CHANGELOG.md

0.0.38 (2024-07-03)

• add tundra.smtp:send for sending emails via SMTP
• change ./pub/configuration/index.dsp to include a refresh all button
• change tundra.content.deliver:mailto to use tundra.smtp:send rather than pub.client:smtp for its SMTP implementation
• change tundra.html:emit to support specifying the orientation used when emitting IData documents
• fix tundra.excel:emit when given a String value to suppress number format exceptions (if the value cannot be parsed as a number using the BigDecimal grammar) and fall back to treating the value as a String
• fix tundra.system:reflect to support both package-level and server-level global variables in Integration Server 10.15
• fix DSP pages to always use light mode/theme to match webMethods page style
• fix scheduling of tundra.support.log:purge for log compression and purging to create and remove against target $self only
• upgrade Bulma CSS framework to v1.0.0
• upgrade Fontawesome icons to v6.5.1

0.0.37 (2023-10-12)

• add ./pub/configuration/ DSP pages for viewing and editing package configuration (note that only server-specific package configuration files are supported for editing via this user interface)
• add tundra.mime.type:text for returning whether a given MIME media type describes a text-based data type
• add tundra.schedule.create:list for creating a list of scheduled tasks from an IData[] document list input
• add tundra.schedule.create:pivot for creating a list of scheduled tasks from a pivoted IData document input
• add tundra.schedule:expedite to run a given scheduled task immediately rather than wait for its next scheduled time
• add tundra.security.hmac:authenticate authenticates a given HMAC authentication code against given content, secret key, and algorithm
• add tundra.security.hmac:calculate calculates an HMAC authentication code for the given content, secret key, and algorithm
• add tundra.security.hmac:verify verifies a given HMAC authentication code against the given content, secret key, and algorithm
• add tundra.service:redirect for registering a service redirection which then routes all subsequent invocations of the given source service to the given target service
• add tundra.support.system.acl:export and tundra.support.system.acl:merge services to support migrating ACLs from one Integration Server instance to another
• add tundra.support.system.global:export and :merge services to support migrating global variables from one Integration Server instance to another
• add Tundra package configuration settings to support auto-starting the service usage and service statistics features at package startup (note: both of the above settings default to false to preserve existing behaviour): feature/service/usage/started == true | false, and feature/service/statistics/started == true | false
• change pub/message/format/list.dsp enabled/disabled icon to improve clarity
• change tundra.bool:format to support formatting arbitrarily specified strings
• change tundra.collection.list.*:arrayify to also return the array length for convenience
• change tundra.collection.map.*:keys to also return the number of keys for convenience
• change tundra.collection.map.*:values to also return the number of values for convenience
• change tundra.content.deliver:http to gzip compress the request body if the following HTTP header is specified: X-Request-Body-Compression: gzip (note that this is an alternative to the standard and already supported approach for HTTP request body compression, which is to specify the HTTP header Content-Encoding: gzip; the X-Request-Body-Compression header, unlike the standard Content-Encoding header, will not be handled automatically by HTTP servers or intermediates, and therefore provides a way to tunnel gzip compressed content through HTTP intermediates to be handled in the software that is ultimately processing the HTTP request)
• change tundra.content.deliver:jms to use a javax.jms.TextMessage when the content type is a text-based data type, otherwise a javax.jms.BytesMessage is used to send the JMS message
• change tundra.content.deliver:sftp to only change directory when required, and to ignore errors when touching the delivered file to set the modification datetime as delivery should be considered successful even if the touch operation does not work for some reason
• change tundra.content.retrieve:jms to support arbitrary pipeline variables being provided to the content processing service
• change tundra.content:deliver file delivery to support specifying the size in bytes of the in memory buffer used when writing the content to the file using the query string parameter buffer
• change tundra.content:deliver parameters to be consistent with other content processing services
• change tundra.content:retrieve to support retrieving JMS message content from a JMS alias topic or queue using a source URI such as: jms://jms-alias-name?queue=jms-queue-name or jms://jms-alias-name?topic=jms-topic-name
• change tundra.content:split parameter naming to be consistent with other content processing services
• change tundra.directory:purge and tundra.directory:gzip to continue processing files even if an exception is encountered, and to throw all encountered exceptions after attempting to process all files rather than failing fast on the first exception encountered
• change tundra.document:deliver parameter naming to be consistent with other content processing services
• change tundra.file:copy and tundra.file:write to support specifying the size in bytes of the in memory buffer used when reading, copying, and writing data to files with the new input parameter $file.buffer.size
• change tundra.http:client $request/method to include patch method in the pick list of methods
• change tundra.http:client implementation to Java to improve performance
• change tundra.http:client to silently support passing trustStore and followRedirect inputs to pub.client:http if they are provided
• change tundra.list.object:sort and tundra.list.string:sort to support sorting in either ascending or descending order
• change tundra.list.string:normalize to use parameter naming consistent with other content processing services
• change tundra.packages:get, tundra.packages:list, and tundra.packages:self to include optional package description in output
• change tundra.schedule:create to not set $schedule.name on the scheduled task's input pipeline unless it is a singleton task
• change tundra.schedule:create to use a new single input parameter $schedule.create, which allows schedules to be created easily from an object stored in a JSON or similar configuration file format that supports hierarchical data structures.
• change tundra.stream:close input parameter name to $content.stream to be consistent with other content processing services
• change tundra.stream:copy parameter names to be consistent with other content processing services
• change tundra.stream:copy to support specifying the size in bytes of the in memory buffer used when copying data between the input and output streams with the new input parameter $stream.buffer.size
• change tundra.stream:normalize parameters to be consistent with other content processing services
• change tundra.stream:transcode parameters to be consistent with other content processing services
• change tundra.string:capitalize input parameter naming to be prefixed with the operation being performed
• change tundra.string:coalesce input parameter naming to be prefixed with the operation being performed
• change tundra.string:concatenate input parameter naming to be prefixed with the operation being performed
• change tundra.string:condense input parameter naming to be prefixed with the operation being performed
• change tundra.string:legalize input parameter naming to be prefixed with the operation being performed
• change tundra.string:lowercase to support converting one or more arbitrarily specified strings to lower case
• change tundra.string:normalize parameter naming to be consistent with other content related services
• change tundra.string:pad to support padding one or more arbitrarily specified strings
• change tundra.string:remove to support removing patterns from one or more arbitrarily specified strings
• change tundra.string:replace to support replacing patterns in one or more arbitrarily specified string values
• change tundra.string:reverse to support reversing one or more arbitrarily specified strings
• change tundra.string:slice parameter naming to be prefixed with the operation being performed
• change tundra.string:split parameter naming to be prefixed with the operation being performed
• change tundra.string:translate to support excluding values with no translation, and to support dedicated reverse default and null values
• change tundra.string:translate to support raising an exception when translation values are missing
• change tundra.string:trim to support trimming one or more arbitrarily specified string values
• change tundra.string:truncate parameter naming to be prefixed with the operation being performed
• change tundra.string:uppercase to support converting one or more arbitrarily specified strings to upper case
• change tundra.support.content.deliver:sftp to use $content.encoding rather than $encoding input parameter
• change tundra.thread:current, tundra.thread:get, and tundra.thread:list to return the Integration Server generated thread dump for the thread in question
• change tundra.uri:emit to not include explicit port in resulting URI string if port specified is the default port for the given scheme
• change tundra.uri:substitute to support both local and global variable substitution
• change tundra:log input parameter $log.context to support any type of object rather than just IData documents
• change thread names used by Tundra thread pools for clarity and simplicity
• change Tundra variable substitution feature to support mustache-template style double-curly brace delimited variables and to support resolving recursive variable references
• fix tundra.bytes:normalize, tundra.stream:normalize, and tundra.string:normalize backwards-compatibility to work correctly when both old $object and new $content input parameters both exist in the pipeline
• fix tundra.content.deliver:jms to use a javax.jms.TextMessage when the content type is a text-based data type, otherwise a javax.jms.BytesMessage is used to send the JMS message
• fix tundra.content.deliver:mailto to correctly use $content.encoding input parameter
• fix tundra.content:deliver branch on $destination to work correctly
• fix tundra.content:translate to not drop TN_parms
• fix tundra.csv:emit to support IData[] arrays where each IData item can specify its elements in different order (previously all items were strictly required to specify elements in the same order)
• fix tundra.csv:emit to work correctly when $content.headings input parameter is specified with column names different to the keys in the provided $document/recordWithNoID IData[] document list
• fix tundra.datetime:format unparseable exception message to be more readable
• fix tundra.list.content:emit backwards compatibility support for $namespaces input parameter
• fix tundra.list.html:emit to work correctly and use parameter naming consistent with other content processing services
• fix tundra.message.trigger:recognize to work correctly rather than immediately throw a java.lang.reflect.InvocationTargetException exception
• fix tundra.schedule:expedite to work correctly on webMethods 10.11, by replacing binary search of task scheduler database columns with linear search
• fix tundra.service:restful to only log the output pipeline once, and to provide the correct original request URI in the log statement
• fix tundra.string:normalize backwards-compatibility to work correctly when $content and $object refer to the same java.io.InputStream object
• fix tundra.string:truncate to allow $truncate.length to be null
• fix tundra.uri:emit to correctly handle query string parameters that include plus characters
• fix tundra.uri:emit to support mailto: URIs that specify a query string but no scheme-specific part (body)
• fix tundra.uri:substitute to correctly handle when $string is null and not throw a NullPointerException
• fix compatibility with webMethods 10.7 by removing log4j dependency
• remove dependency on Bootstrap and jQuery libraries, replacing their use in the Tundra DSP pages instead with Bulma and Fontawesome
• remove deprecated tundra.document.key:lowercase and tundra.document.value:lowercase services, replaced with service redirections to tundra.document:lowercase to continue to provide backwards compatibility
• remove deprecated tundra.document.key:normalize, replaced with service redirection to tundra.document:legalize to continue to provide backwards compatibility
• remove deprecated tundra.document.key:remove and tundra.document.value:remove services, replaced with service redirections to tundra.document:remove to continue to provide backwards compatibility
• remove deprecated tundra.document.key:replace and tundra.document.value:replace services, replaced with service redirections to tundra.document:replace to continue to provide backwards compatibility
• remove deprecated tundra.document.key:trim and tundra.document.value:trim services, replaced with service redirections to tundra.document:trim to continue to provide backwards compatibility
• remove deprecated tundra.document.key:uppercase and tundra.document.value:uppercase services, replaced with service redirections to tundra.document:uppercase to continue to provide backwards compatibility
• remove deprecated tundra.id:generate service, replaced with a service redirection to tundra.uuid:generate to continue to provide backwards compatibility
• remove deprecated tundra.id:normalize service, replaced with a service redirections to tundra.string:legalize to continue to provide backwards compatibility
• remove deprecated tundra.list.document.key:lowercase and tundra.list.document.value:lowercase services, replaced with service redirections to tundra.list.document:lowercase to continue to provide backwards compatibility
• remove deprecated tundra.list.document.key:remove and tundra.list.document.value:remove services, replaced with service redirections to tundra.list.document:remove to continue to provide backwards compatibility
• remove deprecated tundra.list.document.key:replace and tundra.list.document.value:replace services, replaced with service redirections to tundra.list.document:replace to continue to provide backwards compatibility
• remove deprecated tundra.list.document.key:trim and tundra.list.document.value:trim services, replaced with service redirections to tundra.list.document:trim to continue to provide backwards compatibility
• remove deprecated tundra.list.document.key:uppercase and tundra.list.document.value:uppercase services, replaced with service redirections to tundra.list.document:uppercase to continue to provide backwards compatibility
• remove deprecated tundra.list.string:squeeze service, replaced with a service redirection to tundra.string:condense to continue to provide backwards compatibility
• remove deprecated tundra.soap.fault:handle service, replaced with a service redirection to tundra.soap.fault:raise to continue to provide backwards compatibility
• remove deprecated tundra.string:squeeze service, replaced with a service redirection to tundra.string:condense to continue to provide backwards compatibility
• replace tundra.object:convert with tundra.content:convert, and change tundra.content:translate parameter naming to be consistent with other content processing services
• update snakeyaml dependency from v1.18 to v1.23

0.0.36 (2021-07-01)

• add tundra.file:equal for comparing the contents of two files
• add tundra.service:retry for invoking a service and retrying it when the invocation fails
• add tundra.xml.namespace:normalize for normalizing namespace prefixes in the given IData document's keys and/or values to be the prefix with the highest precedence in the given namespace prefix declarations
• change README.md to recommend using git reset rather than git checkout
• change tundra.content:amend input variable names to be consistent with tundra.content:parse and tundra.content:emit
• change tundra.content:deliver ftp delivery to support a rename setting which when true will first write the remote file with a .tmp extension and then rename it, which ensures existing files are not overwritten
• change tundra.datetime:add to support datetime shifting arbitrarily specified datetime strings rather than just one string at a time
• change tundra.datetime:subtract to support datetime shifting arbitrarily specified datetime strings rather than just one string at a time
• change tundra.datetime:within to support start of range being after end of range
• change tundra.deliver:content to return optional $response.context to provide additional context to the delivery response, for example the response status and headers for an HTTP delivery
• change tundra.document:log to support disabling the prefixing of the log statement with the current user and callstack, and to support logging to arbitrarily named server log files
• change tundra.list.content:emit input variable names to be consistent with tundra.content:emit
• change tundra.pipeline:log to support disabling the prefixing of the log statement with the current user and callstack, and to support logging to arbitrarily named server log files
• change tundra.pipeline:sanitize to support the option to recursively sanitize child IData documents and IData[] document lists, and to not remove undeclared fields from documents which have been configured to "Allow unspecified fields"
• change tundra.service:invoke to return $duration truncated to millisecond precision
• change tundra.system:reflect to return the localhost address and unique reference identity and datetime, which are useful for common variable substitution scenarios
• change tundra:log to support disabling the prefixing of the log statement with the current user and callstack, and to support logging to arbitrarily named server log files
• fix tundra.content:amend to normalize namespace prefixes in the given $amendments/key values to be the prefix with the highest precedence in the given $content.namespace namespace declarations, so that the amended keys use the same prefixes as the parsed content
• fix tundra.content:deliver documentation of file: protocol to specify correct input pipeline parameter name $filemode which is used for controlling whether to append / write / create the file
• fix tundra.file:digest to work correctly with file:// URIs
• fix tundra.list.directory:compact to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations
• fix tundra.list.directory:squeeze to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations
• fix tundra.list.document:find to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations
• fix tundra.list.document:match to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations
• fix tundra.list.string:find to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations
• fix tundra.list.string:match to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations
• fix tundra.node:list to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations
• fix tundra.pipeline:clear to not throw java.lang.NullPointerException when $preserve argument is not specified
• fix tundra.support.content.retrieve.file:reflect to use a new empty array when calling List.toArray(T[]) method to take advantage of JVM optimizations

0.0.35 (2020-07-03)

• add tundra.content:transfer for transferring content retrieved from a source location and delivering directly to a destination location
• add tundra.datetime:within for determining if a given datetime is within a given range
• add tundra.file:digest for calculating a message digest for a given file
• add tundra.file:purge for deleting files matching the given filename pattern older than the given duration in the same directory
• add tundra.pipeline:sanitize for dropping undeclared inputs or outputs when compared with the current service's signature
• add tundra.pipeline:validate for imperatively validating the pipeline against the current service's input or output signature
• add tundra.service:restful for marking a service as a REST service, which then automatically handles output serialization, exception handling, and pipeline validation
• add tundra.service:synchronize for performing a thread-synchronized invoke of a service
• add tundra.thread:get to return the thread with the given identity
• add tundra.thread:interrupt to interrupt the thread with the given identity
• add tundra.thread:stop to stop the thread with the given identity (use sparingly, and at your own peril)
• change tundra.base64:decode to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra.base64:encode to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra.bytes:normalize to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra.bytes:transcode to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra.cache.memory:* services to support cache entry expiration
• change tundra.cache.memory:get to also return $cache.expiry.datetime when it is applicable to the cache entry in question
• change tundra.cache.memory:put to return the previously associated value if any and to return a boolean flag explicitly indicating if the attempted put updated the cache
• change tundra.content.deliver:file to set the last modified datetime of the file to the given $content.updated datetime if specified
• change tundra.content.deliver:file to support query string parameter filemode for specifying the file mode to use when calling tundra.file:write
• change tundra.content.deliver:http to use new input arguments to tundra.gzip:compress
• change tundra.content.deliver:jms to provide content schema and content type as JMSMessage properties named Content-Schema and Content-Type respectively
• change tundra.content.deliver:mailto to use the following Integration Server settings if they exist when using the default SMTP server: watt.server.smtpServerPort, watt.server.smtpTransportSecurity, and watt.server.smtpTrustStoreAlias
• change tundra.content.deliver:sftp to set the last modified datetime of the file to the given $content.updated datetime if specified
• change tundra.content.delivery:sap_idoc return message format to be consistent with the other deliver protocol implementations
• change tundra.content.retrieve:file to provide $content.updated to the content processing service
• change tundra.content.retrieve:sftp to provide $content.updated to the content processing service
• change tundra.content:emit implementation to Java for easier reuse by other components and better performance
• change tundra.content:parse implementation to Java for easier reuse by other components and better performance
• change tundra.content:retrieve file protocol handler to support limiting the files retrieved by specifying a new query string parameter updated to only retrieve files within the specified last modified time window
• change tundra.content:retrieve ftp retrieval to support disabling the file rename prior to retrieval by setting the query string parameter rename to false (example: ftp://user:password@server:port/directory/*.txt?rename=false)
• change tundra.content:retrieve sftp protocol handler to support limiting the files retrieved by specifying a new query string parameter updated to only retrieve files within the specified last modified time window
• change tundra.content:retrieve sftp retrieval to support disabling the file rename prior to retrieval by setting the query string parameter rename to false (example: sftp://useralias/./path/*.txt?rename=false)
• change tundra.content:translate to support translation service returning $content which can be used for IData to byte[] conversions
• change tundra.datetime:format to support formatting non-String objects
• change tundra.datetime services to support new named pattern seconds which is the seconds since the unix epoch
• change tundra.document:log implementation to Java for 10x performance improvement, and add support for an optional log message to prefix the logged IData document
• change tundra.document:substitute to support replacing an entire value with a non-String value
• change tundra.document:validate implementation to Java for easier reuse by other components and better performance
• change tundra.file:copy $mode input to be named $file.mode and default to create rather than append, which is now the safest default
• change tundra.file:match $mode input to be named $pattern.mode to remove ambiguity
• change tundra.file:process $mode input to be named $file.mode and add support for create mode which will not overwrite the file if it already exists
• change tundra.file:read $mode input to be named $content.mode, and $encoding to be named $content.encoding to remove ambiguity
• change tundra.file:touch to allow file creation to be optional rather than mandatory
• change tundra.file:touch to support specifying the last modification datetime to use when touching the file
• change tundra.file:write $mode input to be named $file.mode and default to create rather than append, which is now the safest default
• change tundra.gzip:compress to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra.gzip:decompress to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra.html:emit to remove support for table dimension limits
• change tundra.http.response:handle to include HTTP request context (URI, response status code and message, and headers) in exception if applicable
• change tundra.http.response:handle to treat 4xx response codes other than 429 as unrecoverable failures
• change tundra.http:client to default password to an empty string if a user and no password has been specified in the request credentials
• change tundra.list.document:substitute to support replacing an entire value with a non-String value
• change tundra.list.object:difference to return $list.length with the number of items in the returned $list results
• change tundra.list.string:difference to return $list.length with the number of items in the returned $list results
• change tundra.list.string:substitute to support replacing an entire value with a non-String value
• change tundra.mesesage:route to use new tundra.zip:decompress inputs
• change tundra.pipeline:log implementation to Java for 10x performance improvement, and add support for an optional log message to prefix the logged pipeline
• change tundra.pipeline:substitute to support replacing an entire value with a non-String value
• change tundra.schema.content.retrieve:processor schema to optionally support providing the content creation and update datetimes to the content processing service
• change tundra.security.content:digest input and output parameter names for clarity and consistency
• change tundra.service:respond input names to be prefixed with response for clarity
• change tundra.service:respond to make input argument $response.code optional and default to HTTP 200 OK response if not specified
• change tundra.service:respond to rename $encoding input parameter to $content.encoding to remove ambiguity
• change tundra.soap.fault:accept to serialize fault to minified JSON in exception message for brevity
• change tundra.string.base64:decode to use new input parameters when calling tundra.base64:decode
• change tundra.string.base64:enocde to use new input parameters when calling tundra.base64:encode
• change tundra.string:lines to support both CRLF and LF line endings
• change tundra.string:substitute to support replacing an entire value with a non-String value
• change tundra.support.receive:respond to use the thrown exception's innermost cause when determining the HTTP response status code, which affects Tundra/tundra.message:receive and TundraTN/tundra.tn:receive
• change tundra.uri:substitute to also support URI template variable curly bracket delimiters
• change tundra.xml:canonicalize services to use $content.encoding and $content.mode instead of $encoding and $mode to remove ambiguity
• change tundra.xml:emit services to use $content.encoding and $content.mode instead of $encoding and $mode to remove ambiguity
• change tundra.xml:minify services to use $content.encoding and $content.mode instead of $encoding and $mode to remove ambiguity
• change tundra.xml:parse services to use $content.encoding and $content.namespace instead of $encoding and $namespace to remove ambiguity
• change tundra.xpath:exists to use $content.encoding and $content.namespace instead of $encoding and $namespace to remove ambiguity
• change tundra.xpath:get to use $content.encoding and $content.namespace instead of $encoding and $namespace to remove ambiguity
• change tundra.zip:compress to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra.zip:decompress to use $content.mode and $content.encoding inputs rather than $mode and $encoding respectively to remove ambiguity
• change tundra:log implementation to Java for 10x performance improvement
• change Tundra > Service Usage DSP page ./pub/service/usage.dsp to improve performance by moving the sorting of invocations and serializing of pipelines to HTML into the Java service tundra.support.service.usage:get
• change Tundra > Service Usage DSP page ./pub/service/usage.dsp to wrap long values when displaying the input pipeline modal dialog
• change Tundra log purge feature to require an enabled boolean per directory
• change Tundra variable substitution feature to support a new type of substitution for invoking a service dynamically and substituting in the returned output: %invoke:example.employee:get/name?id=1%
• fix tundra.assertion.datetime:unequal to work correctly
• fix tundra.content.retrieve:sftp service comment to be correct and to drop all variables not returned in the output pipeline
• fix tundra.content:parse to default to using the first line as column headings when parsing CSV content
• fix tundra.content:parse validation to throw an unrecoverable exception that halts retries and aborts message processing when applicable
• fix tundra.csv:emit to handle content without a header record correctly
• fix tundra.csv:parse to handle content without a header record correctly
• fix tundra.uri:decode backwards compatibility with deprecated $uri.encoded input parameter

0.0.34 (2019-08-06)

• add tundra.bool:random, tundra.bytes:random, tundra.decimal:random, and tundra.integer:random, for returning cryptographically strong randomly generated numbers
• add tundra.collection.map.*:documentify to convert a Map object to an IData document
• add tundra.configuration:clear and tundra.configuration:refresh for clearing the in-memory cache and refreshing the in-memory cache respectively
• add tundra.document:capitalize and tundra.list.document:capitalize for capitalizing strings in IData documents and IData[] document lists
• add tundra.schedule:runnable to determine if a scheduled task should run or continue to run based on the task's state and the server's task scheduler's state
• add tundra.soap.fault:accept to provide a fault handler that does not thow an exception
• add tundra.string:translate for translating arbitrarily specified string values using a translation table
• add tundra.thread:prioritize for changing the priority of the current thread
• add experimental HTTP logging feature, disabled by default; to enable run tundra.support.http.log:start and then HTTP requests will be logged in ./logs/tundra-http.log, and then to disable run tundra.support.http.log:stop
• change ./pub/service/statistics.dsp to also show the number of successful and failed invocations for each sampled service
• change tundra.assertion.document:equal and tundra.assertion.document:unequal to ignore key position by default
• change tundra.assertion.list.document:equal and tundra.assertion.list.document:unequal to ignore key position by default
• change tundra.configuration:list to return list and document lengths
• change tundra.content:emit to use tundra.xml:encode for escaping values prior to XML serialization, and therefore attributes values will now be escaped in a way that preserves whitespace characters
• change tundra.content:emit use of tundra.xml:encode to support custom XML attribute prefixes
• change tundra.content:retrieve server log statement format to be consistent
• change tundra.csv:emit and tundra.csv:parse to support customizing the escape character, quote character, and quote mode used
• change tundra.csv:emit to support serializing a String[] string list of values in addition to an IData[] document list of records
• change tundra.csv:parse to support the string $null as a way of disabling quoting completely via the quote character input argument
• change tundra.document:amend and dependent services to support merge, create, update, and delete actions on key value pairs
• change tundra.document:condense and tundra.list.document:condense to support condensing keys as well as values
• change tundra.document:emit and tundra.document:parse input argument names to be consistent, and to support XML, JSON, and YAML formats as per the tundra.pipeline:emit and tundra.pipeline:parse services respectively
• change tundra.excel:emit input argument names to be consistent
• change tundra.file:write to support new create mode which will throw an exception if the file already exists
• change tundra.hjson:emit and tundra.hjson:parse input argument names to be consistent
• change tundra.html:emit input argument names to be consistent
• change tundra.http:client to rename $service input to $response.handler to not clash with other more generic uses of the $service variable name
• change tundra.http:client to silently support all pub.client:http inputs
• change tundra.json:emit and tundra.json:parse input argument names to be consistent
• change tundra.json:emit and tundra.json:parse to improve performance by 1000x or more by using the org.glassfish.json implementation classes directly, which avoids the disk access caused by the class loading in the javax.json.spi.JsonProvider.provider() method on every invocation
• change tundra.object:convert to support $mode of base64
• change tundra.pipeline:emit and tundra.pipeline:parse input argument names to be consistent
• change tundra.service:ensure and tundra.service:invoke to support changing the current thread's priority before service invocation
• change tundra.string:capitalize to support capitalizing arbitrarily specified String, String[], and String[][] values
• change tundra.string:concatenate to support sanitizing IData operands
• change tundra.string:condense to support condensing arbitrarily specified String, String[], and String[][] values
• change tundra.string:slice to support slicing arbitrarily specified String, String[], and String[][] values
• change tundra.string:split to support splitting arbitrarily specified strings
• change tundra.user:current to use current invoke state user for implementation
• change tundra.xml:emit and tundra.xml:parse input argument names to be consistent
• change tundra.xml:encode to support custom XML attribute prefixes
• change tundra.yaml:emit and tundra.yaml:parse input argument names to be consistent
• change ./pub/**/*.dsp web page styles to give sortable table headings sortable affordances, such as showing which column has been sorted and in which direction, and changing the mouse cursor to be a pointer when hovering over headings that support sorting via a mouse click, and also update to use a grey colour schema to clash less with the new styling used in Integration Server 10.x
• fix tundra.configuration:get NullPointerException when password cannot be retrieved from password store to be ignored
• fix tundra.duration:multiply to correctly format output using specified pattern
• fix tundra.schedule:create to not modify $schedule input parameter
• fix tundra.string:concatenate to not return anything when given all null input arguments
• rename tundra.soap.fault:handle to tundra.soap.fault:raise

0.0.33 (2018-08-30)

• add tundra.directory:tar for adding files in a given directory to a new tar archive, optionally compressed with gzip
• add tundra.directory:zip for compressing files in a given directory into a zip archive
• add tundra.mime.type:extensions for returning the file extensions associated with a given MIME type
• add Tundra > Service Statistics web dashboard for collecting and viewing service execution duration statistics
• change tundra.content.retrieve:file to purge the archive directory on a different thread to the main thread which is now used exclusively for processing inbound files, and to purge less frequently
• change tundra.directory:ls to return a sorted listing and to also return the length of the listing
• change tundra.document:join to support converting nulls to blanks sanitization
• change tundra.file:write service comment to reflect that this service attempts to create the directories comprising the file path if the path does not already exist before writing the file
• change tundra.http:client to support JSSE and fallback to non-JSSE automatically if TLS 1.1 or higher is not supported by the HTTP server
• change tundra.list.object:join to support converting nulls to blanks sanitization
• change tundra.list.string:join to support converting nulls to blanks sanitization
• change tundra:test to invoke sibling :setup and :teardown services, if they exist alongside the test case service, before and after the test case executes respectively; also now supports executing tests using multiple threads
• fix ./pub/assets/icons to be accessible by Administrators and Developers
• fix tundra.content.retrieve:ftp to not include an extraneous / character between the path and file components of the $content.source argument provided to the content processing service
• fix tundra.content.retrieve:sftp to not include an extraneous / character between the path and file components of the $content.source argument provided to the content processing service
• fix tundra.system:reflect to correctly handle non-String properties
• fix tundra.xml:decode to correctly decode XML character entities
• fix tundra.xml:encode to correctly encode XML character entities; previously this service incorrectly used some HTML character entities which are not supported in XML
• fix the problem reported in Software AG Empower KB article 1780344 "webMethods CloudStreams Server - WmCloudstream package is partially loaded" by removing the dependency on the spring framework to avoid version clashes with other Software AG components that depend on spring, such as webMethods CloudStreams

0.0.32 (2018-04-20)

• add tundra.configuration:peek for inspecting and verifying an on disk package configuration without affecting the cached configuration
• add tundra.datetime:timezone for returning information about the time zone of a given datetime string
• add tundra.document:empty which returns a newly created empty IData document
• add tundra.excel:emit for serializing an IData document to a Microsoft Excel spreadsheet
• add tundra.excel:parse for deserializing a Microsoft Excel spreadsheet to an IData document
• add tundra.list.document:find for filtering an IData[] document list by values associated with a given key containing a given pattern
• add tundra.list.document:match for filtering an IData[] document list by values associated with a given key that match a given pattern
• add tundra.schedule:invoke for manually invoking a scheduled task's service with it's configured input pipeline
• add tundra.string:prefixed to check if a given string starts with a given prefix
• add tundra.string:suffixed to check if a given string ends with a given suffix
• add tundra.string:wrap for breaking a string into lines of a specified length
• change tundra.list.document:group to return list lengths in returned group structure for convenience
• change tundra.list.string:match to improve performance by avoiding reallocation of result arrays
• change tundra.support.http.connect:response to return HTML doctype in response body
• change tundra.xml:validate to throw either a MalformedException if the XML is malformed, or a ValidationException if a schema is specified and the XML is invalid, when $raise? is true
• fix tundra.datetime:add to correctly handle negative non-XML durations
• fix tundra.datetime:parse service comment formatting in Designer
• fix tundra.datetime:subtract to correctly handle negative non-XML durations
• fix tundra.decimal:* services to accept decimals using scientific notation as input, but to always return results normalized as plain decimal strings
• fix tundra.directory:* services to correctly handle File.listFiles() returning null
• fix tundra.list.string:find input name to be $pattern.literal? rather than $literal?
• fix tundra.list.string:match input name to be $pattern.literal? rather than $literal?
• fix tundra.string:squeeze to not return null when input is null or missing
• fix tundra.timezone:* services to correctly handle GMT+hh:mm and GMT-hh:mm time zone IDs
• fix tundra.xml:encode to declare correct input and output variables

0.0.31 (2018-01-09)

• add tundra.document:condense for replacing runs of multiple whitespace characters with a single space
• add tundra.duration:start and tundra.duration:end for using monotonic time to reliably measure elapsed durations
• add tundra.list.document:condense for replacing runs of multiple whitespace characters with a single space
• add tundra.scheduler:pause to pause the scheduled task manager from running any tasks
• add tundra.scheduler:resume to resume the scheduled task manager running tasks
• add tundra.scheduler:status to return the current status of the scheduled task manager
• add tundra.string:build to efficiently build a string using a java.lang.StringBuilder object
• add tundra.string:condense for replacing runs of multiple whitespace characters with a single space
• add tundra.list.integer:add for calculating an addition using each item in a list of integers
• add tundra.list.integer:divide for calculating a division using each item in a list of integers
• add tundra.list.integer:multiply for calculating a multiplication using each item in a list of integers
• add tundra.list.integer:remainder for calculating a division remainder using each item in a list of integers
• add tundra.list.integer:subtract for calculating a subtraction using each item in a list of integers
• add tundra.sap.idoc:identify to assign a DOCNUM to each item in a given IDocList equal to its list index
• add tundra.sap.idoc:length to return the length of a given IDocList
• add tundra.sap.idoc:partition to partition a given IDocList into a new IDocList[] where each item contains at most the given limit
• change tundra.content.deliver:ftp to support relative paths, which can be specified with a path starting with the prefix ./, for example: ftp://example.com/./relative/path/file
• change tundra.content.deliver:sap_idoc to support partitioning an IDoc batch delivery into multiple TIDs using the limit query string value or $limit pipeline argument
• change tundra.content.deliver:sftp to default to treating a URI that does not specify a path as being an absolute path to the root of the file system
• change tundra.content.deliver:sftp to use the given filename with a .tmp suffix when writing the temporary file prior to renaming when the rename option is enabled
• change tundra.content.retrieve:file working and archive file names to use both current datetime (yyyyMMddHHmmssSSS) to provide the file processing datetime, and a newly-generated UUID to provide uniqueness
• change tundra.content.retrieve:ftp to default to treating paths as absolute unless they start with the prefix ./ to indicate that a relative path is being specified
• change tundra.content.retrieve:ftp to rename each file to have a .tmp suffix prior to retrieval
• change tundra.content.retrieve:sftp to default to treating a URI that does not specify a path as being an absolute path to the root of the file system
• change tundra.content.retrieve:sftp to rename each file to have a .tmp suffix prior to retrieval
• change tundra.csv:emit to support CSVs without header rows
• change tundra.csv:parse to support CSVs without header rows
• change tundra.datetime:emit to support java.lang.Number and java.util.Calendar objects, in addition to java.util.Date objects
• change tundra.document:keys to support filtering on associated values, and also return the length of the $keys list as $keys.length
• change tundra.duration:* services to support nanosecond precision
• change tundra.list.string:find to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.list.string:lowercase implementation to Java for better performance
• change tundra.list.string:match to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.list.string:remove to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.list.string:replace to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.list.string:split to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.list.string:squeeze implementation to Java for better performance
• change tundra.list.string:trim implementation to Java for better performance
• change tundra.list.string:uppercase implementation to Java for better performance
• change tundra.mime.type:classify implementation to Java for better performance
• change tundra.string:find to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.string:match to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.string:remove to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.string:replace to support literal patterns and use input arguments consistent with tundra.document regular expression services
• change tundra.string:split to support literal patterns and use input arguments consistent with tundra.document regular expression services
• deprecate tundra.string:squeeze, use tundra.string:condense instead
• fix ./pub/service/usage.dsp rowspan bug for Event Manager threads
• fix tundra.content.deliver:sap_idoc to work around the WmSAP hard limit of 9,999 items in a IDocList by assigning DOCNUM explicitly to be each item's respective list index
• fix tundra.content.deliver:sftp when renaming from a temporary file name and therefore overwriting the destination file, to first delete the destination file before attempting the rename so that if the file already exists it will be overwritten
• fix tundra.string:split input argument to be correctly named $pattern.literal?

0.0.30 (2017-11-03)

• add tundra.integer:format as an alias for tundra.decimal:format
• add tundra.pipeline:squeeze for recursively trimming string values, removing empty string values, arrays, and IData documents, and removing nulls from the pipeline
• add tundra.pipeline:trim for recursively trimming all string values of leading and trailing whitespace in the pipeline
• change tundra.content.retrieve:file to process files in ascending filename order
• change tundra.content:deliver to support sftp delivery on Integration Server versions 9.0 and higher
• change tundra.content:retrieve to support sftp retrieval on Integration Server versions 9.0 and higher
• change tundra.datetime:format to support formatting one or more arbitrarily specified datetime strings in one invocation
• change tundra.decimal:format to support formatting one or more arbitrarily specified datetime strings in one invocation
• change tundra.duration:format to support formatting one or more arbitrarily specified datetime strings in one invocation
• change tundra.html:decode to support decoding one or more arbitrarily specified string values in one invocation
• change tundra.html:encode to support encoding one or more arbitrarily specified string values in one invocation
• change tundra.http.response:accept to drop null response body from $response argument
• change tundra.pipeline:emit to support JSON and YAML content in addition to IData XML content
• change tundra.pipeline:parse to support JSON and YAML content in addition to IData XML content
• change tundra.support.receive:respond to not return exception class in the HTTP response body when handling exceptions
• change tundra.uri:decode to support decoding one or more arbitrarily specified string values in one invocation
• change tundra.uri:emit to support unadorned path strings and files
• change tundra.uri:encode to support encoding one or more arbitrarily specified string values in one invocation
• change tundra.uri:parse to support unadorned path strings and files
• change tundra.xml:decode to support decoding one or more arbitrarily specified string values in one invocation
• change tundra.xml:encode to support encoding one or more arbitrarily specified string values in one invocation
• fix tundra.content.deliver:http to work correctly even if pipeline already contains an IData document named $document
• fix tundra.content.retrieve:ftp to allow arbitrary variables to be passed in the pipeline to the file processing service
• fix tundra.content:retrieve to work correctly for the ftp protocol when the file retrieve fails
• fix tundra.http:client to correctly default $request/method to get when not specified
• fix tundra.support:access to drop unrequired $nodes variable
• fix tundra.uri:emit to work correctly for opaque URIs with a body but no query string

0.0.29 (2017-09-08)

• add tundra.bytes:transcode for converting a byte[] containing text data from one character set to another
• add tundra.document:flip which flips a given IData document so that the keys become the values and the values become the keys
• add tundra.document:legalize for converting to legal Java identifiers either the keys, values, or both keys and values of a given IData document
• add tundra.document:lowercase for converting to lowercase either the keys, values, or both in a given IData document
• add tundra.document:prefix for prepending a given prefix to the keys, or values, or both in a given IData document
• add tundra.document:remove for removing a pattern from either the values, keys, or both keys and values of a given IData document
• add tundra.document:replace for replacing a pattern in either the values, keys, or both keys and values of a given IData document
• add tundra.document:suffix for appending a given suffix to the keys, or values, or both in a given IData document
• add tundra.document:unprefix for removing a given prefix from the keys, or values, or both in a given IData document
• add tundra.document:unsuffix for removing a given suffix from the keys, or values, or both in a given IData document
• add tundra.document:uppercase for converting to uppercase either the keys, values, or both in a given IData document
• add tundra.integer:rebase for converting an integer string from one radix to another
• add tundra.list.document:flip which flips a given IData[] document list so that the keys become the values and the values become the keys
• add tundra.list.document:lowercase for converting to lowercase either the keys, values, or both in a given IData[] document list
• add tundra.list.document:prefix for prepending a given prefix to the keys, or values, or both in a given IData[] document list
• add tundra.list.document:remove for removing a pattern from either the values, keys, or both keys and values of a given IData[] document list
• add tundra.list.document:replace for replacing a pattern in either the values, keys, or both keys and values of a given IData[] document list
• add tundra.list.document:suffix for appending a given suffix to the keys, or values, or both in a given IData[] document list
• add tundra.list.document:trim for trimming leading and trailing whitespace from either the keys, values, or both, in a given IData[] document list
• add tundra.list.document:unprefix for removing a given prefix from the keys, or values, or both in a given IData[] document list
• add tundra.list.document:unsuffix for removing a given suffix from the keys, or values, or both in a given IData[] document list
• add tundra.list.document:uppercase for converting to uppercase either the keys, values, or both in a given IData[] document list
• add tundra.list.exception:raise for throwing a new exception whose message is the concatenation of the messages from the given list of exceptions
• add tundra.list.object:create for creating a new empty array with a given length for the given component class
• add tundra.message:connect to allow a client to test connecting to and authenticating with Integration Server
• add tundra.soap.fault:handle for handling SOAP faults by throwing an exception
• add tundra.stream:transcode for converting a java.io.InputStream containing text data from one character set to another
• add tundra.string:legalize for converting a given string to a legal Java identifier
• add tundra.uuid:generate for generating a new UUID
• change tundra.bytes:normalize to return the $encoding used
• change tundra.collection.list.document:append to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.document:insert to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.document:prepend to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.object:append to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.object:insert to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.object:prepend to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.string:append to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.string:insert to not recurse $items structure and instead only use the top-level objects instead
• change tundra.collection.list.string:prepend to not recurse $items structure and instead only use the top-level objects instead
• change tundra.configuration:get to support resolving variable subsitution %key% substrings against the configuration itself, allowing for configuration values to be built from other configuration values to reduce repetition
• change tundra.content.retrieve:file to attempt to move a file that has failed to process back to its original location and name so that it will be retried next time the service runs, and to return any exceptions encountered while processing in the output pipeline
• change tundra.content.retrieve:file to include $content.length in the input pipeline when calling the file processing service
• change tundra.content.retrieve:file to use a newly generated UUID for the suffix used on files while processing
• change tundra.content:deliver to drop $encoding after call to tundra.xml:emit
• change tundra.content:retrieve to rethrow any exceptions returned by the handler service
• change tundra.content:retrieve to rethrow exceptions encountered during FTP retrievals
• change tundra.content:retrieve to support FTP and FTPS file retrieval
• change tundra.document:get to replace $default.object and $default.string inputs with $value.default
• change tundra.document:get to return $value.list when the value associated with the given $key is a list
• change tundra.document:log to drop unused variables
• change tundra.duration:format to support a list of possible duration patterns used to parse the input duration string, which is useful when the exact pattern is not known
• change tundra.file:process to support specifying whether to rethrow exceptions encountered while processing via $raise? input argument
• change tundra.http.response:accept to drop $encoding after call to tundra.bytes:normalize
• change tundra.http.route implementation to return X-Response-Duration header in HTTP responses
• change tundra.list.object:partition output $results/remainder to be optional
• change tundra.message:receive to extract the response handling code into tundra.support.receive:respond
• change tundra.message:receive to use tundra.service:invoke instead of tundra.service:ensure for a simpler implementation
• change tundra.message:retrieve to support FTP and FTPS file retrieval
• change tundra.object:coalesce to not recurse $operands structure and only work on top-level objects instead
• change tundra.object:coalesce to support arbitrarily specified inputs for coalescing
• change tundra.pipeline:log to drop unused variables
• change tundra.schema.content.retrieve:handler to support handler service optionally returning a list of java.lang.Throwable exception objects thrown while processing the retrieved content
• change tundra.schema.content.retrieve:processor to include new optional input $content.length
• change tundra.service:benchmark to exclude their input arguments from the subsequent invocation pipeline
• change tundra.service:callstack to return $caller.initiator, which is the top-level service in the current call stack
• change tundra.service:defer to exclude their input arguments from the subsequent invocation pipeline
• change tundra.service:ensure to exclude their input arguments from the subsequent invocation pipeline
• change tundra.service:ensure to support catch and finally specific pipelines
• change tundra.service:fork to exclude their input arguments from the subsequent invocation pipeline
• change tundra.service:invoke to exclude their input arguments from the subsequent invocation pipeline
• change tundra.service:invoke to log unthrown exceptions to the error log
• change tundra.stream:normalize to return the $encoding used
• change tundra.string:coalesce to not recurse $operands structure and only work on top-level objects instead
• change tundra.string:coalesce to support arbitrarily specified inputs for coalescing
• change tundra.xml:emit to return the $encoding used
• change tundra:log to drop unused variables
• deprecate tundra.document.key:lowercase, use tundra.document:lowercase instead
• deprecate tundra.document.key:normalize, use tundra.document:legalize instead
• deprecate tundra.document.key:remove, use tundra.document:remove instead
• deprecate tundra.document.key:replace, use tundra.document:replace instead
• deprecate tundra.document.key:trim, use tundra.document:trim instead
• deprecate tundra.document.key:uppercase, use tundra.document:uppercase instead
• deprecate tundra.document.value:lowercase, use tundra.document:lowercase instead
• deprecate tundra.document.value:remove, use tundra.document:remove instead
• deprecate tundra.document.value:replace, use tundra.document:replace instead
• deprecate tundra.document.value:trim, use tundra.document:trim instead
• deprecate tundra.document.value:uppercase, use tundra.document:uppercase instead
• deprecate tundra.id:generate, use tundra.uuid:generate instead
• deprecate tundra.id:normalize, use tundra.string:legalize instead
• deprecate tundra.list.document.key:lowercase, use tundra.list.document:lowercase instead
• deprecate tundra.list.document.key:remove, use tundra.list.document:remove instead
• deprecate tundra.list.document.key:replace, use tundra.list.document:replace instead
• deprecate tundra.list.document.key:trim, use tundra.list.document:trim instead
• deprecate tundra.list.document.key:uppercase, use tundra.list.document:uppercase instead
• deprecate tundra.list.document.value:lowercase, use tundra.list.document:lowercase instead
• deprecate tundra.list.document.value:remove, use tundra.list.document:remove instead
• deprecate tundra.list.document.value:replace, use tundra.list.document:replace instead
• deprecate tundra.list.document.value:trim, use tundra.list.document:trim instead
• deprecate tundra.list.document.value:uppercase, use tundra.list.document:uppercase instead
• fix tundra.content.deliver:http to correctly return the HTTP response body in the $response output parameter
• fix tundra.content.retrieve:ftp to correctly count processed and skipped files
• fix tundra.datetime:add $datetime.pattern to be a picklist
• fix tundra.document:translate to support the service input and output variable names for the input and output IData document being the same
• fix tundra.http.route to not log spurious Access Denied errors
• fix tundra.http:client logged error messages to not include the statement not found in "com.wm.resources.CoreExcpMsgs"
• fix tundra.integer:emit to correctly use the given $radix when emitting the string result
• fix tundra.list.document:group service comment typo
• fix tundra.list.document:partition to not fail output validation when returning an empty remainder list
• fix tundra.list.string:partition to not fail output validation when returning an empty remainder list
• fix tundra.object:instance to not throw ClassNotFoundException when given class name does not exist, and instead return false
• fix tundra.service:ensure to work correctly when no catch and finally pipeline are provided
• fix tundra.string:format service comment formatting to display correctly in Designer
• fix tundra.string:pad service comment formatting
• fix tundra.xml:emit service comment formatting
• fix tundra.yaml:emit missing description in SERVICES.md

0.0.28 (2017-06-05)

• add tundra.cache.memory services which provide in-process on-heap caching of arbitrary key value pairs
• add tundra.collection.map services for document and string maps
• add tundra.decimal:deviate for calculating the standard deviation and related statistics for the given list of decimal numbers
• add tundra.directory:gzip for compressing files in a given directory using the gzip algorithm
• add tundra.document:log for logging a given IData document to the server log as a minified JSON string
• add tundra.file:gzip for compressing a given file with the gzip algorithm
• add tundra.file:zip for compressing a given file with the zip algorithm
• add tundra.list.document:partition for partitioning a given list into a number of lists using the given conditions
• add tundra.list.node:access for granting permissions to a list of nodes
• add tundra.list.object:partition for partitioning a given list into a number of lists using the given conditions
• add tundra.list.string:partition for partitioning a given list into a number of lists using the given conditions
• add tundra.mime.type:classify service which replaces the existing tundra.<type>.mime.type:check services with a single service
• add tundra.support.log:purge feature, for compressing and/or purging log files daily (currently disabled in the package configuration file ./config/package.hjson)
• add tundra.thread:sleep for sleeping the current thread for the given duration
• change tundra.collection.map.*.:get services to return boolean indicating if the given key exists in the given map
• change tundra.directory:purge input argument $duration to be optional
• change tundra.document:trim to include $mode argument which allows for either values, keys, or keys and values to be trimmed
• change tundra.json:emit to support minifying the returned JSON content by removing all extraneous white space
• change tundra.list.document:filter to return the length of the resulting list
• change tundra.list.document:reject to return the length of the resulting list
• change tundra.list.document:take to support taking items from the tail of the given list by specifying a negative count
• change tundra.list.object:filter to return the length of the resulting list
• change tundra.list.object:reject to return the length of the resulting list
• change tundra.list.object:take to support taking items from the tail of the given list by specifying a negative count
• change tundra.list.string:filter to return the length of the resulting list
• change tundra.list.string:reject to return the length of the resulting list
• change tundra.list.string:remove to support literal patterns in addition to regular expression patterns
• change tundra.list.string:take to support taking items from the tail of the given list by specifying a negative count
• change tundra.message:route to support $message being specified as a com.sap.conn.idoc.IDocDocumentList object
• change tundra.message:route to support Content-Encoding header values of base64, gzip, and zip
• change tundra.message:route to support defaulting the routing of an unrecognized message provided as an IData document using pub.publish:publish with the $message.format/name used as the documentTypeName
• change tundra.packages:get to return the union set of all element references for all services in the given package
• change tundra.packages:self to return the union set of all element references for all services in the given package
• change tundra.pipeline:log to log the pipeline as a minified JSON string
• change tundra.service:reflect to return service references and dependents
• change tundra.service:sleep to support multiple duration string formats
• change tundra.string:remove to support literal patterns in addition to regular expression patterns
• change tundra.thread:sleep to support multiple duration string formats
• delete all tundra.<type>.mime.type:check services, now replaced with the tundra.mime.type:classify service
• fix tundra.collection.map.document:get to work correctly
• fix tundra.collection.map.string:get to work correctly
• fix tundra.configuration:all to ignore exceptions thrown by individual configurations when they are attempted to be read from disk and parsed; this ensures that one badly formatted file does not break accessing all other configurations (any exceptions encountered are logged to the server error log)
• fix tundra.configuration:list to ignore exceptions thrown by individual configurations when they are attempted to be read from disk and parsed; this ensures that one badly formatted file does not break accessing all other configurations (any exceptions encountered are logged to the server error log)
• fix tundra.file:copy to not overwite the file when the source and target represent the same canonical path
• fix tundra.html:emit to not throw a java.lang.NullPointerException when serializing an IData[] with null items
• fix tundra.list.object:join $separator argument to be string (it was previously incorrectly constrained to boolean)
• fix tundra.schedule:get to not return null if schedule does not exist; nothing is now returned instead if the schedule does not exist
• fix tundra.yaml:emit to correctly handle top-level objects other than key value maps
• fix tundra.yaml:parse to correctly handle top-level objects other than key value maps
• upgrade commons-csv library from v1.2 to v1.4
• upgrade hjson library from v1.1.4 to v2.1.1
• upgrade snakeyaml library from v1.15 to v1.18

0.0.27 (2017-02-24)

• add tundra.collection.map services for manipulating java.util.Map objects
• add tundra.condition:choose for using the result of evaluating a conditional statement to choose between true and false result arguments
• add tundra.document:trim which trims string values in an IData of leading and trailing whitespace. This service deprecates the existing service tundra.document.value:trim
• add tundra.list.document:merge for merging an IData[] document list into a single IData document
• add tundra.service:fork for invoking a service asynchronously
• change tundra.condition:evaluate to support XPath queries against org.w3c.dom.Node objects that use XML namespaces
• change tundra.content.deliver:http to not parameterise credentials specified in the destination URI, as tundra.http:client now provides this function
• change tundra.datetime:concatenate to support arbitrary datetime patterns and timezones for input and output datetime strings
• change tundra.document:join, tundra.list.object:join, and tundra.list.string:join to support new $sanitization input argument for sanitizing the input document or list being joined by either removing nulls or removing nulls and blanks prior to performing the join
• change tundra.document:merge input argument to be an IData document $operands in which top-level IData values can be provided arbitrarily for merging. The $documents IData[] argument continues to be supported silently for backwards compatibility with existing code
• change tundra.document:split use of tundra.document:merge to use new $operands input argument
• change tundra.document:translate use of tundra.document:merge to use new $operands input argument
• change tundra.http:client to remove default ports (80 for http, 443 for https) from request URI, and to parameterise credentials specified in the authority component of the request URI by removing the credentials from the request URI and instead providing them as input parameters to the call to pub.client:http
• change tundra.http:client use of tundra.document:merge to use new $operands input argument
• change tundra.list.object:join and tundra.list.string:join to allow for a default value to be specified which will be returned when the given list is null, and to treat empty lists the same as null lists
• change tundra.message:route to support specifying publishable document _env values in a message format, which is then used when publishing the document for messages with a route type of publish. Values provided under the key tundra/message/format/route/env in the message format will first have variable substitution performed on them, before being mapped to the _env section of the published document and then published
• change tundra.message:route to support specifying publishable document _properties values in a message format, which is then used when publishing the document to Universal Messaging for messages with a route type of publish. Values provided under the key tundra/message/format/route/properties in the message format will first have variable substitution performed on them, before being mapped to the _properties section of the published document and then published
• change tundra.pipeline:log to prefix logged pipeline with current user in addition to the current call stack
• change tundra.schedule:exists, tundra.schedule:get, and tundra.schedule:list to use Tundra.java project for their implementations
• change tundra.service:benchmark, tundra.service:validate, tundra.service:callstack, tundra.service:reflect, tundra.service:join, tundra.service:ensure to use Tundra.java project for their implementations
• change tundra.service:invoke to use Tundra.java project for its implementation, and deprecate use of this service for asynchronous invocation (tundra.service:fork should now be used instead for this)
• change tundra:log to prefix logged message with current user in addition to the current call stack
• fix tundra.content:emit to drop _properties when serializing a publishable document
• fix tundra.timezone:get to return a GMT offset timezone when given a timezone ID in the form of an offset; previously it would return the first matching timezone with the same offset given, however if the timezone used daylight savings this could result in a different offset returned than the one specified

0.0.26 (2016-10-30)

• add tundra.collection.map services for creating and fetching values from java.util.Map objects
• add tundra.content.deliver:message message protocol handler to route arbitrary content to either the webMethods messaging subsystem via pub.publish:publish, a JMS destination, or a direct service invocation, using tundra.message:route as its implementation
• add tundra.document:uncase for recursively cloning an IData document to a new IData representation with case-insensitive keys
• add tundra.hjson:emit for serializing HJson formatted content
• add tundra.hjson:parse for deserializing HJson formatted content
• add tundra.message.format:clear for removing all message formats from memory
• add tundra.message.trigger:recognize for recognizing the type of publishable document a trigger service was invoked with
• add tundra.service:retryable for making any service automatically retryable by a trigger by automatically converting any thrown exception to be an instance of ISRuntimeException
• add tundra.string:reverse to reverse the order of characters in the given string
• add tundra.ulid:generate for returning newly generated ULID identifers
• add tundra.xml:decode for decoding XML encoded strings: XML entities, such as < and >, are decoded to the appropriate character representation, such as < and >
• add tundra.xml:encode for XML encoding strings: reserved characters in XML, such as < and >, are encoded to the appropriate XML entity, such as < and >, to ensure the XML is rendered correctly by web browsers and other XML rendering software
• add exception logging feature (currently disabled by default; this feature can be enabled in the package configuration file) [experimental feature]
• add support for executing package install/uninstall services automatically (currently disabled by default; this feature can be enabled in the package configuration file) [experimental feature]
• add support for logging received content to ./logs/content.log (currently disabled by default; this feature can be enabled in the package configuration file) [experimental feature]
• change pub/service/usage.dsp to only show pipeline to a depth of 4
• change tundra.configuration:get to always return a configuration IData document even when no configuration files are found for a package
• change tundra.configuration:get to clean up cached configurations for disabled packages
• change tundra.configuration:get to not cache empty configurations in memory
• change tundra.configuration:get to return a mutable deep clone of the cached configuration, to ensure multiple threads do not inadvertently mutate each other's configurations
• change tundra.configuration:get to return the package name the configuration relates to
• change tundra.configuration:get to support Hjson formatted configuration files
• change tundra.configuration:get to support arbitrary XML configuration files
• change tundra.content.deliver:ftp to always logout of server even when an error is encountered
• change tundra.content.deliver:mailto to use new $system/property output from tundra.system:reflect, rather than the deprecated $system/properties
• change tundra.content:retrieve to enable service auditing on error when top-level service
• change tundra.content:retrieve to support customising the input variable name for the retrieved content when passed to the processing service
• change tundra.document:merge to support recursive merging of nested IData documents
• change tundra.document:normalize to support normalizing Trading Networks FixedData objects such as document types correctly
• change tundra.html:emit to support a maximum encoding depth for child IData and IData[] objects
• change tundra.http.response:accept to convert HTTP response header IData keys to be case-insensitive rather than lowercase
• change tundra.http.route:refresh service comment to use file paths relative to the Integration Server instance root
• change tundra.message:receive to accept format input argument to specify the message format to be used to route the message
• change tundra.message:route input arguments to be named consistently with the other tundra.message service arguments
• change tundra.message:route to not delay publishing messages until service success, because this does not work when using a thread pool to route messages
• change tundra.message:route to support additional pipeline variables specified in route record
• change tundra.schedule:* services to support a task name assigned by the task creator, to allow easier identification and hence management of tasks by packages
• change tundra.schedule:create to accept $self as a logical target identifying the local scheduler node, and to always default a null target to $any
• change tundra.schedule:get to return $schedule/target.logical which is set to the value $self when target is equal to the local scheduler node name
• change tundra.schedule:list to return $schedule/target.logical which is set to the value $self when target is equal to the local scheduler node name
• change tundra.system:reflect to silently support $system/properties and $system/directories output variables for backwards-compatibility, and to use a case insensitive IData implementation for environment variables on Microsoft Windows operating systems
• fix tundra.configuration:get to only return configurations for enabled packages
• fix tundra.content.deliver:ftp branch with both switch and evaluate labels true error
• fix tundra.content.deliver:ftp to work correctly when constructing the initial path to cd to prior to the put or append command
• fix tundra.content:deliver service comment to clarify that variables provided in $pipeline take precedence and override their corresponding values in the $destination URI
• fix tundra.directory:ls to throw a meaningful exception when either the directory does not exist, access is denied, or an IO error occurs
• fix tundra.message:receive service comment and README entry typo
• fix tundra.message:route README typo
• fix tundra.node:list to return an empty list when given interface does not exist, and to return the length of the list also
• fix tundra.pipeline:log to correctly prefix the logged pipeline with the full call stack
• fix tundra.schedule:create service comment formatting to display correctly in Designer
• fix tundra.schedule:create service comment to clarify that $schedule/repeat/interval is an XML duration
• fix tundra.schema.message:format service comment typo
• fix tundra.service:ensure to not throw a java.lang.NullPointerException when the current invoke state does not contain any error state
• fix tundra.system:reflect service comment to reflect that the returned $system argument is an immutable IData
• fix tundra.xml:emit to not throw a java.lang.reflect.InvocationTargetException when the Integration Server XSLT implementation does not support javax.xml.transform.TransformerFactory.setFeature method
• fix tundra:log to correctly prefix the logged message with the full call stack
• fix order of Solution menu items to be sorted lexically

0.0.25 (2016-07-05)

• add tundra.collection.list services for working with java.util.List objects
• add tundra.configuration:all service for returning the configurations for all configured packages
• add tundra.configuration:get for reading and returning an IData document representation of package configuration for package-specific and server-specific configuration files
• add tundra.configuration:list services for returning the configurations for all configured packages
• add tundra.document:denormalize for flattening nested child documents and arrays
• add tundra.list.directory:compact to compact a list of directories
• add tundra.list.directory:purge to purge a list of directories
• add tundra.list.directory:remove to delete a list of directories
• add tundra.list.directory:squeeze to squeeze a list of directories
• add tundra.list.document:take for splitting an array into a head with a given count of items and tail with the remaining items
• add tundra.list.file:remove for deleting a list of files
• add tundra.list.object:take for splitting an array into a head with a given count of items and tail with the remaining items
• add tundra.list.string:take for splitting an array into a head with a given count of items and tail with the remaining items
• add tundra.message:receive and supporting services for receiving and recognized arbitrary content and routing to the webMethods messaging layer, JMS destination, or service invocation
• add tundra.message:retrieve for retrieving arbitrary content from a URI and routing to either the webMethods message subsystem, JMS destination, or service invocation
• add tundra.pipeline:denormalize for flattening nested child documents and arrays
• add tundra.scheduler:restart for restarting the Integration Server task scheduler
• add tundra.scheduler:self for returning the node name of the instance of the IS user task scheduler on which the service is run
• add tundra.scheduler:start for starting the Integration Server task scheduler
• add tundra.scheduler:stop for stopping the Integration Server task scheduler
• add tundra.security.acl:create for creating Integration Server ACLs
• add tundra.security.content:digest for calculating a message digest given arbitrary content
• add tundra.xml:parse for parsing XML content into an IData document
• add tundra.xpath:get service for returning the elements associated with an XPath expression
• change tundra.content.deliver:http to support reusing sessions
• change tundra.content:emit to drop _env publishable document section if it exists before serialization
• change tundra.content:retrieve to include retrieve duration in log message
• change tundra.directory:list to support multiple inclusive and exclusive filter patterns
• change tundra.directory:purge to support inclusive and exclusive filename filtering
• change tundra.directory:squeeze to support inclusive and exclusive filename filtering
• change tundra.document:compact to return empty IData document rather than null if all elements get removed
• change tundra.document:get to support resolving XPath expressions against node objects
• change tundra.document:squeeze to return empty IData document rather than null if all elements get removed
• change tundra.document:substitute to support resolving variables against either local variables, global variables, or all variables (global then local)
• change tundra.html:emit implementation to improve performance
• change tundra.http:client to support query string input and additional credential types
• change tundra.http:client to support reusing sessions
• change tundra.json:emit to return pretty printed JSON text and to trim leading and trailing whitespace before returning output
• change tundra.list.document:compact to always return an empty IData[] rather than null if all items were removed
• change tundra.list.document:squeeze implementation to always return an empty IData[] rather than null if all items were removed
• change tundra.list.document:substitute to support resolving variables against either local variables, global variables, or all variables (global then local)
• change tundra.list.string:substitute to support resolving variables against either local variables, global variables, or all variables (global then local)
• change tundra.pipeline:get to support resolving XPath expressions against node objects
• change tundra.pipeline:substitute to support resolving variables against either local variables, global variables, or all variables (global then local)
• change tundra.string:substitute to support resolving variables against either local variables, global variables, or all variables (global then local)
• change tundra.system:reflect to return global variables and use singular names for properties and directories
• change tundra.uri:emit to preserve case in host name just to be safe
• change tundra.uri:emit to treat query string parameter keys as fully-qualified IData keys
• change tundra.uri:parse to preserve case in host name just to be safe
• change tundra.uri:parse to treat query string parameter keys as fully-qualified IData keys
• change tundra.xml:emit to support serializing org.w3c.dom.Node objects, not just org.w3c.dom.Document objects
• change tundra.xpath:exists to addtionally support $content specified as an org.w3c.dom.Node object, or an org.xml.sax.InputSource object
• change tundra:condition:evaluate to support resolving XPath expressions against node objects
• change tundra:test implementation to use new tundra.collection.list services for slight performance improvement
• fix tundra.directory:list to throw a FileNotFoundException rather than a NullPointerException when an IO error occurs while attempting to list the directory
• fix tundra.document:leaves to to work correctly when asked to return leaves that are instances of IData compatible classes by not recursing those classes in this case
• fix tundra.document:stringify NullPointerException by checking if the value to be transformed is null before calling toString()
• fix tundra.list.document:compact to remove nulls from the IData items even when $recurse? is false
• fix tundra.schedule:create to correctly handle datetimes in timezones other than local
• fix tundra.schedule:create to correctly handle when $singleton is true and $schedule/target is $all or all or host on Integration Servers that are not in an Integration Server cluster by only removing existing schedules for the service from the node on which this service is run
• fix tundra.schedule:remove to work correctly when $filter references pipeline variables
• fix tundra.schedule:resume to work correctly when $filter references pipeline variables
• fix tundra.schedule:suspend to work correctly when $filter references pipeline variables
• fix tundra.thread:list output to declare $threads.length argument
• remove all unit tests from package and create new project for these: http://github.com/Permafrost/TundraTest

0.0.24 (2016-01-19)

• add tundra.directory:size for returning the total size in bytes of all files in a given directory
• add tundra.directory:squeeze for reducing the total size in bytes of a given directory by progressively deleting files in order of least recently modified
• add tundra.document:flatten for fetching all values associated with a given list of keys from an IData document and flattening them to a one-dimensional array
• add tundra.packages:self which returns details of the package in which the calling service is stored
• add tundra.pipeline:flatten for collecting all the values associated with a given list of keys and flattening them into a one-dimensional array
• add tundra.string:concatenate for concatenating arbitrarily specified strings, string lists, and string tables optionally separated by a given separator string
• add tundra.string:truncate for truncating all strings in an IData document
• add tundra.uri:substitute for performing variable substitution against the parsed components of a URI string
• change tundra.content:deliver to support $pipeline/$filename for specifying the file name for file deliveries, and if provided this value overrides the value provided in the $destination URI
• change tundra.content:emit to revert the previous change which minifies all serialized XML content, as this legally can change values in the XML, such as collapsing contiguous runs of spaces in an attribute value to a single space, but doing so may cause problems with systems that rely on significant whitespace in values being preserved
• change tundra.content:reject to use double-quotes to delimit service name in exception message/response body
• change tundra.directory:join to support joining multiple children to a parent path
• change tundra.http.response:handle to throw a TransportException that includes the response body content
• change tundra.list.object:insert to ignore null $item arguments: nulls will no longer be inserted into the given list
• change tundra.list.object:prepend to ignore null $item arguments: nulls will no longer be inserted into the given list.
• change tundra.list.object:slice and dependent service's handling of reverse indexing and reverse lengths to be consistent with tundra.string:slice
• fix tundra.content.deliver:http to use $headers/Content-Type as the content type for the HTTP request if provided
• fix tundra.list.object:get and related services to work correctly when using $iteration input argument

0.0.23 (2015-12-02)

• add tundra.directory:compact for recursively deleting empty child directories from a given directory
• add tundra.document:validate for validating an IData document against a document reference or flat file schema
• add tundra.list.datetime:duration for calculating the durations of time for a given list of datetime string pairs
• add tundra.xml:emit for serializing an XML node object to a string, byte array or stream
• change tundra.content.retrieve:file exception message when $source directory does not exist to be more correct by stating that either it does not exist or is unreachable to account for network file shares and problems accessing them across the network
• change tundra.content:deliver HTTP deliveries to include HTTP response headers and status serialized as YAML in $message output argument
• change tundra.content:deliver to support sending emails with no attachments: $content is now allowed be null for mailto: URIs
• change tundra.content:emit to optionally support validating output content
• change tundra.content:parse to optionally support validating input content
• change tundra.content:split to optionally support validating input and output content
• change tundra.content:translate to optionally support validating input and output content
• change tundra.list.content:emit to optionally support validating input and output content
• change tundra.list.content:join to optionally support validating input and output content
• change tundra.list.content:parse to optionally support validating input and output content
• fix tundra.content.retrieve:file skipped files message to be less sure of the reason why
• fix tundra.duration:* services to not throw exception when parsing decimal values
• fix tundra.service:benchmark to correctly measure the performance of services that throw exceptions
• fix tundra.uri:parse to not truncate query string parameter values at the first equals symbol encountered

0.0.22 (2015-11-04)

• add tundra.datetime:today for returning the current date
• add tundra.datetime:tomorrow for returning the current date + 1 day
• add tundra.datetime:yesterday for returning the current date - 1 day
• add tundra.document:nullify for converting string values in an IData document that only contain whitespace characters to null
• add tundra.list.document:blankify for converting all null values to empty strings in an IData[] document list
• add tundra.list.document:normalize for normalizing IData[] document lists
• add tundra.list.document:nullify for converting string values that contain only whitespace characters to null in an IData[] document list
• add tundra.list.object:index for calculating index values for each item in a given list
• add tundra.list.string:blankify for converting null items in a string list to empty strings
• add tundra.list.string:capitalize for capitalizing all strings in a string list
• add tundra.list.string:nullify for converting strings in a string list that only contain whitespace characters to null
• add tundra.list.string:pad for padding each string in a string list
• add tundra.measure:convert for converting values from one unit of measurement to another in the same measurement class
• add tundra.string:format for formatting strings using a java.util.Formatter format string
• add tundra.string:nullify for converting strings that only contain whitespace characters to null
• add tundra.xml:minify for removing comments and extraneous whitespace from XML content
• add experimental support for HTTP compression to Integration Server (not currently enabled by default). This support can be installed by running the following service at startup: tundra.support.http.compression:startup. And it can be uninstalled by running the following service at shutdown: tundra.support.http.compression:shutdown.
• change tundra.content.deliver:http to gzip $content when $headers/Content-Encoding is set to gzip
• change tundra.content:emit to always minify emitted XML content
• change tundra.datetime:* services to support new named pattern datetime.db2 for DB2 timestamp formatted datetimes
• change tundra.document:drop to support dropping a key under all items of an IData[] document list by specifying the key with no array index
• change tundra.document:get to support returning all values associated with a key under an IData[] document list when the key is specified with no array index
• change tundra.document:split to drop the split service input document argument after invocation
• change tundra.document:stringify to treat java.util.Date and java.util.Calendar objects specially: they are now converted to XML datetime strings
• change tundra.document:translate to drop the translation service input document argument after invocation
• change tundra.duration:add to support a variable number of arguments to be added
• change tundra.list.document:group to support multi-level groupings
• change tundra.string:compare to support whitespace insensitive comparisons
• fix tundra.decimal:validate to not throw an exception when $raise? is false
• fix tundra.document:amend to amend a copy of the IData document, rather than the IData document provided itself, so amend conditions are not affected by previous amendments
• fix tundra.document:put to correctly handle null $key argument
• fix tundra.uri:parse to correctly handle opaque URIs containing query strings with values that include URI encoded equal symbols
• upgrade Apache Commons CSV library to v1.2
• upgrade Spring libraries to v4.2.1

0.0.21 (2015-09-10)

• add tundra.document:join for converting an IData document to a human-readable string representation
• change tundra.packages:exists to enable input and output validation
• change tundra.tn.content.retrieve:file to include the archive location of the content being processed in the processing service's pipeline

0.0.20 (2015-08-31)

• add tundra.document.key:remove for removing occurrences of a regular expression pattern from the keys in an IData document
• add tundra.document.value:remove for removing occurrences of a regular expression pattern from the values of an IData document
• add tundra.document:freeze which wraps an IData document in an unmodifiable shell
• add tundra.document:leaves for returning all the leaf values from an IData document
• add tundra.integer:shift for bit shifting integer numbers
• add tundra.list.document.key:remove for removing occurrences of regular expression patterns from the keys of an IData[] document list
• add tundra.list.document.value:remove for removing occurrences of regular expression patterns from the keys of an IData[] document list
• add tundra.list.document:unique for returning only the IData documents that are unique in an IData[] document list
• add tundra.list.string:remove to remove occurrences of a regular expression pattern in a list of strings
• add tundra.packages:exists for querying the existence of an Integration Server package
• add tundra.packages:get for returning information about an Integration Server package
• add tundra.packages:list for returning a list of all Integration Server packages
• add tundra.pipeline:listify for converting a pipeline value into an array
• add tundra.string:remove for removing occurrences of regular expression patterns from a string
• change ./pub/service/usage.dsp web page to include the current datetime
• change tundra.document.key:replace to support $pattern and $replacement being null
• change tundra.document.key:replace to support replacing either the first or all occurrences of the pattern
• change tundra.document.value:replace to support replacing either the first or all occurrences of the pattern
• change tundra.document:copy to support literal keys
• change tundra.document:drop to support literal keys
• change tundra.document:get to support literal keys
• change tundra.document:listify to handle keys that occur multiple times in an IData document by converting all occurrences into a single list
• change tundra.document:put to support literal keys
• change tundra.document:rename to support literal keys
• change tundra.exception:raise to not throw a class cast exception when called with a $exception object that is not an instance of java.lang.Throwable; instead the $exception object will now be ignored if it is not an instance of java.lang.Throwable
• change tundra.list.document.key:replace to support replacing the first or all occurrences of the pattern
• change tundra.list.document.value:replace to support replacing the first or all occurrences of the pattern
• change tundra.list.string:replace to support replacing either the first or all occurrences of the pattern
• change tundra.object:listify to be an alias of tundra.document:listify
• change tundra.pipeline:copy to support literal keys
• change tundra.pipeline:drop to support literal keys
• change tundra.pipeline:get to support literal keys
• change tundra.pipeline:put to support literal keys
• change tundra.pipeline:rename to support literal keys
• change tundra.schedule:create to include a new input parameter $enabled? which determines whether the schedule is created in an active or suspended state
• change tundra.schedule:list to include a new input argument $service which will filter tasks to only those that execute the given service
• change tundra.schedule:remove to include a new input argument $service which will filter tasks to only those that execute the given service
• change tundra.schedule:resume to include a new input argument $service which will filter tasks to only those that execute the given service
• change tundra.schedule:suspend to include a new input argument $service which will filter tasks to only those that execute the given service
• change tundra.string:replace to support replacing either the first or all occurrences of the pattern
• change the following service implementations to use Tundra.java classes and methods:
  • tundra.bool:*
  • tundra.bytes:*
  • tundra.condition:*
  • tundra.csv:*
  • tundra.datetime:*
  • tundra.decimal:*
  • tundra.directory:*
  • tundra.dns:*
  • tundra.document:*
  • tundra.duration:*
  • tundra.exception:*
  • tundra.file:*
  • tundra.gzip:*
  • tundra.html:*
  • tundra.http:*
  • tundra.id:*
  • tundra.integer:*
  • tundra.json:*
  • tundra.list.html:*
  • tundra.mime:*
  • tundra.node:*
  • tundra.session:*
  • tundra.stream:*
  • tundra.string:*
  • tundra.system:*
  • tundra.thread:*
  • tundra.timezone:*
  • tundra.uri:*
  • tundra.user:*
  • tundra.xml:*
  • tundra.xpath:*
  • tundra.yaml:*
  • tundra.zip:*
• fix tundra.document:listify to clone $scope rather than mutating it to prevent inadvertent changes to the IData object mapped to $scope
• fix tundra.list.object:map to return nothing rather than null list when invoked with a null list
• fix tundra.service:benchmark to estimate standard deviation correctly
• upgrade Apache Santuario xmlsec library to v1.5.8

0.0.19 (2015-03-19)

• add tundra.assertion.exception:raised for asserting that an invoked service throws an exception that meets some given criteria
• add tundra.bool:parse for converting a string to a java.lang.Boolean object
• add tundra.content.deliver:ftps service to add support for the FTPS protocol to tundra.content:deliver
• add tundra.html:decode and tundra.html:encode services for HTML-decoding and encoding strings
• add tundra.html:emit for serializing an IData document to an HTML fragment using clean HTML table elements
• add tundra.http.response:status for returning the standard HTTP response message for a given response code
• add tundra.http.route:* services for configuring custom HTTP routes for service invocation
• add tundra.list.document:reject, tundra.list.object:reject, and tundra.list.string:reject for filtering lists to not include items where the given condition evaluates to true
• add tundra.list.html:decode and tundra.list.html:encode services for HTML-decoding and encoding string lists
• add tundra.list.html:emit for converting an IData[] document list to an HTML fragment using a clean HTML table element
• change tundra.bool:emit to allow the $boolean input to be specified as either a string or a java.lang.Boolean object, and to allow $value.true and $value.false inputs to be optional
• change tundra.content:deliver to normalize the input pipeline prior to invoking the tundra.content.deliver delivery protocol implementation service, so that arbitrary input variables can be specified either as complex structures or as flat string variables with 'a/b/c' style keys
• change tundra.datetime:format and related services to support specifying the input time zone to use when parsing a datetime
• change tundra.id:generate to support returning the generated UUID in either the default string representation or in a base64-encoded representation
• change tundra.json:parse to always parse JSON numbers to either java.lang.Long or java.lang.Double objects depending on whether the number is integral or not
• change tundra.list.document:pivot to support pivoting on multiple deeply-nested fully-qualified keys
• change tundra.list.string:normalize implementation from flow to java
• change tundra.schedule:create to include $singleton? input argument, which when true indicates that only one scheduled task should ever exist for this service, and therefore any existing scheduled tasks for this service will be first removed prior to creating a new scheduled task
• change tundra.service:benchmark $message format to use words instead of mathematical symbols for the average, standard deviation, minimum, and maximum value labels
• change tundra.service:benchmark to return the total duration of all executions of the given service
• change tundra:test to simplify implementation by using tundra.string:match and tundra.service:ensure
• upgrade Apache Commons CSV library used for CSV parsing to v1.0.0
• upgrade SnakeYAML library used for YAML parsing to v1.14

0.0.18 (2014-11-17)

• add tundra.list.document:values for returning a list of values associated with a given $key from the respective items in the given IData[] document list
• add tundra.list.string:coalesce for returning the first non-null item in a given string list
• add tundra.string.base64:decode service for convenience when working with base-64 encoded strings in Developer or Designer
• add tundra.string.base64:encode service for convenience when working with base-64 encoded strings in Developer or Designer
• add tundra.string:coalesce for returning the first non-null string argument
• change tundra.content:emit to return $content.type and $encoding, which is useful for knowing the inferred values when not provided as an input arguments
• change tundra.document:get to support optionally specifying a default value to be returned when either the given $key doesn't exist or when it is associated with a null value
• change tundra.duration:* and tundra.list.duration:* services to support specifying different input and output duration patterns
• change tundra.service:ensure to include $exception.service, $exception.package, and the output of WmPublic/pub.flow:getLastError as $exception.info in the $catch service's input pipeline when an exception is thrown by $service
• change default character encoding to be UTF-8 rather than the JVM default charset
• fix tundra.list.document:group to correctly handle grouping key value tuples where the values are null
• fix tundra.xml:validate to not return $errors as null when there are no errors

0.0.17 (2014-10-29)

• add tundra.list.document:group for grouping like items together in an IData[] document list similar to a SQL GROUP BY clause

0.0.16 (2014-10-27)

• add tundra.document:blankify for converting all null values in an IData document to empty strings
• add tundra.list.document:first for returning the first item in a given list
• add tundra.list.document:last for returning the last item from a given list
• add tundra.list.object:first for returning the first item from a list
• add tundra.list.object:last to return the last item from a given list
• add tundra.list.string:first to return the first item in a given list
• add tundra.list.string:last to return the last item from a given list
• add tundra.pipeline:sort for sorting the pipeline's top-level keys in natural ascending order
• add tundra.service:create for creating a new empty flow service in an existing package
• add tundra.service:defer for running services asynchronously on a single dedicated thread; the thread pool is created on first use, instead of at package startup, so that if the service is never used, no thread pool will ever get created
• add tundra.string:blankify which returns an empty string if given a null string, otherwise returns the given string unchanged
• add tundra.string:capitalize for converting the first character in either the first word or all words to upper case
• add tundra.string:characters for returning a list of characters from a given string
• add tundra.string:compare for lexicographically comparing two strings
• add tundra.xml:canonicalize for converting XML content to canonical form
• change tundra.content.deliver:mailto to not compute the from email address if not provided, and instead use the mail.from property returned by tundra.system:reflect
• change tundra.datetime:earlier to support optionally specifying the timezone for the returned datetime
• change tundra.datetime:emit to support optionally specifying the timezone for the returned datetime
• change tundra.datetime:format to support optionally specifying the timezone for the returned datetime
• change tundra.datetime:later to support optionally specifying the timezone for the returned datetime
• change tundra.datetime:now to support optionally specifying the timezone for the returned datetime
• change tundra.decimal:add to include a singleton input decimal to make it easier to use in a transformer with a hard-coded value
• change tundra.decimal:multiply to include a singleton $decimal input variable to make the service easier to use as a transformer
• change tundra.integer:add to include a singleton input integer to make it easier to use in a transformer when adding a hard-coded value
• change tundra.integer:multiply to include a singleton input integer to make it easier to use in a transformer when multiplying by a hard-coded value
• change tundra.json:* services to use latest version of javax.json library, v1.0.4
• change tundra.list.datetime:format to support $patterns.input and $timezone.output arguments
• change tundra.list.document:sort to support sorting integers, decimals, datetimes, and durations, as well as strings; also disallow null $criteria/key input
• change tundra.pipeline:parse to disable output pipeline validation as its not required, and to include URL for the default charset link in the service comment
• change tundra.service:benchmark to return the minimum and maximum durations
• change tundra.service:self to return nothing rather than null, if invoked directly; change implementation to be more efficient
• change tundra.system:reflect to compute a value for the mail.from system property, if it is not already set (value will be computed as "Integration-Server@" + fully-qualified domain name of the server)
• change tundra.timezone:get to support a time zone ID of $default which represents the default or local host time zone
• change default buffer size used by the tundra.bytes:* and tundra.stream:* services from 4096 to 8192 bytes
• change unit tests from to use the tundra.support.test folder instead of the tundra.test folder, as they are not part of the public API of the package
• fix tundra.document:stringify to not return a null value when no input $document is specified
• fix tundra.string:normalize to not return anything when $object is null

0.0.15 (2014-08-27)

• add tundra.csv.mime.type:check which returns true if the given MIME media type is a recognised CSV type
• add tundra.csv:emit for serializing CSV content
• add tundra.csv:parse for deserializing CSV content
• add tundra.datetime:earlier for returning current datetime minus a given duration
• add tundra.datetime:later for returning the current datetime plus a given duration
• add tundra.datetime:maximum for returning the largest datetime in a given list
• add tundra.datetime:minimum for returning the smallest datetime in a given list
• add tundra.decimal:emit for converting Java decimal objects to strings
• add tundra.decimal:parse for converting strings to Java decimal objects
• add tundra.directory:purge for (optionally recursively) deleting files older than a given age from a given directory
• add tundra.directory:reflect for returning meta-data about a given directory
• add tundra.document.key:normalize for converting all keys in an IData document to legal Java identifiers by replacing illegal characters with underscores
• add tundra.document:pivot for iterating over an IData document whose structure is not known
• add tundra.document:stringify for converting all values in a given IData document to strings
• add tundra.file:reflect for returning meta-data about a given file
• add tundra.http.response:accept which is a response handler that accepts all response codes
• add tundra.id:normalize for converting a string to a legal Java identifier by replacing illegal characters with underscores
• add tundra.integer:emit for converting Java integer objects to strings
• add tundra.integer:parse for converting strings to Java integer objects
• add tundra.json.content.type:check which returns true if the given MIME media type is a recognised JSON media type
• add tundra.json:emit for serializing JSON content
• add tundra.json:parse for deserializing JSON content
• add tundra.list.document:keys which returns a union set of all top-level keys for each item in the given IData[] document list
• add tundra.list.document:pivot for converting an IData[] document list to an IData document
• add tundra.list.string:find for checking if the items in the given list include the given regular expression or literal string pattern
• add tundra.list.string:quote for constructing a regular expression that matches any of the literal strings in the given list
• add tundra.service:reflect for returning meta-data about a given service
• add tundra.string:capture for matching a regular expression pattern against a string and returning the associated capturing groups
• add tundra.string:find for partial regular expression and literal pattern matching
• add tundra.string:pad for left and right padding of strings
• add tundra.string:slice for forward and reverse indexed substrings
• add tundra.test unit test services, deprecating separate TundraTest package, as version management across Tundra and TundraTest packages proved too difficult
• add tundra.thread:current which returns the currently executing thread
• add tundra.thread:list which returns all threads currently scheduled on the JVM
• add tundra.uri:normalize for normalizing a URI string
• add tundra.xml.content.type:check which returns true if the given MIME media type is a recognised XML media type
• add tundra.yaml.mime.type:check which returns true if the given MIME media type is a recognised YAML type
• add tundra.yaml:emit for serializing YAML content
• add tundra.yaml:parse for deserializing YAML content
• change tundra.content:deliver to not attempt delivery when no destination is specified
• change tundra.content:deliver to sanitize destination URI scheme when looking for the delivery implementation service by replacing all non-alphanumeric characters with underscores
• change tundra.content:deliver to support delivering IData documents by first serializing the document prior to delivery
• change tundra.content:deliver to support ftp delivery scheme
• change tundra.content:deliver to support jms delivery scheme
• change tundra.content:deliver to support sap+idoc delivery scheme
• change tundra.content:deliver to use charset parameter from $content.type if $encoding is not specified
• change tundra.content:discard to be stateless
• change tundra.content:emit to support serializing CSV, JSON, and YAML content
• change tundra.content:emit, tundra.content:parse, and related services to support XML namespaces
• change tundra.content:parse to support deserializing CSV, JSON, and YAML content
• change tundra.content:reject to be stateless
• change tundra.content:retrieve to append a timestamp to archived file names
• change tundra.content:retrieve to optionally clean up archived files older than a given age
• change tundra.content:retrieve to use a working directory when processing files
• change tundra.datetime:format to accept an array of datetime patterns when the exact pattern is not known
• change tundra.datetime:parse to accept an array of datetime patterns when the exact pattern is not known
• change tundra.decimal:* services to support optional inputs
• change tundra.document:duplicate to return nothing if no input $document is given, rather than null
• change tundra.document:keys to retain key order
• change tundra.document:values to return an array of the nearest ancestor class in common to all values, rather than an Object[]
• change tundra.duration:* services to return nothing if no input was specified, rather than returning null
• change tundra.http.response:handle to use tundra.http.response:accept in its implementation
• change tundra.http:client $request/timeout input to be constrained to an XML duration formatted string [backwards incompatible]
• change tundra.integer:* services to support optional inputs
• change tundra.list.document:resize to support specifying a negative length which results in the given list being truncated by that many items
• change tundra.list.document:sort to support sorting by multiple keys in ascending or descending order
• change tundra.list.object:resize to support specifying a negative length which results in the given list being truncated by that many items
• change tundra.list.service:chain to return total invocation duration
• change tundra.list.string:match to return boolean flags indicating if all, any, or none of the items in the given list matched the given pattern
• change tundra.list.string:match to support literal patterns as well as regular expression patterns
• change tundra.list.string:resize to support specifying a negative length which results in the given list being truncated by that many items
• change tundra.object:reflect to return identity hash code of the object, which can be used to compare object identities
• change tundra.service:benchmark to first validate the service exists before attempting to benchmark it
• change tundra.service:invoke to return invocation duration
• change tundra.string:match to support both regular expression and literal pattern matches
• change tundra.string:quote input argument to be optional
• change tundra.timezone:get services to handle custom datetime patterns
• change tundra.timezone:get to return UTC zone when given timezone ID "Z"
• change tundra.timezone:get to support retrieving a timezone using either a raw millisecond offset value, an XML duration offset value, or an offset specified in the form (+|-)HH:mm
• change tundra.timezone:list services to handle custom datetime patterns
• change tundra.timezone:self services to handle custom datetime patterns
• change tundra.uri:emit to not include user, password, or fragment in resulting URI string if specified as emtpy strings
• fix tundra.content:deliver to correctly support additional HTTP settings in $pipeline
• fix tundra.datetime:add to work correctly with very large durations (larger than a signed 32-bit integer)
• fix tundra.datetime:subtract to work correctly with very large durations (larger than a signed 32-bit integer)
• fix tundra.decimal:average to work correctly when given an empty list
• fix tundra.decimal:maximum to work correctly when given an empty list
• fix tundra.decimal:minimum to work correctly when given an empty list
• fix tundra.gzip:compress to work correctly by first closing the byte array output stream before accessing the underlying byte array
• fix tundra.http:client to correctly support HTTP PUT method
• fix tundra.list.document:sort to copy the input IData[] object before sorting, rather than mutating the original IData[] object
• fix tundra.service:benchmark NaN NumberFormatException thrown when all sample durations were 0 by returning a standard deviation of 0, rather than attempting a division by zero calculation
• fix tundra.timezone:get to return nothing when given timezone ID is not found, rather than returning UTC
• fix tundra.uri:parse to correctly parse opaque URIs that include variable substitution in the query string

0.0.14 (2014-05-28)

• add tundra.content:deliver protocol handlers to the Tundra public API in the folder tundra.content.deliver
• add tundra.list.document:filter to filter a list with a given conditional statement
• add tundra.list.object:filter to filter a list with a given conditional statement
• add tundra.list.string:filter to filter a list with a given conditional statement
• add tundra.schedule:* services for creating, manipulating, and deleting scheduled tasks
• add tundra.service:benchmark for measuring the performance of a given service
• add tundra.string:quote to escape a string value so that it can be used in a regular expression pattern to match the literal value
• change tundra.condition:evaluate to evaluate null and empty conditions to true
• change tundra.condition:evaluate to use the exact same implementation of conditional expressions as used by flow branch steps
• change tundra.datetime:* services to emit XML dates and times without the time zone offset suffix
• change tundra.document:compact to support recursively compacting embedded com.wm.data.IData, com.wm.util.Table, and com.wm.data.IData[] objects
• change tundra.document:get to support retrieving values by indexed key from com.wm.util.Table objects
• change tundra.document:keys to support filtering the returned list of keys with a given regular expression pattern
• change tundra.document:squeeze to remove empty com.wm.data.IData and java.lang.Object[] objects
• change tundra.list.document:compact to support recursively compacting embedded com.wm.data.IData, com.wm.util.Table, and com.wm.data.IData[] objects
• change tundra.list.document:squeeze to remove empty com.wm.data.IData and java.lang.Object[] objects
• change tundra.object:reflect to be able to reflect on a pipeline variable identified by key, which is useful when reflecting on objects that may be lists/array (because mapping them to $object will implicitly dereference the array's zeroth element)
• change tundra.pipeline:get to support retrieving values by indexed key com.wm.util.Table objects
• fix tundra.uri:parse to correctly handle paths that include variable substitution strings

0.0.13 (2014-02-13)

• change tundra.content:discard implementation to use tundra.service:respond
• change tundra.content:reject implementation to use tundra.service:respond

0.0.12 (2014-01-31)

• add tundra.gzip:* services for compressing and decompressing data using the gzip format
• add tundra.service:initiator to check if the calling service is the initiating top-level service for this thread
• add tundra.service:respond for setting a transport-specific response to be returned to the calling process
• add tundra.timezone:* services for inspecting time zone offsets
• add tundra.zip:* services for compressing and decompressing data in the zip format
• change tundra.content:retrieve to allow the archive directory for file protocol to be configurable via a query string parameter
• change tundra.duration:format to correctly parse fractional millisecond values
• change tundra.mime.type:parse to fix typo in service comment

0.0.11 (2013-12-13)

• add tundra.file:copy for copying content from a source file to a target file
• change tundra.content:deliver to fix HTTP and HTTPS delivery compatibility with some web servers by not specifying the user/password in the destination URI when calling pub.client:http; instead the user and password is provided in the auth argument
• change tundra.list.document:join to simplify the default settings to use a single map step
• change tundra.list.document:get to add $iteration input argument, to make the service easier to use from within a loop
• change tundra.list.object:get to add $iteration input argument, to make the service easier to use from within a loop
• change tundra.list.string:get to add $iteration input argument, to make the service easier to use from within a loop
• change tundra.stream:copy to use default buffer size consistently

0.0.10 (2013-11-25)

• add tundra.list.*:grow, :resize and :shrink services for resizing lists
• add tundra.list.object:difference and :intersection, and tundra.list.string:difference and :intersection to return the set difference and intersection of two lists respectively
• add tundra.schema.exception:handler specification describing the service signature required of a tundra.service:ensure $catch service
• change tundra.assertion.list.object:unequal to fix typo in service comment
• change tundra.condition:evaluate to include the ANTLR4 grammar used to generate the parser in the service comments
• change tundra.content:amend and tundra.document:amend to add an optional conditional statement to each amendment to control whether the amendment gets applied
• change tundra.decimal:add to allow nulls in input list, defaulting them to zero
• change tundra.integer:average, :maximum, and :minimum to disallow nulls in input list

0.0.9 (2013-11-05)

• fix tundra.http:client variable substitution bug in the resulting exception message when an unexpected HTTP response code is encountered

0.0.8 (2013-10-11)

• change tundra.content:discard to support being wrapped by another service, and being executed using the /soap/rpc handlers
• change tundra.content:reject to support being wrapped by another service, and being executed using the /soap/rpc handlers

0.0.7 (2013-10-01)

• add tundra.assertion.datetime:equal and :unequal services
• add tundra.bool:* services for manipulating boolean values
• add tundra.content:discard and :reject gateway services for successfully receiving and throwing away and rejecting content respectively
• add tundra.datetime:emit and :parse services for converting between datetime strings and java.util.Date objects
• add tundra.decimal:average, :maximum, and :minimum for calculating the average, maximum, and minimum from a list of decimals
• add tundra.directory:join to construct a single file path from a parent directory and child path or file
• add tundra.directory:ls as an alternative to tundra.directory:list for when performance is a priority over ease of use
• add tundra.document:clear to remove all keys except those specified from the given IData document
• add tundra.document:emit and :parse for encoding/decoding IData objects as XML
• add tundra.document:sort to sort an IData by its keys in natural ascending order
• add tundra.document:squeeze to trim all leading and trailing whitespace, then convert empty strings to null, then compact the IData document by removing all null values
• add tundra.file:match to check if a given file matches a regular expression or wildcard pattern
• add tundra.integer:average, :maximum, and :minimum for calculating the average, maximum, and minimum from a list of integers
• add tundra.list.document:clear to remove all keys except for those specified from each IData document in the given list
• add tundra.pipeline:clear to remove all keys except those specified from the pipeline
• add tundra.pipeline:emit and :parse for encoding/decoding the pipeline as XML
• change tundra.content:deliver to return the delivery response body when applicable
• change tundra.content:emit to HTML-encode values by default when emitting XML
• change tundra.datetime:add, :compare, :duration, and :subtract to support datetimes other than ISO8601
• change tundra.datetime:format so that when the input is null or absent, the output is now guaranteed to be absent rather than null, so that when mapping to an IData structure null elements are not created inadvertently
• change tundra.datetime:format to support JDBC named patterns, and use java.sql.Timestamp to parse and emit JDBC datetime strings
• change tundra.directory:list to improve performance, and fix a NullPointerException when the directory doesn't exist
• change tundra.document:rename to fix bug when old and new keys are the same
• change tundra.document:split to support the input and output arguments being the same for the translation service
• change tundra.document:translate to fix a variable substitution bug in an error message
• change tundra.document:translate to support the input and output arguments being the same for the translation service
• change tundra.http:client to have unambiguous and distinct input ($request) and output ($response) arguments [backwards incompatible]
• change tundra.integer:decrement input argument to be optional so that you don't need to initialize a counter variable to zero prior to calling this service in a loop
• change tundra.integer:increment input argument to be optional so that you don't need to initialize a counter variable to zero prior to calling this service in a loop
• change tundra.uri:parse so that malformed URIs now return more meaningful error messages
• delete all tundra.support.test:* services from Tundra, and moved them to new TundraTest package to keep Tundra as lightweight as possible, and to avoid inadvertently running the test services in production

0.0.6 (2013-07-20)

• add tundra.content:retrieve to fetch content from a source URI and process it with the given content processing service
• add tundra.document:listify to normalize an IData or IData[] item in the pipeline to an IData[]
• add tundra.list.string:match to partition a list of strings into separate lists of matched and unmatched strings by matching each item against the given regular expression
• add tundra.mime.type:* services for parsing and emitting mime types
• add tundra.node:access to set ACL permissions on namespace nodes
• add tundra.object:listify to normalize an Object or Object[] item in the pipeline to an Object[]
• add tundra.string:listify to normalize a String or String[] item in the pipeline to a String[]
• add tundra.system:reflect to return system settings (Java properties, environment settings) and state (Java heap memory usage)
• add tundra.user:current to return the currently logged on user invoking the service
• change tundra.content:deliver to correctly allow any additional variables to be passed in $pipeline
• change tundra.content:deliver to correctly extract the authority component of HTTP URIs
• change tundra.content:deliver to support mailto: URI delivery
• change tundra.directory:list to support wildcard/glob pattern matching, as well as regular expression matching
• change tundra.list.*:drop services to return $list to improve their usefulness
• change tundra.object:instance now always returns $instance?
• change tundra.support:startup to set appropriate ACL permissions on tundra nodes
• change tundra.xml:validate to return $valid? as false when $content is null

0.0.5 (2013-07-03)

• add tundra.xpath:exists to test whether an XPath query exists in an XML document
• add tundra.list.service:chain for calling a list of services sequentially with a shared pipeline
• add tundra.list.service:ensure to provide a try/catch/finally pattern while invoking a list of services sequentially with a shared pipeline
• add tundra.document:first and tundra.document:last to return the first and last {key, value} pair in an IData document
• add tundra.pipeline:first and tundra.pipeline:last to return the first and last {key, value} pair in the pipeline
• add tundra.integer:* services for working with arbitrary precision integers (uses java.math.BigInteger for its implementation)
• add tundra.decimal:* services for working with arbitrary precision decimals (uses java.math.BigDecimal for its implementation)
• add tundra.condition:evaluate for testing conditional predicates against the pipeline or optional scope IData document
• add tundra.service:validate to test whether a named service exists
• add tundra.content:amend and tundra.document:amend for making precise inline edits in XML or flat file content, or IData documents
• change tundra.content:split to support splitting content into disparate schemas

0.0.4 (2013-03-30)

• add tundra.string:squeeze for replacing runs of whitespace characters with a single space character
• add tundra.document:duplicate for making (optionally recursive) copies of IData documents
• add support for fully qualified IData key names a/b/c[0] to relevant tundra.document and tundra.pipeline services
• add tundra.uri:* services for parsing and emitting URI strings
• add tundra.content:deliver and tundra.document:deliver which provides an extensible framework for delivering arbitrary content to a URI destination
• add tundra.dns:resolve to lookup DNS names and IP addresses
• add tundra.dns:localhost to return the name and IP address of the localhost Integration Server is running on
• add tundra.base64:* services for encoding and decoding base64 strings
• add tundra.service:self to return the name of the currently running service
• change tundra.directory and tundra.file services to handle file:// URIs
• rename tundra.file:open to tundra.file:process [backwards incompatible]

0.0.3 (2012-07-28)

• add tundra.content:* services for parsing, emitting, and translating or mapping structured textual data (XML, flat files)
• add lots of useful tundra.document:* services for manipulating IData documents
• add tundra.node:* services for querying Integration Server namespace nodes
• add tundra.pipeline:drop, tundra.pipeline:get, and tundra.pipeline:put for dropping, retrieving, and adding {key, value} pairs from/to the pipeline
• add tundra.string:lines to return all the lines in a string
• add tundra.string:replace to replace all occurrences of a regular expression pattern in a string
• add tundra.string:split to split a string around matches of a regular expression pattern

0.0.2 (2012-07-18)

• add tundra.directory:* for working with directories
• add tundra.file:* for working with files
• add tundra.list.datetime:format for formatting a list of datetime strings
• add tundra.list.duration:format for formatting a list of time duration strings
• add tundra.list.service:invoke for invoking a list of services either synchronously (with an optional level of concurrency) or asynchronously
• add tundra.list.service:join for waiting on a list of asynchronously invoked services to finish
• add tundra.object:stringify for converting any object to a string
• add tundra.pipeline:capture for grabbing a snapshot of the current pipeline as an IData document
• add tundra.pipeline:log for logging the current contents of the pipeline to the server debug log
• add tundra.service:ensure to provide a try/catch/finally pattern for service invocations
• add tundra.service:join for waiting on an asynchronously invoked service to finish
• add tundra.service:sleep for putting the currently executing thread to sleep
• add tundra.string:match for matching strings against regular expression patterns
• add tundra.string:trim for removing leading and trailing whitespace from strings

0.0.1 (2012-06-23)

• add tundra.assertion:* to provide a lightweight unit testing framework
• add tundra.bytes:* for working with byte[] objects
• add tundra.datetime:* for parsing, emitting, and manipulating datetime strings
• add tundra.document:* for manipulating com.wm.data.IData objects (also known as records or documents in Integration Server)
• add tundra.duration:* for parsing, emitting, and manipulating time durations
• add tundra.exception:* for throwing exceptions
• add tundra.id:* for generating unique identifier strings
• add tundra.list:* to provide a comprehensive set of services for enumerating and manipulating lists (com.wm.data.IData[], java.lang.Object[], java.lang.String[])
• add tundra.object:* for reflecting on objects
• add tundra.service:* for invoking services dynamically
• add tundra.stream:* for working with java.io.InputStream and java.io.OutputStream objects
• add tundra.string:* for manipulating strings