twisted.python.filepathView Source (View In Hierarchy)
Implements interfaces: twisted.python.filepath.IFilePath
I am a path on the filesystem that only permits 'downwards' access.
Instantiate me with a pathname (for example, FilePath('/home/myuser/public_html')) and I will attempt to only provide access to files which reside inside that path. I may be a path to a file, a directory, or a file which does not exist.
The correct way to use me is to instantiate me, and then do ALL filesystem access through me. In other words, do not import the 'os' module; if you need to open a file, call my 'open' method. If you need to list a directory, call my 'path' method.
Even if you pass me a relative path, I will convert that to an absolute path internally.
Note: although time-related methods do return floating-point results,
they may still be only second resolution depending on the platform and the
last value passed to
If you want greater-than-second precision, call
os.stat_float_times(True), or use Python 2.5.
Greater-than-second precision is only available in Windows on Python2.5 and
The type of
path when instantiating decides the mode of the
FilePath(b"/") will return a
FilePath(u"/") will return a
FilePath("/") will return a
Python 2, and a
|Instance Variable||alwaysCreate||When opening this file, only succeed if the file does not already exist. (type:
|Instance Variable||path||The path from which 'downward' traversal is permitted. (type:
|Instance Variable||statinfo 0||(WARNING: statinfo is deprecated as of Twisted 15.0.0 and will become a
private attribute) The currently cached status information about the file
on the filesystem that this
|Method||__init__||Convert a path string to an absolute path if necessary and initialize
|Method||__getstate__||Support serialization by discarding cached
|Method||sep||Return a filesystem separator.|
|Method||child||Create and return a new
|Method||preauthChild||Use me if
|Method||childSearchPreauth||Return my first existing child with a name in
|Method||siblingExtensionSearch||Attempt to return a path with my name, given multiple possible extensions.|
|Method||realpath||Returns the absolute target as a
|Method||siblingExtension||Attempt to return a path with my name, given the extension at
|Method||linkTo||Creates a symlink to self to at the path in the
|Method||open||Open this file using
|Method||restat||Re-calculate cached effects of 'stat'. To refresh information on this path after you know the filesystem may have changed, call this method.|
|Method||changed||Clear any cached information about the state of this path on disk.|
|Method||chmod||Changes the permissions on self, if possible. Propagates errors from
|Method||getsize||Retrieve the size of this file in bytes.|
|Method||getModificationTime||Retrieve the time of last access from this file.|
|Method||getStatusChangeTime||Retrieve the time of the last status change for this file.|
|Method||getAccessTime||Retrieve the time that this file was last accessed.|
|Method||getInodeNumber||Retrieve the file serial number, also called inode number, which distinguishes this file from all other files on the same device.|
|Method||getDevice||Retrieves the device containing the file. The inode number and device number together uniquely identify the file, but the device number is not necessarily consistent across reboots or system crashes.|
|Method||getNumberOfHardLinks||Retrieves the number of hard links to the file.|
|Method||getUserID||Returns the user ID of the file's owner.|
|Method||getGroupID||Returns the group ID of the file.|
|Method||getPermissions||Returns the permissions of the file. Should also work on Windows, however, those permissions may not be what is expected in Windows.|
|Method||exists||Check if this
|Method||isdir||Check if this
|Method||isfile||Check if this file path refers to a regular file.|
|Method||isBlockDevice||Returns whether the underlying path is a block device.|
|Method||isSocket||Returns whether the underlying path is a socket.|
|Method||islink||Check if this
|Method||isabs||Check if this
|Method||listdir||List the base names of the direct children of this
|Method||splitext||Split the file path into a pair
|Method||touch||Updates the access and last modification times of the file at this file path to the current time. Also creates the file if it does not already exist.|
|Method||remove||Removes the file or directory that is represented by self. If
|Method||makedirs||Create all directories not yet existing in
|Method||globChildren||Assuming I am representing a directory, return a list of FilePaths representing my children that match the given pattern.|
|Method||basename||Retrieve the final component of the file path's path (everything after the final path separator).|
|Method||dirname||Retrieve all of the components of the
|Method||parent||A file path for the directory containing the file at this file path.|
|Method||setContent||Replace the file at this path with a new file that contains the given bytes, trying to avoid data-loss in the meanwhile.|
|Method||createDirectory||Create the directory the
|Method||create||Exclusively create a file, only if this file previously did not exist.|
|Method||temporarySibling||Construct a path referring to a sibling of this path.|
|Method||copyTo||Copies self to destination.|
|Method||moveTo||Move self to destination - basically renaming self to whatever destination is named.|
|Method||statinfo||FilePath.statinfo is deprecated.|
|Method||_asBytesPath||Return the path of this
|Method||_asTextPath||Return the path of this
Inherited from AbstractFilePath:
|Method||getContent||Retrieve the contents of the file at this path.|
|Method||parents||Retrieve an iterator of all the ancestors of this path.|
|Method||children||List the children of this path object.|
|Method||walk||Yield myself, then each of my children, and each of those children's children in turn.|
|Method||descendant||Retrieve a child or child's child of this path.|
|Method||segmentsFrom||Return a list of segments between a child and its ancestor.|
|Method||__hash__||Hash the same as another
|Method||getmtime||Deprecated. Use getModificationTime instead.|
|Method||getatime||Deprecated. Use getAccessTime instead.|
|Method||getctime||Deprecated. Use getStatusChangeTime instead.|
FilePathpoints to. This attribute is
Noneif the file is in an indeterminate state (either this
FilePathhas not yet had cause to call
FilePath.changedindicated that new information is required), 0 if
stat()was called and returned an error (i.e. the path did not exist when
stat()was called), or a
stat_resultobject that describes the last known status of the underlying file (or directory, as the case may be). Trust me when I tell you that you do not want to use this attribute. Instead, use the methods on
FilePathwhich give you information about it, like
getModificationTime(), and so on. (type:
Support serialization by discarding cached
results and returning everything else.
Return a filesystem separator.
|Returns||The native filesystem separator. (type: The same type as |
Create and return a new
representing a path contained by
|Parameters||path||The base name of the new |
|Returns||The child path. (type: |
|Raises||InsecurePath||If the result of combining this path with |
Return my first existing child with a name in
paths is expected to be a list of *pre-secured* path
fragments; in most cases this will be specified by a system administrator
and not an arbitrary user.
If no appropriately-named children exist, this will return
Attempt to return a path with my name, given multiple possible extensions.
Each extension in
exts will be tested and the first path
which exists will be returned. If no path exists,
will be returned. If
'' is in
exts, then if the
file referred to by this path exists,
self will be
The extension '*' has a magic meaning, which means "any path that
self.path + '.' is acceptable".
Returns the absolute target as a
self is a link, self otherwise.
The absolute link is the ultimate file or directory the link refers to
(for instance, if the link refers to another link, and another...). If the
filesystem does not support symlinks, or if the link is cyclical, raises a
in that it does not resolve link names in the middle (ex. /x/y/z, y is a
link to w - realpath on z will return /x/y/z, not /x/w/z).
|Raises||LinkError||if links are not supported or links are cyclical.|
Creates a symlink to self to at the path in the
|Parameters||linkFilePath||a FilePath representing the link to be created. (type: |
Open this file using
mode or for writing if
In all cases the file is opened in binary mode, so it is not necessary
|Parameters||mode||The mode to open the file in. Default is |
|Returns||An open |
Re-calculate cached effects of 'stat'. To refresh information on this path after you know the filesystem may have changed, call this method.
|Parameters||reraise||a boolean. If true, re-raise exceptions from |
Clear any cached information about the state of this path on disk.
Retrieve the file serial number, also called inode number, which distinguishes this file from all other files on the same device.
|Returns||a number representing the file serial number (type: |
|Raises||NotImplementedError||if the platform is Windows, since the inode number would be a dummy value for all files in Windows|
Retrieves the device containing the file. The inode number and device number together uniquely identify the file, but the device number is not necessarily consistent across reboots or system crashes.
|Returns||a number representing the device (type: |
|Raises||NotImplementedError||if the platform is Windows, since the device number would be 0 for all partitions on a Windows platform|
Retrieves the number of hard links to the file.
This count keeps track of how many directories have entries for this file. If the count is ever decremented to zero then the file itself is discarded as soon as no process still holds it open. Symbolic links are not counted in the total.
|Returns||the number of hard links to the file (type: |
|Raises||NotImplementedError||if the platform is Windows, since Windows doesn't maintain a link count for
directories, and |
Returns the user ID of the file's owner.
|Returns||the user ID of the file's owner (type: |
|Raises||NotImplementedError||if the platform is Windows, since the UID is always 0 on Windows|
Returns the group ID of the file.
|Returns||the group ID of the file (type: |
|Raises||NotImplementedError||if the platform is Windows, since the GID is always 0 on windows|
Returns the permissions of the file. Should also work on Windows, however, those permissions may not be what is expected in Windows.
|Returns||the permissions for the file (type: |
Returns whether the underlying path is a block device.
Returns whether the underlying path is a socket.
List the base names of the direct children of this
|Raises||Anything the platform |
Updates the access and last modification times of the file at this file path to the current time. Also creates the file if it does not already exist.
|Raises||Exception||if unable to create or modify the last modification time of the file.|
Removes the file or directory that is represented by self. If
self.path is a directory, recursively remove all its children
before removing the directory. If it's a file or link, just delete it.
Replace the file at this path with a new file that contains the given bytes, trying to avoid data-loss in the meanwhile.
On UNIX-like platforms, this method does its best to ensure that by the time this method returns, either the old contents or the new contents of the file will be present at this path for subsequent readers regardless of premature device removal, program crash, or power loss, making the following assumptions:
On most versions of Windows there is no atomic
(see http://bit.ly/win32-overwrite for more information), so
this method is slightly less helpful. There is a small window where the
file at this path may be deleted before the new file is moved to replace
it: however, the new file will be fully written and flushed beforehand so
in the unlikely event that there is a crash at that point, it should be
possible for the user to manually recover the new version of their data.
In the future, Twisted will support atomic file moves on those versions of
Windows which do support them: see Twisted
This method should be safe for use by multiple concurrent processes, but note that it is not easy to predict which process's contents will ultimately end up on disk if they invoke this method at close to the same time.
|Parameters||content||The desired contents of the file at this path. (type: |
|ext||An extension to append to the temporary filename used to store the bytes
while they are being written. This can be used to make sure that temporary
files can be identified by their suffix, for cleanup in case of crashes. (type: |
Exclusively create a file, only if this file previously did not exist.
|Returns||A file-like object opened from this path.|
Construct a path referring to a sibling of this path.
The resulting path will be unpredictable, so that other subprocesses should neither accidentally attempt to refer to the same path before it is created, nor they should other processes be able to guess its name in advance.
|Parameters||extension||A suffix to append to the created filename. (Note that if you want an
extension with a '.' you must include the '.' yourself.) (type: |
|Returns||a path object with the given extension suffix, |
Copies self to destination.
If self doesn't exist, an OSError is raised.
If self is a directory, this method copies its children (but not itself) recursively to destination - if destination does not exist as a directory, this method creates it. If destination is a file, an IOError will be raised.
If self is a file, this method copies it to destination. If destination is a file, this method overwrites it. If destination is a directory, an IOError will be raised.
If self is a link (and followLinks is False), self will be copied over as a new symlink with the same target as returned by os.readlink. That means that if it is absolute, both the old and new symlink will link to the same thing. If it's relative, then perhaps not (and it's also possible that this relative link will be broken).
File/directory permissions and ownership will NOT be copied over.
If followLinks is True, symlinks are followed so that they're treated as their targets. In other words, if self is a link, the link's target will be copied. If destination is a link, self will be copied to the destination's target (the actual destination will be destination's target). Symlinks under self (if self is a directory) will be followed and its target's children be copied recursively.
If followLinks is False, symlinks will be copied over as symlinks.
|Parameters||destination||the destination (a FilePath) to which self should be copied|
|followLinks||whether symlinks in self should be treated as links or as their targets|
Move self to destination - basically renaming self to whatever destination is named.
If destination is an already-existing directory, moves all children to destination if destination is empty. If destination is a non-empty directory, or destination is a file, an OSError will be raised.
If moving between filesystems, self needs to be copied, and everything that applies to copyTo applies to moveTo.
|Parameters||destination||the destination (a FilePath) to which self should be copied|
|followLinks||whether symlinks in self should be treated as links or as their targets (only applicable when moving between filesystems)|