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.

import-trades.ts is the second step in the data pipeline. It reads every trades_page_*.html file that download-trades.ts saved in tmp/, parses the HTML trade tables using regex, and inserts each row as a structured document into the trade-results MongoDB collection. A SHA1 hash derived from each row’s content acts as a unique key so the script is fully idempotent — re-running it never creates duplicates.

​
How It Works

​
Reading HTML Pages

The script scans tmp/ for files matching trades_page_*.html and sorts them numerically (page 1 before page 2, and so on). If no matching files exist the script exits with an error, so always run download-trades.ts first.

​
Parsing Trade Table Rows

Each file’s HTML is walked with a <tr> regex. For every row the script extracts the inner text of each <td>, strips HTML tags, and collapses whitespace. Rows with fewer than 10 cells are skipped (headers, footers, empty rows). The ten fields extracted per row map directly to ITradeDto:
FieldDescription
timeTrade timestamp — parsed from Russian locale text (see below)
symbolISIN code of the security, e.g. UZ7011340005
issuerIssuer name, e.g. <Hamkorbank> ATB
securityTypeSecurity type, e.g. Простые акции (common shares)
marketMarket sector, e.g. STK, BON
platformTrading session, e.g. G1, G2
tradePricePrice per security in the trade
quantityNumber of securities traded
volumeTotal trade value in UZS (tradePrice × quantity)
hashSHA1 deduplication key (generated, not parsed from HTML)

​
Russian Date Parsing

UZSE renders timestamps in Russian locale. The script contains a full month-name lookup table and parses strings in the format: 
17 апреля 2026, 16:02
All twelve Russian genitive month names are handled (января through декабря), producing a UTC Date object for each trade.

​
SHA1 Deduplication

Before inserting a row the script computes a SHA1 hash over the concatenated string: 
symbol | time.toISOString() | tradePrice | quantity | volume | pageIndex | rowIndex | urlKey
where urlKey is derived from the begin, end, and search_key query parameters embedded in the page HTML. This hash is stored as the hash field and is backed by a unique index in MongoDB — any document with an already-seen hash is rejected with error code 11000 (duplicate key) and counted as skipped rather than inserted.

​
Usage

npx tsx scripts/import-trades.ts
The script takes no arguments. It automatically reads all trades_page_*.html files from tmp/.

​
Idempotency

Running import-trades.ts more than once against the same HTML files is safe. The console output at the end of every run reports both counters: 
Done. Inserted: 342, skipped (duplicates): 0
On a re-run over already-imported data: 
Done. Inserted: 0, skipped (duplicates): 342
After all pages have been processed, the script automatically deletes every .html file from tmp/ and prints Tmp cleaned. — so you will not be able to re-import from the same downloaded files without re-running download-trades.ts.

​
MongoDB Collection

Imported trade documents are stored in the trade-results collection. Each document has the following structure (from Trade.schema.ts): 
{
  time:         Date,    // trade timestamp
  symbol:       string,  // ISIN code
  issuer:       string,  // issuer name
  securityType: string,  // e.g. "Простые акции"
  market:       string,  // e.g. "STK"
  platform:     string,  // e.g. "G1"
  tradePrice:   number,
  quantity:     number,
  volume:       number,  // UZS
  hash:         string,  // unique SHA1 key
}
Indexed fields: time, symbol, and a compound { symbol, time } index for efficient range queries. The hash field carries a unique index.
Always run download-trades.ts immediately before import-trades.ts for the same date range. The batch scripts (fetch.sh / fetch.bat) pair every download with an import automatically, which is the recommended workflow for large history imports.

​
Next Step

Build OHLCV Candlesticks

Aggregate all imported trades into continuous OHLCV candlestick series across 11 timeframes, with automatic gap-filling for intraday minutes and non-trading days.

Build docs developers (and LLMs) love

Get started for freeTalk to us