alex/yomichan
1
Fork 0
Code Releases Activity

Moving large files to CSV format, deleting unused kradfile

Browse Source
This commit is contained in:
Alex Yatskov 2016-03-31 20:03:39 -07:00
parent b97e75ba32
commit 7eadff3457
4 changed files with 54 additions and 87 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
build_dict.sh
View File

@ -3,7 +3,6 @@
KANJIDIC=util/data/kanjidic
EDICT=util/data/edict
ENAMDICT=util/data/enamdict
KRADFILE=util/data/kradfile
DICT_DIR=ext/jp/data
DICT_DIR=ext/bg/data
util/compile.py --kanjidic $KANJIDIC --kradfile $KRADFILE --edict $EDICT --enamdict $ENAMDICT $DICT_DIR
util/compile.py --kanjidic $KANJIDIC --edict $EDICT --enamdict $ENAMDICT $DICT_DIR

67
ext/bg/dictionary.js
View File

@ -19,43 +19,33 @@
class Dictionary {
constructor() {
this.termDicts = [];
this.kanjiDicts = [];
this.terms = [];
this.termIndices = {};
this.kanji = [];
this.kanjiIndices = {};
}
addTermDict(termDict) {
this.termDicts.push(termDict);
addTermDict(terms) {
let index = this.terms.length;
for (const [e, r, g, t] in terms) {
this.storeIndex(this.termIndices, e, index);
this.storeIndex(this.termIndices, r, index++);
this.terms.push([e, r, g, t]);
}
}
addKanjiDict(kanjiDict) {
this.kanjiDicts.push(kanjiDict);
addKanjiDict(kanji) {
let index = this.kanji.length;
for (const [c, k, o, g] in kanji) {
this.storeIndex(this.kanjiIndices, c, index++);
this.kanji.push([c, k, o, g]);
}
}
findTerm(term) {
let results = [];
for (let dict of this.termDicts) {
results = results.concat(this.findTermInDict(term, dict));
}
return results;
}
findKanji(kanji) {
const results = [];
for (let dict of this.kanjiDicts) {
const result = this.findKanjiInDict(kanji, dict);
if (result !== null) {
results.push(result);
}
}
return results;
}
findTermInDict(term, dict) {
return (dict.indices[term] || []).map(index => {
const [e, r, g, t] = dict.defs[index];
return (this.termIndices[term] || []).map(index => {
const [e, r, g, t] = this.terms[index];
return {
id: index,
expression: e,
@ -66,12 +56,8 @@ class Dictionary {
});
}
findKanjiInDict(kanji, dict) {
const def = dict.defs[kanji];
if (def === null) {
return null;
}
findKanji(kanji) {
return (this.kanjiIndices[kanji] || []).map(index => {
const [c, k, o, g] = def;
return {
id: kanji.charCodeAt(0),
@ -80,5 +66,14 @@ class Dictionary {
onyomi: o,
glossary: g
};
});
}
storeIndex(indices, term, index) {
if (term.length > 0) {
const indices = this.termIndices[term] || [];
indices.push(term);
this.termIndices[term] = indices;
}
}
}

4
ext/client.js
View File

@ -27,7 +27,9 @@ class Client {
this.popup.classList.add('yomichan-popup');
this.popup.addEventListener('mousedown', (e) => e.stopPropagation());
this.popup.addEventListener('scroll', (e) => e.stopPropagation());
document.body.appendChild(this.popup);
const base = document.body.appendChild('div');
base.createShadowRoot().appendChild(this.popup);
chrome.runtime.onMessage.addListener(this.onMessage.bind(this));
window.addEventListener('mousedown', this.onMouseDown.bind(this));

51
util/compile.py
View File

@ -18,7 +18,6 @@
import codecs
import json
import optparse
import os.path
import re
@ -111,7 +110,7 @@ def load_definitions(path):
def parse_kanji_dic(path):
results = {}
results = []
for line in load_definitions(path):
segments = line.split()
@ -119,32 +118,20 @@ def parse_kanji_dic(path):
kunyomi = ', '.join(filter(lambda x: filter(is_hiragana, x), segments[1:]))
onyomi = ', '.join(filter(lambda x: filter(is_katakana, x), segments[1:]))
glossary = '; '.join(re.findall('\{([^\}]+)\}', line))
results[character] = (kunyomi, onyomi, glossary)
return results
def parse_krad_file(path):
results = {}
for line in load_definitions(path):
segments = line.split(' ')
character = segments[0]
radicals = ' '.join(segments[2:])
results[character] = radicals;
results.append((character, kunyomi, onyomi, glossary))
return results
def parse_edict(path):
defs = []
results = []
for line in load_definitions(path):
segments = line.split('/')
expression = segments[0].split(' ')
term = expression[0]
match = re.search('\[([^\]]+)\]', expression[1])
reading = None if match is None else match.group(1)
reading = '' if match is None else match.group(1)
glossary = '; '.join(filter(lambda x: len(x) > 0, segments[1:]))
glossary = re.sub('\(\d+\)\s*', '', glossary)
@ -156,30 +143,21 @@ def parse_edict(path):
tags = set(tags).intersection(PARSED_TAGS)
tags = ' '.join(tags)
defs.append((term, reading, glossary, tags))
results.append((term, reading, glossary, tags))
indices = {}
for i, d in enumerate(defs):
for key in d[:2]:
if key is not None:
values = indices.get(key, [])
values.append(i)
indices[key] = values
return {'defs': defs, 'indices': indices}
return results[1:]
def build_dict(output_dir, input_file, parser):
if input_file is not None:
base = os.path.splitext(os.path.basename(input_file))[0]
with open(os.path.join(output_dir, base) + '.json', 'w') as fp:
# json.dump(parser(input_file), fp, sort_keys=True, indent=4, separators=(',', ': '))
json.dump(parser(input_file), fp)
with codecs.open(os.path.join(output_dir, base) + '.csv', 'wb', encoding='utf-8') as fp:
for d in parser(input_file):
fp.write('\t'.join(d) + '\n')
def build(dict_dir, kanjidic, kradfile, edict, enamdict):
def build(dict_dir, kanjidic, edict, enamdict):
build_dict(dict_dir, kanjidic, parse_kanji_dic)
build_dict(dict_dir, kradfile, parse_krad_file)
build_dict(dict_dir, edict, parse_edict)
build_dict(dict_dir, enamdict, parse_edict)
@ -187,7 +165,6 @@ def build(dict_dir, kanjidic, kradfile, edict, enamdict):
def main():
parser = optparse.OptionParser()
parser.add_option('--kanjidic', dest='kanjidic')
parser.add_option('--kradfile', dest='kradfile')
parser.add_option('--edict', dest='edict')
parser.add_option('--enamdict', dest='enamdict')
@ -196,13 +173,7 @@ def main():
if len(args) == 0:
parser.print_help()
else:
build(
args[0],
options.kanjidic,
options.kradfile,
options.edict,
options.enamdict
)
build(args[0], options.kanjidic, options.edict, options.enamdict)
if __name__ == '__main__':