Repository.scan(): avoid iterating same repo segments/objects repeatedly

segments: avoid some listdir()
objects: avoid repeatedly re-reading object headers / seeking in segment

Fixes #1610.

src/borg/repository.py

@@ -775,12 +775,10 @@ class Repository:
             self.index = self.open_index(transaction_id)
         at_start = marker is None
         # smallest valid seg is <uint32> 0, smallest valid offs is <uint32> 8
-        marker_segment, marker_offset = (0, 0) if at_start else self.index[marker]
+        start_segment, start_offset = (0, 0) if at_start else self.index[marker]
         result = []
-        for segment, filename in self.io.segment_iterator():
-            if segment < marker_segment:
-                continue
-            obj_iterator = self.io.iter_objects(segment, read_data=False, include_data=False)
+        for segment, filename in self.io.segment_iterator(start_segment):
+            obj_iterator = self.io.iter_objects(segment, start_offset, read_data=False, include_data=False)
             while True:
                 try:
                     tag, id, offset, size = next(obj_iterator)
@@ -788,7 +786,11 @@ class Repository:
                     # either end-of-segment or an error - we can not seek to objects at
                     # higher offsets than one that has an error in the header fields
                     break
-                if segment == marker_segment and offset <= marker_offset:
+                if start_offset > 0:
+                    # we are using a marker and the marker points to the last object we have already
+                    # returned in the previous scan() call - thus, we need to skip this one object.
+                    # also, for the next segment, we need to start at offset 0.
+                    start_offset = 0
                     continue
                 if tag == TAG_PUT and (segment, offset) == self.index.get(id):
                     # we have found an existing and current object
@@ -886,14 +888,25 @@ class LoggedIO:
             os.posix_fadvise(fd.fileno(), 0, 0, os.POSIX_FADV_DONTNEED)
         fd.close()
 
-    def segment_iterator(self, reverse=False):
+    def segment_iterator(self, segment=None, reverse=False):
+        if segment is None:
+            segment = 0 if not reverse else 2 ** 32 - 1
         data_path = os.path.join(self.path, 'data')
-        dirs = sorted((dir for dir in os.listdir(data_path) if dir.isdigit()), key=int, reverse=reverse)
+        start_segment_dir = segment // self.segments_per_dir
+        dirs = os.listdir(data_path)
+        if not reverse:
+            dirs = [dir for dir in dirs if dir.isdigit() and int(dir) >= start_segment_dir]
+        else:
+            dirs = [dir for dir in dirs if dir.isdigit() and int(dir) <= start_segment_dir]
+        dirs = sorted(dirs, key=int, reverse=reverse)
         for dir in dirs:
             filenames = os.listdir(os.path.join(data_path, dir))
-            sorted_filenames = sorted((filename for filename in filenames
-                                       if filename.isdigit()), key=int, reverse=reverse)
-            for filename in sorted_filenames:
+            if not reverse:
+                filenames = [filename for filename in filenames if filename.isdigit() and int(filename) >= segment]
+            else:
+                filenames = [filename for filename in filenames if filename.isdigit() and int(filename) <= segment]
+            filenames = sorted(filenames, key=int, reverse=reverse)
+            for filename in filenames:
                 yield int(filename), os.path.join(data_path, dir, filename)
 
     def get_latest_segment(self):
@@ -999,7 +1012,7 @@ class LoggedIO:
     def segment_size(self, segment):
         return os.path.getsize(self.segment_filename(segment))
 
-    def iter_objects(self, segment, include_data=False, read_data=True):
+    def iter_objects(self, segment, offset=0, include_data=False, read_data=True):
         """
         Return object iterator for *segment*.
 
@@ -1009,10 +1022,14 @@ class LoggedIO:
         The iterator returns four-tuples of (tag, key, offset, data|size).
         """
         fd = self.get_fd(segment)
-        fd.seek(0)
-        if fd.read(MAGIC_LEN) != MAGIC:
-            raise IntegrityError('Invalid segment magic [segment {}, offset {}]'.format(segment, 0))
-        offset = MAGIC_LEN
+        fd.seek(offset)
+        if offset == 0:
+            # we are touching this segment for the first time, check the MAGIC.
+            # Repository.scan() calls us with segment > 0 when it continues an ongoing iteration
+            # from a marker position - but then we have checked the magic before already.
+            if fd.read(MAGIC_LEN) != MAGIC:
+                raise IntegrityError('Invalid segment magic [segment {}, offset {}]'.format(segment, 0))
+            offset = MAGIC_LEN
         header = fd.read(self.header_fmt.size)
         while header:
             size, tag, key, data = self._read(fd, self.header_fmt, header, segment, offset,