问题
I am working with jGit in Java and I have managed to clone an entire repository. But I can not find a way to download a single file inside a repository. I have tried:
- Change the URL specifying the path of the file.
- Change the URL by specifying a subdirectory.
Neither of them worked.
My code (clone whole repository) is as follows:
public File cloneRepository(String url, String path) throws GitAPIException {
File download = new File (path);
if (download.exists()){
System.out.println("Error");
return null;
}
else{
CloneCommand clone = Git.cloneRepository()
.setURI(url)
.setDirectory(download);
try (Git repositorio = clone.call()) {
return repositorio.getRepository().getWorkTree();
}
}
}
How could it be changed to download a single file?
回答1:
you just need to walk the file tree to find file you need Here is a code sample
// Clone remote repository
CloneCommand cloneCommand = Git.cloneRepository();
cloneCommand.setDirectory(localRepository.toFile());
cloneCommand.setBare(false);
cloneCommand.setURI(remoteRepository.toString());
cloneCommand.setTransportConfigCallback(transportConfigCallback);
Repository repository = cloneCommand.call().getRepository();
ObjectId ref = repository.resolve("your commit or branch");
// Resolve commit hash
RevWalk revWalk = new RevWalk(repository);
RevCommit revCommit = revWalk.parseCommit(ref);
Commit commit = new Commit(revCommit);
TreeWalk tree = new TreeWalk(repository);
tree.addTree(revCommit.getTree());
tree.setRecursive(true);
// Set path filter, to show files on matching path
tree.setFilter(PathFilter.create("path to file"));
try (ObjectReader objectReader = repository.newObjectReader()) {
while (tree.next()) {
String fileName = tree.getPathString();
byte[] fileBytes = objectReader.open(tree.getObjectId(0)).getBytes();
}
}
return paths;
来源:https://stackoverflow.com/questions/48619755/how-to-clone-a-single-file-with-jgit