R Client for the MTurk Requester API
GetAssignments()
now attempts to convert non-character variables to an appropriate R vector type via type.convert(, as.is = TRUE)
.GetStatistic()
and RequesterReport()
, including "NumberAssignmentsAvailable", "NumberAssignmentsAccepted", "NumberAssignmentsReturned", "TotalFeePayout", and "TotalRewardAndFeePayout". This is reflected in the return value of ListStatistics()
.? MTurkR
. (#93)GrantBonus()
allows the reason
argument to contain newlines and tabs. (#109)CreateQualificationType()
is called with validate.answerkey = TRUE
. (#117, h/t Darrell Penta)BulkCreateFromHITLayout()
that caused the function to fail. (#114, h/t Tyler Scott)GetBonuses()
. (#113, h/t Andrew Brown)credentials()
function. (h/t Kim Gross)GenerateExternalQuestion
now sets a default value of frame.height = 400
.ListQualificationTypes()
. (#104)credentials()
, and options("MTurkR.keypair")
are completely deprecated. A warning will now be issued if trying to supply credentials in this way. AWS credentials should be specified in environment variables AWS_ACCESS_KEY_ID
and AWS_SECRET_ACCESS_KEY
. Messages are triggered on package load if environment variables are not set. (#102)HITStatus()
will now display any non-empty RequesterAnnotation field for a HIT. (#101)print
, which was removed in v0.5, has been fully removed. Use verbose
instead. (#97)mturkrhelp
and APIReference
have been removed. (#96)MTurkR.Wizard
has been removed from MTurkR and has been recreated as a separate package called MTurkRGUI. (#95)HITStatus
now reports condensed column names better suited for narrow displays.response.group
arguments passed to HIT-related functions. (#89)authenticate
, which was deprecated in v0.5, is removed.credentials
, AWS credentials should be specified via environment variables AWS_ACCESS_KEY_ID
and AWS_SECRET_ACCESS_KEY
. credentials
will be deprecated.SufficientFunds()
reflects the new commission structure effective July 21, 2015. (#87)BulkCreateFromTemplate
and BulkCreateFromURLs
now have a frame.height
argument with a default of 450.GenerateHITReviewPolicy
and GenerateAssignmentReviewPolicy
was expanded and tested, and a related HTMLQuestion HIT template is now installed.request()
. (h/t Sean Murphy, #106)bringToTop(-1)
. This is fixed. (h/t Carolina Kuepper-Tetzel)MTurkR 0.6.5 has been sent to CRAN. In addition to various bug fixes, this release includes two major sets of updates:
BulkCreate
, BulkCreateFromTemplate
, BulkCreateFromURLs
, and BulkCreateFromHITLayout
. These functions should greatly simplify managing large, multi-HIT projects (e.g., for coding, content analysis, categorization, transcription, etc.). See package documentation and a tutorial on the MTurkR wiki for full details.This release primarily updates the graphical wizard (MTurkR.Wizard
) for interacting with the MTurk API using a point-and-click interface. The interface has been significantly overhauled to improve its visual appearance by both updating layouts manually and converting to "themed tk" widgets (supported by tcl/tk v8.5). The overhaul also includes support for additional MTurkR features, documentation on the MTurkR wiki, and better handling of HITTypeId, HITId, and QualificationTypeId values.
The release also includes a few other changes:
ContactWorkers
(when batch = TRUE
) that contacted workers multiple times.CreateHIT
and CreateQualificationType
and the inclusion of template files for better understanding the HTMLQuestion, QuestionForm, and AnswerKey data structures used by MTurk.This release includes a large number of bug fixes, documentation corrections, and a large number of user-visible changes to function arguments meant to substantially simplify package use. Some highlights:
GenerateQualificationRequirement
now supports new comparators (given MTurk API update on 2014-07-17): "DoesNotExist", "In", and "NotIn". The "DoesNotExist" comparator allows requesters to effectively block workers who do not have, e.g., a requester-defined qualification. The "In" and "NotIn" comparators allow a boolean OR logical in QualificationRequirements (e.g., allowing a worker to be from the US or the UK) rather than having to create two separate HITs with different Locale restrictions. See documentation for details.signature
argument to request
now defaults to NULL
and is generated automatically (authenticate
is therefore deprecated). ParseErrorCodes
is removed (now used internally in request
). Arguments can be passed to request
via a new ...
argument in all functions.options('MTurkR.print')
is deleted and replaced by options('MTurkR.verbose')
. All functions will, temporarily, accept print
arguments. All code should be updated to reflect the new terminology moving forward.*ToDataFrame
have been replaced with as.data.frame
S3 methods, which are handled by a global as.data.frame.MTurkResponse
function.This release includes several smaller bug fixes and a few larger enhancements, described below.
GenerateReviewPolicy
is removed in lieu of two new functions: GenerateHITReviewPolicy
and GenerateAssignmentReviewPolicy
, both of which have a simplified function API and include extensive examples in the documentation. The use of CreateHIT
remains unchanged.HITStatus
, ApproveAllAssignments
, ChangeHITType
, DisposeHIT
, DisableHIT
, ExpireHIT
, ExtendHIT
, GetAssignment
, GetBonuses
, SetHITAsReviewing
now include an optional annotation
argument that can be used to apply the function to all HITs of a given batch. For example, to add two assignments to every HIT in a batch, simply find the batch number from the RUI and pass it as follows: ExtendHIT(annotation = "BatchId:78382;", add.assignments = 2)
.wizard.simple
and graphical mturkr.wizard
interactive interfaces for MTurkR.NULL
on certain failures where dataframes have always been returned on successes.See NEWS and GitHub commit logs for full details.