从使用多个异步调用的函数返回

Question

So the basic idea was to write a method that will scrap webpage to get JSON data that contains rating of a product. Then call this method multipletimes over few domains (.de, .uk, .fr, .nl etc) to collect all ratings.

So I ended up with scrapWebPage method which scraps single page:

const scrapWebPage = async (countryAppData, productNumber) => {
    const shopUrl = `https://www.shopExample.${countryAppData.countryCode}/?q=${productNumber}`
    const avoidCORSUrl = 'https://allorigins.me/get?url=' + shopUrl + '&callback=?'

    return await axios
        .get(avoidCORSUrl, {xmlMode: false, normalizeWhitespace: true})
        .then(response => {
            const $ = cheerio.load(response.data)
            let scrapedWebPageJson

            contentForParsing = $("script").get().children[0].data                   
            scrapedWebPageJson = JSON.parse(contentForParsing)

            return scrapedWebPageJson
        })
}

scrapWebPage also contains some parsing to return some JSON data I want - it resolves correctly (tested this) and returns Promise.

But then I'd like to call this method over multiple domains so I created getProductDataFromManyDomains :

const getProductDataFromManyDomains = (productNum) => {
    let prodData = {
        reviews: []
    }

    const appCountries = [
        {countryCode: 'nl'}, 
        {countryCode: 'pl'},
        {countryCode: 'de'}
    ]

    appCountries.forEach(async countryApp => {
        let countryData = {}

        let parsedWebPage = await scrapWebPage(countryApp, productNum)

        countryData.countryCode  = countryApp.countryCode
        countryData.ratingCount  = parsedWebPage.aggregateRating.ratingCount
        countryData.ratingValue  = parsedWebPage.aggregateRating.ratingValue
        countryData.reviews      = parsedWebPage.reviews   

        prodData.reviews.push(countryData)
    })

    return prodData
}

And now I receive prodData before populating... while I'd like to receive actual data (populated prodData ).

I'm not sure how I should construct this getProductDataFromManyDomains method to actually return data and not prodData before populating. Is that possible? Or what is good pattern here to deal with stuff like that?

Answer 1

Use a for loop instead of .forEach() . The for loop will pause for await, the .forEach() loop will not. This is because the async callback you pass to .forEach() will return a promise, but .forEach() is not designed to do anything with that promise so it does not wait for it to resolve before continuing the loop, but a for loop using await does.

Then, getProductDataFromManyDomains() will need to be async and will return a promise with your final result.

async function getProductDataFromManyDomains(productNum) {
    let prodData = {
        reviews: []
    }

    const appCountries = [
        {countryCode: 'nl'}, 
        {countryCode: 'pl'},
        {countryCode: 'de'}
    ]

    for (let countryApp of appCountries) {
        let countryData = {}

        let parsedWebPage = await scrapWebPage(countryApp, productNum)

        countryData.countryCode  = countryApp.countryCode
        countryData.ratingCount  = parsedWebPage.aggregateRating.ratingCount
        countryData.ratingValue  = parsedWebPage.aggregateRating.ratingValue
        countryData.reviews      = parsedWebPage.reviews   

        prodData.reviews.push(countryData)
    })

    // this will be the resolved value of the promise that
    //   getProductDataFromManyDomains() returns
    return prodData;
}

// usage
getProductDataFromManyDomains(productNum).then(result => {
    console.log(result);
});

You could also run your multiple requests in parallel rather than one at a time, but since you originally attempted to make your code do them one at a time, I showed you how to do that.

If you wanted to do them in parallel, you would just accumulate the promises in an array and use Promise.all() to know when they are all done and you would not await the request.

Here's a version of the code that runs the requests in parallel, using .map() and Promise.all() :

function getProductDataFromManyDomains(productNum) {
    let prodData = {
        reviews: []
    }

    const appCountries = [
        {countryCode: 'nl'}, 
        {countryCode: 'pl'},
        {countryCode: 'de'}
    ]

    return Promise.all(appCounteries.map(countryApp => {

        return scrapWebPage(countryApp, productNum).then(parsedWebPage => {
            let countryData = {}
            countryData.countryCode  = countryApp.countryCode
            countryData.ratingCount  = parsedWebPage.aggregateRating.ratingCount
            countryData.ratingValue  = parsedWebPage.aggregateRating.ratingValue
            countryData.reviews      = parsedWebPage.reviews 
            return countryData;         
        });
    })).then(results => {
        // put results into prodData and make that the resolved value
        prodData.reviews = results;
        return prodData;
    });
}

getProductDataFromManyDomains(productNum).then(result => {
    console.log(result);
});