interface: plugins.abstract.token_detail.AbstractTokenDetail

type: optional

Token detail plug-in is used for attaching (an external) information to any token in a concordance. Typically, this can be used to attach dictionaries, encyclopediae to individual tokens, named entities etc.

The plug-in is composed of three main general components:

1. backend represents an adapter communicating with an (external) service
2. client frontend visually interprets the data provided by the backend,
3. server frontend exports backend data to be readable by the client frontend.

In general it is expected to be possible to mix these (especially backend vs. frontend) in different ways - e.g. RawHtmlFrontend is probably usable along with any backend producing raw HTML output.

Please note that in case of this plug-in the key to customization lies in frontends and backends. It means that in case you need a special functionality, it will be probably enough to extend this plug-in by an empty class and add your frontend or backend (depending on what needs to be customized).

<kontext>
...
  <plugins>
    ...
    <corparch>
      <module>default_corparch</module>
      <js_module>ucnkCorparch</js_module>
      <file>/path/to/kontext/conf/corplist.xml</file>
    </corparch>
    <token_detail>
      <module>default_token_detail</module>
      <js_module>defaultTokenDetail</js_module>
      <backends>/path/to/kontext/conf/token-detail-providers.json</backends>
    </token_detail>
  </plugins>
</kontext>

[
  {
    "ident": "wiktionary_for_ic_9_en",
    "heading": {
      "en_US": "Wiktionary",
      "cs_CZ": "Wiktionary"
    },
    "backend": "plugins.default_token_detail.backends.HTTPBackend",
    "frontend": "plugins.default_token_detail.frontends.RawHtmlFrontend",
    "conf": {
      "server": "en.wiktionary.org",
      "path": "/w/index.php?title={lemma}&action=render",
      "ssl": true,
      "port": 443
    }
  },
  {
    "ident": "other_provider",
    "...": "..... etc ..."
  }
]

In a corpora configuration file (corplist.xml), a corpus we want to be used along with this service must define a token detail provider:

<kontext>
  <corplist>
    <corpus ident="intercorp_v9_en">  
      <token_detail>
        <provider>wiktionary_for_ic_9_en</provider>
        <provider>other_provider</provider>
      </token_detail>
    </corpus>
    ...
  </corplist>
</kontext>