store: lazily get file size on demand for the fncache case
We don't have the information in the first place, so we can avoid querying the
file system inconditionnaly for use case we don't needs it.
This change requires the StoreFile class to be passed a vfs to retrieve the
file_size if needed.
In the non-fncache case, we already have the information from file system
walking, so we keep it and use it.
--- a/mercurial/store.py Mon May 15 08:57:45 2023 +0200
+++ b/mercurial/store.py Mon May 15 08:58:01 2023 +0200
@@ -520,9 +520,17 @@
"""a file matching an entry"""
unencoded_path = attr.ib()
- file_size = attr.ib()
+ _file_size = attr.ib(default=False)
is_volatile = attr.ib(default=False)
+ def file_size(self, vfs):
+ if self._file_size is not None:
+ return self._file_size
+ try:
+ return vfs.stat(self.unencoded_path).st_size
+ except FileNotFoundError:
+ return 0
+
class basicstore:
'''base class for local repository stores'''
@@ -900,16 +908,12 @@
# However the fncache might contains such file added by
# previous version of Mercurial.
continue
- try:
- yield RevlogStoreEntry(
- unencoded_path=f,
- revlog_type=FILEFLAGS_FILELOG,
- is_revlog_main=bool(t & FILEFLAGS_REVLOG_MAIN),
- is_volatile=bool(t & FILEFLAGS_VOLATILE),
- file_size=self.getsize(ef),
- )
- except FileNotFoundError:
- pass
+ yield RevlogStoreEntry(
+ unencoded_path=f,
+ revlog_type=FILEFLAGS_FILELOG,
+ is_revlog_main=bool(t & FILEFLAGS_REVLOG_MAIN),
+ is_volatile=bool(t & FILEFLAGS_VOLATILE),
+ )
def copylist(self):
d = (
--- a/mercurial/streamclone.py Mon May 15 08:57:45 2023 +0200
+++ b/mercurial/streamclone.py Mon May 15 08:58:01 2023 +0200
@@ -271,9 +271,10 @@
repo.ui.debug(b'scanning\n')
for entry in _walkstreamfiles(repo):
for f in entry.files():
- if f.file_size:
- entries.append((f.unencoded_path, f.file_size))
- total_bytes += f.file_size
+ file_size = f.file_size(repo.store.vfs)
+ if file_size:
+ entries.append((f.unencoded_path, file_size))
+ total_bytes += file_size
_test_sync_point_walk_1(repo)
_test_sync_point_walk_2(repo)
@@ -680,12 +681,13 @@
for entry in _walkstreamfiles(repo, matcher):
for f in entry.files():
- if f.file_size:
+ file_size = f.file_size(repo.store.vfs)
+ if file_size:
ft = _fileappend
if f.is_volatile:
ft = _filefull
- entries.append((_srcstore, f.unencoded_path, ft, f.file_size))
- totalfilesize += f.file_size
+ entries.append((_srcstore, f.unencoded_path, ft, file_size))
+ totalfilesize += file_size
for name in _walkstreamfullstorefiles(repo):
if repo.svfs.exists(name):
totalfilesize += repo.svfs.lstat(name).st_size
--- a/mercurial/verify.py Mon May 15 08:57:45 2023 +0200
+++ b/mercurial/verify.py Mon May 15 08:58:01 2023 +0200
@@ -410,7 +410,7 @@
for entry in repo.store.datafiles(undecodable=undecodable):
for file_ in entry.files():
f = file_.unencoded_path
- size = file_.file_size
+ size = file_.file_size(repo.store.vfs)
if (size > 0 or not revlogv1) and f.startswith(b'meta/'):
storefiles.add(_normpath(f))
subdirs.add(os.path.dirname(f))
@@ -477,7 +477,7 @@
undecodable = []
for entry in repo.store.datafiles(undecodable=undecodable):
for file_ in entry.files():
- size = file_.file_size
+ size = file_.file_size(repo.store.vfs)
f = file_.unencoded_path
if (size > 0 or not revlogv1) and f.startswith(b'data/'):
storefiles.add(_normpath(f))