Metacat

changed log labels per trac 843

fixed bug 818

updated the metacatpopulator to use the d1 eml parser to create system metadata in a smarter fashion.

better log message

more debugging for objectInfo problem

checking for a null objectformat

checking for a null objectformat

checking for a null objectformat

backtracking

add some error output

add some error output

added a check for bad data into listObjects

commented out a println

updated DB scripts for D1 schemas

changed improper use of getFS() to a /. getFS() should never be used with a URL since it will return \ on windows which is not a valid separator in a URL.

moved the dataone types to the metacat.properties file

updated for new dataone jars

changes to work with new dataone jars

include MetacatHandlerPlugin interface so that other projects [semtools] can extend the actions handled by Metcat without editing the codebase directly. The SemtoolsPlugin is the prototype for this and can be registered using the Metacat admin interface. Note that the plugin implementation class and supporting jars must be included in the webapp in order for the request to be handled correctly (the semtools build is injecting these as part of it's build process).

added explicit 200 codes to ResourceHandler

added better logging and http code setting

fixed bug where the query cache was not getting reset by a REST insert/query combo.

[merged from 1.9.3 branch] security fix: never unregister the "public" user session

re-added some code inadvertently removed

merged with version from 1.9.3 branch to fix bug with character encodings

added a todo

new todos from code review with roger

added TODOs and cleaned up the code a bit

do not read from the inputstream before passing it off to the SAXParser

fixed mime multipart problems so that roger can try to use the d1client

fixed error where insert/update errors were not getting caught as exceptions

fixed bug with logger where localid didn't get appended when inserting a data file

reformatting logs for robert

added another logging statement

added replicate log statements with the guid and localId

fixed update bug I introduced yesterday while fixing replication

hopefully fixed bug with systemmetadata replication

debug statements in dbsaxhandler

still working on getting guid problems fixed with replication

fixed identifier manager problem with getting guid for systememtadaa doc

fixing problems with replication and systemmetadata

set the content type to text/xml for the rest interface

fixed yet another null sessionData bug

Fixed spelling error.

Set content type on listObjects to be XML so that it is recognized by clients.
Reformatted method to use original file conventions.

Reformatted code to fix indenting.

fixed more access control problems. also fixed extraneous white space that was getting put into the listobject output stream

fixed problems with access control on sysmeta

fix for set access bug

fixes for the access control setting for sysmeta

added functionality to set access permissions to system metadata the same as the document that it describes

fixed major bug in replication where the document info was being truncated due to a poorly implemented sax parser

hopefully fixed get bug where session was null

fixing another null auth token bug

changing d1 log format

added code to make sure the localId and guid get added to get and create d1 logs

removing println

fixed more NPE problems with sessionData. hopefully got them all this time

fix for null session bug

hudson tests are now running. needed date conversions. still more work to do with systemmetadata date format.

updated for date format that supports time zones correctly.

changed date comparison to compare to gmt only

fixed NPE with sessionData

added some new error handling to work around this issue https://trac.dataone.org/ticket/706

fixed bug in access control where the sessionid was not correctly passed along so changes of privileges were not being made

fixed bug where null session data was not treated properly.

Correct log warning message - not "Account" but "Accession" number

fixing date formatting and timezone issues

fixed bug with log processing

fixed some bugs in time handling for listObjects and getLogRecords

bug fix to handle nulls

fixed bug in getLogRecords where the wrong id was getting set as the identifier

added new exception handling to serialize number format exceptions for rest calls

fixed listObject bugs, added params to the listObjects rest interface, consolidated the date format passed in params

fixed query for listObjects so that paging works as advertised.

formatting

fixed error with listObjects and get.

got the crud test to pass on knb-mn, now need to figure out why the rest service still isn't working

still looking for listObjects problem

more debug for another instance

adding some debug info

fixed bug with upper case comparison

changed getSessionData so that it will respond property to public tokens

MetacatPopulator is now working. It allows you to do a query from one metacat instance, then insert any returned docs into another metacat instance

cleanup of 'fixed' ResourceHandler

tracked down streaming bug. now need to figure out a better way to fix it.

added getValue() to the guid so that it prints the guid and not an object id

added a DataOneLogger for event notifications on the CN. The logger is called DataOneLogger and can be managed in the log4j.properties file

finished getLogRecords. Fixed some date bugs in metacat.

adding getlogrecords to the rest interface.

changed getLogRecords to use new classes for log events

fixed bug in metacat that allowed attribute data to be appended to element data in returnfields

fixed bugs associated with crud access control changes.

implementing getLogRecords

added EventLog statements to all public methods in CrudService in preparation for addin the getLogRecords() method

added better debugging error messages for CN dev errors

added generateMissingSystemMetadata to the ResourceHandler so it is now available via a REST call.

made the test functional so that it checks that missing system metadata will be created.

mostly done with generating systemmetadata from legacy knb docs. need to write a better test and enable this through the ResourceHandler then it will be done.

working on getting systemmetadata creation working for legacy knb objects