SQL基础教程（第2版） ([日] MICK)（Z-Library）

【One-Line Pitch】 A patient, hands-on introduction to SQL and relational databases that takes you from installing PostgreSQL to writing window functions and joining tables. Best for complete beginners, self-taught learners who want systematic grounding, and instructors looking for a ready-made teaching text. 【Book Arc】 - **Opening (~0%–12%)**: Sets up the environment and mental model — what a database is, why a DBMS beats files and spreadsheets, installing PostgreSQL, creating a learning database, and the basic rules of writing SQL (constants, quoting, statement terminators). - **Early (~12%–32%)**: The querying core — SELECT/FROM, aliases, DISTINCT, WHERE, arithmetic and comparison operators, NULL handling, logical operators and three-valued logic, then aggregation (COUNT/SUM/AVG), GROUP BY, HAVING, and ORDER BY. - **Middle (~32%–52%)**: Data manipulation and complex queries — INSERT (including INSERT…SELECT), DELETE/TRUNCATE, UPDATE, then views, subqueries, scalar subqueries, and correlated subqueries with their scoping rules. - **Late (~52% onward)**: Functions, predicates, and CASE expressions, followed by joining multiple tables (inner, outer, cross, three-table joins) and advanced processing such as window functions and GROUPING operators (ROLLUP/CUBE/GROUPING SETS). - **Ending**: Connecting a database to an application, closing the loop from standalone SQL practice to real programmatic use. 【Key Takeaways】 - **A DBMS solves concrete problems files can't** (Opening): concurrent multi-user access, handling millions of rows, automation, and protection against accidents — the motivation for learning SQL at all. - **SELECT is the heart of SQL** (Early): master SELECT/FROM, WHERE, DISTINCT, and aliases and you've cleared the biggest hurdle; everything else builds on this. - **NULL is not zero and not empty** (Early): arithmetic with NULL yields NULL, comparisons with NULL don't return true/false, and you must use IS NULL / IS NOT NULL — a recurring source of beginner bugs. - **SQL logic is three-valued** (Early): conditions evaluate to TRUE, FALSE, or UNKNOWN, and truth tables govern how NOT/AND/OR combine them. - **Aggregation excludes NULL automatically** (Early): SUM/AVG/COUNT ignore NULLs, which is why "NULL in arithmetic" and "NULL in aggregation" behave differently. - **Clause order is fixed** (Early–Middle): SELECT → FROM → WHERE → GROUP BY → HAVING → ORDER BY; aliases defined in SELECT aren't visible to GROUP BY, and ORDER BY belongs at the end. - **Views store queries, not data** (Middle): a view is a saved SELECT that materializes a temporary table on read; it has limits (no ORDER BY in definition, restricted updatability). - **Subqueries come in flavors** (Middle): scalar subqueries fit anywhere a single value fits; correlated subqueries re-execute per row and require the correlation condition inside the subquery, not outside. 【Reading Tips】 - Deep-read Chapters 1–3 (setup, SELECT, aggregation); this is the foundation everything else assumes, and the exercises here pay off most. - Skim the PostgreSQL installation walkthrough if you already have a working database, but don't skip the "create a learning database" step — the book's examples depend on it. - Treat NULL and three-valued logic as a dedicated study topic; re-read those sections and test edge cases yourself rather than trusting intuition. - Work the end-of-chapter exercises before checking the appendix answers; the book is explicitly designed for hands-on practice. - When you reach views, subqueries, and joins, draw the execution order on paper — the book visualizes correlated subquery execution, and replicating that habit helps. 【Coverage Limits】 This guide is synthesized from stratified excerpts covering roughly the first half of the book plus table-of-contents glimpses of later chapters; detailed content on functions, joins, window functions, and application connectivity is only partially represented, so specifics there are inferred from headings rather than full text.