shutil
-- --High-level file operations 高级的文件操作模块。
os模块提供了对目录或者文件的新建/删除/查看文件属性，还提供了对文件以及目录的路径操作。比如说：绝对路径，父目录…… 但是，os文件的操作还应该包含移动 复制 打包 压缩 解压等操作，这些os模块都没有提供。
而本章所讲的shutil则就是对os中文件操作的补充。--移动 复制 打包 压缩 解压，
shutil功能：
1 shutil.copyfileobj(fsrc, fdst[, length=16*1024]) #copy文件内容到另一个文件，可以copy指定大小的内容
1 #先来看看其源代码。 2 def copyfileobj(fsrc, fdst, length=16*1024): 3 """copy data from file-like object fsrc to file-like object fdst""" 4 while 1: 5 buf = fsrc.read(length) 6 if not buf: 7 break 8 fdst.write(buf) 9 10 #注意！ 在其中fsrc，fdst都是文件对象，都需要打开后才能进行复制操作 11 import shutil 12 f1=open('name','r') 13 f2=open('name_copy','w+') 14 shutil.copyfileobj(f1,f2,length=16*1024)
2 shutil.copyfile(src,dst) #copy文件内容，是不是感觉上面的文件复制很麻烦？还需要自己手动用open函数打开文件，在这里就不需要了，事实上，copyfile调用了copyfileobj
1 def copyfile(src, dst, *, follow_symlinks=True): 2 """Copy data from src to dst. 3 4 If follow_symlinks is not set and src is a symbolic link, a new 5 symlink will be created instead of copying the file it points to. 6 7 """ 8 if _samefile(src, dst): 9 raise SameFileError("{!r} and {!r} are the same file".format(src, dst)) 10 11 for fn in [src, dst]: 12 try: 13 st = os.stat(fn) 14 except OSError: 15 # File most likely does not exist 16 pass 17 else: 18 # XXX What about other special files? (sockets, devices...) 19 if stat.S_ISFIFO(st.st_mode): 20 raise SpecialFileError("`%s` is a named pipe" % fn) 21 22 if not follow_symlinks and os.path.islink(src): 23 os.symlink(os.readlink(src), dst) 24 else: 25 with open(src, 'rb') as fsrc: 26 with open(dst, 'wb') as fdst: 27 copyfileobj(fsrc, fdst) 28 return dst 29 30 查看源代码
1 shutil.copyfile('name','name_copy_2') 2 #一句就可以实现复制文件内容
3 shutil.copymode(src,dst) #仅copy权限，不更改文件内容，组和用户。
1 def copymode(src, dst, *, follow_symlinks=True): 2 """Copy mode bits from src to dst. 3 4 If follow_symlinks is not set, symlinks aren't followed if and only 5 if both `src` and `dst` are symlinks. If `lchmod` isn't available 6 (e.g. Linux) this method does nothing. 7 8 """ 9 if not follow_symlinks and os.path.islink(src) and os.path.islink(dst): 10 if hasattr(os, 'lchmod'): 11 stat_func, chmod_func = os.lstat, os.lchmod 12 else: 13 return 14 elif hasattr(os, 'chmod'): 15 stat_func, chmod_func = os.stat, os.chmod 16 else: 17 return 18 19 st = stat_func(src) 20 chmod_func(dst, stat.S_IMODE(st.st_mode)) 21 22 查看源代码
1 def copymode(src, dst, *, follow_symlinks=True): 2 """Copy mode bits from src to dst. 3 4 If follow_symlinks is not set, symlinks aren't followed if and only 5 if both `src` and `dst` are symlinks. If `lchmod` isn't available 6 (e.g. Linux) this method does nothing. 7 8 """ 9 if not follow_symlinks and os.path.islink(src) and os.path.islink(dst): 10 if hasattr(os, 'lchmod'): 11 stat_func, chmod_func = os.lstat, os.lchmod 12 else: 13 return 14 elif hasattr(os, 'chmod'): 15 stat_func, chmod_func = os.stat, os.chmod 16 else: 17 return 18 19 st = stat_func(src) 20 chmod_func(dst, stat.S_IMODE(st.st_mode)) 21 22 查看源代码
4 shutil.copystat(src,dst) #复制所有的状态信息，包括权限，组，用户，时间等
1 def copystat(src, dst, *, follow_symlinks=True): 2 """Copy all stat info (mode bits, atime, mtime, flags) from src to dst. 3 4 If the optional flag `follow_symlinks` is not set, symlinks aren't followed if and 5 only if both `src` and `dst` are symlinks. 6 7 """ 8 def _nop(*args, ns=None, follow_symlinks=None): 9 pass 10 11 # follow symlinks (aka don't not follow symlinks) 12 follow = follow_symlinks or not (os.path.islink(src) and os.path.islink(dst)) 13 if follow: 14 # use the real function if it exists 15 def lookup(name): 16 return getattr(os, name, _nop) 17 else: 18 # use the real function only if it exists 19 # *and* it supports follow_symlinks 20 def lookup(name): 21 fn = getattr(os, name, _nop) 22 if fn in os.supports_follow_symlinks: 23 return fn 24 return _nop 25 26 st = lookup("stat")(src, follow_symlinks=follow) 27 mode = stat.S_IMODE(st.st_mode) 28 lookup("utime")(dst, ns=(st.st_atime_ns, st.st_mtime_ns), 29 follow_symlinks=follow) 30 try: 31 lookup("chmod")(dst, mode, follow_symlinks=follow) 32 except NotImplementedError: 33 # if we got a NotImplementedError, it's because 34 # * follow_symlinks=False, 35 # * lchown() is unavailable, and 36 # * either 37 # * fchownat() is unavailable or 38 # * fchownat() doesn't implement AT_SYMLINK_NOFOLLOW. 39 # (it returned ENOSUP.) 40 # therefore we're out of options--we simply cannot chown the 41 # symlink. give up, suppress the error. 42 # (which is what shutil always did in this circumstance.) 43 pass 44 if hasattr(st, 'st_flags'): 45 try: 46 lookup("chflags")(dst, st.st_flags, follow_symlinks=follow) 47 except OSError as why: 48 for err in 'EOPNOTSUPP', 'ENOTSUP': 49 if hasattr(errno, err) and why.errno == getattr(errno, err): 50 break 51 else: 52 raise 53 _copyxattr(src, dst, follow_symlinks=follow) 54 55 查看源代码
5 shutil.copy(src,dst) #复制文件的内容以及权限，先copyfile后copymode
1 def copy(src, dst, *, follow_symlinks=True): 2 """Copy data and mode bits ("cp src dst"). Return the file's destination. 3 4 The destination may be a directory. 5 6 If follow_symlinks is false, symlinks won't be followed. This 7 resembles GNU's "cp -P src dst". 8 9 If source and destination are the same file, a SameFileError will be 10 raised. 11 12 """ 13 if os.path.isdir(dst): 14 dst = os.path.join(dst, os.path.basename(src)) 15 copyfile(src, dst, follow_symlinks=follow_symlinks) 16 copymode(src, dst, follow_symlinks=follow_symlinks) 17 return dst 18 19 查看源代码
6 shutil.copy2(src,dst) #复制文件的内容以及文件的所有状态信息。先copyfile后copystat
1 def copy2(src, dst, *, follow_symlinks=True): 2 """Copy data and all stat info ("cp -p src dst"). Return the file's 3 destination." 4 5 The destination may be a directory. 6 7 If follow_symlinks is false, symlinks won't be followed. This 8 resembles GNU's "cp -P src dst". 9 10 """ 11 if os.path.isdir(dst): 12 dst = os.path.join(dst, os.path.basename(src)) 13 copyfile(src, dst, follow_symlinks=follow_symlinks) 14 copystat(src, dst, follow_symlinks=follow_symlinks) 15 return dst 16 17 查看源代码
7 shutil.copytree(src, dst, symlinks=False, ignore=None, copy_function=copy2,ignore_dangling_symlinks=False) #递归的复制文件内容及状态信息
1 def copytree(src, dst, symlinks=False, ignore=None, copy_function=copy2, 2 ignore_dangling_symlinks=False): 3 """Recursively copy a directory tree. 4 5 The destination directory must not already exist. 6 If exception(s) occur, an Error is raised with a list of reasons. 7 8 If the optional symlinks flag is true, symbolic links in the 9 source tree result in symbolic links in the destination tree; if 10 it is false, the contents of the files pointed to by symbolic 11 links are copied. If the file pointed by the symlink doesn't 12 exist, an exception will be added in the list of errors raised in 13 an Error exception at the end of the copy process. 14 15 You can set the optional ignore_dangling_symlinks flag to true if you 16 want to silence this exception. Notice that this has no effect on 17 platforms that don't support os.symlink. 18 19 The optional ignore argument is a callable. If given, it 20 is called with the `src` parameter, which is the directory 21 being visited by copytree(), and `names` which is the list of 22 `src` contents, as returned by os.listdir(): 23 24 callable(src, names) -> ignored_names 25 26 Since copytree() is called recursively, the callable will be 27 called once for each directory that is copied. It returns a 28 list of names relative to the `src` directory that should 29 not be copied. 30 31 The optional copy_function argument is a callable that will be used 32 to copy each file. It will be called with the source path and the 33 destination path as arguments. By default, copy2() is used, but any 34 function that supports the same signature (like copy()) can be used. 35 36 """ 37 names = os.listdir(src) 38 if ignore is not None: 39 ignored_names = ignore(src, names) 40 else: 41 ignored_names = set() 42 43 os.makedirs(dst) 44 errors = [] 45 for name in names: 46 if name in ignored_names: 47 continue 48 srcname = os.path.join(src, name) 49 dstname = os.path.join(dst, name) 50 try: 51 if os.path.islink(srcname): 52 linkto = os.readlink(srcname) 53 if symlinks: 54 # We can't just leave it to `copy_function` because legacy 55 # code with a custom `copy_function` may rely on copytree 56 # doing the right thing. 57 os.symlink(linkto, dstname) 58 copystat(srcname, dstname, follow_symlinks=not symlinks) 59 else: 60 # ignore dangling symlink if the flag is on 61 if not os.path.exists(linkto) and ignore_dangling_symlinks: 62 continue 63 # otherwise let the copy occurs. copy2 will raise an error 64 if os.path.isdir(srcname): 65 copytree(srcname, dstname, symlinks, ignore, 66 copy_function) 67 else: 68 copy_function(srcname, dstname) 69 elif os.path.isdir(srcname): 70 copytree(srcname, dstname, symlinks, ignore, copy_function) 71 else: 72 # Will raise a SpecialFileError for unsupported file types 73 copy_function(srcname, dstname) 74 # catch the Error from the recursive copytree so that we can 75 # continue with other files 76 except Error as err: 77 errors.extend(err.args[0]) 78 except OSError as why: 79 errors.append((srcname, dstname, str(why))) 80 try: 81 copystat(src, dst) 82 except OSError as why: 83 # Copying file access times may fail on Windows 84 if getattr(why, 'winerror', None) is None: 85 errors.append((src, dst, str(why))) 86 if errors: 87 raise Error(errors) 88 return dst 89 90 # version vulnerable to race conditions 91 92 查看源代码
1 def copytree(src, dst, symlinks=False, ignore=None, copy_function=copy2, 2 ignore_dangling_symlinks=False): 3 """Recursively copy a directory tree. 4 5 The destination directory must not already exist. 6 If exception(s) occur, an Error is raised with a list of reasons. 7 8 If the optional symlinks flag is true, symbolic links in the 9 source tree result in symbolic links in the destination tree; if 10 it is false, the contents of the files pointed to by symbolic 11 links are copied. If the file pointed by the symlink doesn't 12 exist, an exception will be added in the list of errors raised in 13 an Error exception at the end of the copy process. 14 15 You can set the optional ignore_dangling_symlinks flag to true if you 16 want to silence this exception. Notice that this has no effect on 17 platforms that don't support os.symlink. 18 19 The optional ignore argument is a callable. If given, it 20 is called with the `src` parameter, which is the directory 21 being visited by copytree(), and `names` which is the list of 22 `src` contents, as returned by os.listdir(): 23 24 callable(src, names) -> ignored_names 25 26 Since copytree() is called recursively, the callable will be 27 called once for each directory that is copied. It returns a 28 list of names relative to the `src` directory that should 29 not be copied. 30 31 The optional copy_function argument is a callable that will be used 32 to copy each file. It will be called with the source path and the 33 destination path as arguments. By default, copy2() is used, but any 34 function that supports the same signature (like copy()) can be used. 35 36 """ 37 names = os.listdir(src) 38 if ignore is not None: 39 ignored_names = ignore(src, names) 40 else: 41 ignored_names = set() 42 43 os.makedirs(dst) 44 errors = [] 45 for name in names: 46 if name in ignored_names: 47 continue 48 srcname = os.path.join(src, name) 49 dstname = os.path.join(dst, name) 50 try: 51 if os.path.islink(srcname): 52 linkto = os.readlink(srcname) 53 if symlinks: 54 # We can't just leave it to `copy_function` because legacy 55 # code with a custom `copy_function` may rely on copytree 56 # doing the right thing. 57 os.symlink(linkto, dstname) 58 copystat(srcname, dstname, follow_symlinks=not symlinks) 59 else: 60 # ignore dangling symlink if the flag is on 61 if not os.path.exists(linkto) and ignore_dangling_symlinks: 62 continue 63 # otherwise let the copy occurs. copy2 will raise an error 64 if os.path.isdir(srcname): 65 copytree(srcname, dstname, symlinks, ignore, 66 copy_function) 67 else: 68 copy_function(srcname, dstname) 69 elif os.path.isdir(srcname): 70 copytree(srcname, dstname, symlinks, ignore, copy_function) 71 else: 72 # Will raise a SpecialFileError for unsupported file types 73 copy_function(srcname, dstname) 74 # catch the Error from the recursive copytree so that we can 75 # continue with other files 76 except Error as err: 77 errors.extend(err.args[0]) 78 except OSError as why: 79 errors.append((srcname, dstname, str(why))) 80 try: 81 copystat(src, dst) 82 except OSError as why: 83 # Copying file access times may fail on Windows 84 if getattr(why, 'winerror', None) is None: 85 errors.append((src, dst, str(why))) 86 if errors: 87 raise Error(errors) 88 return dst 89 90 # version vulnerable to race conditions 91 92 查看源代码
8 shutil.rmtree(path, ignore_errors=False, onerror=None) #递归地删除文件
1 def rmtree(path, ignore_errors=False, onerror=None): 2 """Recursively delete a directory tree. 3 4 If ignore_errors is set, errors are ignored; otherwise, if onerror 5 is set, it is called to handle the error with arguments (func, 6 path, exc_info) where func is platform and implementation dependent; 7 path is the argument to that function that caused it to fail; and 8 exc_info is a tuple returned by sys.exc_info(). If ignore_errors 9 is false and onerror is None, an exception is raised. 10 11 """ 12 if ignore_errors: 13 def onerror(*args): 14 pass 15 elif onerror is None: 16 def onerror(*args): 17 raise 18 if _use_fd_functions: 19 # While the unsafe rmtree works fine on bytes, the fd based does not. 20 if isinstance(path, bytes): 21 path = os.fsdecode(path) 22 # Note: To guard against symlink races, we use the standard 23 # lstat()/open()/fstat() trick. 24 try: 25 orig_st = os.lstat(path) 26 except Exception: 27 onerror(os.lstat, path, sys.exc_info()) 28 return 29 try: 30 fd = os.open(path, os.O_RDONLY) 31 except Exception: 32 onerror(os.lstat, path, sys.exc_info()) 33 return 34 try: 35 if os.path.samestat(orig_st, os.fstat(fd)): 36 _rmtree_safe_fd(fd, path, onerror) 37 try: 38 os.rmdir(path) 39 except OSError: 40 onerror(os.rmdir, path, sys.exc_info()) 41 else: 42 try: 43 # symlinks to directories are forbidden, see bug #1669 44 raise OSError("Cannot call rmtree on a symbolic link") 45 except OSError: 46 onerror(os.path.islink, path, sys.exc_info()) 47 finally: 48 os.close(fd) 49 else: 50 return _rmtree_unsafe(path, onerror) 51 52 # Allow introspection of whether or not the hardening against symlink 53 # attacks is supported on the current platform 54 rmtree.avoids_symlink_attacks = _use_fd_functions 55 56 查看源代码
9 shutil.move(src, dst) #递归的移动文件
1 def move(src, dst): 2 """Recursively move a file or directory to another location. This is 3 similar to the Unix "mv" command. Return the file or directory's 4 destination. 5 6 If the destination is a directory or a symlink to a directory, the source 7 is moved inside the directory. The destination path must not already 8 exist. 9 10 If the destination already exists but is not a directory, it may be 11 overwritten depending on os.rename() semantics. 12 13 If the destination is on our current filesystem, then rename() is used. 14 Otherwise, src is copied to the destination and then removed. Symlinks are 15 recreated under the new name if os.rename() fails because of cross 16 filesystem renames. 17 18 A lot more could be done here... A look at a mv.c shows a lot of 19 the issues this implementation glosses over. 20 21 """ 22 real_dst = dst 23 if os.path.isdir(dst): 24 if _samefile(src, dst): 25 # We might be on a case insensitive filesystem, 26 # perform the rename anyway. 27 os.rename(src, dst) 28 return 29 30 real_dst = os.path.join(dst, _basename(src)) 31 if os.path.exists(real_dst): 32 raise Error("Destination path '%s' already exists" % real_dst) 33 try: 34 os.rename(src, real_dst) 35 except OSError: 36 if os.path.islink(src): 37 linkto = os.readlink(src) 38 os.symlink(linkto, real_dst) 39 os.unlink(src) 40 elif os.path.isdir(src): 41 if _destinsrc(src, dst): 42 raise Error("Cannot move a directory '%s' into itself '%s'." % (src, dst)) 43 copytree(src, real_dst, symlinks=True) 44 rmtree(src) 45 else: 46 copy2(src, real_dst) 47 os.unlink(src) 48 return real_dst 49 50 查看源代码
10 make_archive(base_name, format, root_dir=None, base_dir=None, verbose=0,dry_run=0, owner=None, group=None, logger=None) #压缩打包
1 def make_archive(base_name, format, root_dir=None, base_dir=None, verbose=0, 2 dry_run=0, owner=None, group=None, logger=None): 3 """Create an archive file (eg. zip or tar). 4 5 'base_name' is the name of the file to create, minus any format-specific 6 extension; 'format' is the archive format: one of "zip", "tar", "bztar" 7 or "gztar". 8 9 'root_dir' is a directory that will be the root directory of the 10 archive; ie. we typically chdir into 'root_dir' before creating the 11 archive. 'base_dir' is the directory where we start archiving from; 12 ie. 'base_dir' will be the common prefix of all files and 13 directories in the archive. 'root_dir' and 'base_dir' both default 14 to the current directory. Returns the name of the archive file. 15 16 'owner' and 'group' are used when creating a tar archive. By default, 17 uses the current owner and group. 18 """ 19 save_cwd = os.getcwd() 20 if root_dir is not None: 21 if logger is not None: 22 logger.debug("changing into '%s'", root_dir) 23 base_name = os.path.abspath(base_name) 24 if not dry_run: 25 os.chdir(root_dir) 26 27 if base_dir is None: 28 base_dir = os.curdir 29 30 kwargs = {'dry_run': dry_run, 'logger': logger} 31 32 try: 33 format_info = _ARCHIVE_FORMATS[format] 34 except KeyError: 35 raise ValueError("unknown archive format '%s'" % format) 36 37 func = format_info[0] 38 for arg, val in format_info[1]: 39 kwargs[arg] = val 40 41 if format != 'zip': 42 kwargs['owner'] = owner 43 kwargs['group'] = group 44 45 try: 46 filename = func(base_name, base_dir, **kwargs) 47 finally: 48 if root_dir is not None: 49 if logger is not None: 50 logger.debug("changing back to '%s'", save_cwd) 51 os.chdir(save_cwd) 52 53 return filename 54 55 查看源代码
base_name： 压缩打包后的文件名或者路径名
format： 压缩或者打包格式 "zip", "tar", "bztar"or "gztar"
root_dir : 将哪个目录或者文件打包（也就是源文件）
1 >>> shutil.make_archive('tarball','gztar',root_dir='copytree_test') 2 3 [root@slyoyo python_test]# ls -l 4 total 12 5 drwxr-xr-x. 3 root root 4096 May 14 19:36 copytree_copy 6 drwxr-xr-x. 3 root root 4096 May 14 19:36 copytree_test 7 -rw-r--r--. 1 root root 0 May 14 21:12 tarball.tar.gz 8 -rw-r--r--. 1 python python 79 May 14 05:17 test1 9 -rw-r--r--. 1 root root 0 May 14 19:10 test2 10 11 实例