PostgreSQL 手册
简体中文(ZH-CN) English(EN) 繁体中文(ZH-TW) 日本語(JA) 한국어(KO) Melayu(MS) Français(FR) Deutsch(DE)
directory search
前言 何为PostgreSQL？ PostgreSQL简史 格式约定 更多信息 臭虫汇报指导 I. 教程 章1. 从头开始 1.1. 安装 1.2. 体系基本概念 1.3. 创建一个数据库 1.4. 访问数据库 章2. SQL语言 2.1. 介绍 2.2. 概念 2.3. 创建新表 2.4. 向表中添加行 2.5. 查询一个表 2.6. 表间链接 2.7. 聚集函数 2.8. 更新 2.9. 删除 章3. 高级特性 3.1. 介绍 3.2. 视图 3.3. 外键 3.4. 事务 3.5. 窗口函数 3.6. 继承 3.7. 结论 II. SQL语言 章4. SQL语法 4.1. 词法结构 4.2. 值表达式 4.3. 调用函数 章5. 数据定义 5.1. 表的基本概念 5.2. 缺省值 5.3. 约束 5.4. 系统字段 5.5. 修改表 5.6. 权限 5.7. 模式 5.8. 继承 5.9. 分区 5.10. 其它数据库对象 5.11. 依赖性跟踪 章 6. 数据操作 6.1. 插入数据 6.2. 更新数据 6.3. 删除数据 章7. 查询 7.1. 概述 7.2. 表表达式 7.3. 选择列表 7.4. 组合查询 7.5. 行排序 7.6. LIMIT和OFFSET 7.7. VALUES列表 7.8. WITH的查询（公用表表达式） 章8. 数据类型 8.1. 数值类型 8.2. 货币类型 8.3. 字符类型 8.4. 二进制数据类型 8.5. 日期/时间类型 8.6. 布尔类型 8.7. 枚举类型 8.8. 几何类型 8.9. 网络地址类型 8.10. 位串类型 8.11. 文本搜索类型 8.12. UUID类型 8.13. XML类型 8.14. 数组 8.15. 复合类型 8.16. 对象标识符类型 8.17. 伪类型 章 9. 函数和操作符 9.1. 逻辑操作符 9.2. 比较操作符 9.3. 数学函数和操作符 9.4. 字符串函数和操作符 9.5. 二进制字符串函数和操作符 9.6. 位串函数和操作符 9.7. 模式匹配 9.8. 数据类型格式化函数 9.9. 时间/日期函数和操作符 9.10. 支持枚举函数 9.11. 几何函数和操作符 9.12. 网络地址函数和操作符 9.13. 文本检索函数和操作符 9.14. XML函数 9.15. 序列操作函数 9.16. 条件表达式 9.17. 数组函数和操作符 9.18. 聚合函数 9.19. 窗口函数 9.20. 子查询表达式 9.21. 行和数组比较 9.22. 返回集合的函数 9.23. 系统信息函数 9.24. 系统管理函数 9.25. 触发器函数 章10. 类型转换 10.3. 函数 10.2. 操作符 10.1. 概述 10.4. 值存储 10.5. UNION 章11. 索引 11.1. 介绍 11.2. 索引类型 11.3. 多字段索引 11.4. 索引和ORDER BY 11.5. 组合多个索引 11.6. 唯一索引 11.7. 表达式上的索引 11.8. 部分索引 11.9. 操作类和操作簇 11.10. 检查索引的使用 章12. Full Text Search 12.1. Introduction 12.2. Tables and Indexes 12.3. Controlling Text Search 12.4. Additional Features 12.5. Parsers 12.6. Dictionaries 12.7. Configuration Example 12.8. Testing and Debugging Text Search 12.9. GiST and GIN Index Types 12.10. psql Support 12.11. Limitations 12.12. Migration from Pre-8.3 Text Search 章13. 并发控制 13.1. 介绍 13.2. 事务隔离 13.3. 明确锁定 13.4. 应用层数据完整性检查 13.5. 锁和索引 章14. 性能提升技巧 14.1. 使用EXPLAIN 14.2. 规划器使用的统计信息 14.3. 用明确的JOIN语句控制规划器 14.4. 向数据库中添加记录 14.5. 非持久性设置 III. 服务器管理 章15. 安装指导 15.1. 简版 15.2. 要求 15.3. 获取源码 15.4. 升级 15.5. 安装过程 15.6. 安装后的设置 15.7. 支持的平台 15.8. 特殊平台的要求 章16. Installation from Source Code on Windows 16.1. Building with Visual C++ or the Platform SDK 16.2. Building libpq with Visual C++ or Borland C++ 章17. 服务器安装和操作 17.1. PostgreSQL用户帐户 17.2. 创建数据库集群 17.3. 启动数据库服务器 17.4. 管理内核资源 17.5. 关闭服务 17.6. 防止服务器欺骗 17.7. 加密选项 17.8. 用SSL进行安全的TCP/IP连接 17.9. Secure TCP/IP Connections with SSH Tunnels 章18. 服务器配置 18.1. 设置参数 18.2. 文件位置 18.3. 连接和认证 18.4. 资源消耗 18.5. 预写式日志 18.6. 查询规划 18.7. 错误报告和日志 18.8. 运行时统计 18.9. 自动清理 18.10. 客户端连接缺省 18.12. 版本和平台兼容性 18.11. 锁管理 18.13. 预置选项 18.14. 自定义的选项 18.15. 开发人员选项 18.16. 短选项 章19. 用户认证 19.1. pg_hba.conf 文件 19.2. 用户名映射 19.3. 认证方法 19.4. 用户认证 章20. 数据库角色和权限 20.1. 数据库角色 20.2. 角色属性 20.3. 权限 20.4. 角色成员 20.5. 函数和触发器 章21. 管理数据库 21.1. 概述 21.2. 创建一个数据库 21.3. 临时库 21.4. 数据库配置 21.5. 删除数据库 21.6. 表空间 章22. 本土化 22.1. 区域支持 22.2. 字符集支持 章23. 日常数据库维护工作 23.1. Routine Vacuuming日常清理 23.2. 经常重建索引 23.3. 日志文件维护 章24. 备份和恢复 24.1. SQL转储 24.2. 文件系统级别的备份 24.3. 在线备份以及即时恢复(PITR) 24.4. 版本间迁移 章25. 高可用性与负载均衡，复制 25.1. 不同解决方案的比较 25.2. 日志传送备份服务器 25.3. 失效切换 25.4. 日志传送的替代方法 25.5. 热备 章26. 恢复配置 26.1. 归档恢复设置 26.2. 恢复目标设置 26.3. 备服务器设置 章27. 监控数据库的活动 27.1. 标准Unix工具 27.2. 统计收集器 27.3. 查看锁 27.4. 动态跟踪 章28. 监控磁盘使用情况 28.1. 判断磁盘的使用量 28.2. 磁盘满导致的失效 章29. 可靠性和预写式日志 29.1. 可靠性 29.2. 预写式日志(WAL) 29.3. 异步提交 29.4. WAL配置 29.5. WAL内部 章30. Regression Tests 30.1. Running the Tests 30.2. Test Evaluation 30.3. Variant Comparison Files 30.4. Test Coverage Examination IV. 客户端接口 章31. libpq-C库 31.1. 数据库联接函数 31.2. 连接状态函数 31.3. 命令执行函数 31.4. 异步命令处理 31.5. 取消正在处理的查询 31.6. 捷径接口 31.7. 异步通知 31.8. 与COPY命令相关的函数 31.9. Control Functions 控制函数 31.10. 其他函数 31.11. 注意信息处理 31.12. 事件系统 31.13. 环境变量 31.14. 口令文件 31.15. 连接服务的文件 31.16. LDAP查找连接参数 31.17. SSL支持 31.18. 在多线程程序里的行为 31.19. 制作libpq程序 31.20. 例子程序 章32. 大对象 32.1. 介绍 32.2. 实现特点 32.3. 客户端接口 32.4. 服务器端函数 32.5. 例子程序 章33. ECPG - Embedded SQL in C 33.1. The Concept 33.2. Connecting to the Database Server 33.3. Closing a Connection 33.4. Running SQL Commands 33.5. Choosing a Connection 33.6. Using Host Variables 33.7. Dynamic SQL 33.8. pgtypes library 33.9. Using Descriptor Areas 33.10. Informix compatibility mode 33.11. Error Handling 33.12. Preprocessor directives 33.13. Processing Embedded SQL Programs 33.14. Library Functions 33.15. Internals 章34. 信息模式 34.1. 关于这个模式 34.2. 数据类型 34.3. information_schema_catalog_name 34.4. administrable_role_authorizations 34.5. applicable_roles 34.6. attributes 34.7. check_constraint_routine_usage 34.8. check_constraints 34.9. column_domain_usage 34.10. column_privileges 34.11. column_udt_usage 34.12. 字段 34.13. constraint_column_usage 34.14. constraint_table_usage 34.15. data_type_privileges 34.16. domain_constraints 34.18. domains 34.17. domain_udt_usage 34.19. element_types 34.20. enabled_roles 34.21. foreign_data_wrapper_options 34.22. foreign_data_wrappers 34.23. foreign_server_options 34.24. foreign_servers 34.25. key_column_usage 34.26. parameters 34.27. referential_constraints 34.28. role_column_grants 34.29. role_routine_grants 34.30. role_table_grants 34.31. role_usage_grants 34.32. routine_privileges 34.33. routines 34.34. schemata 34.35. sequences 34.36. sql_features 34.37. sql_implementation_info 34.38. sql_languages 34.39. sql_packages 34.40. sql_parts 34.41. sql_sizing 34.42. sql_sizing_profiles 34.43. table_constraints 34.44. table_privileges 34.45. tables 34.46. triggered_update_columns 34.47. 触发器 34.48. usage_privileges 34.49. user_mapping_options 34.50. user_mappings 34.51. view_column_usage 34.52. view_routine_usage 34.53. view_table_usage 34.54. 视图 V. 服务器端编程 章35. 扩展SQL 35.1. 扩展性是如何实现的 35.2. PostgreSQL类型系统 35.3. User-Defined Functions 35.4. Query Language (SQL) Functions 35.5. Function Overloading 35.6. Function Volatility Categories 35.7. Procedural Language Functions 35.8. Internal Functions 35.9. C-Language Functions 35.10. User-Defined Aggregates 35.11. User-Defined Types 35.12. User-Defined Operators 35.13. Operator Optimization Information 35.14. Interfacing Extensions To Indexes 35.15. 用C++扩展 章36. 触发器 36.1. 触发器行为概述 36.3. 用 C 写触发器 36.2. 数据改变的可视性 36.4. 一个完整的例子 章37. 规则系统 37.1. The Query Tree 37.2. 视图和规则系统 37.3. 在INSERT，UPDATE和DELETE上的规则 37.4. 规则和权限 37.5. 规则和命令状态 37.6. 规则与触发器得比较 章38. Procedural Languages 38.1. Installing Procedural Languages 章39. PL/pgSQL - SQL过程语言 39.1. 概述 39.2. PL/pgSQL的结构 39.3. 声明 39.4. 表达式 39.5. 基本语句 39.6. 控制结构 39.7. 游标 39.8. 错误和消息 39.9. 触发器过程 39.10. PL/pgSQL Under the Hood 39.11. 开发PL/pgSQL的一些提示 39.12. 从OraclePL/SQL 进行移植 章40. PL/Tcl - Tcl Procedural Language 40.1. Overview 40.2. PL/Tcl Functions and Arguments 40.3. Data Values in PL/Tcl 40.4. Global Data in PL/Tcl 40.5. Database Access from PL/Tcl 40.6. Trigger Procedures in PL/Tcl 40.7. Modules and the unknown command 40.8. Tcl Procedure Names 章41. PL/Perl - Perl Procedural Language 41.1. PL/Perl Functions and Arguments 41.2. Data Values in PL/Perl 41.3. Built-in Functions 41.4. Global Values in PL/Perl 41.6. PL/Perl Triggers 41.5. Trusted and Untrusted PL/Perl 41.7. PL/Perl Under the Hood 章42. PL/Python - Python Procedural Language 42.1. Python 2 vs. Python 3 42.2. PL/Python Functions 42.3. Data Values 42.4. Sharing Data 42.5. Anonymous Code Blocks 42.6. Trigger Functions 42.7. Database Access 42.8. Utility Functions 42.9. Environment Variables 章43. Server Programming Interface 43.1. Interface Functions Spi-spi-connect Spi-spi-finish Spi-spi-push Spi-spi-pop Spi-spi-execute Spi-spi-exec Spi-spi-execute-with-args Spi-spi-prepare Spi-spi-prepare-cursor Spi-spi-prepare-params Spi-spi-getargcount Spi-spi-getargtypeid Spi-spi-is-cursor-plan Spi-spi-execute-plan Spi-spi-execute-plan-with-paramlist Spi-spi-execp Spi-spi-cursor-open Spi-spi-cursor-open-with-args Spi-spi-cursor-open-with-paramlist Spi-spi-cursor-find Spi-spi-cursor-fetch Spi-spi-cursor-move Spi-spi-scroll-cursor-fetch Spi-spi-scroll-cursor-move Spi-spi-cursor-close Spi-spi-saveplan 43.2. Interface Support Functions Spi-spi-fname Spi-spi-fnumber Spi-spi-getvalue Spi-spi-getbinval Spi-spi-gettype Spi-spi-gettypeid Spi-spi-getrelname Spi-spi-getnspname 43.3. Memory Management Spi-spi-palloc Spi-realloc Spi-spi-pfree Spi-spi-copytuple Spi-spi-returntuple Spi-spi-modifytuple Spi-spi-freetuple Spi-spi-freetupletable Spi-spi-freeplan 43.4. Visibility of Data Changes 43.5. Examples VI. 参考手册 I. SQL命令 Sql-abort Sql-alteraggregate Sql-alterconversion Sql-alterdatabase Sql-alterdefaultprivileges Sql-alterdomain Sql-alterforeigndatawrapper Sql-alterfunction Sql-altergroup Sql-alterindex Sql-alterlanguage Sql-alterlargeobject Sql-alteroperator Sql-alteropclass Sql-alteropfamily Sql-alterrole Sql-alterschema Sql-altersequence Sql-alterserver Sql-altertable Sql-altertablespace Sql-altertsconfig Sql-altertsdictionary Sql-altertsparser Sql-altertstemplate Sql-altertrigger Sql-altertype Sql-alteruser Sql-alterusermapping Sql-alterview Sql-analyze Sql-begin Sql-checkpoint Sql-close Sql-cluster Sql-comment Sql-commit Sql-commit-prepared Sql-copy Sql-createaggregate Sql-createcast Sql-createconstraint Sql-createconversion Sql-createdatabase Sql-createdomain Sql-createforeigndatawrapper Sql-createfunction Sql-creategroup Sql-createindex Sql-createlanguage Sql-createoperator Sql-createopclass Sql-createopfamily Sql-createrole Sql-createrule Sql-createschema Sql-createsequence Sql-createserver Sql-createtable Sql-createtableas Sql-createtablespace Sql-createtsconfig Sql-createtsdictionary Sql-createtsparser Sql-createtstemplate Sql-createtrigger Sql-createtype Sql-createuser Sql-createusermapping Sql-createview Sql-deallocate Sql-declare Sql-delete Sql-discard Sql-do Sql-dropaggregate Sql-dropcast Sql-dropconversion Sql-dropdatabase Sql-dropdomain Sql-dropforeigndatawrapper Sql-dropfunction Sql-dropgroup Sql-dropindex Sql-droplanguage Sql-dropoperator Sql-dropopclass Sql-dropopfamily Sql-drop-owned Sql-droprole Sql-droprule Sql-dropschema Sql-dropsequence Sql-dropserver Sql-droptable Sql-droptablespace Sql-droptsconfig Sql-droptsdictionary Sql-droptsparser Sql-droptstemplate Sql-droptrigger Sql-droptype Sql-dropuser Sql-dropusermapping Sql-dropview Sql-end Sql-execute Sql-explain Sql-fetch Sql-grant Sql-insert Sql-listen Sql-load Sql-lock Sql-move Sql-notify Sql-prepare Sql-prepare-transaction Sql-reassign-owned Sql-reindex Sql-release-savepoint Sql-reset Sql-revoke Sql-rollback Sql-rollback-prepared Sql-rollback-to Sql-savepoint Sql-select Sql-selectinto Sql-set Sql-set-constraints Sql-set-role Sql-set-session-authorization Sql-set-transaction Sql-show Sql-start-transaction Sql-truncate Sql-unlisten Sql-update Sql-vacuum Sql-values II. 客户端应用程序 App-clusterdb App-createdb App-createlang App-createuser App-dropdb App-droplang App-dropuser App-ecpg App-pgconfig App-pgdump App-pg-dumpall App-pgrestore App-psql App-reindexdb App-vacuumdb III. PostgreSQL服务器应用程序 App-initdb App-pgcontroldata App-pg-ctl App-pgresetxlog App-postgres App-postmaster VII. 内部 章44. PostgreSQL内部概览 44.1. 查询路径 44.2. 连接是如何建立起来的 44.3. 分析器阶段 44.4. ThePostgreSQL规则系统 44.5. 规划器/优化器 44.6. 执行器 章45. 系统表 45.1. 概述 45.2. pg_aggregate 45.3. pg_am 45.4. pg_amop 45.5. pg_amproc 45.6. pg_attrdef 45.7. pg_attribute 45.8. pg_authid 45.9. pg_auth_members 45.10. pg_cast 45.11. pg_class 45.12. pg_constraint 45.13. pg_conversion 45.14. pg_database 45.15. pg_db_role_setting 45.16. pg_default_acl 45.17. pg_depend 45.18. pg_description 45.19. pg_enum 45.20. pg_foreign_data_wrapper 45.21. pg_foreign_server 45.22. pg_index 45.23. pg_inherits 45.24. pg_language 45.25. pg_largeobject 45.26. pg_largeobject_metadata 45.27. pg_namespace 45.28. pg_opclass 45.29. pg_operator 45.30. pg_opfamily 45.31. pg_pltemplate 45.32. pg_proc 45.33. pg_rewrite 45.34. pg_shdepend 45.35. pg_shdescription 45.36. pg_statistic 45.37. pg_tablespace 45.38. pg_trigger 45.39. pg_ts_config 45.40. pg_ts_config_map 45.41. pg_ts_dict 45.42. pg_ts_parser 45.43. pg_ts_template 45.44. pg_type 45.45. pg_user_mapping 45.46. System Views 45.47. pg_cursors 45.48. pg_group 45.49. pg_indexes 45.50. pg_locks 45.51. pg_prepared_statements 45.52. pg_prepared_xacts 45.53. pg_roles 45.54. pg_rules 45.55. pg_settings 45.56. pg_shadow 45.57. pg_stats 45.58. pg_tables 45.59. pg_timezone_abbrevs 45.60. pg_timezone_names 45.61. pg_user 45.62. pg_user_mappings 45.63. pg_views 章46. Frontend/Backend Protocol 46.1. Overview 46.2. Message Flow 46.3. Streaming Replication Protocol 46.4. Message Data Types 46.5. Message Formats 46.6. Error and Notice Message Fields 46.7. Summary of Changes since Protocol 2.0 47. PostgreSQL Coding Conventions 47.1. Formatting 47.2. Reporting Errors Within the Server 47.3. Error Message Style Guide 章48. Native Language Support 48.1. For the Translator 48.2. For the Programmer 章49. Writing A Procedural Language Handler 章50. Genetic Query Optimizer 50.1. Query Handling as a Complex Optimization Problem 50.2. Genetic Algorithms 50.3. Genetic Query Optimization (GEQO) in PostgreSQL 50.4. Further Reading 章51. 索引访问方法接口定义 51.1. 索引的系统表记录 51.2. 索引访问方法函数 51.3. 索引扫描 51.4. 索引锁的考量 51.5. 索引唯一性检查 51.6. 索引开销估计函数 章52. GiST Indexes 52.1. Introduction 52.2. Extensibility 52.3. Implementation 52.4. Examples 52.5. Crash Recovery 章53. GIN Indexes 53.1. Introduction 53.2. Extensibility 53.3. Implementation 53.4. GIN tips and tricks 53.5. Limitations 53.6. Examples 章54. 数据库物理存储 54.1. 数据库文件布局 54.2. TOAST 54.3. 自由空间映射 54.4. 可见映射 54.5. 数据库分页文件 章55. BKI后端接口 55.1. BKI 文件格式 55.2. BKI命令 55.3. 系统初始化的BKI文件的结构 55.4. 例子 章56. 规划器如何使用统计信息 56.1. 行预期的例子 VIII. 附录 A. PostgreSQL错误代码 B. 日期/时间支持 B.1. 日期/时间输入解析 B.2. 日期/时间关键字 B.3. 日期/时间配置文件 B.4. 日期单位的历史 C. SQL关键字 D. SQL Conformance D.1. Supported Features D.2. Unsupported Features E. Release Notes Release-0-01 Release-0-02 Release-0-03 Release-1-0 Release-1-01 Release-1-02 Release-1-09 Release-6-0 Release-6-1 Release-6-1-1 Release-6-2 Release-6-2-1 Release-6-3 Release-6-3-1 Release-6-3-2 Release-6-4 Release-6-4-1 Release-6-4-2 Release-6-5 Release-6-5-1 Release-6-5-2 Release-6-5-3 Release-7-0 Release-7-0-1 Release-7-0-2 Release-7-0-3 Release-7-1 Release-7-1-1 Release-7-1-2 Release-7-1-3 Release-7-2 Release-7-2-1 Release-7-2-2 Release-7-2-3 Release-7-2-4 Release-7-2-5 Release-7-2-6 Release-7-2-7 Release-7-2-8 Release-7-3 Release-7-3-1 Release-7-3-10 Release-7-3-11 Release-7-3-12 Release-7-3-13 Release-7-3-14 Release-7-3-15 Release-7-3-16 Release-7-3-17 Release-7-3-18 Release-7-3-19 Release-7-3-2 Release-7-3-20 Release-7-3-21 Release-7-3-3 Release-7-3-4 Release-7-3-5 Release-7-3-6 Release-7-3-7 Release-7-3-8 Release-7-3-9 Release-7-4 Release-7-4-1 Release-7-4-10 Release-7-4-11 Release-7-4-12 Release-7-4-13 Release-7-4-14 Release-7-4-15 Release-7-4-16 Release-7-4-17 Release-7-4-18 Release-7-4-19 Release-7-4-2 Release-7-4-20 Release-7-4-21 Release-7-4-22 Release-7-4-23 Release-7-4-24 Release-7-4-25 Release-7-4-26 Release-7-4-27 Release-7-4-28 Release-7-4-29 Release-7-4-3 Release-7-4-30 Release-7-4-4 Release-7-4-5 Release-7-4-6 Release-7-4-7 Release-7-4-8 Release-7-4-9 Release-8-0 Release-8-0-1 Release-8-0-10 Release-8-0-11 Release-8-0-12 Release-8-0-13 Release-8-0-14 Release-8-0-15 Release-8-0-16 Release-8-0-17 Release-8-0-18 Release-8-0-19 Release-8-0-2 Release-8-0-20 Release-8-0-21 Release-8-0-22 Release-8-0-23 Release-8-0-24 Release-8-0-25 Release-8-0-26 Release-8-0-3 Release-8-0-4 Release-8-0-5 Release-8-0-6 Release-8-0-7 Release-8-0-8 Release-8-0-9 Release-8-1 Release-8-1-1 Release-8-1-10 Release-8-1-11 Release-8-1-12 Release-8-1-13 Release-8-1-14 Release-8-1-15 Release-8-1-16 Release-8-1-17 Release-8-1-18 Release-8-1-19 Release-8-1-2 Release-8-1-20 Release-8-1-21 Release-8-1-22 Release-8-1-23 Release-8-1-3 Release-8-1-4 Release-8-1-5 Release-8-1-6 Release-8-1-7 Release-8-1-8 Release-8-1-9 Release-8-2 Release-8-2-1 Release-8-2-10 Release-8-2-11 Release-8-2-12 Release-8-2-13 Release-8-2-14 Release-8-2-15 Release-8-2-16 Release-8-2-17 Release-8-2-18 Release-8-2-19 Release-8-2-2 Release-8-2-20 Release-8-2-21 Release-8-2-3 Release-8-2-4 Release-8-2-5 Release-8-2-6 Release-8-2-7 Release-8-2-8 Release-8-2-9 Release-8-3 Release-8-3-1 Release-8-3-10 Release-8-3-11 Release-8-3-12 Release-8-3-13 Release-8-3-14 Release-8-3-15 Release-8-3-2 Release-8-3-3 Release-8-3-4 Release-8-3-5 Release-8-3-6 Release-8-3-7 Release-8-3-8 Release-8-3-9 Release-8-4 Release-8-4-1 Release-8-4-2 Release-8-4-3 Release-8-4-4 Release-8-4-5 Release-8-4-6 Release-8-4-7 Release-8-4-8 Release-9-0 Release-9-0-1 Release-9-0-2 Release-9-0-3 Release-9-0-4 F. 额外提供的模块 F.1. adminpack F.2. auto_explain F.3. btree_gin F.4. btree_gist F.5. chkpass F.6. citext F.7. cube F.8. dblink Contrib-dblink-connect Contrib-dblink-connect-u Contrib-dblink-disconnect Contrib-dblink Contrib-dblink-exec Contrib-dblink-open Contrib-dblink-fetch Contrib-dblink-close Contrib-dblink-get-connections Contrib-dblink-error-message Contrib-dblink-send-query Contrib-dblink-is-busy Contrib-dblink-get-notify Contrib-dblink-get-result Contrib-dblink-cancel-query Contrib-dblink-get-pkey Contrib-dblink-build-sql-insert Contrib-dblink-build-sql-delete Contrib-dblink-build-sql-update F.9. dict_int F.10. dict_xsyn F.11. earthdistance F.12. fuzzystrmatch F.13. hstore F.14. intagg F.15. intarray F.16. isn F.17. lo F.18. ltree F.19. oid2name F.20. pageinspect F.21. passwordcheck F.22. pg_archivecleanup F.23. pgbench F.24. pg_buffercache F.25. pgcrypto F.26. pg_freespacemap F.27. pgrowlocks F.28. pg_standby F.29. pg_stat_statements F.30. pgstattuple F.31. pg_trgm F.32. pg_upgrade F.33. seg F.34. spi F.35. sslinfo F.36. tablefunc F.37. test_parser F.38. tsearch2 F.39. unaccent F.40. uuid-ossp F.41. vacuumlo F.42. xml2 G. 外部项目 G.1. 客户端接口 G.2. 过程语言 G.3. 扩展 H. The Source Code Repository H.1. Getting The Source Via Git I. 文档 I.1. DocBook I.2. 工具集 I.3. 制作文档 I.4. 文档写作 I.5. 风格指导 J. 首字母缩略词 参考书目 Bookindex Index

©
This document uses PHP Chinese website manual Release

characters

E.165. Release 7.2

Release date: 2002-02-04

E.165.1. Overview

This release improves PostgreSQL for use in high-volume applications.

Major changes in this release:

VACUUM

Vacuuming no longer locks tables, thus allowing normal user access during the vacuum. A new VACUUM FULL command does old-style vacuum by locking the table and shrinking the on-disk copy of the table.

Transactions

There is no longer a problem with installations that exceed four billion transactions.

OIDs

OIDs are now optional. Users can now create tables without OIDs for cases where OID usage is excessive.

Optimizer

The system now computes histogram column statistics during ANALYZE, allowing much better optimizer choices.

Security

A new MD5 encryption option allows more secure storage and transfer of passwords. A new Unix-domain socket authentication option is available on Linux and BSD systems.

Statistics

Administrators can use the new table access statistics module to get fine-grained information about table and index usage.

Internationalization

Program and library messages can now be displayed in several languages.

E.165.2. Migration to Version 7.2

A dump/restore using pg_dump is required for those wishing to migrate data from any previous release.

Observe the following incompatibilities:

  • The semantics of the VACUUM command have changed in this release. You might wish to update your maintenance procedures accordingly.

  • In this release, comparisons using = NULL will always return false (or NULL, more precisely). Previous releases automatically transformed this syntax to IS NULL. The old behavior can be re-enabled using a postgresql.conf parameter.

  • The pg_hba.conf and pg_ident.conf configuration is now only reloaded after receiving a SIGHUP signal, not with each connection.

  • The function octet_length() now returns the uncompressed data length.

  • The date/time value 'current' is no longer available. You will need to rewrite your applications.

  • The timestamp(), time(), and interval() functions are no longer available. Instead of timestamp(), use timestamp 'string' or CAST.

The SELECT ... LIMIT #,# syntax will be removed in the next release. You should change your queries to use separate LIMIT and OFFSET clauses, e.g. LIMIT 10 OFFSET 20.

E.165.3. Changes

E.165.3.1. Server Operation

  • Create temporary files in a separate directory (Bruce)

  • Delete orphaned temporary files on postmaster startup (Bruce)

  • Added unique indexes to some system tables (Tom)

  • System table operator reorganization (Oleg Bartunov, Teodor Sigaev, Tom)

  • Renamed pg_log to pg_clog (Tom)

  • Enable SIGTERM, SIGQUIT to kill backends (Jan)

  • Removed compile-time limit on number of backends (Tom)

  • Better cleanup for semaphore resource failure (Tatsuo, Tom)

  • Allow safe transaction ID wraparound (Tom)

  • Removed OIDs from some system tables (Tom)

  • Removed "triggered data change violation" error check (Tom)

  • SPI portal creation of prepared/saved plans (Jan)

  • Allow SPI column functions to work for system columns (Tom)

  • Long value compression improvement (Tom)

  • Statistics collector for table, index access (Jan)

  • Truncate extra-long sequence names to a reasonable value (Tom)

  • Measure transaction times in milliseconds (Thomas)

  • Fix TID sequential scans (Hiroshi)

  • Superuser ID now fixed at 1 (Peter E)

  • New pg_ctl "reload" option (Tom)

E.165.3.2. Performance

  • Optimizer improvements (Tom)

  • New histogram column statistics for optimizer (Tom)

  • Reuse write-ahead log files rather than discarding them (Tom)

  • Cache improvements (Tom)

  • IS NULL, IS NOT NULL optimizer improvement (Tom)

  • Improve lock manager to reduce lock contention (Tom)

  • Keep relcache entries for index access support functions (Tom)

  • Allow better selectivity with NaN and infinities in NUMERIC (Tom)

  • R-tree performance improvements (Kenneth Been)

  • B-tree splits more efficient (Tom)

E.165.3.3. Privileges

  • Change UPDATE, DELETE privileges to be distinct (Peter E)

  • New REFERENCES, TRIGGER privileges (Peter E)

  • Allow GRANT/REVOKE to/from more than one user at a time (Peter E)

  • New has_table_privilege() function (Joe Conway)

  • Allow non-superuser to vacuum database (Tom)

  • New SET SESSION AUTHORIZATION command (Peter E)

  • Fix bug in privilege modifications on newly created tables (Tom)

  • Disallow access to pg_statistic for non-superuser, add user-accessible views (Tom)

E.165.3.4. Client Authentication

  • Fork postmaster before doing authentication to prevent hangs (Peter E)

  • Add ident authentication over Unix domain sockets on Linux, *BSD (Helge Bahmann, Oliver Elphick, Teodor Sigaev, Bruce)

  • Add a password authentication method that uses MD5 encryption (Bruce)

  • Allow encryption of stored passwords using MD5 (Bruce)

  • PAM authentication (Dominic J. Eidson)

  • Load pg_hba.conf and pg_ident.conf only on startup and SIGHUP (Bruce)

E.165.3.5. Server Configuration

  • Interpretation of some time zone abbreviations as Australian rather than North American now settable at run time (Bruce)

  • New parameter to set default transaction isolation level (Peter E)

  • New parameter to enable conversion of "expr = NULL" into "expr IS NULL", off by default (Peter E)

  • New parameter to control memory usage by VACUUM (Tom)

  • New parameter to set client authentication timeout (Tom)

  • New parameter to set maximum number of open files (Tom)

E.165.3.6. Queries

  • Statements added by INSERT rules now execute after the INSERT (Jan)

  • Prevent unadorned relation names in target list (Bruce)

  • NULLs now sort after all normal values in ORDER BY (Tom)

  • New IS UNKNOWN, IS NOT UNKNOWN Boolean tests (Tom)

  • New SHARE UPDATE EXCLUSIVE lock mode (Tom)

  • New EXPLAIN ANALYZE command that shows run times and row counts (Martijn van Oosterhout)

  • Fix problem with LIMIT and subqueries (Tom)

  • Fix for LIMIT, DISTINCT ON pushed into subqueries (Tom)

  • Fix nested EXCEPT/INTERSECT (Tom)

E.165.3.7. Schema Manipulation

  • Fix SERIAL in temporary tables (Bruce)

  • Allow temporary sequences (Bruce)

  • Sequences now use int8 internally (Tom)

  • New SERIAL8 creates int8 columns with sequences, default still SERIAL4 (Tom)

  • Make OIDs optional using WITHOUT OIDS (Tom)

  • Add %TYPE syntax to CREATE TYPE (Ian Lance Taylor)

  • Add ALTER TABLE / DROP CONSTRAINT for CHECK constraints (Christopher Kings-Lynne)

  • New CREATE OR REPLACE FUNCTION to alter existing function (preserving the function OID) (Gavin Sherry)

  • Add ALTER TABLE / ADD [ UNIQUE | PRIMARY ] (Christopher Kings-Lynne)

  • Allow column renaming in views

  • Make ALTER TABLE / RENAME COLUMN update column names of indexes (Brent Verner)

  • Fix for ALTER TABLE / ADD CONSTRAINT ... CHECK with inherited tables (Stephan Szabo)

  • ALTER TABLE RENAME update foreign-key trigger arguments correctly (Brent Verner)

  • DROP AGGREGATE and COMMENT ON AGGREGATE now accept an aggtype (Tom)

  • Add automatic return type data casting for SQL functions (Tom)

  • Allow GiST indexes to handle NULLs and multikey indexes (Oleg Bartunov, Teodor Sigaev, Tom)

  • Enable partial indexes (Martijn van Oosterhout)

E.165.3.8. Utility Commands

  • Add RESET ALL, SHOW ALL (Marko Kreen)

  • CREATE/ALTER USER/GROUP now allow options in any order (Vince)

  • Add LOCK A, B, C functionality (Neil Padgett)

  • New ENCRYPTED/UNENCRYPTED option to CREATE/ALTER USER (Bruce)

  • New light-weight VACUUM does not lock table; old semantics are available as VACUUM FULL (Tom)

  • Disable COPY TO/FROM on views (Bruce)

  • COPY DELIMITERS string must be exactly one character (Tom)

  • VACUUM warning about index tuples fewer than heap now only appears when appropriate (Martijn van Oosterhout)

  • Fix privilege checks for CREATE INDEX (Tom)

  • Disallow inappropriate use of CREATE/DROP INDEX/TRIGGER/VIEW (Tom)

E.165.3.9. Data Types and Functions

  • SUM(), AVG(), COUNT() now uses int8 internally for speed (Tom)

  • Add convert(), convert2() (Tatsuo)

  • New function bit_length() (Peter E)

  • Make the "n" in CHAR(n)/VARCHAR(n) represents letters, not bytes (Tatsuo)

  • CHAR(), VARCHAR() now reject strings that are too long (Peter E)

  • BIT VARYING now rejects bit strings that are too long (Peter E)

  • BIT now rejects bit strings that do not match declared size (Peter E)

  • INET, CIDR text conversion functions (Alex Pilosov)

  • INET, CIDR operators << and <<= indexable (Alex Pilosov)

  • Bytea \### now requires valid three digit octal number

  • Bytea comparison improvements, now supports =, <>, >, >=, <, and <=

  • Bytea now supports B-tree indexes

  • Bytea now supports LIKE, LIKE...ESCAPE, NOT LIKE, NOT LIKE...ESCAPE

  • Bytea now supports concatenation

  • New bytea functions: position, substring, trim, btrim, and length

  • New encode() function mode, "escaped", converts minimally escaped bytea to/from text

  • Add pg_database_encoding_max_length() (Tatsuo)

  • Add pg_client_encoding() function (Tatsuo)

  • now() returns time with millisecond precision (Thomas)

  • New TIMESTAMP WITHOUT TIMEZONE data type (Thomas)

  • Add ISO date/time specification with "T", yyyy-mm-ddThh:mm:ss (Thomas)

  • New xid/int comparison functions (Hiroshi)

  • Add precision to TIME, TIMESTAMP, and INTERVAL data types (Thomas)

  • Modify type coercion logic to attempt binary-compatible functions first (Tom)

  • New encode() function installed by default (Marko Kreen)

  • Improved to_*() conversion functions (Karel Zak)

  • Optimize LIKE/ILIKE when using single-byte encodings (Tatsuo)

  • New functions in contrib/pgcrypto: crypt(), hmac(), encrypt(), gen_salt() (Marko Kreen)

  • Correct description of translate() function (Bruce)

  • Add INTERVAL argument for SET TIME ZONE (Thomas)

  • Add INTERVAL YEAR TO MONTH (etc.) syntax (Thomas)

  • Optimize length functions when using single-byte encodings (Tatsuo)

  • Fix path_inter, path_distance, path_length, dist_ppath to handle closed paths (Curtis Barrett, Tom)

  • octet_length(text) now returns non-compressed length (Tatsuo, Bruce)

  • Handle "July" full name in date/time literals (Greg Sabino Mullane)

  • Some datatype() function calls now evaluated differently

  • Add support for Julian and ISO time specifications (Thomas)

E.165.3.10. Internationalization

  • National language support in psql, pg_dump, libpq, and server (Peter E)

  • Message translations in Chinese (simplified, traditional), Czech, French, German, Hungarian, Russian, Swedish (Peter E, Serguei A. Mokhov, Karel Zak, Weiping He, Zhenbang Wei, Kovacs Zoltan)

  • Make trim, ltrim, rtrim, btrim, lpad, rpad, translate multibyte aware (Tatsuo)

  • Add LATIN5,6,7,8,9,10 support (Tatsuo)

  • Add ISO 8859-5,6,7,8 support (Tatsuo)

  • Correct LATIN5 to mean ISO-8859-9, not ISO-8859-5 (Tatsuo)

  • Make mic2ascii() non-ASCII aware (Tatsuo)

  • Reject invalid multibyte character sequences (Tatsuo)

E.165.3.11. PL/pgSQL

  • Now uses portals for SELECT loops, allowing huge result sets (Jan)

  • CURSOR and REFCURSOR support (Jan)

  • Can now return open cursors (Jan)

  • Add ELSEIF (Klaus Reger)

  • Improve PL/pgSQL error reporting, including location of error (Tom)

  • Allow IS or FOR key words in cursor declaration, for compatibility (Bruce)

  • Fix for SELECT ... FOR UPDATE (Tom)

  • Fix for PERFORM returning multiple rows (Tom)

  • Make PL/pgSQL use the server's type coercion code (Tom)

  • Memory leak fix (Jan, Tom)

  • Make trailing semicolon optional (Tom)

E.165.3.12. PL/Perl

  • New untrusted PL/Perl (Alex Pilosov)

  • PL/Perl is now built on some platforms even if libperl is not shared (Peter E)

E.165.3.13. PL/Tcl

  • Now reports errorInfo (Vsevolod Lobko)

  • Add spi_lastoid function (bob@redivi.com)

E.165.3.14. PL/Python

  • ...is new (Andrew Bosma)

E.165.3.15. psql

  • \d displays indexes in unique, primary groupings (Christopher Kings-Lynne)

  • Allow trailing semicolons in backslash commands (Greg Sabino Mullane)

  • Read password from /dev/tty if possible

  • Force new password prompt when changing user and database (Tatsuo, Tom)

  • Format the correct number of columns for Unicode (Patrice)

E.165.3.16. libpq

  • New function PQescapeString() to escape quotes in command strings (Florian Weimer)

  • New function PQescapeBytea() escapes binary strings for use as SQL string literals

E.165.3.17. JDBC

  • Return OID of INSERT (Ken K)

  • Handle more data types (Ken K)

  • Handle single quotes and newlines in strings (Ken K)

  • Handle NULL variables (Ken K)

  • Fix for time zone handling (Barry Lind)

  • Improved Druid support

  • Allow eight-bit characters with non-multibyte server (Barry Lind)

  • Support BIT, BINARY types (Ned Wolpert)

  • Reduce memory usage (Michael Stephens, Dave Cramer)

  • Update DatabaseMetaData (Peter E)

  • Add DatabaseMetaData.getCatalogs() (Peter E)

  • Encoding fixes (Anders Bengtsson)

  • Get/setCatalog methods (Jason Davies)

  • DatabaseMetaData.getColumns() now returns column defaults (Jason Davies)

  • DatabaseMetaData.getColumns() performance improvement (Jeroen van Vianen)

  • Some JDBC1 and JDBC2 merging (Anders Bengtsson)

  • Transaction performance improvements (Barry Lind)

  • Array fixes (Greg Zoller)

  • Serialize addition

  • Fix batch processing (Rene Pijlman)

  • ExecSQL method reorganization (Anders Bengtsson)

  • GetColumn() fixes (Jeroen van Vianen)

  • Fix isWriteable() function (Rene Pijlman)

  • Improved passage of JDBC2 conformance tests (Rene Pijlman)

  • Add bytea type capability (Barry Lind)

  • Add isNullable() (Rene Pijlman)

  • JDBC date/time test suite fixes (Liam Stewart)

  • Fix for SELECT 'id' AS xxx FROM table (Dave Cramer)

  • Fix DatabaseMetaData to show precision properly (Mark Lillywhite)

  • New getImported/getExported keys (Jason Davies)

  • MD5 password encryption support (Jeremy Wohl)

  • Fix to actually use type cache (Ned Wolpert)

E.165.3.18. ODBC

  • Remove query size limit (Hiroshi)

  • Remove text field size limit (Hiroshi)

  • Fix for SQLPrimaryKeys in multibyte mode (Hiroshi)

  • Allow ODBC procedure calls (Hiroshi)

  • Improve boolean handing (Aidan Mountford)

  • Most configuration options now settable via DSN (Hiroshi)

  • Multibyte, performance fixes (Hiroshi)

  • Allow driver to be used with iODBC or unixODBC (Peter E)

  • MD5 password encryption support (Bruce)

  • Add more compatibility functions to odbc.sql (Peter E)

E.165.3.19. ECPG

  • EXECUTE ... INTO implemented (Christof Petig)

  • Multiple row descriptor support (e.g. CARDINALITY) (Christof Petig)

  • Fix for GRANT parameters (Lee Kindness)

  • Fix INITIALLY DEFERRED bug

  • Various bug fixes (Michael, Christof Petig)

  • Auto allocation for indicator variable arrays (int *ind_p=NULL)

  • Auto allocation for string arrays (char **foo_pp=NULL)

  • ECPGfree_auto_mem fixed

  • All function names with external linkage are now prefixed by ECPG

  • Fixes for arrays of structures (Michael)

E.165.3.20. Misc. Interfaces

  • Python fix fetchone() (Gerhard Haring)

  • Use UTF, Unicode in Tcl where appropriate (Vsevolod Lobko, Reinhard Max)

  • Add Tcl COPY TO/FROM (ljb)

  • Prevent output of default index op class in pg_dump (Tom)

  • Fix libpgeasy memory leak (Bruce)

E.165.3.21. Build and Install

  • Configure, dynamic loader, and shared library fixes (Peter E)

  • Fixes in QNX 4 port (Bernd Tegge)

  • Fixes in Cygwin and Windows ports (Jason Tishler, Gerhard Haring, Dmitry Yurtaev, Darko Prenosil, Mikhail Terekhov)

  • Fix for Windows socket communication failures (Magnus, Mikhail Terekhov)

  • Hurd compile fix (Oliver Elphick)

  • BeOS fixes (Cyril Velter)

  • Remove configure --enable-unicode-conversion, now enabled by multibyte (Tatsuo)

  • AIX fixes (Tatsuo, Andreas)

  • Fix parallel make (Peter E)

  • Install SQL language manual pages into OS-specific directories (Peter E)

  • Rename config.h to pg_config.h (Peter E)

  • Reorganize installation layout of header files (Peter E)

E.165.3.22. Source Code

  • Remove SEP_CHAR (Bruce)

  • New GUC hooks (Tom)

  • Merge GUC and command line handling (Marko Kreen)

  • Remove EXTEND INDEX (Martijn van Oosterhout, Tom)

  • New pgjindent utility to indent java code (Bruce)

  • Remove define of true/false when compiling under C++ (Leandro Fanzone, Tom)

  • pgindent fixes (Bruce, Tom)

  • Replace strcasecmp() with strcmp() where appropriate (Peter E)

  • Dynahash portability improvements (Tom)

  • Add 'volatile' usage in spinlock structures

  • Improve signal handling logic (Tom)

E.165.3.23. Contrib

  • New contrib/rtree_gist (Oleg Bartunov, Teodor Sigaev)

  • New contrib/tsearch full-text indexing (Oleg, Teodor Sigaev)

  • Add contrib/dblink for remote database access (Joe Conway)

  • contrib/ora2pg Oracle conversion utility (Gilles Darold)

  • contrib/xml XML conversion utility (John Gray)

  • contrib/fulltextindex fixes (Christopher Kings-Lynne)

  • New contrib/fuzzystrmatch with levenshtein and metaphone, soundex merged (Joe Conway)

  • Add contrib/intarray boolean queries, binary search, fixes (Oleg Bartunov)

  • New pg_upgrade utility (Bruce)

  • Add new pg_resetxlog options (Bruce, Tom)

Previous article： Next article：