Parsing file formats, databases, data mining, and enrichment.
Splitting a MySQL dump by size corrupts it. Measured on a 560 MB dump in August 2026, with the chunk that half-imports, a splitter that does not break statements, current tool status and prices, and timings for six import routes.
URL parsing across two incompatible standards, RFC 3986 and the WHATWG URL Standard: the same inputs run through urllib.parse, yarl, furl, Node and PHP on 13 August 2026, with measured throughput and the crawler consequences.
Load web scraping results into SQL Server: row-by-row INSERT, fast_executemany, BULK INSERT, SqlBulkCopy, SSIS and the new mssql-python bulkcopy, plus the 2,100-parameter and 1,000-row ceilings that break loads. Checked against Microsoft docs on 13 August 2026.
What data mining is, how the CRISP-DM cycle really runs, the core techniques, and what the tools cost: versions and prices read from vendor pages on 13 August 2026, plus a timed apriori-versus-FP-growth run.
XML parsing in Python, Node, PHP, Java and Go, rechecked August 2026: namespace traps that return nothing, DOM versus streaming memory measured on a 32 MB feed, and which parsers still open external entities by default.