Open BernhardValenti opened 3 years ago
This is because the JMdict dictionary data doesn't list the part-of-speech for nouns. It lists n
and adj-na
as tags, but this is not the same as part-of-speech.
updated my templates to:
{{#*inline "part-of-speech-pretty"}}
{{~#if (op "===" . "v1")~}}Ichidan verb
{{~else if (op "===" . "v5")~}}Godan verb
{{~else if (op "===" . "vk")~}}Kuru verb
{{~else if (op "===" . "vs")~}}Suru verb
{{~else if (op "===" . "vz")~}}Zuru verb
{{~else if (op "===" . "adj-i")~}}I-adjective
{{~else if (op "===" . "adj-na")~}}Na-adjective
{{~else if (op "===" . "n")~}}Noun
{{~else~}}{{.}}
{{~/if~}}
{{/inline}}
{{#*inline "part-of-speech"}}
{{~#scope~}}
{{~#if (op "!==" definition.type "kanji")~}}
{{~#set "first" true}}{{/set~}}
{{~#each definition.expressions~}}
{{~#each wordClasses~}}
{{~#unless (get (concat "used_" .))~}}
{{~> part-of-speech-pretty . ~}}
{{~#unless (get "first")}}, {{/unless~}}
{{~#set (concat "used_" .) true~}}{{~/set~}}
{{~#set "first" false~}}{{~/set~}}
{{~/unless~}}
{{~/each~}}
{{~/each~}}
{{~#if (get "first")~}}
{{#each definition.definitionTags}}
{{~#unless (get (concat "used_" .))~}}
{{~> part-of-speech-pretty name ~}}
{{~#unless (get "first")}}, {{/unless~}}
{{~#set (concat "used_" .) true~}}{{~/set~}}
{{~#set "first" false~}}{{~/set~}}
{{~/unless~}}
{{/each}}
{{~/if~}}
{{~#if (get "first")~}}Unknown{{~/if~}}
{{~/if~}}
{{~/scope~}}
{{/inline}}
This is because the JMdict dictionary data doesn't list the part-of-speech for nouns. It lists
n
andadj-na
as tags, but this is not the same as part-of-speech.
Just to be clear, the proper JMdict XML source data does distinguish part-of-speech tags from all other miscellaneous tags.
In the JMdict dictionary file produced for Yomichan by yomichan-import, the part-of-speech field only contains a limited and modified subset of those tags. From what I understand, these values are used behind-the-scenes for de-conjugating words into their dictionary forms so that they may be queried by yomichan. Part-of-speech tags that are not used for de-conjugation are not added to this part-of-speech list. However, all of the part-of-speech information is still added properly to the definition tags of each term; they're just mixed in with all the other miscellaneous and usage-domain tags that are displayed to the user in the glossary.
Based on my understanding of how this works, I don't think this {part-of-speech}
handlebar should even exist. All of this information already exists in a complete form within the {glossary}
field. The part-of-speech of a given word can also vary depending on the sense in which it is used. For example, 亜 can be a prefix or a noun. The {part-of-speech}
handlebar could be updated to return a list of all possible parts-of-speech for an expression, but why? That information isn't as useful and could even be confusing without the corresponding sense context (which exists in the glossary).
The part-of-speech tags for JMdict entries that are displayed in yomichan are each contained within a <span>
node with a data-category="partOfSpeech"
attribute, so maybe something could be done with that if a user needs a way to extract or query the data.
Description trying to add
{part-of-speech}
to my cards, but the template returnsUnknown
for nouns and na-adjectives. i-adjectives and verbs seem to return the proper values.eg:
雨
and元気
dictionary is JMdict, and in the yomichan browser it does show the
n
for noun, andadj-na
for na-adjectives.Browser version chrome 92
Yomichan version 21.7.31.2