yield用于在函数内部停止其执行临时暂停其执行并将控制权转移给函数的调用方。
首先，你发布的代码并没有产生你所说的错误，因为在编写问题时，你实际上在函数体中缩进了yeild关键字。
但是你的程序将不会按预期运行，因为它是语义错误的。你需要在for循环中进一步缩进yield。

import scrapy

class FlashDealsSpider(scrapy.Spider):
    name = 'flash_deals'
    allowed_domains = ['www.geekbuying.com']
    start_urls = ['https://www.geekbuying.com/deals/categorydeals']

    def parse(self, response):
        items = response.xpath("//div[@class='flash_li']")
        for product in items:
            product_name = product.xpath(".//a[@class='flash_li_link']/text()").get()
            product_sale_price = product.xpath(" .//div[@class='flash_li_price']/span/text()").get()
            product_org_price = product.xpath(".//div[@class='flash_li_price']/del/text()").get()
            product_url = product.xpath(".//a[@class='flash_li_link']/@href").get()
            discount =  product.xpath(".//div[@class='category_li_off']/text()").get()
             yield {
                    'name': product_name,
                    'sale_price': product_sale_price,
                    'orginal_price': product_org_price,
                    'url': product_url,
                    'discount': discount
             }

        next_page = response.xpath("//a[@class='next']/@href").get()
        if next_page:
            yield response.follow(url=next_page, callback=self.parse)