changeset 50512:a32d739b0ffb

store: make `walk` return an entry for phase if requested so

Instead of having dedicated code in the streamclone code, we should have the
store deal with advertising the data it contains.

--- a/hgext/remotefilelog/remotefilelogserver.py	Mon May 22 10:20:24 2023 +0100
+++ b/hgext/remotefilelog/remotefilelogserver.py	Sun May 21 02:15:04 2023 +0200
@@ -145,7 +145,7 @@
     )
 
     # don't clone filelogs to shallow clients
-    def _walkstreamfiles(orig, repo, matcher=None):
+    def _walkstreamfiles(orig, repo, matcher=None, phase=False):
         if state.shallowremote:
             # if we are shallow ourselves, stream our local commits
             if shallowutil.isenabled(repo):
@@ -200,7 +200,7 @@
                 _(b"Cannot clone from a shallow repo to a full repo.")
             )
         else:
-            for x in orig(repo, matcher):
+            for x in orig(repo, matcher, phase=phase):
                 yield x
 
     extensions.wrapfunction(streamclone, b'_walkstreamfiles', _walkstreamfiles)

--- a/mercurial/store.py	Mon May 22 10:20:24 2023 +0100
+++ b/mercurial/store.py	Sun May 21 02:15:04 2023 +0200
@@ -685,7 +685,7 @@
                     details=file_details,
                 )
 
-    def top_entries(self) -> Generator[BaseStoreEntry, None, None]:
+    def top_entries(self, phase=False) -> Generator[BaseStoreEntry, None, None]:
         files = reversed(self._walk(b'', False))
 
         changelogs = collections.defaultdict(dict)
@@ -725,11 +725,18 @@
                     target_id=b'',
                     details=file_details,
                 )
+        if phase and self.vfs.exists(b'phaseroots'):
+            yield SimpleStoreEntry(
+                entry_path=b'phaseroots',
+                is_volatile=True,
+            )
 
-    def walk(self, matcher=None) -> Generator[BaseStoreEntry, None, None]:
+    def walk(
+        self, matcher=None, phase=False
+    ) -> Generator[BaseStoreEntry, None, None]:
         """return files related to data storage (ie: revlogs)
 
-        yields (file_type, unencoded, size)
+        yields instance from BaseStoreEntry subclasses
 
         if a matcher is passed, storage files of only those tracked paths
         are passed with matches the matcher
@@ -737,7 +744,7 @@
         # yield data files first
         for x in self.data_entries(matcher):
             yield x
-        for x in self.top_entries():
+        for x in self.top_entries(phase=phase):
             yield x
 
     def copylist(self):

--- a/mercurial/streamclone.py	Mon May 22 10:20:24 2023 +0100
+++ b/mercurial/streamclone.py	Sun May 21 02:15:04 2023 +0200
@@ -241,8 +241,8 @@
 
 
 # This is it's own function so extensions can override it.
-def _walkstreamfiles(repo, matcher=None):
-    return repo.store.walk(matcher)
+def _walkstreamfiles(repo, matcher=None, phase=False):
+    return repo.store.walk(matcher, phase=phase)
 
 
 def generatev1(repo):
@@ -679,7 +679,8 @@
     if includes or excludes:
         matcher = narrowspec.match(repo.root, includes, excludes)
 
-    for entry in _walkstreamfiles(repo, matcher):
+    phase = not repo.publishing()
+    for entry in _walkstreamfiles(repo, matcher, phase=phase):
         for f in entry.files():
             file_size = f.file_size(repo.store.vfs)
             if file_size:
@@ -688,10 +689,6 @@
                     ft = _filefull
                 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
-            entries.append((_srcstore, name, _filefull, None))
     if includeobsmarkers and repo.svfs.exists(b'obsstore'):
         totalfilesize += repo.svfs.lstat(b'obsstore').st_size
         entries.append((_srcstore, b'obsstore', _filefull, None))