Kevin Xiang Li

See title.

动态更改menu/page_size时，调用apply_schema报错

38

我在开发一个韩文输入法，目标是平常输入韩文的时候不显示menu (page_size = 0)，menu只在反查汉字的时候显示。基于[PR135](https://github.com/hchunhui/librime-lua/pull/135)，我写了一个简单版本的processor，当按下F10键时会改变朙月拼音的page_size。不过`env.engine.apply_schema`报错说attempt to call a nil value (field 'apply_schema')，`env.engine:apply_schema`也有类似报错。 报错message: ``` E0930 14:06:07.148494 44123520 lua_gears.cc:150] LuaProcessor::ProcessKeyEvent error(2): /Users/kevin/Library/Rime/lua/page_size_processor.lua:20: attempt to call a nil value (field 'apply_schema') ``` processor代码如下：...

Building flutter_web_browser 0.17.1 gives the following warning: Note: /opt/homebrew/Caskroom/flutter/3.3.8/flutter/.pub-cache/hosted/pub.dartlang.org/flutter_web_browser-0.17.1/android/src/main/java/dev/vbonnet/flutterwebbrowser/FlutterWebBrowserPlugin.java uses or overrides a deprecated API. Note: Recompile with -Xlint:deprecation for details.

Thoughts on automating the addition of canto_unique characters and phrases

9

Because we don't have a reliable Cantonese tokenizer, all the phrases added to the classifier need to satisfy the following property, in addition to not being shared by Mandarin: **Frequency...

Thanks for developing this awesome crate! It'd great if the underlying dot representation can be shared with another great library called [`graphviz-rust`](https://github.com/besok/graphviz-rust). Sharing the representation can combine the intuitive interface...

**Describe the issue** MFA 3.0.0 deprecated phone_set and the `mfa train` command starts to offer an alternative `phone_groups_path` option. However, `mfa validate` still only offers the deprecated `phone_set` but no...

Trouble building on macOS (my workarounds and suggestions)

4

This library looks awesome but I had trouble building fstalign on macOS sonoma (the issues likely apply to M1 machines running ventura as well). I will outline the steps I...

添加phrase dict的支持

2

https://github.com/mozillazg/rust-pinyin/issues/56 加入了一个PinyinPhraseIter，用户提供一个词语串（已分词），拿到每个词语相应的拼音。如果一个词语中含有不支持的字符，整个词语的注音为None，不然就生成一个Some(vec![..])，vec里面是词语中每个字的注音。 ```rust # #[cfg(feature = "plain")] { use pinyin::{ToPinyin, Pinyin}; let mut iter = ["薄荷", "是", "便宜货"].iter().to_pinyin(); let mut next_plain = || iter.next().map(|ps| ps.map(|ps| ps.iter().map(|p| Pinyin::plain(*p)).collect::())); assert_eq!(next_plain(), Some(Some(vec!["bo", "he"])));...

**Title** Support self-hosting ML models **Describe the feature request** ML models are currently downloaded from dl.google.com, which is blocked in countries like China. It would be very helpful if this...