1688 Excel Scraper Tool
₹12500-37500 INR
Posted 1 minute ago
•
Ends in 6 days
₹12500-37500 INR
Paid on delivery
I need a small, no-cost scraper that my data-entry colleagues can run without touching code. Their task will be to upload a single Excel file that lists one or more [login to view URL] product URLs. When they press “Run”, the tool must create a fresh workbook where every product colour gets its own row. For each colour please capture the full listing details: title, current price, every available size with its stock count, weight, the live product link itself and every gallery-image URL. All values must line up with the column order in the sample file I will provide. Robust error handling is critical. If a page turns up blank, blocked or otherwise unreadable, the script should retry automatically; if the second attempt still fails, mark that row as failed and record the exact reason so we can review it later. Technical notes • Python is preferred, using Selenium, Playwright or another free library. • Package the script as a simple double-clickable .exe or light GUI so non-technical staff can operate it. • Make sure the scraper copes with dynamically loaded content, lazy images and basic anti-bot checks on 1688. Deliverables • Clean, well-commented Python source code plus a ready-to-run executable. • User guide (one-page PDF or README) that shows the steps for uploading the URL file and receiving the output. • A demonstration run on fifty sample products to confirm formatting and error logging. • Brief notes on updating selectors if 1688 changes its layout. Once the tool meets these requirements I will test it with a larger batch of links and sign off.
Project ID: 40812777
About the project
12 proposals
Open for bidding
Remote project
Active 2 secs ago
Place your bid
Bid amount Email addressBenefits of bidding on Freelancer
Set your budget and timeframe
It's free to sign up and bid on jobs
12 freelancers are bidding on average ₹21,128 INR for this job


Your 1688 colour-row scraper for non-technical data-entry staff — here is exactly how I will build it. Playwright will handle the scraping because 1688 leans heavily on JavaScript rendering and lazy-loaded gallery images. Selenium struggles with both on this platform, so Playwright with realistic scroll timing is the right call for getting live image URLs to actually appear before capture. For every colour variant I will extract title, current price, sizes with stock counts, weight, live product link, and all gallery image URLs, written row by row into a fresh workbook matching your sample file's column order exactly. On any blank or blocked page the script retries once automatically. If the second attempt still fails, it marks that row with the exact failure reason so your team can review it without guessing — your words, not mine. One problem worth flagging that your brief does not cover: 1688 serves different page structures for logged-in versus guest sessions. I will build in a cookie-injection option so the tool can run authenticated if guest access gets throttled on larger batches. Deliverables: clean commented source code, a double-clickable exe requiring no Python installation, a one-page user guide, a demo on 50 products, and notes on updating selectors if 1688 changes layout. All within 3 days. Does your sample output file already have fixed column headers I must match exactly, or should I propose the order?


Hi, I will deliver a double‑clickable .exe that reads an Excel of 1688 URLs, outputs a new workbook with a row per colour capturing title, price, sizes, stock, weight, link and image URLs, and logs retries. I can send a free sample run now


Hello, I'd package this as a small Windows app your data-entry team can double-click, drop in one Excel of 1688 product links, hit Run, and get a fresh workbook back with one row per colour. No code on their side. Technically I'd lean on Playwright (or Selenium if you prefer) so the page actually finishes loading before we grab anything. That covers dynamic content, lazy gallery images, and the usual light anti-bot friction on 1688. For each colour we pull title, current price, every size with stock, weight, the live link, and every gallery image URL, columns matching the sample file you send. If a page comes back blank, blocked, or unreadable, the tool retries once on its own. Still failing? That row is marked failed with the exact reason so you can review later without digging through logs. Deliverables: clean commented Python, a ready .exe, a one-page user guide, a demo run on fifty sample products, and short notes on updating selectors if 1688 changes layout. Once that looks solid on your end, you can throw a bigger batch at it for sign-off.


The part that'll actually break this tool over time isn't the scraping logic — it's 1688's dynamically loaded content and anti-bot checks, which means Selenium/Playwright needs proper wait conditions (not fixed sleeps) for lazy-loaded images and price/stock data, plus retry logic that distinguishes "page genuinely blocked" from "just loaded slowly." My approach: Playwright (better at handling dynamic content and slightly more resilient to basic bot detection than plain Selenium) reading the uploaded Excel URL list, scraping each listing with explicit waits for color/size/stock elements, writing one row per color variant matching your sample column order exactly. Failed pages get one automatic retry; if that fails too, the row gets marked failed with the specific error reason (timeout, blocked, element not found) rather than a generic failure flag, so your team can triage efficiently. Packaged as a standalone .exe (PyInstaller) with a minimal GUI — upload file, click Run, get output workbook — no Python installation needed on their end. I'd include notes on which CSS selectors to update if 1688 changes its layout, since that's the main long-term maintenance point for this kind of tool. Demo run on 50 sample products to confirm formatting/error logging before you sign off, as specified.
内容来自 Freelancer News,版权归原作者所有。
阅读原文 ↗