Metadata-Version: 2.1
Name: stream-sqlite
Version: 0.0.0
Summary: Python function to extract all the rows from a SQLite database, without loading the entire file into memory or disk
Home-page: https://github.com/uktrade/stream-sqlite
Author: Department for International Trade
Author-email: sre@digital.trade.gov.uk
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Topic :: Database
Requires-Python: >=3.5.0
Description-Content-Type: text/markdown
License-File: LICENSE

# stream-sqlite

It can read a table from sqlite page(s)
Hardcoded database path for convenience

Missing:
Parsing sql in sqlite_master to identify the table name and the starting page
no checks for consistency between record size and space allocated to the record
record overflow
handling of NO ROWID tables. For the moment, I always output the row index at the beginning of the record
TESTS


Inefficient:
use recursion to read a page
several useless seek and read in the routines



Wrong:
reading of varint using more than 2 bytes (?)
use hardcoded page size

