* Add incremental SQLite history SDK for automated pipelines. Extract sqlite history helpers into a dedicated module and expose update_history APIs that resume from existing MAX(time) values instead of re-fetching fixed date ranges. Co-authored-by: Cursor <cursoragent@cursor.com> * Fix incremental history deals and stale rate view cleanup. Fetch account events once during incremental updates, drop stale rate_* views when timeframes change, and avoid SQLite variable limits on wide frames. Co-authored-by: Cursor <cursoragent@cursor.com> * Fix incremental deal filtering edge cases Co-authored-by: Cursor <cursoragent@cursor.com> * Address PR review feedback for incremental SQLite history. Make rate views collision-free, batch incremental resume queries, scope deduplication to appended boundaries, validate before opening MT5, use atomic SQLite transactions, and expand docs/tests for the new helpers. Co-authored-by: Cursor <cursoragent@cursor.com> * Document collect-history SQLite schema with ER diagram. Co-authored-by: Cursor <cursoragent@cursor.com> * Fix account-event filtering and drop legacy rates resume. Account events must follow only account_event_start, not per-symbol trade cursors. Require normalized rates schema and fail fast when timeframe is missing. Co-authored-by: Cursor <cursoragent@cursor.com> * Validate normalized rates schema before incremental resume. Require symbol, timeframe, and time on existing rates tables with clear ValueError messages, and add regression tests for malformed schemas. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com>
4.5 KiB
SQLite History Module
::: mt5cli.sqlite_history
collect-history schema
The collect-history command (and the matching collect_history SDK function) writes
selected MT5 datasets into one SQLite database. Each dataset becomes a table; column
names and types mirror the pdmt5 DataFrame schema for that export, with two additions:
symbolis prepended on every table.timeframeis prepended onratesso appended runs at different bar sizes stay distinguishable.
SQLite does not declare foreign keys. Rows are linked logically by symbol, time
windows, and (for deals) position_id / order. Duplicate rows are removed on
append using dataset-specific keys (for example ticket on history tables, or
(symbol, timeframe, time) on rates).
Optional views are created when --with-views is set and the history-deals dataset
was written.
Entity-relationship diagram
Sample layout for a full collection with --with-views:
erDiagram
rates {
TEXT symbol "dedup key"
INTEGER timeframe "dedup key"
TEXT time "dedup key"
REAL open
REAL high
REAL low
REAL close
INTEGER tick_volume
INTEGER spread
INTEGER real_volume
}
ticks {
TEXT symbol "dedup key"
TEXT time "dedup key"
INTEGER time_msc "dedup key (preferred)"
REAL bid
REAL ask
REAL last
INTEGER volume
INTEGER flags
REAL volume_real
}
history_orders {
INTEGER ticket "dedup key"
TEXT symbol
TEXT time
INTEGER type
INTEGER state
REAL volume_initial
REAL price_open
REAL price_current
INTEGER magic
}
history_deals {
INTEGER ticket "dedup key"
INTEGER order
INTEGER position_id "groups position view"
TEXT symbol
TEXT time
INTEGER type "0/1 trade, else cash event"
INTEGER entry "0 IN, 1 OUT, 2 INOUT, 3 OUT_BY"
REAL volume
REAL price
REAL profit
REAL commission
REAL swap
REAL fee
}
cash_events {
INTEGER ticket
TEXT symbol
TEXT time
INTEGER type
REAL profit
}
positions_reconstructed {
INTEGER position_id
TEXT symbol
TEXT open_time
TEXT close_time
INTEGER direction
REAL volume_open
REAL volume_close
REAL volume_reversal
REAL open_price
REAL close_price
REAL total_profit
INTEGER reversal_count
INTEGER deals_count
}
rates ||--o{ history_deals : "symbol (logical)"
ticks ||--o{ history_deals : "symbol (logical)"
history_orders ||--o{ history_deals : "order ~ ticket (logical)"
history_deals ||--|| cash_events : "VIEW: type NOT IN (0,1)"
history_deals ||--o{ positions_reconstructed : "VIEW: GROUP BY position_id"
Tables and views
| Object | Kind | Source | Notes |
|---|---|---|---|
rates |
table | copy_rates_range |
Indexed on (symbol, timeframe, time) when columns exist. |
ticks |
table | copy_ticks_range |
Indexed on (symbol, time) when columns exist. |
history_orders |
table | history_orders_get |
Fetched per --symbol, then concatenated. |
history_deals |
table | history_deals_get |
Fetched per --symbol, then concatenated. Indexed on (position_id, symbol) when present. |
cash_events |
view | history_deals |
Non-trade deal types (deposits, balance ops, etc.). Requires type column. |
positions_reconstructed |
view | history_deals |
One row per closed position_id; volume-weighted prices and reversal stats. |
Column sets can vary with terminal and pdmt5 version. Views are skipped with a warning when required columns are missing.
Incremental collection
The update_history SDK path uses the same base tables and optional
cash_events / positions_reconstructed views. It additionally maintains
rate_<symbol>__<timeframe> compatibility views when create_rate_views=True.