Usage
track.start(dir="rdatadir", pos = 1, envir = as.environment(pos), create = TRUE, clobber = c("no", "files", "variables", "vars", "var"), discardMissing = FALSE, cache = NULL, cachePolicy = NULL, options = NULL, RDataSuffix = NULL, auto = NULL, readonly = FALSE, lockEnv = FALSE, check.Last = TRUE, autoCheckSize = 1e6, verbose = TRUE)
track.stop(pos = 1, envir = as.environment(pos), all = FALSE, stop.on.error = FALSE, keepVars = FALSE, sessionEnd = FALSE, verbose = TRUE, detach = TRUE, callFrom = NULL)
track.rescan(pos = 1, envir = as.environment(pos), discardMissing = FALSE, forgetModified = FALSE, level = c("low", "high"), dryRun = FALSE, verbose = TRUE)
track.Last()
Arguments
dir
The directory where tracking data is stored
pos
The search path position of the environment being tracked
(default is 1 for the global environment)
envir
The environment being tracked. This is an alternate way
(to the use of pos=)
of specifying the environment being tracked, but should be rarely needed.
create
If TRUE, create the tracking directory if it
doesn't exist
clobber
Controls action taken when there are objects of the
same name in the
tracking directory and in envir: no means stop (unless
objects are have the same values and are small); files means use the
version from the tracking directory; and variables,
vars or var means
use the version in envir (and overwrite the version in the
tracking directory). See also argument autoCheckSize.
discardMissing
Discard all information about objects whose save
file is missing?
cache
Should objects be keep cached in memory? Default is TRUE. This option is a shorthand way of supplying options=list(cache=...).
cachePolicy
Policy for keeping cached objects in memory.
Default is tltPurge, for which cached objects are removed
from memory at the end of a top-level task. This option is a shorthand way of supplying options=list(cachePolicy=...).
options
Option values (as a list) to be used for this tracking database. See track.options().
all
If TRUE, all tracked environment are unlinked
auto
If TRUE, automatically track new variables and
deleted variables in the environment (through use of a task
callback). If auto==NULL, take the value from
getOptions("global.track.options")$autoTrack, and if that is
NULL, use TRUE
readonly
Logical flag indicating whether the tracking db should be attached in
a readonly mode. The global environment (pos=1 in the search
path) cannot be tracked in a readonly mode.
stop.on.error
If FALSE, failures to unlink a tracking
environment are ignored, though a warning message is printed
keepVars
If FALSE, all tracked variables are removed and
will be no longer accessible. If TRUE, tracked variables will
be left as ordinary variables in the environment, as well as remaining
in files.
detach
If TRUE, the environment attached to the search
path (in a position other than 2) will be detached after stopping
tracking, IF it was created by track.attach() and if there are no
variables left remaining in the environment after removing
all tracked variables. If detach="force", the attached
environment will be removed even if there are variables remaining in
it (though not if it was not created by track.attach).
callFrom
A character string used in a message saying where
track.stop() was called from.
forgetModified
If TRUE, discard the versions of objects that are
modified and in memory
RDataSuffix
The suffix to use for RData files. This
should not normally need to be specified.
lockEnv
Should the environment be locked for a readonly
tracking environment? The default is FALSE because locking the
environment is irreversible, and it prevents rescanning or caching
(because can't delete or add bindings)
check.Last
By default, a warning is issued if the .Last
function in the track package is masked by any other
.Last function. Supplying check.Last=FALSE
inhibits this check and warning.
autoCheckSize
If there are objects with the same name in an
existing tracking db being attached, and the environment, and clobber='no',
track.start() will check whether the objects are the same
(using identical()). If they are the same,
track.start() will proceed. This check is only performed if
the total size of the objects, and the files, is less than
autoCheckSize bytes.
level
Should the rescan be done at a high or low level: a high
level just stops tracking and restarts it; a low level tries to
individually bring the environment in line with the file database.
dryRun
If TRUE, no changes are actualy made, but
messages are printed describing what changes would be made.
sessionEnd
If TRUE, this is a call at the end of a
session and no recovery from errors is possible -- just try to
as best as can to save objects as appropriate.
verbose
print a message about what directory is being tracked?