#!/usr/bin/env python # # Copyright 2011-2015 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 or the MIT license # , at your # option. This file may not be copied, modified, or distributed # except according to those terms. # This script uses the following Unicode tables: # - PropertyValueAliases.txt # - ScriptExtensions.txt # - Scripts.txt # # Since this should not require frequent updates, we just store this # out-of-line and check the unicode.rs file into git. import fileinput, re, os, sys preamble = '''// Copyright 2012-2018 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 or the MIT license // , at your // option. This file may not be copied, modified, or distributed // except according to those terms. // NOTE: The following code was generated by "scripts/unicode.py", do not edit directly #![allow(missing_docs, non_upper_case_globals, non_snake_case)] pub use tables_impl::*; #[rustfmt::skip] mod tables_impl { use crate::ScriptExtension; ''' # Close `mod impl {` ending=''' } ''' UNICODE_VERSION = (16, 0, 0) UNICODE_VERSION_NUMBER = "%s.%s.%s" %UNICODE_VERSION def escape_char(c): return "'\\u{%x}'" % c def fetch(f): if not os.path.exists(os.path.basename(f)): if "emoji" in f: os.system("curl -O https://www.unicode.org/Public/emoji/%s.%s/%s" % (UNICODE_VERSION[0], UNICODE_VERSION[1], f)) else: os.system("curl -O https://www.unicode.org/Public/%s/ucd/%s" % (UNICODE_VERSION_NUMBER, f)) if not os.path.exists(os.path.basename(f)): sys.stderr.write("cannot load %s" % f) exit(1) def group_cats(cats): cats_out = {} for cat in cats: cats_out[cat] = group_cat(cats[cat]) return cats_out def aliases(): """ Fetch the shorthand aliases for each longhand Script name """ fetch("PropertyValueAliases.txt") longforms = {} shortforms = {} re1 = re.compile(r"^ *sc *; *(\w+) *; *(\w+)") for line in fileinput.input(os.path.basename("PropertyValueAliases.txt")): m = re1.match(line) if m: l = m.group(2).strip() s = m.group(1).strip() assert(s not in longforms) assert(l not in shortforms) longforms[s] = l shortforms[l] = s else: continue return (longforms, shortforms) 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) # Implementation from unicode-segmentation def load_properties(f, interestingprops): fetch(f) props = {} # Note: these regexes are different from those in unicode-segmentation, # becase we need to handle spaces here re1 = re.compile(r"^ *([0-9A-F]+) *; *([^#]+) *#") re2 = re.compile(r"^ *([0-9A-F]+)\.\.([0-9A-F]+) *; *([^#]+) *#") for line in fileinput.input(os.path.basename(f)): prop = None d_lo = 0 d_hi = 0 m = re1.match(line) if m: d_lo = m.group(1) d_hi = m.group(1) prop = m.group(2).strip() else: m = re2.match(line) if m: d_lo = m.group(1) d_hi = m.group(2) prop = m.group(3).strip() else: continue if interestingprops and prop not in interestingprops: continue d_lo = int(d_lo, 16) d_hi = int(d_hi, 16) if prop not in props: props[prop] = [] props[prop].append((d_lo, d_hi)) return props # Implementation from unicode-segmentation def emit_table(f, name, t_data, t_type = "&'static [(char, char)]", is_pub=True, pfun=lambda x: "(%s,%s)" % (escape_char(x[0]), escape_char(x[1])), is_const=True): pub_string = "const" if not is_const: pub_string = "let" if is_pub: pub_string = "pub " + pub_string f.write(" %s %s: %s = &[\n" % (pub_string, name, t_type)) data = "" first = True for dat in t_data: if not first: data += "," first = False data += pfun(dat) format_table_content(f, data, 8) f.write("\n ];\n\n") def emit_search(f): f.write(""" pub fn bsearch_range_value_table(c: char, r: &'static [(char, char, T)]) -> Option { use core::cmp::Ordering::{Equal, Less, Greater}; match r.binary_search_by(|&(lo, hi, _)| { if lo <= c && c <= hi { Equal } else if hi < c { Less } else { Greater } }) { Ok(idx) => { let (_, _, cat) = r[idx]; Some(cat) } Err(_) => None } } #[inline] pub fn get_script(c: char) -> Option