CLI output name and actual output name do not match

[kkristof200]

amazon-buddy products -k 'fixie bike' -n 100 --random-ua true --sort true --min-rating 4.5 --filetype json Outputs -> Result was saved to: products(fixie bike)_1593862189487

Actual file name: products(fixie bike)_1593862189473.json

There seems to be an inconsistency at the timestamps used for the file name.

[kkristof200]

get fileName() {
    switch (this._scrapeType) {
        case 'products':
            return `${this._scrapeType}(${this._keyword})_${Date.now()}`;
        case 'reviews':
        case 'asin':
            return `${this._scrapeType}(${this._asin})_${Date.now()}`;
        default:
            throw new Error(`Unknow scraping type: ${this._scrapeType}`);
    }
}

This method returns a different name (different timestamp since 'Date.now()' is called). So it won't be consistent during a session.

I'd add

this._fileName = null;

to the constructor and convert the 'get fileName()' method to

get fileName() {
    if (this._fileName != null) {
        return this._fileName;
    }

    switch (this._scrapeType) {
        case 'products':
            this._fileName = `${this._scrapeType}(${this._keyword})_${Date.now()}`;
        case 'reviews':
        case 'asin':
            this._fileName = `${this._scrapeType}(${this._asin})_${Date.now()}`;
        default:
            throw new Error(`Unknow scraping type: ${this._scrapeType}`);
    }

    return this.fileName;
}

If lazy inits/vars are available in js, then it would solve the issue without extra created variables. I'm not very familiar with js so the syntax might not be on point.

[kkristof200]

other solution would be to create a date object on init, like

this.sessionStartTs = Date.now();

and rewrite 'get fileName()' like this

get fileName() {
    switch (this._scrapeType) {
        case 'products':
            return `${this._scrapeType}(${this._keyword})_${this.sessionStartTs}`;
        case 'reviews':
        case 'asin':
            return `${this._scrapeType}(${this._asin})_${this.sessionStartTs}`;
        default:
            throw new Error(`Unknow scraping type: ${this._scrapeType}`);
    }
}

[kkristof200]

All of the above would work only if a new 'AmazonScraper' is created for every new scrape. If that not is the case, the fileName, or timestamp needs to be reset at every new scrape call.

[drawrowfly]

fixed in latest update