From de850738b2dadd8898dcdd19f97a2a1f4a48fea3 Mon Sep 17 00:00:00 2001
From: Slavek Kabrda <bkabrda@redhat.com>
Date: Nov 15 2018 07:17:07 +0000
Subject: Allow running 'git gc' explicitly after every object-adding git operation


---

diff --git a/doc/configuration.rst b/doc/configuration.rst
index 2467079..4355b16 100644
--- a/doc/configuration.rst
+++ b/doc/configuration.rst
@@ -1383,6 +1383,31 @@ information.
 Defaults to: ``None``
 
 
+GIT_GARBAGE_COLLECT
+~~~~~~~~~~~~~~~~~~~
+
+This configuration key allows for explicit running of ``git gc --auto``
+after every operation that adds new objects to any git repository -
+that is after pushing and merging. The reason for having this functionality
+in Pagure is that gc is not guaranteed to be run by git after every
+object-adding operation.
+
+The garbage collection run by Pagure will respect git settings, so you
+can tweak ``gc.auto`` and ``gc.autoPackLimit`` to your liking
+and that will have immediate effect on the task that runs the garbage
+collection. These values can be configured system-wide in ``/etc/gitconfig``.
+See https://git-scm.com/docs/git-gc#git-gc---auto for more details.
+
+This is especially useful if repositories are stored on NFS (or similar
+network storage), where file metadata access is expensive - having unpacked
+objects in repositories requires *a lot* of metadata reads.
+
+Note that the garbage collection is only run on repos that are not on
+repoSpanner.
+
+Defaults to: ``False``
+
+
 CELERY_CONFIG
 ~~~~~~~~~~~~~
 
diff --git a/pagure/default_config.py b/pagure/default_config.py
index 940eb68..960e509 100644
--- a/pagure/default_config.py
+++ b/pagure/default_config.py
@@ -206,6 +206,13 @@ GL_RC = None
 GL_BINDIR = None
 
 
+# Whether or not to run "git gc --auto" after every change to a project
+# This will only run for projects not on repospanner and will use
+# default git config values
+# See https://git-scm.com/docs/git-gc#git-gc---auto for more details
+GIT_GARBAGE_COLLECT = False
+
+
 # SMTP settings
 SMTP_SERVER = "localhost"
 SMTP_PORT = 25
diff --git a/pagure/hooks/default.py b/pagure/hooks/default.py
index 1623126..b492b1f 100644
--- a/pagure/hooks/default.py
+++ b/pagure/hooks/default.py
@@ -275,6 +275,11 @@ class DefaultRunner(BaseRunner):
             parent.namespace,
             parent.user.user if parent.is_fork else None,
         )
+        if not project.is_on_repospanner and \
+                _config.get("GIT_GARBAGE_COLLECT", False):
+            pagure.lib.tasks.git_garbage_collect.delay(
+                project.repopath("main")
+            )
 
 
 class Default(BaseHook):
diff --git a/pagure/lib/tasks.py b/pagure/lib/tasks.py
index 0977ca4..feb6b6e 100644
--- a/pagure/lib/tasks.py
+++ b/pagure/lib/tasks.py
@@ -1151,3 +1151,15 @@ def pull_request_ready_branch(self, session, namespace, name, user):
             )
             del (branches[pr.branch_from])
     return {"new_branch": branches, "branch_w_pr": branches_pr}
+
+
+@conn.task(queue=pagure_config.get("MEDIUM_CELERY_QUEUE", None), bind=True)
+@pagure_task
+def git_garbage_collect(self, session, repopath):
+    # libgit2 doesn't support "git gc" and probably never will:
+    # https://github.com/libgit2/libgit2/issues/3247
+    _log.info("Running 'git gc --auto' for repo %s", repopath)
+    subprocess.check_output(
+        ["git", "gc", "--auto", "-q"],
+        cwd=repopath,
+    )