Use zip2zip to copy zipfiles

ZipDelete() works by copying every non-deleted entry to a new zipfile.
Current implementation uses python's zipfile module to perform entry
copying, which is inefficient, as every entry must be decompressed and
then re-compressed. Instead, use zip2zip which avoid re-compression.

Improvement: deleting META/dynamic_partitions_info.txt from
raven-target_files-9465001.zip improved from 500+ seconds to 13 seconds.

Change-Id: I0548255bc29380303314763f6d81e74bf3dbb76e
This commit is contained in:
Kelvin Zhang
2023-02-15 16:52:46 -08:00
parent 19961b53c4
commit c8ff84b219

View File

@@ -302,6 +302,8 @@ def RunAndCheckOutput(args, verbose=None, **kwargs):
Raises:
ExternalError: On non-zero exit from the command.
"""
if verbose is None:
verbose = OPTIONS.verbose
proc = Run(args, verbose=verbose, **kwargs)
output, _ = proc.communicate()
if output is None:
@@ -2893,13 +2895,12 @@ def ZipDelete(zip_filename, entries, force=False):
fd, new_zipfile = tempfile.mkstemp(dir=os.path.dirname(zip_filename))
os.close(fd)
cmd = ["zip2zip", "-i", zip_filename, "-o", new_zipfile]
for entry in entries:
cmd.append("-x")
cmd.append(entry)
RunAndCheckOutput(cmd)
with zipfile.ZipFile(new_zipfile, 'w') as zout:
for item in zin.infolist():
if item.filename in entries:
continue
buffer = zin.read(item.filename)
zout.writestr(item, buffer)
os.replace(new_zipfile, zip_filename)