_table_content(f, data, 8)
+    f.write("\n    ];\n\n")
+
+    f.write("    // Compatibility decompositions\n")
+    f.write("    static compatibility_table : &'static [(char, &'static [char])] = &[\n")
+    data = ""
+    first = True
+    for char in compat_keys:
+        if not first:
+            data += ","
+        first = False
+        data += "(%s,&[" % escape_char(char)
+        first2 = True
+        for d in compat[char]:
+            if not first2:
+                data += ","
+            first2 = False
+            data += escape_char(d)
+        data += "])"
+    format_table_content(f, data, 8)
+    f.write("\n    ];\n\n")
+
+    f.write("    static combining_class_table : &'static [(char, char, u8)] = &[\n")
+    ix = 0
+    for pair in combine:
+        f.write(ch_prefix(ix))
+        f.write("(%s, %s, %s)" % (escape_char(pair[0]), escape_char(pair[1]), pair[2]))
+        ix += 1
+    f.write("\n    ];\n")
+
+    f.write("    pub fn canonical(c: char, i: &fn(char)) "
+        + "{ d(c, i, false); }\n\n")
+    f.write("    pub fn compatibility(c: char, i: &fn(char)) "
+            +"{ d(c, i, true); }\n\n")
+    f.write("    pub fn canonical_combining_class(c: char) -> u8 {\n"
+        + "        bsearch_range_value_table(c, combining_class_table)\n"
+        + "    }\n\n")
+    f.write("    fn d(c: char, i: &fn(char), k: bool) {\n")
+    f.write("        use iterator::Iterator;\n");
 
-    f.write("          _ { }\n")
-    f.write("        }\n\n")
+    f.write("        if c <= '\\x7f' { i(c); return; }\n")
+
+    # First check the canonical decompositions
+    f.write("""
+        match bsearch_table(c, canonical_table) {
+            Some(canon) => {
+                for x in canon.iter() {
+                    d(*x, |b| i(b), k);
+                }
+                return;
+            }
+            None => ()
+        }\n\n""")
 
     # Bottom out if we're not doing compat.
-    f.write("        if !k { i(c); ret; }\n\n ")
+    f.write("        if !k { i(c); return; }\n")
 
     # Then check the compatibility decompositions
-    f.write("        // Compatibility decomposition\n")
-    f.write("        alt c {\n")
-    for char in compat_keys:
-        f.write("          %s {\n" % escape_char(char))
-        for d in compat[char]:
-            f.write("            d(%s, i, k);\n"
-                    % escape_char(d))
-        f.write("          }\n")
-
-    f.write("          _ { }\n")
-    f.write("        }\n\n")
+    f.write("""
+        match bsearch_table(c, compatibility_table) {
+            Some(compat) => {
+                for x in compat.iter() {
+                    d(*x, |b| i(b), k);
+                }
+                return;
+            }
+            None => ()
+        }\n\n""")
 
     # Finally bottom out.
     f.write("        i(c);\n")
@@ -234,7 +351,7 @@ def emit_decomp_module(f, canon, compat):
         os.remove(i);
 rf = open(r, "w")
 
-(canon_decomp, compat_decomp, gencats) = load_unicode_data("UnicodeData.txt")
+(canon_decomp, compat_decomp, gencats, combines) = load_unicode_data("UnicodeData.txt")
 
 # Preamble
 rf.write('''// Copyright 2012-2013 The Rust Project Developers. See the COPYRIGHT
@@ -256,7 +373,7 @@ def emit_decomp_module(f, canon, compat):
 
 emit_property_module(rf, "general_category", gencats)
 
-#emit_decomp_module(rf, canon_decomp, compat_decomp)
+emit_decomp_module(rf, canon_decomp, compat_decomp, combines)
 
 derived = load_derived_core_properties("DerivedCoreProperties.txt")
 emit_property_module(rf, "derived_property", derived)

branches/try2/src/libextra/fileinput.rs

@@ -27,7 +27,7 @@ reset once it has been finished, so attempting to iterate on `[None,
 None]` will only take input once unless `io::stdin().seek(0, SeekSet)`
 is called between.
 
-The `pathify` function handles converting a list of file paths as
+The `make_path_option_vec` function handles converting a list of file paths as
 strings to the appropriate format, including the (optional) conversion
 of `"-"` to `stdin`.
 
@@ -42,7 +42,7 @@ to handle any `FileInput` structs. E.g. a simple `cat` program
 
 or a program that numbers lines after concatenating two files
 
-    for input_vec_state(pathify([~"a.txt", ~"b.txt"])) |line, state| {
+    for input_vec_state(make_path_option_vec([~"a.txt", ~"b.txt"])) |line, state| {
         io::println(fmt!("%u: %s", state.line_num,
                                    line));
     }
@@ -145,8 +145,14 @@ struct FileInput_ {
     previous_was_newline: bool
 }
 
-// XXX: remove this when Reader has &mut self. Should be removable via
-// "self.fi." -> "self." and renaming FileInput_. Documentation above
+
+// FIXME #5723: remove this when Reader has &mut self.
+// Removing it would mean giving read_byte in the Reader impl for
+// FileInput &mut self, which in turn means giving most of the
+// io::Reader trait methods &mut self. That can't be done right now
+// because of io::with_bytes_reader and #5723.
+// Should be removable via
+// "self.fi" -> "self." and renaming FileInput_. Documentation above
 // will likely have to be updated to use `let mut in = ...`.
 pub struct FileInput  {
     fi: @mut FileInput_
@@ -194,7 +200,7 @@ impl FileInput {
     */
     pub fn from_args() -> FileInput {
         let args = os::args();
-        let pathed = pathify(args.tail(), true);
+        let pathed = make_path_option_vec(args.tail(), true);
         FileInput::from_vec(pathed)
     }
 
@@ -351,8 +357,7 @@ Convert a list of strings to an appropriate form for a `FileInput`
 instance. `stdin_hyphen` controls whether `-` represents `stdin` or
 a literal `-`.
 */
-// XXX: stupid, unclear name
-pub fn pathify(vec: &[~str], stdin_hyphen : bool) -> ~[Option<Path>] {
+pub fn make_path_option_vec(vec: &[~str], stdin_hyphen : bool) -> ~[Option<Path>] {
     vec.iter().map(|str| {
         if stdin_hyphen && "-" == *str {
             None
@@ -410,7 +415,7 @@ pub fn input_vec_state(files: ~[Option<Path>],
 #[cfg(test)]
 mod test {
 
-    use super::{FileInput, pathify, input_vec, input_vec_state};
+    use super::{FileInput, make_path_option_vec, input_vec, input_vec_state};
 
     use std::io;
     use std::uint;
@@ -426,22 +431,22 @@ mod test {
     }
 
     #[test]
-    fn test_pathify() {
+    fn test_make_path_option_vec() {
         let strs = [~"some/path",
                     ~"some/other/path"];
         let paths = ~[Some(Path("some/path")),
                       Some(Path("some/other/path"))];
 
-        assert_eq!(pathify(strs, true), paths.clone());
-        assert_eq!(pathify(strs, false), paths);
+        assert_eq!(make_path_option_vec(strs, true), paths.clone());
+        assert_eq!(make_path_option_vec(strs, false), paths);
 
-        assert_eq!(pathify([~"-"], true), ~[None]);
-        assert_eq!(pathify([~"-"], false), ~[Some(Path("-"))]);
+        assert_eq!(make_path_option_vec([~"-"], true), ~[None]);
+        assert_eq!(make_path_option_vec([~"-"], false), ~[Some(Path("-"))]);
     }
 
     #[test]
     fn test_fileinput_read_byte() {
-        let filenames = pathify(vec::from_fn(
+        let filenames = make_path_option_vec(vec::from_fn(
             3,
             |i| fmt!("tmp/lib-fileinput-test-fileinput-read-byte-%u.tmp", i)), true);
 
@@ -471,7 +476,7 @@ mod test {
 
     #[test]
     fn test_fileinput_read() {
-        let filenames = pathify(vec::from_fn(
+        let filenames = make_path_option_vec(vec::from_fn(
             3,
             |i| fmt!("tmp/lib-fileinput-test-fileinput-read-%u.tmp", i)), true);
 
@@ -492,7 +497,7 @@ mod test {
     #[test]
     fn test_input_vec() {
         let mut all_lines = ~[];
-        let filenames = pathify(vec::from_fn(
+        let filenames = make_path_option_vec(vec::from_fn(
             3,
             |i| fmt!("tmp/lib-fileinput-test-input-vec-%u.tmp", i)), true);
 
@@ -514,7 +519,7 @@ mod test {
 
     #[test]
     fn test_input_vec_state() {
-        let filenames = pathify(vec::from_fn(
+        let filenames = make_path_option_vec(vec::from_fn(
             3,
             |i| fmt!("tmp/lib-fileinput-test-input-vec-state-%u.tmp", i)),true);
 
@@ -536,7 +541,7 @@ mod test {
 
     #[test]
     fn test_empty_files() {
-        let filenames = pathify(vec::from_fn(
+        let filenames = make_path_option_vec(vec::from_fn(
             3,
             |i| fmt!("tmp/lib-fileinput-test-empty-files-%u.tmp", i)),true);
 
@@ -583,7 +588,7 @@ mod test {
 
     #[test]
     fn test_next_file() {
-        let filenames = pathify(vec::from_fn(
+        let filenames = make_path_option_vec(vec::from_fn(
             3,
             |i| fmt!("tmp/lib-fileinput-test-next-file-%u.tmp", i)),true);
 
@@ -614,7 +619,7 @@ mod test {
     #[test]
     #[should_fail]
     fn test_input_vec_missing_file() {
-        do input_vec(pathify([~"this/file/doesnt/exist"], true)) |line| {
+        do input_vec(make_path_option_vec([~"this/file/doesnt/exist"], true)) |line| {
             println(line);
             true
         };