# Unique code of this Data Store Server. Not more than 40 characters.
data-store-server-code = DSS2

data-folder = targets/playground/data

# The root directory of the data store
storeroot-dir = ${data-folder}/store

# The directory where the command queue file is located; defaults to storeroot-dir 
commandqueue-dir =

# Port
port = 8890

# Session timeout in minutes
session-timeout = 720

# Path to the keystore
keystore.path = ../datastore_server/dist/etc/openBIS.keystore

# Password of the keystore
keystore.password = changeit

# Key password of the keystore
keystore.key-password = changeit

# The check interval (in seconds)
check-interval = 5

# The time-out for clean up work in the shutdown sequence (in seconds).
# Note that that the maximal time for the shutdown sequence to complete can be as large 
# as twice this time.
# Remark: On a network file system, it is not recommended to turn this value to something 
# lower than 180.
shutdown-timeout = 2

# If free disk space goes below value defined here, a notification email will be sent.
# Value must be specified in kilobytes (1048576 = 1024 * 1024 = 1GB). If no high water mark is
# specified or if value is negative, the system will not be watching.
highwater-mark = -1

# If a data set is successfully registered it sends out an email to the registrator. 
# If this property is not specified, no email is sent to the registrator. This property
# does not affect the mails which are sent, when the data set could not be registered.
notify-successful-registration = false

# The URL of the openBIS server
server-url = http://localhost:8888/openbis

# The username to use when contacting the openBIS server
username = etlserver

# The password to use when contacting the openBIS server
password = etlserver

# The base URL for Web client access.
download-url = https://localhost:${port}

# SMTP properties (must start with 'mail' to be considered). 
mail.smtp.host = file://targets/email
mail.from = datastore_server@localhost

# ---------------- Database parameters

basysbio-databaseEngineCode = postgresql
basysbio-basicDatabaseName = basysbio
basysbio-databaseKind = dev
basysbio-readOnlyGroup = basysbio_readonly
basysbio-readWriteGroup = basysbio_readwrite
basysbio-scriptFolder = source/sql

# ---------------- Timing parameters for file system operations on remote shares.

# Time (in seconds) to wait for any file system operation to finish. Operations exceeding this 
# timeout will be terminated. 
timeout = 60
# Number of times that a timed out operation will be tried again (0 means: every file system 
# operation will only ever be performed once).
max-retries = 11
# Time (in seconds) to wait after an operation has been timed out before re-trying.  
failure-interval = 10 

# The period of no write access that needs to pass before an incoming data item is considered 
# complete and ready to be processed (in seconds) [default: 300]. 
# Valid only when auto-detection method is used to determine if an incoming data are ready to be processed.
quiet-period = 10

#maintenance-plugins = post-registration-upload

post-registration-upload.class = eu.basysbio.cisd.dss.PostRegistrationDatabaseUploadTask
post-registration-upload.interval = 1440000
post-registration-upload.database.databaseEngineCode = ${basysbio-databaseEngineCode}
post-registration-upload.database.basicDatabaseName = ${basysbio-basicDatabaseName}
post-registration-upload.database.databaseKind = ${basysbio-databaseKind}
post-registration-upload.database.readOnlyGroup = ${basysbio-readOnlyGroup}
post-registration-upload.database.readWriteGroup = ${basysbio-readWriteGroup}
post-registration-upload.database.scriptFolder = ${basysbio-scriptFolder}

# Data Set Validators
data-set-validators = time-series-validator, LCA-MTP-PCAV-time-series-validator

LCA-MTP-PCAV-time-series-validator.data-set-type = LCA_MTP_PCAV_TIME_SERIES
LCA-MTP-PCAV-time-series-validator.path-patterns = *
LCA-MTP-PCAV-time-series-validator.exclude-path-patterns = request.properties
LCA-MTP-PCAV-time-series-validator.columns = id, no-of-replicates, confidence-level, data-column
LCA-MTP-PCAV-time-series-validator.id.header-pattern = BSB_ID
LCA-MTP-PCAV-time-series-validator.id.mandatory = true
LCA-MTP-PCAV-time-series-validator.id.order = 1
LCA-MTP-PCAV-time-series-validator.id.value-type = unique
LCA-MTP-PCAV-time-series-validator.id.value-pattern = BBA[0-9]+
LCA-MTP-PCAV-time-series-validator.no-of-replicates.header-pattern = no of replicates
LCA-MTP-PCAV-time-series-validator.no-of-replicates.order = 2
LCA-MTP-PCAV-time-series-validator.no-of-replicates.value-type = string
LCA-MTP-PCAV-time-series-validator.no-of-replicates.value-pattern = [0-9]+
LCA-MTP-PCAV-time-series-validator.confidence-level.header-pattern = confidence level
LCA-MTP-PCAV-time-series-validator.confidence-level.order = 3
LCA-MTP-PCAV-time-series-validator.confidence-level.value-type = string
LCA-MTP-PCAV-time-series-validator.confidence-level.value-pattern = [A-Z]
LCA-MTP-PCAV-time-series-validator.data-column.header-validator = eu.basysbio.cisd.dss.DataColumnHeaderValidator
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.elements = experiment-code, cultivation-method, biological-replicate-code, \
			time-point, time-point-type, technical-replicate-code, cel-loc, data-set-type, value-type, scale, bi-id, cg
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.experiment-code.type = vocabulary 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.experiment-code.terms = Ma, GI, GM, MG 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.cultivation-method.type = vocabulary 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.cultivation-method.terms = BR, SF, MW, MS 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.biological-replicate-code.type = string 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.biological-replicate-code.pattern = BN 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.time-point.type = integer 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.time-point-type.type = vocabulary 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.time-point-type.terms = EX, IN, SI 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.technical-replicate-code.type = string 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.technical-replicate-code.pattern = T[0-9]+(_T[0-9]+)* 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.cel-loc.type = vocabulary 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.cel-loc.terms = CE, ES, ME, CY, NC 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.data-set-type.type = vocabulary 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.data-set-type.terms = LcaMtpPaCav 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.value-type.type = string  
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.value-type.pattern = Average\\[U\\]  
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.scale.type = vocabulary 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.scale.terms = LIN, Lin, Log2, Log10, Ln 
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.bi-id.type = string  
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.bi-id.pattern = NB  
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.cg.type = string  
LCA-MTP-PCAV-time-series-validator.data-column.header-validator.cg.pattern = NC  
LCA-MTP-PCAV-time-series-validator.data-column.can-define-multiple-columns = true
LCA-MTP-PCAV-time-series-validator.data-column.value-type = numeric

time-point-data-set-types = AgilentRNA2color, MetaboliteLCMS

time-series-validator.data-set-type = TIME_SERIES
time-series-validator.path-patterns = *
time-series-validator.exclude-path-patterns = request.properties
time-series-validator.columns = id, human-readable, controlled-gene, data-column
time-series-validator.id.header-pattern = CompoundID|GeneLocus|Abs
time-series-validator.id.mandatory = true
time-series-validator.id.order = 1
time-series-validator.id.value-validator = ch.systemsx.cisd.etlserver.validation.HeaderBasedValueValidatorFactory
time-series-validator.id.header-types = compound, gene-locus, abs
time-series-validator.id.compound.header-pattern = CompoundID
time-series-validator.id.compound.value-type = unique
time-series-validator.id.compound.value-pattern = BSBME:[0-9]+(BSBME:[0-9])*|CHEBI:[0-9]+(_CHEBI:[0-9]+)*
time-series-validator.id.gene-locus.header-pattern = GeneLocus
time-series-validator.id.gene-locus.value-type = unique
time-series-validator.id.gene-locus.value-pattern = BSU[0-9]+|BSU_misc_RNA_[0-9]+|VMG_[0-9]+_[0-9]+|VMG_[0-9]+_[0-9]+_c
time-series-validator.id.abs.header-pattern = Abs
time-series-validator.id.abs.value-type = unique
time-series-validator.id.abs.value-pattern = OD600
time-series-validator.human-readable.header-pattern = HumanReadable
time-series-validator.human-readable.order = 2
time-series-validator.controlled-gene.header-pattern = ControlledGene
time-series-validator.controlled-gene.order = 3
time-series-validator.data-column.header-validator = eu.basysbio.cisd.dss.DataColumnHeaderValidator
time-series-validator.data-column.header-validator.elements = experiment-code, cultivation-method, biological-replicate-code, \
			time-point, time-point-type, technical-replicate-code, cel-loc, data-set-type, value-type, scale, bi-id, cg
time-series-validator.data-column.header-validator.experiment-code.type = vocabulary 
time-series-validator.data-column.header-validator.experiment-code.terms = Ma, GI, GM, MG 
time-series-validator.data-column.header-validator.cultivation-method.type = vocabulary 
time-series-validator.data-column.header-validator.cultivation-method.terms = BR, SF, MW, MS 
time-series-validator.data-column.header-validator.biological-replicate-code.type = string 
time-series-validator.data-column.header-validator.biological-replicate-code.pattern = B[0-9]+(_B[0-9]+)* 
time-series-validator.data-column.header-validator.time-point.type = integer 
time-series-validator.data-column.header-validator.time-point-type.type = vocabulary 
time-series-validator.data-column.header-validator.time-point-type.terms = EX, IN, SI 
time-series-validator.data-column.header-validator.technical-replicate-code.type = string 
time-series-validator.data-column.header-validator.technical-replicate-code.pattern = T[0-9]+(_T[0-9]+)* 
time-series-validator.data-column.header-validator.cel-loc.type = vocabulary 
time-series-validator.data-column.header-validator.cel-loc.terms = CE, ES, ME, CY, NC 
time-series-validator.data-column.header-validator.data-set-type.type = vocabulary 
time-series-validator.data-column.header-validator.data-set-type.terms = ${time-point-data-set-types} 
time-series-validator.data-column.header-validator.value-type.type = string  
time-series-validator.data-column.header-validator.value-type.pattern = (Value|Mean|Median|Std|Var|Error|Iqr)\\[(mM|Ratio1|RatioCs)\\]  
time-series-validator.data-column.header-validator.scale.type = vocabulary 
time-series-validator.data-column.header-validator.scale.terms = LIN, Log2, Log10, Ln 
time-series-validator.data-column.header-validator.bi-id.type = string  
time-series-validator.data-column.header-validator.bi-id.pattern = NB|BBA[0-9]*  
time-series-validator.data-column.header-validator.cg.type = string  
time-series-validator.data-column.header-validator.cg.pattern = NC|BSU[0-9]+_[0-9]  
time-series-validator.data-column.can-define-multiple-columns = true
time-series-validator.data-column.value-type = numeric


# Comma separated names of processing threads. Each thread should have configuration properties prefixed with its name.
# E.g. 'code-extractor' property for the thread 'my-etl' should be specified as 'my-etl.code-extractor'
inputs = main-thread

# ---------------------------------------------------------------------------
# main thread configuration
# ---------------------------------------------------------------------------
# The directory to watch for incoming data.
main-thread.incoming-dir = ${data-folder}/incoming
main-thread.incoming-data-completeness-condition = auto-detection
main-thread.delete-unidentified = true
main-thread.data-set-info-extractor = eu.basysbio.cisd.dss.DataSetInfoExtractor
main-thread.type-extractor = ch.systemsx.cisd.etlserver.cifex.CifexTypeExtractor
main-thread.storage-processor = ch.systemsx.cisd.etlserver.CifexStorageProcessor
main-thread.storage-processor.keep-file-regex = .*(?<!request\.properties)$ 
main-thread.storage-processor.processor = eu.basysbio.cisd.dss.StorageProcessor
main-thread.storage-processor.processor.processor = ch.systemsx.cisd.etlserver.DefaultStorageProcessor
main-thread.storage-processor.processor.time-series-data-set-drop-box-path = ${main-thread.incoming-dir}
main-thread.storage-processor.processor.time-point-data-set-drop-box-path = ${main-thread.dataset-handler.incoming-dir}
main-thread.storage-processor.processor.time-point-data-set-file-name-separator = .
main-thread.storage-processor.processor.data-set-properties-file-name = data-set.properties
main-thread.storage-processor.processor.translation.data-set-types = ${time-point-data-set-types}, LcaMtpPaCav
main-thread.storage-processor.processor.translation.AgilentRNA2color = AGILENT_RNA_2_COLOR
main-thread.storage-processor.processor.translation.MetaboliteLCMS = METABOLITE_LCMS
main-thread.storage-processor.processor.translation.LcaMtpPaCav = LCA_MTP_PA_CAV
main-thread.storage-processor.processor.database.databaseEngineCode = ${basysbio-databaseEngineCode}
main-thread.storage-processor.processor.database.basicDatabaseName = ${basysbio-basicDatabaseName}
main-thread.storage-processor.processor.database.databaseKind = ${basysbio-databaseKind}
main-thread.storage-processor.processor.database.readOnlyGroup = ${basysbio-readOnlyGroup}
main-thread.storage-processor.processor.database.readWriteGroup = ${basysbio-readWriteGroup}
main-thread.storage-processor.processor.database.scriptFolder = ${basysbio-scriptFolder}

main-thread.dataset-handler = eu.basysbio.cisd.dss.TimeSeriesAndTimePointDataSetHandler
main-thread.dataset-handler.incoming-dir = ${data-folder}/incoming-time-point
main-thread.dataset-handler.incoming-data-completeness-condition = auto-detection
main-thread.dataset-handler.data-set-info-extractor = ch.systemsx.cisd.etlserver.DefaultDataSetInfoExtractor
main-thread.dataset-handler.data-set-info-extractor.strip-file-extension = true
main-thread.dataset-handler.data-set-info-extractor.entity-separator = ${main-thread.storage-processor.processor.time-point-data-set-file-name-separator}
main-thread.dataset-handler.data-set-info-extractor.index-of-sample-code = 0
main-thread.dataset-handler.data-set-info-extractor.space-code = TEST
main-thread.dataset-handler.data-set-info-extractor.data-set-properties-file-name = data-set.properties
main-thread.dataset-handler.type-extractor = eu.basysbio.cisd.dss.TypeExtractor
main-thread.dataset-handler.type-extractor.file-format-type = PROPRIETARY
main-thread.dataset-handler.type-extractor.locator-type = RELATIVE_LOCATION
main-thread.dataset-handler.type-extractor.is-measured = true
main-thread.dataset-handler.storage-processor = ch.systemsx.cisd.etlserver.DefaultStorageProcessor