Difference between revisions of "Indexes"

From LongJump Support Wiki
imported>Aeric
imported>Aeric
Line 59: Line 59:
</syntaxhighlight>
</syntaxhighlight>


===Caution===
===Effects of Operations===


;Creating a new index on existing data:
;Creating a new Unique Key index on existing data:
:*If you create a new unique index on existing data and the data has no duplicate entries for that combination of fields, the index is created successfully.
:*If the data has no duplicate entries for that combination of fields, the index is created successfully.
:*If the data has duplicate entries for that combination of fields, then index creation fails.
:*If the data has duplicate entries for that combination of fields, index creation fails.
:*The check for duplicate entries ''includes'' entries in the [[Recycle Bin]], since they could be restored at a later date.
:*The check for duplicate entries ''includes'' entries in the [[Recycle Bin]], since they could be restored at a later date.


;Adding new record to the object:
;Adding a new record to an object with a Unique Key index:
:*If there is no existing record with same values for fields selected in the unique index, the record is added successfully.
:*If no existing record has the same values for the fields that make up the unique index, the record is added successfully.
:*If there is an existing record with same values for fields selected in the unique index, an error occurs.
:*If there is an existing record with the same values for those fields, an error occurs.


;Updating a record:
;Updating a record to change Unique Key values:
:*If there exists another record with values to be updated for fields selected in the unique index of the current record, then it will give message “Operation results in unique index violation ".
:*If no other record exists with those values, the update succeeds.
:*If there exists no record with the updated fields for the unique index fields, then the record is updated successfully.
:*If another record exists with those values, the update fails.


;Deleting a record:When a record is deleted, corresponding index entries are deleted, as well.
;Deleting a record:When a record is deleted, corresponding index entries are deleted, as well.
;Restoring a record:If there exists another record with the same values for index fields (that is not part of recycle bin) then it will give a message "Operation results in unique index violation". Otherwise, the operation will be successful.
;Adding Index to an Object:The addition of unique index to an object will not check for the uniqueness of existing records.


==Working with Indexes==
==Working with Indexes==

Revision as of 22:01, 12 July 2011

Designer > Data > Indexes

Indexes are used to access database information efficiently. Once an index is supplied, the database can go directly to the record and retrieve it, without having to search for it.

About Indexes

Lock-tiny.gif

Users that have the Customize Objects permission can create and modify Indexes in Objects 

In addition to speeding up data retrieval, indexes can be used to:

  • Prevent Duplicate Records: When you define a Unique Key, a record that has the same key-field values as an existing record cannot be added to the database.
  • Streamline Data Searches: When you create a (non-unique) Index Key for an Object, you can make searches run faster by selecting records with a particular key (or set of keys). That way, you pare down the data set to minimize the number of records that are actually searched.

How It Works

An index key uniquely identifies a record. In effect, it defines an address for that record. Once they system has that address, it can go straight to the record.

If you are looking for a friend in a city, one option is to go door to door looking for that person. That process is analogous to a search. Searches are thorough, and they're great when you want to retrieve a collection of records. But they're not very efficient.

Another option is to go the phone book, and find your friend's address. The phone book is an index. The person's name is the key you use to look up their address. And once you have their address, you can go straight to your friend's house.

Choosing Fields for Indexes

You can select up to three fields in an object, which are combined to create an index key.

Most field types are available for use as components in an index, with the following exceptions - these field types are not eligible as components in an index:

  • Multiline text
  • Rich text area
  • File
  • Image
  • Custom components

Guidelines

  • You can create multiple index keys for an object.
  • Any regular Index Key can be selected as a Unique Key.
  • The collection of values in an Index Key may contain duplicates. The collection of values in a Unique Key cannot.
  • Multiple Unique Keys can be created.
  • After an index is created, only the Unique Key checkbox can be modified. The selected fields cannot be changed. (The key cannot be edited. A new index field can be created, instead.)
  • The sum of the length (maximum storage length) of the fields selected for the index cannot exceed 1024 bytes.

Choosing fields for index keys takes some consideration.

Index Keys: You should select fields that distinguish the record from others in the database, and choose to combine typical "demographic" fields. For example:

 City
 State
 Industry+Zip Code
 Product+State
 Opportunity+Close Date
 URL

Unique Keys: Because the resulting value must be unique, generic "demographic" fields are generally not suitable. The following list includes the kinds of field combinations you should consider for a Unique Index Key:

 Email Address
 Contact Name+Product+Date Purchased
 Account Number

Effects of Operations

Creating a new Unique Key index on existing data
  • If the data has no duplicate entries for that combination of fields, the index is created successfully.
  • If the data has duplicate entries for that combination of fields, index creation fails.
  • The check for duplicate entries includes entries in the Recycle Bin, since they could be restored at a later date.
Adding a new record to an object with a Unique Key index
  • If no existing record has the same values for the fields that make up the unique index, the record is added successfully.
  • If there is an existing record with the same values for those fields, an error occurs.
Updating a record to change Unique Key values
  • If no other record exists with those values, the update succeeds.
  • If another record exists with those values, the update fails.
Deleting a record
When a record is deleted, corresponding index entries are deleted, as well.

Working with Indexes

Add an Index

  1. Click Designer > Data > Objects > {object}
  2. Click the Index tab
  3. Click the [Add Index] button to add an index, and add this information:
    Name
    Name is a Text field. Name takes only alphanumeric characters, and underscore.
    Enable Unique Key
    Enable unique Key is a check box. Check this box if you need to make the combination of selected fields unique.
    Select Fields to be used in Index
    This contains two combo boxes. Move one or more fields from left combo box to right combo box. The items in the right combo box will be used to create the index.
  4. Click the [Save] Button to create the index with given values

Delete an Index

To delete an existing index:

  1. Click Designer > Data > Objects > {object}
  2. Click the Index tab
  3. Click the [Delete] button