Skip to content

Latest commit

 

History

History
77 lines (51 loc) · 3 KB

tvtvropes.md

File metadata and controls

77 lines (51 loc) · 3 KB

TV Tropes

Unfortnately, there're no public APIs avaiable in TVTropes, https://tvtropes.org/pmwiki/posts.php?discussion=14313163830A95025400, https://tvtropes.org/pmwiki/posts.php?discussion=14478420050A97600200. So we need to get contents in other way.

Basic page API

https://tvtropes.org/pmwiki/pmwiki.php/{GROUPNAME}/{TITLE}

For example,

  • https://tvtropes.org/pmwiki/pmwiki.php/Literature/BlackDaggerBrotherhood will be { GROUPNAME: "Literature", TITLE: "BlackDaggerBrotherhood" }
  • https://tvtropes.org/pmwiki/pmwiki.php/Main/ClipboardOfAuthority will be { GROUPNAME: "Main", TITLE: "ClipboardOfAuthority" }
  • https://tvtropes.org/pmwiki/pmwiki.php/SoYouWantTo/BreakTheCutie will be { GROUPNAME: "SoYouWantTo", TITLE: "BreakTheCutie" }
  • https://tvtropes.org/pmwiki/pmwiki.php/VideoExamples/Hell will be { GROUPNAME: "VideoExamples", TITLE: "Hell" }

Get content of a page

https://tvtropes.org/pmwiki/context.php?groupname={GROUPNAME}&title={TITLE}

For example, the Literature work, Black Dagger Brotherhood will be: https://tvtropes.org/pmwiki/context.php?groupname=Literature&title=BlackDaggerBrotherhood

Exceptions:

VideoExamples is not available here.

https://tvtropes.org/pmwiki/context.php?groupname=VideoExamples&title=Hell

Fornately we can use document.querySelectorAll("._pmvv-foot-scrollwrapper a") here. And dataset can get all of infomations!

Get search list: Google search

https://cse.google.com/cse/element/v1?&num={PAGE}&start={START}&cx={CX}&cse_tok={TOKEN}&callback={CALLBACK}&q={QUERY}&safe={SAFE}

  • PAGE: How many the page shows here. 10 in the site.
  • START: The article searching action start with. 0, 10, or 20...
  • CALLBACK: Which callback function should start? Google use something like google.search.cse.api3879. But can we use JSON.parse?
  • QUERY: The keyword you are searching for.
  • SAFE: I guess it's safe search. So let's off.
  • TOKEN: https://cse.google.com/cse/cse.js?cx={CX} -- Then find "cse_token": "...", in the file.
  • CX: We are asking for something in #search-box...
<div id="search-box">
    <form class="search" action="/pmwiki/search_result.php">
        <input type="hidden" name="cx" value="...">
</form>

Steps

So we firstly need:

CX: document.querySelector("input[name='cx']").value

then we request: https://cse.google.com/cse/cse.js?cx={CX}

And texts.match(/"cse_token":\s*"([^"]+)"/);.

Now we get {TOKEN} and we can combine other scripts.

Get discussion

https://tvtropes.org/pmwiki/posts.php?discussion={ID}

Notes: Get JSON

const fn = (responseText) => {
    const jsonRegex = /\{.*\}/s;
    const jsonMatch = responseText.match(jsonRegex);
    if (jsonMatch) {
        const jsonObject = jsonMatch[0].replace(/,\s*([}\]])/g, "$1"); // remove last comma
        return JSON.parse(jsonObject);
    }
    return {};
}