How to use the fathom-web.dom function in fathom-web

// Initial tests of this are pretty innaccurate; lots to learn to be able to tweak the rules and use it well
const rules = ruleset(
    rule(
        dom('p,div,li,blockquote,h1,h2,h3,h4,h5,h6'),
        props(scoreByLength).type('paragraphish'),
    ),
    rule(
        type('paragraphish'),
        score(fnode => {
            const paragraphishNote = fnode.noteFor('paragraphish')
            return paragraphishNote
                ? (1 - linkDensity(fnode, paragraphishNote.inlineLength)) * 1.5
                : (1 - linkDensity(fnode)) * 1.5
        }),
    ),
    rule(dom('p'), score(4.5).type('paragraphish')),
    rule(
        type('paragraphish').bestCluster({
            splittingDistance: 3,
            differentDepthCost: 6.5,
            differentTagCost: 2,
            sameTagCost: 0.5,
            strideCost: 0,
        }),
        out('content').allThrough(domSort),
    ),
)

export default rules

/**
       * Title rules
       */
      // consider all eligible h1 elements
      rule(dom('h1').when(this.isEligibleTitle.bind(this)), type('title')),
      // better score based on y-axis proximity to max scoring image element
      rule(type('title'), score(this.isNearImageTopOrBottom.bind(this)), {name: 'isNearImageTopOrBottom'}),
      // return title element(s) with max score
      rule(type('title').max(), out('title')),

      /**
       * Price rules
       */
      // 72% by itself, at [4, 4, 4, 4...]!:
      // consider all eligible span and h2 elements
      rule(dom('span, h2').when(this.isEligiblePrice.bind(this)), type('price')),
      // check if the element has a '$' in its innerText
      rule(type('price'), score(this.hasDollarSign.bind(this)), {name: 'hasDollarSign'}),
      // better score the closer the element is to the top of the page
      rule(type('price'), score(this.isAboveTheFold.bind(this)), {name: 'isAboveTheFoldPrice'}),
      // check if the id has "price" in it
      rule(type('price'), score(this.hasPriceInID.bind(this)), {name: 'hasPriceInID'}),
      rule(type('price'), score(this.hasPriceInParentID.bind(this)), {name: 'hasPriceInParentID'}),
      // check if any class names have "price" in them
      rule(type('price'), score(this.hasPriceInClassName.bind(this)), {name: 'hasPriceInClassName'}),
      rule(type('price'), score(this.hasPriceInParentClassName.bind(this)), {name: 'hasPriceInParentClassName'}),
      // better score for larger font size
      rule(type('price'), score(this.fontIsBig.bind(this)), {name: 'fontIsBig'}),
      // better score based on x-axis proximity to max scoring image element
      rule(type('price'), score(this.isNearImage.bind(this)), {name: 'isNearImage'}),
      // check if innerText has a price pattern
      rule(type('price'), score(this.hasPriceishPattern.bind(this)), {name: 'hasPriceishPattern'}),

const hasAncestor = (tagName, scoreIfHas) => fnode => {
    const lowerTag = tagName.toLowerCase();
    for (let element = fnode.element, parent;
        (parent = element.parentNode) != null &&
            parent.nodeType === parent.ELEMENT_NODE;
        element = parent) {
        if (element.tagName.toLowerCase() === lowerTag)
            return scoreIfHas;
    }
    return 1;
};

const rules = ruleset(
    // Isolate the actual blog post body text. Based on Fathom's example
    // Readability rules
    rule(dom('p,li,ol,ul,code,blockquote,pre,h1,h2,h3,h4,h5,h6'),
        props(scoreByLength).type('paragraphish')),
    rule(type('paragraphish'), score(byInverseLinkDensity)),
    rule(dom('p'), score(4.5).type('paragraphish')),

    // Tweaks for this particular blog
    rule(type('paragraphish'), score(hasAncestor('article', 10))),
    rule(dom('.entry-summary p'), score(0).type('paragraphish')),
    rule(dom('figure'), props(scoreByImageSize).type('paragraphish')),
    rule(dom('.jetpack-video-wrapper'), props(() => ({
        score: 100,
        note: {length: 1},
    })).type('paragraphish')),

    // Find the best cluster of paragraph-ish nodes
    rule(
        type('paragraphish').bestCluster({

// bonus for non-extreme aspect ratios, to filter out banners or nav elements
      // TODO: Meant to make this a penalty, but it turns out to work as is.
      // Try as a penalty.
      rule(type('image'), score(this.hasSquareAspectRatio.bind(this)), {name: 'hasSquareAspectRatio'}),
      // no background images, even ones that have reasonable aspect ratios
      // TODO: If necessary, also look at parents. I've seen them say
      // "background" in their IDs as well.
      rule(type('image'), score(this.hasBackgroundInID.bind(this)), {name: 'hasBackgroundInID'}),
      // return image element(s) with max score
      rule(type('image').max(), out('image')),

      /**
       * Title rules
       */
      // consider all eligible h1 elements
      rule(dom('h1').when(this.isEligibleTitle.bind(this)), type('title')),
      // better score based on y-axis proximity to max scoring image element
      rule(type('title'), score(this.isNearImageTopOrBottom.bind(this)), {name: 'isNearImageTopOrBottom'}),
      // return title element(s) with max score
      rule(type('title').max(), out('title')),

      /**
       * Price rules
       */
      // 72% by itself, at [4, 4, 4, 4...]!:
      // consider all eligible span and h2 elements
      rule(dom('span, h2').when(this.isEligiblePrice.bind(this)), type('price')),
      // check if the element has a '$' in its innerText
      rule(type('price'), score(this.hasDollarSign.bind(this)), {name: 'hasDollarSign'}),
      // better score the closer the element is to the top of the page
      rule(type('price'), score(this.isAboveTheFold.bind(this)), {name: 'isAboveTheFoldPrice'}),
      // check if the id has "price" in it

const hasAncestor = (tagName, scoreIfHas) => fnode => {
    const lowerTag = tagName.toLowerCase();
    for (let element = fnode.element, parent;
        (parent = element.parentNode) != null &&
            parent.nodeType === parent.ELEMENT_NODE;
        element = parent) {
        if (element.tagName.toLowerCase() === lowerTag)
            return scoreIfHas;
    }
    return 1;
};

const rules = ruleset(
    // Isolate the actual blog post body text. Based on Fathom's example
    // Readability rules
    rule(dom('p,li,ol,ul,code,blockquote,pre,h1,h2,h3,h4,h5,h6'),
        props(scoreByLength).type('paragraphish')),
    rule(type('paragraphish'), score(byInverseLinkDensity)),
    rule(dom('p'), score(4.5).type('paragraphish')),

    // Tweaks for this particular blog
    rule(type('paragraphish'), score(hasAncestor('article', 10))),
    rule(dom('.entry-summary p'), score(0).type('paragraphish')),
    rule(dom('figure'), props(scoreByImageSize).type('paragraphish')),

    // Find the best cluster of paragraph-ish nodes
    rule(
        type('paragraphish').bestCluster({
            splittingDistance: 3,
            differentDepthCost: 6.5,
            differentTagCost: 2,
            sameTagCost: 0.5,