Sections without intro texts

[DanielSWolf]

Consider the following script:

var wtf = require("wtf_wikipedia");

const text = `
==English==

===Etymology===
{{suffix|abbreviate|ory|lang=en}}

===Pronunciation===
* {{a|GenAm}} {{IPA|/əˈbɹi.vi.əˌtɔɹ.i/|lang=en}}

===Adjective===
{{en-adj}}

# Serving or tending to [[abbreviate]]; [[shortening]]; [[abridging]]. {{defdate|Mid 19<sup>th</sup> century.}}<ref name=SOED>{{R:SOED5|page=3}}</ref>

====Translations====
{{trans-top|serving or tending to abbreviate; shortening; abridging}}
* Dutch: {{t+|nl|afkortend}}, {{t+|nl|verkortend}}
{{trans-mid}}
* Portuguese: {{t|pt|abreviatório}}, {{t|pt|abreviativo}}
{{trans-bottom}}

===References===
<references/>
`;

const document = wtf(text, { title: 'abbreviatory' });
console.log(JSON.stringify(document.json(), null, 2));

This document (which is taken from Wiktionary) has a single top-level section with the title "English". Nested within that top-level section are four second-level sections.

Here's the output:

{
  "title": "abbreviatory",
  "citations": [
    {
      "template": "citation",
      "type": "inline",
      "data": {},
      "inline": {
        "text": "{{R:SOED5|page=3}}"
      }
    }
  ],
  "sections": [
    {
      "title": "",
      "depth": 0,
      "sentences": [
        {
          "text": "==English=="
        }
      ]
    },
    {
      "title": "Etymology",
      "depth": 1,
      "sentences": [],
      "templates": [
        {
          "template": "suffix",
          "data": {
            "lang": "en"
          }
        }
      ]
    },
    {
      "title": "Pronunciation",
      "depth": 1,
      "sentences": [],
      "templates": [
        {
          "template": "ipa",
          "lang": null,
          "ipa": "/əˈbɹi.vi.əˌtɔɹ.i/lang=en"
        }
      ]
    },
    {
      "title": "Adjective",
      "depth": 1,
      "sentences": [
        {
          "text": "# Serving or tending to abbreviate; shortening; abridging.",
          "links": [
            {
              "page": "abbreviate"
            },
            {
              "page": "shortening"
            },
            {
              "page": "abridging"
            }
          ]
        }
      ],
      "templates": [
        {
          "template": "citation",
          "type": "inline",
          "data": {},
          "inline": {
            "text": "{{R:SOED5|page=3}}"
          }
        }
      ]
    },
    {
      "title": "Translations",
      "depth": 2,
      "sentences": [],
      "templates": [
        {
          "template": "t+",
          "data": [
            "nl",
            "afkortend"
          ]
        },
        {
          "template": "t+",
          "data": [
            "nl",
            "verkortend"
          ]
        }
      ],
      "tables": [
        [
          {
            "text": "Dutch: ,"
          },
          {
            "text": "Portuguese: abreviatório, abreviativo"
          }
        ]
      ]
    },
    {
      "title": "References",
      "depth": 1,
      "sentences": []
    }
  ]
}

Here's what strikes me as odd:

• All sections and subsections are returned as a flat array. I would have expected the top-level array to contain only a single section. How do I get only the top-level sections for a document?
• The title property for the first section is empty. Shouldn't that be "English"? In turn, its sentences property contains "text": "==English==".

[spencermountain]

hey Daniel, yeah you're correct about the mis-parsing of the section title. I'll check that out.
you can use the depth property to find sections of a specific level. There are also .children() .parent() and so on methods, please check out the api

I played around with the idea of nested-sections, but it got really messy.
will look at the ==english== issue asap
cheers

[DanielSWolf]

Thanks for the clarification about nested sections. I'm looking forward to the fixed title parsing!

[spencermountain]

this ==English== bug is now fixed in 6.0.0
cheers