![]() That is, only a and c, but not b, are installed as a consequence of running: sudo apt-get install a c Concrete Example ![]() Then, since a's dependency on v is satisfied by c, b is not installed. In contrast, suppose you were to run this command instead of those: sudo apt-get install a c That is, all three packages were installed as a result of running: sudo apt-get install a Suppose you then run: sudo apt-get install cĪfter that, a, b, and c are all installed. ![]() Suppose without loss of generality that it picks b. But when all these conditions are the same, the decision will be made the same each time. Which one it picks is the result of a computation it performs that may be affected by what packages you have installed, what packages are available, what version of APT you are using, and how it is configured. Then APT will select b or c to satisfy the dependency. Now suppose you run: sudo apt-get install a Suppose b does not depend on c, c does not depend on b, and none of those packages are installed already. Suppose you wish to install package a, which depends on a virtual package v, and v can be satisfied by package b or package c and in no other way. More specifically, this happens because there can be more than one alternative that satisfies a dependency. I'm asking mostly dependencies-topo-sort related However, the effects can differ even when all operations succeed, due to different dependency resolution. The main practical difference is probably the one karel has described. Running 1 command that specifies N packages to be installed will sometimes have a different effect from running N commands each of which specifies 1 package to be installed, even when the actual packages specified are the same in each case. Yes, different packages can be selected to satisfy dependencies.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |