GerhardHoffmann/ATBUpdateTool
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Replaced 'git clone' with 'git clone --filter=blob:none' to speed

Browse Source 
up cloning of customer-repository.
This commit is contained in:
Gerhard Hoffmann 2023-10-18 11:36:13 +02:00
parent 258d883a51
commit 22f25e5251
1 changed files with 50 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

53
git/git_client.cpp
View File

@ -26,12 +26,54 @@ GitClient::GitClient(QString const &customerNrStr,
}
bool GitClient::gitCloneCustomerRepository() {
QString gitCommand("git clone ");
/* Blobless clone
==============
When using the --filter=blob:none option, the initial git clone will
download all reachable commits and trees, and only download the blobs
for commits when you do a git checkout. This includes the first checkout
inside the git clone operation.
The important thing to notice is that we have a copy of every blob at
HEAD but the blobs in the history are not present. If your repository
has a deep history full of large blobs, then this option can
significantly reduce your git clone times. The commit and tree data is
still present, so any subsequent git checkout only needs to download
the missing blobs. The Git client knows how to batch these requests to
ask the server only for the missing blobs.
Further, when running git fetch in a blobless clone, the server only
sends the new commits and trees. The new blobs are downloaded only
after a git checkout. Note that git pull runs git fetch and then git
merge, so it will download the necessary blobs during the git merge
command.
When using a blobless clone, you will trigger a blob download whenever
you need the contents of a file, but you will not need one if you only
need the OID (object-id) of a file. This means that git log can detect
which commits changed a given path without needing to download extra
data.
This means that blobless clones can perform commands like git
merge-base, git log, or even git log -- <path> with the same performance
as a full clone.
Commands like git diff or git blame <path> require the contents of the
paths to compute diffs, so these will trigger blob downloads the first
time they are run. However, the good news is that after that you will
have those blobs in your repository and do not need to download them a
second time. Most developers only need to run git blame on a small
number of files, so this tradeoff of a slightly slower git blame command
is worth the faster clone and fetch times.
Note: git v2.18 does not support treeless clones: --filter=tree:0.
*/
QString gitCommand("git clone --filter=blob:none ");
gitCommand += m_repositoryPath;
Command c(gitCommand);
qInfo() << "IN CURRENT WD" << m_workingDirectory
<< "CLONE" << m_repositoryPath << "...";
<< "CLONE VIA COMMAND" << gitCommand;
if (c.execute(m_workingDirectory)) { // execute the command in wd
QString const result = c.getCommandResult();
@ -47,8 +89,13 @@ bool GitClient::gitCloneCustomerRepository() {
}
}
}
Utils::printCriticalErrorMsg(
QString("ERROR CLONE RESULT HAS WRONG FORMAT. rcc=%1 CLONE_RESULT=%2")
.arg(re.captureCount())
.arg(result));
return false;
}
Utils::printCriticalErrorMsg(QString("ERROR CLONE RESULT HAS WRONG FORMAT. CLONE_RESULT=") + result);
Utils::printCriticalErrorMsg("ERROR CLONE RESULT IS EMPTY");
}
return false;
}