aboutsummaryrefslogtreecommitdiff
path: root/ext
diff options
context:
space:
mode:
authorsiikamiika <siikamiika@users.noreply.github.com>2019-11-05 15:56:45 +0200
committersiikamiika <siikamiika@users.noreply.github.com>2019-11-23 17:45:44 +0200
commit955e131f9673e006556bc2c5e0b3551a614ccc48 (patch)
treedcabce9d7e6008ef490d7e6316eae99008db1c70 /ext
parent17003189888694da51d840dcf0464355bb342a4c (diff)
add parser selection options
Diffstat (limited to 'ext')
-rw-r--r--ext/bg/js/api.js37
-rw-r--r--ext/bg/js/mecab.js2
-rw-r--r--ext/bg/js/options.js5
-rw-r--r--ext/bg/js/search-query-parser.js40
-rw-r--r--ext/bg/js/settings.js6
-rw-r--r--ext/bg/settings.html29
6 files changed, 87 insertions, 32 deletions
diff --git a/ext/bg/js/api.js b/ext/bg/js/api.js
index 2ab01af3..967bded7 100644
--- a/ext/bg/js/api.js
+++ b/ext/bg/js/api.js
@@ -109,25 +109,30 @@ async function apiTextParseMecab(text, optionsContext) {
const options = await apiOptionsGet(optionsContext);
const mecab = utilBackend().mecab;
- const results = [];
- for (const parsedLine of await mecab.parseText(text)) {
- for (const {expression, reading, source} of parsedLine) {
- const term = [];
- if (expression && reading) {
- for (const {text, furigana} of jpDistributeFuriganaInflected(
- expression,
- jpKatakanaToHiragana(reading),
- source
- )) {
- // can't use 'furigana' in templates
- term.push({text, reading: furigana});
+ const results = {};
+ const rawResults = await mecab.parseText(text);
+ for (const mecabName in rawResults) {
+ const result = [];
+ for (const parsedLine of rawResults[mecabName]) {
+ for (const {expression, reading, source} of parsedLine) {
+ const term = [];
+ if (expression && reading) {
+ for (const {text, furigana} of jpDistributeFuriganaInflected(
+ expression,
+ jpKatakanaToHiragana(reading),
+ source
+ )) {
+ // can't use 'furigana' in templates
+ term.push({text, reading: furigana});
+ }
+ } else {
+ term.push({text: source});
}
- } else {
- term.push({text: source});
+ result.push(term);
}
- results.push(term);
+ result.push([{text: '\n'}]);
}
- results.push([{text: '\n'}]);
+ results[mecabName] = result;
}
return results;
}
diff --git a/ext/bg/js/mecab.js b/ext/bg/js/mecab.js
index 14f68393..fba9b2eb 100644
--- a/ext/bg/js/mecab.js
+++ b/ext/bg/js/mecab.js
@@ -60,4 +60,4 @@ class Mecab {
}
}
-Mecab.timeout = 1000;
+Mecab.timeout = 5000;
diff --git a/ext/bg/js/options.js b/ext/bg/js/options.js
index be1ccfbb..f1bafaf9 100644
--- a/ext/bg/js/options.js
+++ b/ext/bg/js/options.js
@@ -311,6 +311,11 @@ function profileOptionsCreateDefaults() {
dictionaries: {},
+ parsing: {
+ enableScanningParser: true,
+ enableMecabParser: false
+ },
+
anki: {
enable: false,
server: 'http://127.0.0.1:8765',
diff --git a/ext/bg/js/search-query-parser.js b/ext/bg/js/search-query-parser.js
index 1cf00425..81eb18c3 100644
--- a/ext/bg/js/search-query-parser.js
+++ b/ext/bg/js/search-query-parser.js
@@ -86,22 +86,32 @@ class QueryParser {
this.search.setSpinnerVisible(true);
await this.setPreview(text);
- // const results = await apiTextParse(text, this.search.getOptionsContext());
- const results = await apiTextParseMecab(text, this.search.getOptionsContext());
-
- const content = await apiTemplateRender('query-parser.html', {
- terms: results.map((term) => {
- return term.filter(part => part.text.trim()).map((part) => {
- return {
- text: Array.from(part.text),
- reading: part.reading,
- raw: !part.reading || !part.reading.trim(),
- };
- });
- })
- });
+ const results = {};
+ if (this.search.options.parsing.enableScanningParser) {
+ results['scan'] = await apiTextParse(text, this.search.getOptionsContext());
+ }
+ if (this.search.options.parsing.enableMecabParser) {
+ let mecabResults = await apiTextParseMecab(text, this.search.getOptionsContext());
+ for (const mecabDictName in mecabResults) {
+ results[`mecab-${mecabDictName}`] = mecabResults[mecabDictName];
+ }
+ }
- this.queryParser.innerHTML = content;
+ const contents = await Promise.all(Object.values(results).map(async result => {
+ return await apiTemplateRender('query-parser.html', {
+ terms: result.map((term) => {
+ return term.filter(part => part.text.trim()).map((part) => {
+ return {
+ text: Array.from(part.text),
+ reading: part.reading,
+ raw: !part.reading || !part.reading.trim(),
+ };
+ });
+ })
+ });
+ }));
+
+ this.queryParser.innerHTML = contents.join('<hr>');
this.queryParser.querySelectorAll('.query-parser-char').forEach((charElement) => {
this.activateScanning(charElement);
diff --git a/ext/bg/js/settings.js b/ext/bg/js/settings.js
index e562c54e..f4fe032a 100644
--- a/ext/bg/js/settings.js
+++ b/ext/bg/js/settings.js
@@ -64,6 +64,9 @@ async function formRead(options) {
options.scanning.modifier = $('#scan-modifier-key').val();
options.scanning.popupNestingMaxDepth = parseInt($('#popup-nesting-max-depth').val(), 10);
+ options.parsing.enableScanningParser = $('#parsing-scan-enable').prop('checked');
+ options.parsing.enableMecabParser = $('#parsing-mecab-enable').prop('checked');
+
const optionsAnkiEnableOld = options.anki.enable;
options.anki.enable = $('#anki-enable').prop('checked');
options.anki.tags = utilBackgroundIsolate($('#card-tags').val().split(/[,; ]+/));
@@ -126,6 +129,9 @@ async function formWrite(options) {
$('#scan-modifier-key').val(options.scanning.modifier);
$('#popup-nesting-max-depth').val(options.scanning.popupNestingMaxDepth);
+ $('#parsing-scan-enable').prop('checked', options.parsing.enableScanningParser);
+ $('#parsing-mecab-enable').prop('checked', options.parsing.enableMecabParser);
+
$('#anki-enable').prop('checked', options.anki.enable);
$('#card-tags').val(options.anki.tags.join(' '));
$('#sentence-detection-extent').val(options.anki.sentenceExt);
diff --git a/ext/bg/settings.html b/ext/bg/settings.html
index bdcc11d3..8505567b 100644
--- a/ext/bg/settings.html
+++ b/ext/bg/settings.html
@@ -587,6 +587,35 @@
</div></div>
</div>
+ <div id="text-parsing">
+ <h3>Text Parsing Options</h3>
+
+ <p class="help-block">
+ Yomichan can attempt to parse entire sentences or longer text blocks on the search page,
+ adding furigana above words and a small space between words.
+ </p>
+
+ <p class="help-block">
+ Two types of parsers are supported. The first one, enabled by default, works using the built-in
+ scanning functionality by automatically advancing in the sentence after a matching word.
+ </p>
+
+ <p class="help-block">
+ The second type is an external program called <a href="https://en.wikipedia.org/wiki/MeCab" target="_blank" rel="noopener">MeCab</a>
+ that uses its own dictionaries and a special parsing algorithm. To get it working, you must first
+ install it and <a href="https://github.com/siikamiika/yomichan-mecab-installer" target="_blank" rel="noopener">a native messaging component</a>
+ that acts as a bridge between the program and Yomichan.
+ </p>
+
+ <div class="checkbox">
+ <label><input type="checkbox" id="parsing-scan-enable"> Enable text parsing using installed dictionaries</label>
+ </div>
+
+ <div class="checkbox">
+ <label><input type="checkbox" id="parsing-mecab-enable"> Enable text parsing using MeCab</label>
+ </div>
+ </div>
+
<div>
<div>
<img src="/mixed/img/spinner.gif" class="pull-right" id="anki-spinner" alt>