I create branches for each new function. Then I combine them in the trunk after sending them to the test site. This means that the trunk is unstable.
I wonder if there is a better approach.
EDIT After reading the comment, I understand that I should have indicated that this is a web application. So the website for testing each branch seems a little difficult to maintain.
I am using the following:
branches: not always compiled, not always stable
trunk: always compiled, not always stable
tags: always compiled, always stable
live release.
, ,
,
, , . , , , ( , ) .
. . , , . - , . . .
/ /
, "" ... .
"", , , - . , , .
- , . , , , , . , , , . Else, , , , . , , - , - .
, , .
? , , , ,
? ?
?
Otherwise, a distributed version control system such as Mercurial might be a way forward. With this, you can define new repositories for each function, as well as for testing, and click on the trunk repository.
Source: https://habr.com/ru/post/1717235/More articles:using cookies with twisted.web.client - pythonHow do I go through a hidden div in Prototype, like in jquery? - javascriptVS2008 on Win7 64-bit: debugging Windows service - .netPass Delphi to an external Delphi function from C # - setGood design project J2ME - designPeriod-supported date and time library for C # / VB.NET? - c #Why jQuery.browser.version returns 1.9.1.2 for Firefox 3.5.2 - jqueryUsing AcceptChanges with SQL Server - c #Get CSS style from PHP - domSubmitting Object Content Through RIA Services - .netAll Articles