SVN-based method for automatic statistics of configuration audit information

Abstract

The invention discloses an SVN-based method for automatic statistics of configuration audit information. The method comprises the steps as follows: 1) inquiring version repository information on an SVN server in an appointed statistical time period and selecting original journal files corresponding to all user change operations between the current statistical time period and the last statistical time period, and making a copy of the selected original journal files; 2) separately analyzing each copied original journal file to obtain change operation information including the account and time of the change operation, the file path of the change operation and the file operation state; 3) performing statistics on the change operation information obtained from each original journal file to obtain change operations and operation corresponding relations in the current statistical time period and storing the obtained information in a statistical database. The SVN-based method for automatic statistics of configuration audit information of the invention is easy to realize, wide in applied range, high in statistical efficiency and flexible in statistical pattern without depending on user permission.

Description

technical field [0001] The invention relates to the technical field of configuration management, in particular to a method for automatically counting configuration audit information based on SVN. Background technique [0002] Configuration management refers to a series of measures to control and regulate software products and their development process and life cycle through technical or administrative means. Get the precise configuration information of the product. An important part of configuration management is tracking who made each change, when and why, so that problems can be located quickly. [0003] At present, the configuration management of most software products is mainly implemented through Subversion (SVN for short). SVN will generate a version every time a change is submitted, and the records of changes are scattered in various log files. The current SVN-based configuration management audit method mainly uses the log export function provided by SVN itself to e...

AI Technical Summary

Problems solved by technology

For example, when the SVN repository creates a trunk folder, it only assigns read and write permissions to the creator user by default, while other users do not have read permissions to the trunk folder. If you use other users to apply for reading SVN logs, you cannot Get full SVN log

Usually, configuration management does not have a fixed administrator user, so it is impossible to obtain the full version of the log at any time;

[0005] 2) In actual application, the statistical workload is heavy and the statistical efficiency is low

For projects with strong confidentiality, users with global access are not allowed, so it is necessary to use multiple accounts with different permissions to perform multiple export logs to obtain complete SVN logs; for large-scale software (such as millions of lines of code) and For a large team (such as hundreds of users), when it is necessary to review and count the configuration management information of SVN, or to track the code submission status of a product in a certain period of time, it is only possible to manually export all SVN related log information, and then process it, which requires a lot of time and human resources; for logs with a long span, SVN needs a lot of time to extract and parse;

[0006] 3) The statistical method is fixed, and only one method of statistics can be performed each time according to the needs of the applicant. When the statistical applicant needs to update the statistical conditions, it is necessary to re-export a variety of qualified log files for statistical analysis

Images