Memory-managed wrapper type.

Type class for types which can be safely cast to File, for instance with toFile.

Cast to File, for types for which this is known to be safe. For general casts, use castTo.

Gets an output stream for appending data to the file. If the file doesn't already exist it is created.

By default files created are generally readable by everyone, but if you pass G_FILE_CREATE_PRIVATE in flags the file will be made readable only to the current user, to the level that is supported on the target filesystem.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Some file systems don't allow all file names, and may return an IOErrorEnumInvalidFilename error. If the file is a directory the IOErrorEnumIsDirectory error will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on.

Asynchronously opens file for appending.

For more details, see fileAppendTo which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileAppendToFinish to get the result of the operation.

Finishes an asynchronous file append operation started with fileAppendToAsync.

Prepares the file attribute query string for copying to file.

This function prepares an attribute query string to be passed to fileQueryInfo to get a list of attributes normally copied with the file (see fileCopyAttributes for the detailed description). This function is used by the implementation of fileCopyAttributes and is useful when one needs to query and set the attributes in two stages (e.g., for recursive move of a directory).

Since: 2.68

Copies the file source to the location specified by destination. Can not handle recursive copies of directories.

If the flag G_FILE_COPY_OVERWRITE is specified an already existing destination file is overwritten.

If the flag G_FILE_COPY_NOFOLLOW_SYMLINKS is specified then symlinks will be copied as symlinks, otherwise the target of the source symlink will be copied.

If the flag G_FILE_COPY_ALL_METADATA is specified then all the metadata that is possible to copy is copied, not just the default subset (which, for instance, does not include the owner, see FileInfo).

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If progressCallback is not Nothing, then the operation can be monitored by setting this to a FileProgressCallback function. progressCallbackData will be passed to this function. It is guaranteed that this callback will be called after all data has been transferred with the total number of bytes copied during the operation.

If the source file does not exist, then the IOErrorEnumNotFound error is returned, independent on the status of the destination.

If G_FILE_COPY_OVERWRITE is not specified and the target exists, then the error IOErrorEnumExists is returned.

If trying to overwrite a file over a directory, the IOErrorEnumIsDirectory error is returned. If trying to overwrite a directory with a directory the IOErrorEnumWouldMerge error is returned.

If the source is a directory and the target does not exist, or G_FILE_COPY_OVERWRITE is specified and the target is a file, then the IOErrorEnumWouldRecurse error is returned.

If you are interested in copying the File object itself (not the on-disk file), see fileDup.

Copies the file attributes from source to destination.

Normally only a subset of the file attributes are copied, those that are copies in a normal file copy operation (which for instance does not include e.g. owner). However if G_FILE_COPY_ALL_METADATA is specified in flags, then all the metadata that is possible to copy is copied. This is useful when implementing move by copy + delete source.

Finishes copying the file started with fileCopyAsync.

Creates a new file and returns an output stream for writing to it. The file must not already exist.

By default files created are generally readable by everyone, but if you pass G_FILE_CREATE_PRIVATE in flags the file will be made readable only to the current user, to the level that is supported on the target filesystem.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If a file or directory with this name already exists the IOErrorEnumExists error will be returned. Some file systems don't allow all file names, and may return an IOErrorEnumInvalidFilename error, and if the name is to long IOErrorEnumFilenameTooLong will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on.

Asynchronously creates a new file and returns an output stream for writing to it. The file must not already exist.

For more details, see fileCreate which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileCreateFinish to get the result of the operation.

Finishes an asynchronous file create operation started with fileCreateAsync.

Creates a new file and returns a stream for reading and writing to it. The file must not already exist.

By default files created are generally readable by everyone, but if you pass G_FILE_CREATE_PRIVATE in flags the file will be made readable only to the current user, to the level that is supported on the target filesystem.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If a file or directory with this name already exists, the IOErrorEnumExists error will be returned. Some file systems don't allow all file names, and may return an IOErrorEnumInvalidFilename error, and if the name is too long, IOErrorEnumFilenameTooLong will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on.

Note that in many non-local file cases read and write streams are not supported, so make sure you really need to do read and write streaming, rather than just opening for reading or writing.

Since: 2.22

Asynchronously creates a new file and returns a stream for reading and writing to it. The file must not already exist.

For more details, see fileCreateReadwrite which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileCreateReadwriteFinish to get the result of the operation.

Since: 2.22

Finishes an asynchronous file create operation started with fileCreateReadwriteAsync.

Since: 2.22

Deletes a file. If the file is a directory, it will only be deleted if it is empty. This has the same semantics as unlink.

If file doesn’t exist, IOErrorEnumNotFound will be returned. This allows for deletion to be implemented avoiding time-of-check to time-of-use races: > >g_autoptr(GError) local_error = NULL; >if (!g_file_delete (my_file, my_cancellable, &local_error) && > !g_error_matches (local_error, G_IO_ERROR, G_IO_ERROR_NOT_FOUND)) > { > // deletion failed for some reason other than the file not existing: > // so report the error > g_warning ("Failed to delete %s: %s", > g_file_peek_path (my_file), local_error->message); > }

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Asynchronously delete a file. If the file is a directory, it will only be deleted if it is empty. This has the same semantics as unlink.

Since: 2.34

Finishes deleting a file started with fileDeleteAsync.

Since: 2.34

Duplicates a File handle. This operation does not duplicate the actual file or directory represented by the File; see fileCopy if attempting to copy a file.

fileDup is useful when a second handle is needed to the same underlying file, for use in a separate thread (tFile is not thread-safe). For use within the same thread, use objectRef to increment the existing object’s reference count.

This call does no blocking I/O.

Starts an asynchronous eject on a mountable. When this operation has completed, callback will be called with userUser data, and the operation can be finalized with fileEjectMountableFinish.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Finishes an asynchronous eject operation started by fileEjectMountable.

Starts an asynchronous eject on a mountable. When this operation has completed, callback will be called with userUser data, and the operation can be finalized with fileEjectMountableWithOperationFinish.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Since: 2.22

Finishes an asynchronous eject operation started by fileEjectMountableWithOperation.

Since: 2.22

Gets the requested information about the files in a directory. The result is a FileEnumerator object that will give out FileInfo objects for all the files in the directory.

The attributes value is a string that specifies the file attributes that should be gathered. It is not an error if it's not possible to read a particular requested attribute from a file - it just won't be set. attributes should be a comma-separated list of attributes or attribute wildcards. The wildcard "*" means all attributes, and a wildcard like "standard::*" means all attributes in the standard namespace. An example attribute query be "standard::*,owner[user](#g:signal:user)". The standard attributes are available as defines, like FILE_ATTRIBUTE_STANDARD_NAME.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If the file does not exist, the IOErrorEnumNotFound error will be returned. If the file is not a directory, the IOErrorEnumNotDirectory error will be returned. Other errors are possible too.

Asynchronously gets the requested information about the files in a directory. The result is a FileEnumerator object that will give out FileInfo objects for all the files in the directory.

For more details, see fileEnumerateChildren which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileEnumerateChildrenFinish to get the result of the operation.

Finishes an async enumerate children operation. See fileEnumerateChildrenAsync.

Checks if the two given GFiles refer to the same file.

Note that two GFiles that differ can still refer to the same file on the filesystem due to various forms of filename aliasing.

This call does no blocking I/O.

Gets a Mount for the File.

Mount is returned only for user interesting locations, see VolumeMonitor. If the FileIface for file does not have a mount, error will be set to IOErrorEnumNotFound and Nothing will be returned.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Asynchronously gets the mount for the file.

For more details, see fileFindEnclosingMount which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileFindEnclosingMountFinish to get the result of the operation.

Finishes an asynchronous find mount request. See fileFindEnclosingMountAsync.

Gets the base name (the last component of the path) for a given File.

If called for the top level of a system (such as the filesystem root or a uri like sftp://host/) it will return a single directory separator (and on Windows, possibly a drive letter).

The base name is a byte string (not UTF-8). It has no defined encoding or rules other than it may not contain zero bytes. If you want to use filenames in a user interface you should use the display name that you can get by requesting the FILE_ATTRIBUTE_STANDARD_DISPLAY_NAME attribute with fileQueryInfo.

This call does no blocking I/O.

Gets a child of file with basename equal to name.

Note that the file with that specific name might not exist, but you can still have a File that points to it. You can use this for instance to create that file.

This call does no blocking I/O.

Gets the child of file for a given displayName (i.e. a UTF-8 version of the name). If this function fails, it returns Nothing and error will be set. This is very useful when constructing a File for a new file and the user entered the filename in the user interface, for instance when you select a directory and type a filename in the file selector.

This call does no blocking I/O.

Gets the parent directory for the file. If the file represents the root directory of the file system, then Nothing will be returned.

This call does no blocking I/O.

Gets the parse name of the file. A parse name is a UTF-8 string that describes the file such that one can get the File back using fileParseName.

This is generally used to show the File as a nice full-pathname kind of string in a user interface, like in a location entry.

For local files with names that can safely be converted to UTF-8 the pathname is used, otherwise the IRI is used (a form of URI that allows UTF-8 characters unescaped).

This call does no blocking I/O.

Gets the local pathname for File, if one exists. If non-Nothing, this is guaranteed to be an absolute, canonical path. It might contain symlinks.

This call does no blocking I/O.

Gets the path for descendant relative to parent.

This call does no blocking I/O.

Gets the URI for the file.

This call does no blocking I/O.

Gets the URI scheme for a File. RFC 3986 decodes the scheme as: > >URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]

Common schemes include "file", "http", "ftp", etc.

The scheme can be different from the one used to construct the File, in that it might be replaced with one that is logically equivalent to the File.

This call does no blocking I/O.

Checks if file has a parent, and optionally, if it is parent.

If parent is Nothing then this function returns True if file has any parent at all. If parent is non-Nothing then True is only returned if file is an immediate child of parent.

Since: 2.24

Checks whether file has the prefix specified by prefix.

In other words, if the names of initial elements of file's pathname match prefix. Only full pathname elements are matched, so a path like /foo is not considered a prefix of /foobar, only of /foo/bar.

A File is not a prefix of itself. If you want to check for equality, use fileEqual.

This call does no I/O, as it works purely on names. As such it can sometimes return False even if file is inside a prefix (from a filesystem point of view), because the prefix of file is an alias of prefix.

Checks to see if a File has a given URI scheme.

This call does no blocking I/O.

Creates a hash value for a File.

This call does no blocking I/O.

Checks to see if a file is native to the platform.

A native file is one expressed in the platform-native filename format, e.g. "C:\Windows" or "/usr/bin/". This does not mean the file is local, as it might be on a locally mounted remote filesystem.

On some systems non-native files may be available using the native filesystem via a userspace filesystem (FUSE), in these cases this call will return False, but fileGetPath will still return a native path.

This call does no blocking I/O.

Loads the contents of file and returns it as Bytes.

If file is a resource:// based URI, the resulting bytes will reference the embedded resource instead of a copy. Otherwise, this is equivalent to calling fileLoadContents and bytesNewTake.

For resources, etagOut will be set to Nothing.

The data contained in the resulting Bytes is always zero-terminated, but this is not included in the Bytes length. The resulting Bytes should be freed with bytesUnref when no longer in use.

Since: 2.56

Asynchronously loads the contents of file as Bytes.

If file is a resource:// based URI, the resulting bytes will reference the embedded resource instead of a copy. Otherwise, this is equivalent to calling fileLoadContentsAsync and bytesNewTake.

callback should call fileLoadBytesFinish to get the result of this asynchronous operation.

See fileLoadBytes for more information.

Since: 2.56

Completes an asynchronous request to fileLoadBytesAsync.

For resources, etagOut will be set to Nothing.

The data contained in the resulting Bytes is always zero-terminated, but this is not included in the Bytes length. The resulting Bytes should be freed with bytesUnref when no longer in use.

See fileLoadBytes for more information.

Since: 2.56

Loads the content of the file into memory. The data is always zero-terminated, but this is not included in the resultant length. The returned contents should be freed with free when no longer needed.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Starts an asynchronous load of the file's contents.

For more details, see fileLoadContents which is the synchronous version of this call.

When the load operation has completed, callback will be called with user data. To finish the operation, call fileLoadContentsFinish with the AsyncResult returned by the callback.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Finishes an asynchronous load of the file's contents. The contents are placed in contents, and length is set to the size of the contents string. The contents should be freed with free when no longer needed. If etagOut is present, it will be set to the new entity tag for the file.

Finishes an asynchronous partial load operation that was started with g_file_load_partial_contents_async(). The data is always zero-terminated, but this is not included in the resultant length. The returned contents should be freed with free when no longer needed.

Creates a directory. Note that this will only create a child directory of the immediate parent directory of the path or URI given by the File. To recursively create directories, see fileMakeDirectoryWithParents. This function will fail if the parent directory does not exist, setting error to IOErrorEnumNotFound. If the file system doesn't support creating directories, this function will fail, setting error to IOErrorEnumNotSupported.

For a local File the newly created directory will have the default (current) ownership and permissions of the current process.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Asynchronously creates a directory.

Since: 2.38

Finishes an asynchronous directory creation, started with fileMakeDirectoryAsync.

Since: 2.38

Creates a directory and any parent directories that may not exist similar to 'mkdir -p'. If the file system does not support creating directories, this function will fail, setting error to IOErrorEnumNotSupported. If the directory itself already exists, this function will fail setting error to IOErrorEnumExists, unlike the similar mkdirWithParents.

For a local File the newly created directories will have the default (current) ownership and permissions of the current process.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Since: 2.18

Creates a symbolic link named file which contains the string symlinkValue.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Collects the results from an earlier call to g_file_measure_disk_usage_async(). See g_file_measure_disk_usage() for more information.

Since: 2.38

Obtains a file or directory monitor for the given file, depending on the type of the file.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Since: 2.18

Obtains a directory monitor for the given file. This may fail if directory monitoring is not supported.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

It does not make sense for flags to contain FileMonitorFlagsWatchHardLinks, since hard links can not be made to directories. It is not possible to monitor all the files in a directory for changes made via hard links; if you want to do this then you must register individual watches with fileMonitor.

Obtains a file monitor for the given file. If no file notification mechanism exists, then regular polling of the file is used.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If flags contains FileMonitorFlagsWatchHardLinks then the monitor will also attempt to report changes made to the file via another filename (ie, a hard link). Without this flag, you can only rely on changes made through the filename contained in file to be reported. Using this flag may result in an increase in resource usage, and may not have any effect depending on the FileMonitor backend and/or filesystem type.

Starts a mountOperation, mounting the volume that contains the file location.

When this operation has completed, callback will be called with userUser data, and the operation can be finalized with fileMountEnclosingVolumeFinish.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Finishes a mount operation started by fileMountEnclosingVolume.

Mounts a file of type G_FILE_TYPE_MOUNTABLE. Using mountOperation, you can request callbacks when, for instance, passwords are needed during authentication.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

When the operation is finished, callback will be called. You can then call fileMountMountableFinish to get the result of the operation.

Finishes a mount operation. See fileMountMountable for details.

Finish an asynchronous mount operation that was started with fileMountMountable.

Tries to move the file or directory source to the location specified by destination. If native move operations are supported then this is used, otherwise a copy + delete fallback is used. The native implementation may support moving directories (for instance on moves inside the same filesystem), but the fallback code does not.

If the flag G_FILE_COPY_OVERWRITE is specified an already existing destination file is overwritten.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If progressCallback is not Nothing, then the operation can be monitored by setting this to a FileProgressCallback function. progressCallbackData will be passed to this function. It is guaranteed that this callback will be called after all data has been transferred with the total number of bytes copied during the operation.

If the source file does not exist, then the IOErrorEnumNotFound error is returned, independent on the status of the destination.

If G_FILE_COPY_OVERWRITE is not specified and the target exists, then the error IOErrorEnumExists is returned.

If trying to overwrite a file over a directory, the IOErrorEnumIsDirectory error is returned. If trying to overwrite a directory with a directory the IOErrorEnumWouldMerge error is returned.

If the source is a directory and the target does not exist, or G_FILE_COPY_OVERWRITE is specified and the target is a file, then the IOErrorEnumWouldRecurse error may be returned (if the native move operation isn't available).

Creates a File with the given argument from the command line. The value of arg can be either a URI, an absolute path or a relative path resolved relative to the current working directory. This operation never fails, but the returned object might not support any I/O operation if arg points to a malformed path.

Note that on Windows, this function expects its argument to be in UTF-8 -- not the system code page. This means that you should not use this function with string from argv as it is passed to main(). g_win32_get_command_line() will return a UTF-8 version of the commandline. Application also uses UTF-8 but applicationCommandLineCreateFileForArg may be more useful for you there. It is also always possible to use this function with OptionContext arguments of type OptionArgFilename.

Creates a File with the given argument from the command line.

This function is similar to fileNewForCommandlineArg except that it allows for passing the current working directory as an argument instead of using the current working directory of the process.

This is useful if the commandline argument was given in a context other than the invocation of the current process.

See also applicationCommandLineCreateFileForArg.

Since: 2.36

Constructs a File for a given path. This operation never fails, but the returned object might not support any I/O operation if path is malformed.

Constructs a File for a given URI. This operation never fails, but the returned object might not support any I/O operation if uri is malformed or if the uri type is not supported.

Opens a file in the preferred directory for temporary files (as returned by getTmpDir) and returns a File and FileIOStream pointing to it.

tmpl should be a string in the GLib file name encoding containing a sequence of six 'X' characters, and containing no directory components. If it is Nothing, a default template is used.

Unlike the other File constructors, this will return Nothing if a temporary file could not be created.

Since: 2.32

Opens an existing file for reading and writing. The result is a FileIOStream that can be used to read and write the contents of the file.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If the file does not exist, the IOErrorEnumNotFound error will be returned. If the file is a directory, the IOErrorEnumIsDirectory error will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on. Note that in many non-local file cases read and write streams are not supported, so make sure you really need to do read and write streaming, rather than just opening for reading or writing.

Since: 2.22

Asynchronously opens file for reading and writing.

For more details, see fileOpenReadwrite which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileOpenReadwriteFinish to get the result of the operation.

Since: 2.22

Finishes an asynchronous file read operation started with fileOpenReadwriteAsync.

Since: 2.22

Constructs a File with the given parseName (i.e. something given by fileGetParseName). This operation never fails, but the returned object might not support any I/O operation if the parseName cannot be parsed.

Exactly like fileGetPath, but caches the result via g_object_set_qdata_full(). This is useful for example in C applications which mix g_file_* APIs with native ones. It also avoids an extra duplicated string when possible, so will be generally more efficient.

This call does no blocking I/O.

Since: 2.56

Polls a file of type G_FILE_TYPE_MOUNTABLE.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

When the operation is finished, callback will be called. You can then call fileMountMountableFinish to get the result of the operation.

Since: 2.22

Finishes a poll operation. See filePollMountable for details.

Finish an asynchronous poll operation that was polled with filePollMountable.

Since: 2.22

Returns the AppInfo that is registered as the default application to handle the file specified by file.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Async version of fileQueryDefaultHandler.

Since: 2.60

Finishes a fileQueryDefaultHandlerAsync operation.

Since: 2.60

Utility function to check if a particular file exists. This is implemented using fileQueryInfo and as such does blocking I/O.

Note that in many cases it is racy to first check for file existence and then execute something based on the outcome of that, because the file might have been created or removed in between the operations. The general approach to handling that is to not check, but just do the operation and handle the errors as they come.

As an example of race-free checking, take the case of reading a file, and if it doesn't exist, creating it. There are two racy versions: read it, and on error create it; and: check if it exists, if not create it. These can both result in two processes creating the file (with perhaps a partially written file as the result). The correct approach is to always try to create the file with fileCreate which will either atomically create the file or fail with a IOErrorEnumExists error.

However, in many cases an existence check is useful in a user interface, for instance to make a menu item sensitive/insensitive, so that you don't have to fool users that something is possible and then just show an error dialog. If you do this, you should make sure to also handle the errors that can happen due to races when you execute the operation.

Utility function to inspect the FileType of a file. This is implemented using fileQueryInfo and as such does blocking I/O.

The primary use case of this method is to check if a file is a regular file, directory, or symlink.

Since: 2.18

Similar to fileQueryInfo, but obtains information about the filesystem the file is on, rather than the file itself. For instance the amount of space available and the type of the filesystem.

The attributes value is a string that specifies the attributes that should be gathered. It is not an error if it's not possible to read a particular requested attribute from a file - it just won't be set. attributes should be a comma-separated list of attributes or attribute wildcards. The wildcard "*" means all attributes, and a wildcard like "filesystem::*" means all attributes in the filesystem namespace. The standard namespace for filesystem attributes is "filesystem". Common attributes of interest are FILE_ATTRIBUTE_FILESYSTEM_SIZE (the total size of the filesystem in bytes), FILE_ATTRIBUTE_FILESYSTEM_FREE (number of bytes available), and FILE_ATTRIBUTE_FILESYSTEM_TYPE (type of the filesystem).

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If the file does not exist, the IOErrorEnumNotFound error will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on.

Asynchronously gets the requested information about the filesystem that the specified file is on. The result is a FileInfo object that contains key-value attributes (such as type or size for the file).

For more details, see fileQueryFilesystemInfo which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileQueryInfoFinish to get the result of the operation.

Finishes an asynchronous filesystem info query. See fileQueryFilesystemInfoAsync.

Gets the requested information about specified file. The result is a FileInfo object that contains key-value attributes (such as the type or size of the file).

The attributes value is a string that specifies the file attributes that should be gathered. It is not an error if it's not possible to read a particular requested attribute from a file - it just won't be set. attributes should be a comma-separated list of attributes or attribute wildcards. The wildcard "*" means all attributes, and a wildcard like "standard::*" means all attributes in the standard namespace. An example attribute query be "standard::*,owner[user](#g:signal:user)". The standard attributes are available as defines, like FILE_ATTRIBUTE_STANDARD_NAME.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

For symlinks, normally the information about the target of the symlink is returned, rather than information about the symlink itself. However if you pass G_FILE_QUERY_INFO_NOFOLLOW_SYMLINKS in flags the information about the symlink itself will be returned. Also, for symlinks that point to non-existing files the information about the symlink itself will be returned.

If the file does not exist, the IOErrorEnumNotFound error will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on.

Asynchronously gets the requested information about specified file. The result is a FileInfo object that contains key-value attributes (such as type or size for the file).

For more details, see fileQueryInfo which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileQueryInfoFinish to get the result of the operation.

Finishes an asynchronous file info query. See fileQueryInfoAsync.

Obtain the list of settable attributes for the file.

Returns the type and full attribute name of all the attributes that can be set on this file. This doesn't mean setting it will always succeed though, you might get an access failure, or some specific file may not support a specific attribute.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Obtain the list of attribute namespaces where new attributes can be created by a user. An example of this is extended attributes (in the "xattr" namespace).

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

Opens a file for reading. The result is a FileInputStream that can be used to read the contents of the file.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If the file does not exist, the IOErrorEnumNotFound error will be returned. If the file is a directory, the IOErrorEnumIsDirectory error will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on.

Asynchronously opens file for reading.

For more details, see fileRead which is the synchronous version of this call.

When the operation is finished, callback will be called. You can then call fileReadFinish to get the result of the operation.

Finishes an asynchronous file read operation started with fileReadAsync.

Returns an output stream for overwriting the file, possibly creating a backup copy of the file first. If the file doesn't exist, it will be created.

This will try to replace the file in the safest way possible so that any errors during the writing will not affect an already existing copy of the file. For instance, for local files it may write to a temporary file and then atomically rename over the destination when the stream is closed.

By default files created are generally readable by everyone, but if you pass G_FILE_CREATE_PRIVATE in flags the file will be made readable only to the current user, to the level that is supported on the target filesystem.

If cancellable is not Nothing, then the operation can be cancelled by triggering the cancellable object from another thread. If the operation was cancelled, the error IOErrorEnumCancelled will be returned.

If you pass in a non-Nothing etag value and file already exists, then this value is compared to the current entity tag of the file, and if they differ an IOErrorEnumWrongEtag error is returned. This generally means that the file has been changed since you last read it. You can get the new etag from fileOutputStreamGetEtag after you've finished writing and closed the FileOutputStream. When you load a new file you can use fileInputStreamQueryInfo to get the etag of the file.

If makeBackup is True, this function will attempt to make a backup of the current file before overwriting it. If this fails a IOErrorEnumCantCreateBackup error will be returned. If you want to replace anyway, try again with makeBackup set to False.

If the file is a directory the IOErrorEnumIsDirectory error will be returned, and if the file is some other form of non-regular file then a IOErrorEnumNotRegularFile error will be returned. Some file systems don't allow all file names, and may return an IOErrorEnumInvalidFilename error, and if the name is to long IOErrorEnumFilenameTooLong will be returned. Other errors are possible too, and depend on what kind of filesystem the file is on.