Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/backtest-kit/uzse-backtest-app/llms.txt

Use this file to discover all available pages before exploring further.

download-trades.ts is the first step in the data pipeline. It launches a headless Chromium browser via Playwright, navigates to the UZSE trade results page at uzse.uz/trade_results, and iterates through every paginated result for a given symbol and date range — saving each page as a raw HTML file in the local tmp/ directory, ready for import.

​
How It Works

The script uses Playwright’s Chromium engine in headless mode so it can execute the JavaScript-rendered UZSE trade results page just as a real browser would. Pagination detection is handled by getLastPage(), which scans the first page’s HTML for a class="last next" anchor tag and reads its page= query parameter. This tells the script exactly how many pages to fetch before stopping. Retry logic is built into every page fetch. If a navigation times out (the networkidle wait exceeds 30 seconds), the script warns and retries automatically:
  • Retries: 10 attempts per page
  • Retry delay: 30 seconds between attempts
  • Timeout per attempt: 30 seconds
Tmp directory management: at start-up the script creates tmp/ if it does not exist and clears any pre-existing .html files from a previous run before fetching fresh pages.

​
Usage

npx tsx scripts/download-trades.ts <symbol> <begin> <end> [mktId]

​
Parameters

​
symbol
string
required
The ISIN code of the security to download trades for.
Example: UZ7011340005
​
begin
string
required
Start date of the query window in DD.MM.YYYY format.
Example: 01.03.2026
​
end
string
required
End date of the query window in DD.MM.YYYY format.
Example: 31.03.2026
​
mktId
string
default:"STK"
Market sector filter. Defaults to STK (equities) when omitted.
Common values: STK (stocks), BON (bonds).

​
Examples

Download all trades for March 2026: 
npx tsx scripts/download-trades.ts UZ7011340005 01.03.2026 31.03.2026
Download a two-day window with the market sector made explicit: 
npx tsx scripts/download-trades.ts UZ7011340005 17.04.2026 18.04.2026 STK

​
Output

Each paginated result is written to the tmp/ directory as a numbered HTML file: 
tmp/
  trades_page_1.html
  trades_page_2.html
  trades_page_3.html
  ...
The script prints the total page count when it begins and logs progress for every page beyond the first: 
Total pages: 14
Downloaded page 2/14
Downloaded page 3/14
...
Done. HTML saved to /path/to/tmp
The tmp/ directory is wiped of all .html files at the start of every run and again automatically after import-trades.ts has finished processing them. Never store files you want to keep inside tmp/.
Wide date ranges can produce dozens or hundreds of pages. The script retries each page up to 10 times with a 30-second delay on timeout, so a slow or rate-limited connection can make the process take considerably longer than expected. Prefer month-by-month date ranges (as the batch scripts do) to keep runs manageable.

​
Next Step

Import Trades into MongoDB

Parse the downloaded HTML files, extract structured trade records, and insert them into MongoDB with SHA1-based deduplication.

Build docs developers (and LLMs) love

Get started for freeTalk to us