Page tree
Skip to end of metadata
Go to start of metadata
The following fixes are available to be set.

MIME_CONVERT

This no longer seems to be in use.

REPLACE_ITEM_NULL

This extension allows developers to use Document.replaceItemValue(String, null) to remove the Item

CREATE_DB

Extends the use of DbDirectory.createDatabase(String) to return the existing Database object for the specified path or create a new database at the specified path and return the Database object relating to that.

REMOVE_ITEM

Normal behaviour is to remove the first Item with the specified name. This switch removes all Items with the specified name.

APPEND_ITEM_VALUE

Normal use of Document.appendItemValue() appends a new item of the same name with the new value.
This switch changes the behaviour to work in a more intuitive manner, appending the value to the existing Item.

VIEW_UPDATE_OFF

Ensures any method accessing view entries calls View.setAutoUpdate(false) before doing anything.

FORCE_JAVA_DATES

See Tommy Valand''s blog post. This method ensures Java dates are always returned when interacting with Document Items or ViewEntry column values.

MIME_BLOCK_ITEM_INTERFACE

Just writes a warning to the console if Items are accessed while MIME Items are still open. This fix has been downgraded from blocking to just warning against access

DOC_UNID_NULLS

Normal behaviour of Database.getDocumentByUNID() is to throw an error if the UNID cannot be found. With this fix, that method will now just return a null object and suppress the error.

VIEWENTRY_RETURN_CONSTANT_VALUES

Ensures ViewColumn.getColumnValuesIndex() and ViewColumn.getColumnValuesIndex() include view columns whose value is a constant value rather than a field value or result of an @formula.

ODA_NAMES

Uses org.openntf.domino.impl.NameODA class whenever methods would normally use lotus.domino.Name. This class can be stored in Java obejcts and uses String manipulation to convert between formats.

PEDANTIC_GC_TRACKING

Instructs the automatic garbage collector to maintain an internal tracker for C++ backend-object IDs and only recycle when no further references remain (causes a performance hit).
NOTE: This fix is not auto-enabled with "khan" mode in XPages

FORCE_HEX_LOWER_CASE

Hex values such as replicaid, unid and noteid are forced to lower case before being returned. By default string comparisons are case-sensitive, so comparing may be problematic. This will workaround that problem.

BLOCK_NTH_DOCUMENT

This is designed to prevent using getNthDocument() which has performance impacts. However, XPages repeat controls need to use this when passing a DocumentCollection as the source. As a result, this fix is not auto-enabled with "khan" mode in XPages.

  • No labels