Kindy Lin

项目里用到一个 200M+ 的词典，同样初始化和解析时间都很长，粗略试了下 [marisa_trie](https://github.com/pytries/marisa-trie)，貌似是可以降低内存占用和加载效率的。marisa_trie 的 dump 文件尺寸极小，且支持 mmap。 dict.txt : ~260M /tmp/jieba.cache : ~1.3G my_trie.marisa : ~32M 实际用的话，应该还需要把一些使用 `FREQ` 的地方按照 marisa_trie 对应方法调整一下即可。 ```py def gen_pfdict_mtrie(): f = get_module_res(DEFAULT_DICT_NAME) info =...

config in Pulumi.yaml is called **project level configuration**, and doc at [concepts/config - Project Level Configuration](https://www.pulumi.com/docs/concepts/config/#project-level-configuration) spec here [concepts/projects/project-file/#config-options](https://www.pulumi.com/docs/concepts/projects/project-file/#config-options) I also been confused by the difference schema of project level...

@jarwin123 sorry, but I'm not maintainer, please ask question at: * Homepage: http://software.schmorp.de/pkg/libev * Mailinglist: [email protected] * http://lists.schmorp.de/cgi-bin/mailman/listinfo/libev

Hi @justinvp, thanks for the reply, it helps a lot. --- one more question: if user config some field/value as secret in stack config, should pulumi mark the value as...

those old JIRA search API just closed. 😭

I'll open a PR to fix it. - [ ] make it works again - [x] check all languages, not only based-on en-US cc @crazywoola

@crazywoola please take a look this pr, for i18n checker.

> This page seems to be public available. I think this page should be disabled for production environment. @crazywoola do we have some examples for this use case?

@crazywoola I create a folder `dev-only` and move i18n-checker there. this path `/dev-only/**` can only access in development environments.

seems @blocknote/core have no break changes between v0.19.2 and v0.23.6, I'll try that version~