I have archive.zip with two files: hello.txt and world.txt
I want to overwrite hello.txt file with new one with that code:
import zipfile
z = zipfile.ZipFile('archive.zip','a')
z.write('hello.txt')
z.close()
but it won't overwrite file, somehow it creates another instance of hello.txt — take a look at winzip screenshot:
Since there is no smth like zipfile.remove(), what's the best way to handle this problem?
There's no way to do that with python zipfile module. You have to create a new zip file and recompress everything again from the first file, plus the new modified file.
Below is some code to do just that. But note that it isn't efficient, since it decompresses and then recompresses all data.
import tempfile
import zipfile
import shutil
import os
def remove_from_zip(zipfname, *filenames):
tempdir = tempfile.mkdtemp()
try:
tempname = os.path.join(tempdir, 'new.zip')
with zipfile.ZipFile(zipfname, 'r') as zipread:
with zipfile.ZipFile(tempname, 'w') as zipwrite:
for item in zipread.infolist():
if item.filename not in filenames:
data = zipread.read(item.filename)
zipwrite.writestr(item, data)
shutil.move(tempname, zipfname)
finally:
shutil.rmtree(tempdir)
Usage:
remove_from_zip('archive.zip', 'hello.txt')
with zipfile.ZipFile('archive.zip', 'a') as z:
z.write('hello.txt')
Building on nosklo's answer.
UpdateableZipFile A class that inherits from ZipFile, maintians the same interface but adds the ability to overwrite files (via writestr or write) and removing files.
import os
import shutil
import tempfile
from zipfile import ZipFile, ZIP_STORED, ZipInfo
class UpdateableZipFile(ZipFile):
"""
Add delete (via remove_file) and update (via writestr and write methods)
To enable update features use UpdateableZipFile with the 'with statement',
Upon __exit__ (if updates were applied) a new zip file will override the exiting one with the updates
"""
class DeleteMarker(object):
pass
def __init__(self, file, mode="r", compression=ZIP_STORED, allowZip64=False):
# Init base
super(UpdateableZipFile, self).__init__(file, mode=mode,
compression=compression,
allowZip64=allowZip64)
# track file to override in zip
self._replace = {}
# Whether the with statement was called
self._allow_updates = False
def writestr(self, zinfo_or_arcname, bytes, compress_type=None):
if isinstance(zinfo_or_arcname, ZipInfo):
name = zinfo_or_arcname.filename
else:
name = zinfo_or_arcname
# If the file exits, and needs to be overridden,
# mark the entry, and create a temp-file for it
# we allow this only if the with statement is used
if self._allow_updates and name in self.namelist():
temp_file = self._replace[name] = self._replace.get(name,
tempfile.TemporaryFile())
temp_file.write(bytes)
# Otherwise just act normally
else:
super(UpdateableZipFile, self).writestr(zinfo_or_arcname,
bytes, compress_type=compress_type)
def write(self, filename, arcname=None, compress_type=None):
arcname = arcname or filename
# If the file exits, and needs to be overridden,
# mark the entry, and create a temp-file for it
# we allow this only if the with statement is used
if self._allow_updates and arcname in self.namelist():
temp_file = self._replace[arcname] = self._replace.get(arcname,
tempfile.TemporaryFile())
with open(filename, "rb") as source:
shutil.copyfileobj(source, temp_file)
# Otherwise just act normally
else:
super(UpdateableZipFile, self).write(filename,
arcname=arcname, compress_type=compress_type)
def __enter__(self):
# Allow updates
self._allow_updates = True
return self
def __exit__(self, exc_type, exc_val, exc_tb):
# call base to close zip file, organically
try:
super(UpdateableZipFile, self).__exit__(exc_type, exc_val, exc_tb)
if len(self._replace) > 0:
self._rebuild_zip()
finally:
# In case rebuild zip failed,
# be sure to still release all the temp files
self._close_all_temp_files()
self._allow_updates = False
def _close_all_temp_files(self):
for temp_file in self._replace.itervalues():
if hasattr(temp_file, 'close'):
temp_file.close()
def remove_file(self, path):
self._replace[path] = self.DeleteMarker()
def _rebuild_zip(self):
tempdir = tempfile.mkdtemp()
try:
temp_zip_path = os.path.join(tempdir, 'new.zip')
with ZipFile(self.filename, 'r') as zip_read:
# Create new zip with assigned properties
with ZipFile(temp_zip_path, 'w', compression=self.compression,
allowZip64=self._allowZip64) as zip_write:
for item in zip_read.infolist():
# Check if the file should be replaced / or deleted
replacement = self._replace.get(item.filename, None)
# If marked for deletion, do not copy file to new zipfile
if isinstance(replacement, self.DeleteMarker):
del self._replace[item.filename]
continue
# If marked for replacement, copy temp_file, instead of old file
elif replacement is not None:
del self._replace[item.filename]
# Write replacement to archive,
# and then close it (deleting the temp file)
replacement.seek(0)
data = replacement.read()
replacement.close()
else:
data = zip_read.read(item.filename)
zip_write.writestr(item, data)
# Override the archive with the updated one
shutil.move(temp_zip_path, self.filename)
finally:
shutil.rmtree(tempdir)
usage example:
with UpdateableZipFile("C:\Temp\Test2.docx", "a") as o:
# Overwrite a file with a string
o.writestr("word/document.xml", "Some data")
# exclude an exiting file from the zip
o.remove_file("word/fontTable.xml")
# Write a new file (with no conflict) to the zp
o.writestr("new_file", "more data")
# Overwrite a file with a file
o.write(r"C:\Temp\example.png", "word/settings.xml")
Based on this answer here's a quick and dirty way to monkey patch stock zipfile to support file deletion (while we waiting for it being accepted for python:main):
from zipfile import ZipFile, ZipInfo
from operator import attrgetter
import functools
def enable_zip_remove(func):
def _zipfile_remove_member(self, member):
# get a sorted filelist by header offset, in case the dir order
# doesn't match the actual entry order
fp = self.fp
entry_offset = 0
filelist = sorted(self.filelist, key=attrgetter('header_offset'))
for i in range(len(filelist)):
info = filelist[i]
# find the target member
if info.header_offset < member.header_offset:
continue
# get the total size of the entry
entry_size = None
if i == len(filelist) - 1:
entry_size = self.start_dir - info.header_offset
else:
entry_size = filelist[i + 1].header_offset - info.header_offset
# found the member, set the entry offset
if member == info:
entry_offset = entry_size
continue
# Move entry
# read the actual entry data
fp.seek(info.header_offset)
entry_data = fp.read(entry_size)
# update the header
info.header_offset -= entry_offset
# write the entry to the new position
fp.seek(info.header_offset)
fp.write(entry_data)
fp.flush()
# update state
self.start_dir -= entry_offset
self.filelist.remove(member)
del self.NameToInfo[member.filename]
self._didModify = True
# seek to the start of the central dir
fp.seek(self.start_dir)
def zipfile_remove(self, member):
"""Remove a file from the archive. The archive must be open with mode 'a'"""
if self.mode != 'a':
raise RuntimeError("remove() requires mode 'a'")
if not self.fp:
raise ValueError(
"Attempt to write to ZIP archive that was already closed")
if self._writing:
raise ValueError(
"Can't write to ZIP archive while an open writing handle exists."
)
# Make sure we have an info object
if isinstance(member, ZipInfo):
# 'member' is already an info object
zinfo = member
else:
# get the info object
zinfo = self.getinfo(member)
return self._zipfile_remove_member(zinfo)
#functools.wraps(func)
def wrapper(*args, **kwargs):
if not hasattr(ZipFile, "remove"):
setattr(ZipFile, "_zipfile_remove_member", _zipfile_remove_member)
setattr(ZipFile, "remove", zipfile_remove)
return func(*args, **kwargs)
return wrapper
Usage:
#enable_zip_remove
def replace_zip_file():
with ZipFile("archive.zip", "a") as z:
z.remove("hello.txt")
z.write("hello.txt")
P.S. NSFW
My solution is similar to the other answers but uses SQLite to manage the intermediate files and provides __getitem__, __setitem__ and __delitem__ for an easy interface.
By default the db is in-memory but you can provide a temp file path if you have a zip larger than available memory.
And of course SQLite is built into Python and faster than the file system
import sqlite3
import subprocess
import zipfile
from pathlib import Path
from sql import CREATE_TABLE, DELETE_FILE, INSERT_FILE, SELECT_CONTENT
class EditableZip:
"""Intended to make editing files inside zip archive easy, this class is capable of loading files
from a zip file into a sqlite database, facilitates editing/removing/adding files, and saving
to a zip.
The database can be in-memory (default) or in a temporary on disk file if
temp_db_path is provided.
If an on-disk file is used, EditableZip.close can be called to remove the file or EditableZip
can be used as a context manager.
If auto_save is set to True and an initial zip_path was provided then the file will
be overwritten when EditableZip closes. If you wish to save to a different file,
or no zip_path is used in instantiation, auto_save can take a file path.
Files can be added by item assignment
with EditableZip(auto_save="example.zip") as ez:
ez["thing.txt"] = "stuff"
# empty dir
ez["empty/"] = None
Assignment accepts Non-text files as bytes.
EditableZip is subscriptable. If the subscript is a path in the db, the data will be returned.
EditableZip.files can be used to iterate over files in the db.
"""
def __init__(
self,
zip_path: None | str | Path = None,
temp_db_path: None | Path = None,
auto_save: bool | str | Path = False,
):
self.temp_db_path, self.auto_save, self.file_path = (
temp_db_path,
auto_save,
zip_path,
)
self.db = sqlite3.connect(
str(temp_db_path if temp_db_path is not None else ":memory:")
)
self.db.execute(CREATE_TABLE)
if self.file_path:
self.load(self.file_path)
#property
def files(self):
"Returns a generator of all file paths in the database."
try:
return (
i[0] for i in self.db.execute("SELECT file_path FROM files").fetchall()
)
except TypeError:
return None
def load(self, zip_path: str | Path) -> None:
"Add all files from zip at zip_path to db."
with zipfile.ZipFile(zip_path, mode="r") as archive:
for item in archive.infolist():
self[item.filename] = (
None if item.filename[-1] == "/" else archive.read(item)
)
def save(self, zip_path: None | str | Path) -> Path:
"Save all files from db to zip at zip_path."
zip_path = self.file_path if zip_path is None else zip_path
with zipfile.ZipFile(zip_path, "w") as archive:
for file in self.files:
if file_data := self.fetch(file):
archive.writestr(file, file_data)
else:
archive.writestr(zipfile.ZipInfo(file), "")
return zip_path
def close(self):
"Auto save if applicable and close + remove db."
if self.auto_save:
self.save(
zip_path=self.auto_save
if isinstance(self.auto_save, (str, Path))
else None
)
self.db.close()
if isinstance(self.temp_db_path, Path):
self.temp_db_path.unlink(missing_ok=True)
def fetch(self, file_path: str) -> bytes:
"Get content of db file for file_path."
try:
return self.db.execute(SELECT_CONTENT, {"file_path": file_path}).fetchone()[
0
]
except TypeError:
return None
def __getitem__(self, key):
result = self.fetch(key)
try:
return result.decode("utf-8")
except AttributeError:
return result
def __setitem__(self, file_path, content: str | bytes):
if isinstance(content, str):
content = content.encode("utf-8")
self.db.execute(
INSERT_FILE,
{"file_path": file_path, "file_content": content},
)
def __delitem__(self, file_path):
self.db.execute(DELETE_FILE, {"file_path": file_path})
def __enter__(self):
return self
def __exit__(self, exc_type, exc_value, traceback):
self.close()
if __name__ == "__main__":
# A use case: editing epub files.
# File source:
# https://archiveofourown.org/downloads/13795605/Victoria%20Potter%20and%20the.epub?updated_at=1650231615
file_path = Path("Victoria Potter and the.epub")
new_file = (file_path.parent / (file_path.stem + "- lowercase")).with_suffix(
file_path.suffix
)
# Create a copy of the epub with all letters lowercase
with EditableZip(zip_path=file_path, auto_save=new_file) as ez:
for file in ez.files:
if Path(file).suffix in [".html", ".xhtml"]:
ez[file] = ez[file].lower()
Reference: Delete file from zipfile with the ZipFile Module
In short,
You can take the code from https://github.com/python/cpython/blob/659eb048cc9cac73c46349eb29845bc5cd630f09/Lib/zipfile.py and create a separate file from it. After that just reference it from your project instead of built-in python library: import myproject.zipfile as zipfile.
Usage:
with zipfile.ZipFile(f"archive.zip", "a") as z:
z.remove(f"firstfile.txt")
I am trying to open a file in a class and close it on exit in this manner.
class PlanetaryImage(object):
#classmethod
def open(cls, filename):
with open(filename, 'rb') as fp:
return cls(fp, filename)
def __init__(self, stream, filename=None, memory_layout='DISK'):
self.filename = filename
self._parse_data(stream)
def _parse_data(self, stream):
data_stream = stream
try:
if self.data_filename is not None:
dirpath = os.path.dirname(self.filename)
data_file = os.path.abspath(
os.path.join(dirpath, self.data_filename))
data_stream = open(data_file, 'rb')
data_stream.seek(self.start_byte)
if self.format in self.BAND_STORAGE_TYPE:
return getattr(self, self.BAND_STORAGE_TYPE[self.format])(data_stream)
raise Exception('Unkown format (%s)' % self.format)
finally:
data_stream.close()
There are certain cases where I am having to use open one more file in _parse_data function. I wanted to use with but the if statements make it difficult. Any suggestions on how to make the try section more pythonic.
There's no reason for _parse_data to try to open a file. It should be the caller's responsibility to either use PlanetaryImage.open with a file name or to provide an open file handle to __init__. _parse_data should do just one thing: parse the data from its stream argument.
class PlanetaryImage(object):
#classmethod
def open(cls, filename):
with open(filename, 'rb') as fp:
return cls(fp, filename)
def __init__(self, stream, memory_layout='DISK'):
self._parse_data(stream)
def _parse_data(self, data_stream):
try:
data_stream.seek(self.start_byte)
if self.format in self.BAND_STORAGE_TYPE:
return getattr(self, self.BAND_STORAGE_TYPE[self.format])(data_stream)
raise Exception('Unkown format (%s)' % self.format)
finally:
data_stream.close()
Now, there are simply two options for using the class:
with open(filename, 'rb') as fp:
x = PlanetaryImage(fp)
...
or
x = PlanetaryImage(filename)
....
I am trying to copy my console log to a file when running a script utilizing a snippet found on this link. I tried to customize it by adding a timestamp utilizing strftime from the time module, but the snippet now adds a timestamp to both the start of a new row and the end:
2014-12-10 12:15:35: Working on local page 12014-12-10 12:15:35:
What did I do wrong? How would I fix this so that the timestamp is only shown at the start of a newline?
from time import strftime
class copyConsoleToFile(object):
""" Enables logging of console output to a file, use
>> tlogger = copyConsoleToFile('logfile.txt', 'w')
at the start of the code to start logging.
"""
def __init__(self, name, mode):
self.file = open(name, mode)
self.stdout = sys.stdout
sys.stdout = self
def close(self):
if self.stdout is not None:
sys.stdout = self.stdout
self.stdout = None
if self.file is not None:
self.file.close()
self.file = None
def write(self, data):
self.file.write(strftime("%Y-%m-%d %H:%M:%S") + ': ' + data)
self.stdout.write(data)
def flush(self):
self.file.flush()
self.stdout.flush()
def __del__(self):
self.close()
As stated in the comments, it seems that a sub-process adds an extra write to each line. I suggest to remove the non-useful extra characters:
with open('logfile.txt','r+') as fopen:
string = ""
for line in fopen.readlines():
string = string + line[:-23] + "\n"
with open('logfile.txt','w') as fopen:
fopen.write(string)
The code was adapted from: https://stackoverflow.com/a/45594783/1751393
After reading some texts regarding creation of files under python, i decided to create this class which creates a new file on a directory, and creating a backup on the other directory if the file already exists (and if it's older than x hours )
The main reason i opened this question is to know if this is a correct way to write a class using try/except correctly, because actually i'm getting a little confused about the preference of using try/except instead if/elses.
Bellow, the working example:
import os
import datetime
class CreateXML():
def __init__(self, path, filename):
self.path = path
self.bkp_path = "%s\\backup" % path
self.filename = filename
self.bkp_file = "%s.previous" % filename
self.create_check = datetime.datetime.now()-datetime.timedelta(hours=-8)
#staticmethod
def create_dir(path):
try:
os.makedirs(path)
return True
except:
return False
#staticmethod
def file_check(file):
try:
open(file)
return True
except:
return False
def create_file(self, target_dir, target_file):
try:
target = "%s\\%s" % (target_dir, target_file)
open(target, 'w')
except:
return False
def start_creation(self):
try:
# Check if file exists
if self.file_check("%s\\%s" % (self.path, self.filename)):
self.create_dir(self.bkp_path)
creation = os.path.getmtime("%s\\%s" % (self.path, self.filename))
fcdata = datetime.datetime.fromtimestamp(creation)
# File exists and its older than 8 hours
if fcdata < self.create_check:
bkp_file_path = "%s\\%s " % (self.bkp_path, self.bkp_file)
new_file_path = "%s\\%s " % (self.path, self.filename)
# If backup file exists, erase current backup file
# Move existing file to backup and create new file.
if self.file_check("%s\\%s" % (self.bkp_path, self.bkp_file)):
os.remove(bkp_file_path)
os.rename(new_file_path, bkp_file_path)
self.create_file(self.bkp_path, self.bkp_file)
#No backup file, create new one.
else:
self.create_file(self.bkp_path, self.bkp_file)
else:
# Fresh creation
self.create_dir(self.path)
self.create_file(self.path, self.filename)
except OSError, e:
print e
if __name__ == '__main__':
path = 'c:\\tempdata'
filename = 'somefile.txt'
cx = CreateXML(path, filename)
cx.start_creation()
So, basically the real question here is:
-With the example above, the usage of try/except is correct?
-It's correct to perform the validations using try/except to check if file or directory allready exists? instead using a simplified version like this one:
import os
# Simple method of doing it
path = 'c:\\tempdata'
filename = 'somefile.txt'
bkp_path = 'c:\\tempdata\\backup'
bkp_file = 'somefile.txt.bkp'
new_file_path = "%s\\%s" % (path, filename)
bkp_file_path = "%s\\%s" % (bkp_path, bkp_file)
if not os.path.exists(path):
print "create path"
os.makedirs(bkp_path)
if not os.path.isfile(new_file_path):
print "create new file"
open(new_file_path, 'w')
else:
print"file exists, moving to backup folder"
#check if backup file exists
if not os.path.isfile(bkp_file_path):
print "New backup file created"
open(bkp_file_path, 'w')
else:
print "backup exists, removing backup, backup the current, and creating newfile"
os.remove(bkp_file_path)
os.rename(new_file_path, bkp_file_path)
open(bkp_file_path, 'w')
-If the usage of try/except is correct, its recomended write an a big class to create a file if it's possible to write a short version of it?
Please do not close this tread, since i'm really confused about what is the "most correct pythonic way to do it".
Thanks in advance.
class Downloader(threading.Thread):
def __init__(self, priority_level, output_function):
self.IDs = self.load_IDs(priority_level)
self.sleep_interval = self.gen_sleep(priority_level)
self.output = output_function
self.name = '[Downloader::%s]'%(str(priority_level))
self.output('[Downloader] New downloader created (prio: %s)!'%(str(priority_level))
def load_IDs(self, prio):
filename = 'id_prio%s.data'%str(prio)
ID_file = open(filename, 'r')
ID_data = ID_file.read()
ID_file.close()
temp = open(filename, 'w')
temp.write('\n')
temp.close()
IDs = [line.split(':') for line in ID_data.split('\n') if ID != '']
return IDs
[MORE CODE...]
For some reason, I get the following error:
File "pastebin_rip_2.py", line 40
def load_IDs(self, prio):
^
SyntaxError: invalid syntax
What am I doing wrong? for a moment I considered the issue might have been the placing of init, because when I moved it to the end of the Downloader class it worked fine (which doesn't make sense?). Well, Downloader did, anyway. Instead I got a message complaining about the class after Downloader.
I really don't see what's wrong. Help?
(Entire code: http://snipt.org/xkky)
I think you're missing a closing paren at this line:
self.output('[Downloader] New downloader created (prio: %s)!'%(str(priority_level))