Oracle环境中UNIX Shell脚本编程指南

需积分: 0 174 浏览量更新于2024-07-24 收藏 73KB PDF 举报

本文档深入探讨了在Oracle环境中进行UNIX Shell脚本编程的相关知识。作者William A. Ducat, 作为Ambassador, Inc. 的大使，以其丰富的经验，引领读者了解如何在UNIX系统中有效利用Shell脚本进行操作和管理Oracle环境。以下是文档的主要内容概要： 1. **介绍**：文章首先介绍了文档的目的，即为Oracle用户和系统管理员提供实用的Shell脚本编写技巧，以便在日常运维和自动化任务中提高效率。 2. **约定与规范**：文档概述了在编写Shell脚本时应遵循的最佳实践，包括变量命名、注释和代码组织方式，确保脚本的可读性和一致性。 3. **基本构建块**：这部分介绍了Shell编程的基本元素，如命令、变量、运算符和控制结构，这些都是脚本执行的核心组成部分。 4. **标准输入/输出/错误流（STDOUT, STDERR, STDIN）**：作者详细讲解了这些流在脚本中的作用，以及如何通过管道（`|`）和重定向（`>`和`>>`）来管理和处理它们。 5. **环境变量**：这部分讨论了如何使用和管理环境变量，以及它们如何影响脚本的行为和与Oracle系统的交互。 6. **脚本编写**：作者给出了创建和运行Shell脚本的基础知识，包括如何编写脚本、保存和执行它们，并强调了测试和调试的重要性。 7. **参数传递**：讨论了如何接收和处理命令行参数，这对于实现脚本的灵活性和可配置性至关重要。 8. **流程控制**：涵盖了条件语句（if...else...fi）、循环（for, while, until）等控制结构，这些是编写复杂逻辑脚本的关键。 9. **识别进程**：讲解了如何使用进程ID（PID）和其他标识符来跟踪和管理脚本运行的进程。 10. **实践示例**：文中包含多个实例，展示了实际场景下的密码管理、嵌入式SQL操作、自定义函数的实现等，帮助读者理解概念并掌握应用。 11. **密码管理**：介绍了一种密码设置脚本（`setup_passwords.ksh`），演示了如何安全地处理密码和敏感信息。 12. **嵌入式SQL**：展示了如何将SQL查询融入Shell脚本中，以简化数据库操作。 13. **嵌入式子程序（函数）**：讲解了如何定义和调用用户自定义函数，提升脚本的复用性和可维护性。 14. **参数传递**：进一步深入参数传递，讨论了不同类型的参数传递方式，以及如何处理参数的验证和错误处理。 15. **全局函数**：介绍了可以跨脚本使用的全局函数，这些函数有助于组织和封装常用功能。 16. **结论**：总结了全文的关键点，并强调了Shell脚本在Oracle环境中的重要性和持续学习的价值。 17. **附录**：包含了一些实用的代码片段和示例图，如将标准错误重定向到文件或屏幕、标准输入的处理等，供读者参考。通过阅读这份文档，读者不仅能掌握在Oracle环境中编写和管理Shell脚本的基础知识，还能学习到一些高级技巧，提升自己的技能水平。

UNIX Shell Scripting for Oracle Ducat

www.rmoug.org RMOUG Training Days 2001

STDIN

Standard input typically comes from one of three sources. The first is interactively from the keyboard, the second is from an

input file, and the third is from the output of another command. When writing a script to be run in batch mode, keyboard

entry is usually not an option. For instance, if a script is require to enter Sql*Plus and execute an update statement, how can it

be done? One possibility would be to add an “@” command to the startup of Sql*Plus, but maybe a single file is desired.

One answer is to have Sql*Plus take it’s input from the script instead of the keyboard. Redirecting STDIN can do this.

sqlplus scott/tiger –s <<!!

select * from global_name;

quit;

Figure 3 - Redirecting STDIN in a script

In this example, the “<<” redirector indicates that an alternate STDIN follows. In this case, “!!” is used to bracket the starting

and ending of the input. There is nothing special about the “!!” characters, and any bracketing characters can be used as long

as they follow the following rules; The first set of bracketing characters must immediately follow the “<<” redirector, and

nothing can follow the characters on the line. To end the standard input, the characters must exist on a line all by themselves.

The most common problem when coding these brackets occurs when spaces are entered before or after the bracketing

characters. When this happens, the redirection does not happen properly, and the results can be very confusing.

vi hint

I have never found a reason to end lines with spaces in a script, and the following command

can be used to remove all trailing spaces from a file when using vi: :1,$s/ *$//

The second method of changing standard input is to pipe “|” the results of one command into another command. For instance,

the sort command will take a file name as input and display the sorted results to the screen, but other methods are possible.

The following two commands accomplish the same results:

sort myfile

cat myfile | sort

In the first example, sort takes “myfile” as the input, but in the second example, the results of the cat command (which

simply displays a file) are sent as input to the sort command. While the second example may seem more complicated, it

opens up a whole world of possibilities. For instance, let’s assume we have a file containing the results of a series of Oracle

queries, and we want to display a sorted list of “ORA” errors contained in the file. In addition, we want to align the output so

each “ORA-“ string is indented in a manner that places the “O” in column 4. The following command would do this:

grep “ORA-“ myfile | sed –e ‘s/^β*//;s/^/βββ/’ | sort

Let’s take this command apart. The first portion is the grep. Oracle errors generally start with “ORA-“, and we use grep to

extract all of these lines from “myfile”. When Oracle generates the error messages, the resulting lines can be indented all over

the place. We want them to be consistent for display purposes. The sed command is used to take the results of the grep

command and format them. Sed is a stream editor, which applies VI like commands to each line of input. In our case, we are

going to apply two substitute commands (separated by the “;” character) to each line. The first substitution, (s/^β*//),

will find the beginning of line followed by any number of spaces, and remove the spaces. This will result in all the ORA-

lines being left justified. The second substitution will add 3 spaces to the beginning of each line. Next, the results are sent to

sort where they are displayed on the screen in sorted order. Of course, we could pipe the results to a file using the “>”

redirector, or the results could be displayed to the screen and sent to a file by piping the results to the “tee” command. The

options are really only bound by your imagination. In fact, UNIX commands can be thought of much like a box of

剩余21页未读，继续阅读

jinzhizhulove

粉丝: 0
资源: 3

Oracle环境中UNIX Shell脚本编程指南

Unix Oracle Shell Scripting

Unix.Shell编程

Unix_Shell_Programming

shell scripts PPT document

shell方式重启weblogic

Oracle Solaris 11.3 Managing Remote Systems-30

php.ini-development

Oracle专业人员的Shell脚本编程指南

Oracle DBA的Shell脚本实战

Oracle专家的Shell脚本编程指南

最新资源