AlphabeticIndex


public final class AlphabeticIndex
extends Object implements Iterable<Bucket<V>>

java.lang.Object
   ↳ android.icu.text.AlphabeticIndex<V>


AlphabeticIndex supports the creation of a UI index appropriate for a given language. It can support either direct use, or use with a client that doesn't support localized collation. The following is an example of what an index might look like in a UI:

  ... A B C D E F G H I J K L M N O P Q R S T U V W X Y Z  ...

  A
     Addison
     Albertson
     Azensky
  B
     Baecker
  ...
 
The class can generate a list of labels for use as a UI "index", that is, a list of clickable characters (or character sequences) that allow the user to see a segment (bucket) of a larger "target" list. That is, each label corresponds to a bucket in the target list, where everything in the bucket is greater than or equal to the character (according to the locale's collation). Strings can be added to the index; they will be in sorted order in the right bucket.

The class also supports having buckets for strings before the first (underflow), after the last (overflow), and between scripts (inflow). For example, if the index is constructed with labels for Russian and English, Greek characters would fall into an inflow bucket between the other two scripts.

Note: If you expect to have a lot of ASCII or Latin characters as well as characters from the user's language, then it is a good idea to call addLabels(ULocale.English).

Direct Use

The following shows an example of building an index directly. The "show..." methods below are just to illustrate usage.

 // Create a simple index where the values for the strings are Integers, and add the strings

 AlphabeticIndex<Integer> index = new AlphabeticIndex<Integer>(desiredLocale).addLabels(additionalLocale);
 int counter = 0;
 for (String item : test) {
     index.addRecord(item, counter++);
 }
 ...
 // Show index at top. We could skip or gray out empty buckets

 for (AlphabeticIndex.Bucket<Integer> bucket : index) {
     if (showAll || bucket.size() != 0) {
         showLabelAtTop(UI, bucket.getLabel());
     }
 }
  ...
 // Show the buckets with their contents, skipping empty buckets

 for (AlphabeticIndex.Bucket<Integer> bucket : index) {
     if (bucket.size() != 0) {
         showLabelInList(UI, bucket.getLabel());
         for (AlphabeticIndex.Record<Integer> item : bucket) {
             showIndexedItem(UI, item.getName(), item.getData());
         }
 
The caller can build different UIs using this class. For example, an index character could be omitted or grayed-out if its bucket is empty. Small buckets could also be combined based on size, such as:
 ... A-F G-N O-Z ...
 

Client Support

Callers can also use the AlphabeticIndex.ImmutableIndex, or the AlphabeticIndex itself, to support sorting on a client that doesn't support AlphabeticIndex functionality.

The ImmutableIndex is both immutable and thread-safe. The corresponding AlphabeticIndex methods are not thread-safe because they "lazily" build the index buckets.

  • ImmutableIndex.getBucket(index) provides random access to all buckets and their labels and label types.
  • AlphabeticIndex.getBucketLabels() or the bucket iterator on either class can be used to get a list of the labels, such as "...", "A", "B",..., and send that list to the client.
  • When the client has a new name, it sends that name to the server. The server needs to call the following methods, and communicate the bucketIndex and collationKey back to the client.
     int bucketIndex = index.getBucketIndex(name);
     String label = immutableIndex.getBucket(bucketIndex).getLabel();  // optional
     RawCollationKey collationKey = collator.getRawCollationKey(name, null);
     
  • The client would put the name (and associated information) into its bucket for bucketIndex. The collationKey is a sequence of bytes that can be compared with a binary compare, and produce the right localized result.

Summary

Nested classes

class AlphabeticIndex.Bucket<V>

An index "bucket" with a label string and type. 

class AlphabeticIndex.ImmutableIndex<V>

Immutable, thread-safe version of AlphabeticIndex

class AlphabeticIndex.Record<V>

A (name, data) pair, to be sorted by name into one of the index buckets. 

Public constructors

AlphabeticIndex(ULocale locale)

Create the index object.

AlphabeticIndex(Locale locale)

Create the index object.

AlphabeticIndex(RuleBasedCollator collator)

Create an AlphabeticIndex that uses a specific collator.

Public methods

AlphabeticIndex<V> addLabels(ULocale... additions)

Add more index characters (aside from what are in the locale)

AlphabeticIndex<V> addLabels(UnicodeSet additions)

Add more index characters (aside from what are in the locale)

AlphabeticIndex<V> addLabels(Locale... additions)

Add more index characters (aside from what are in the locale)

AlphabeticIndex<V> addRecord(CharSequence name, V data)

Add a record (name and data) to the index.

ImmutableIndex<V> buildImmutableIndex()

Builds an immutable, thread-safe version of this instance, without data records.

AlphabeticIndex<V> clearRecords()

Clear the index.

int getBucketCount()

Return the number of buckets in the index.

int getBucketIndex(CharSequence name)

Get the bucket number for the given name.

List<String> getBucketLabels()

Get the labels.

RuleBasedCollator getCollator()

Get a clone of the collator used internally.

String getInflowLabel()

Get the default label used for abbreviated buckets between other labels.

int getMaxLabelCount()

Get the limit on the number of labels in the index.

String getOverflowLabel()

Get the default label used in the IndexCharacters' locale for overflow, eg the first item in: ...

int getRecordCount()

Return the number of records in the index: that is, the total number of distinct <name,data> pairs added with addRecord(...), over all the buckets.

String getUnderflowLabel()

Get the default label used in the IndexCharacters' locale for underflow, eg the last item in: X Y Z ...

Iterator<Bucket<V>> iterator()

Return an iterator over the buckets.

AlphabeticIndex<V> setInflowLabel(String inflowLabel)

Set the inflowLabel label

AlphabeticIndex<V> setMaxLabelCount(int maxLabelCount)

Set a limit on the number of labels in the index.

AlphabeticIndex<V> setOverflowLabel(String overflowLabel)

Set the overflow label

AlphabeticIndex<V> setUnderflowLabel(String underflowLabel)

Set the underflowLabel label

Inherited methods

Public constructors

AlphabeticIndex

Added in API level 24
public AlphabeticIndex (ULocale locale)

Create the index object.

Parameters
locale ULocale: The locale for the index.

AlphabeticIndex

Added in API level 24
public AlphabeticIndex (Locale locale)

Create the index object.

Parameters
locale Locale: The locale for the index.

AlphabeticIndex

Added in API level 24
public AlphabeticIndex (RuleBasedCollator collator)

Create an AlphabeticIndex that uses a specific collator.

The index will be created with no labels; the addLabels() function must be called after creation to add the desired labels to the index.

The index will work directly with the supplied collator. If the caller will need to continue working with the collator it should be cloned first, so that the collator provided to the AlphabeticIndex remains unchanged after creation of the index.

Parameters
collator RuleBasedCollator: The collator to use to order the contents of this index.

Public methods

addLabels

Added in API level 24
public AlphabeticIndex<V> addLabels (ULocale... additions)

Add more index characters (aside from what are in the locale)

Parameters
additions ULocale: additional characters to add to the index, such as those in Swedish.

Returns
AlphabeticIndex<V> this, for chaining

addLabels

Added in API level 24
public AlphabeticIndex<V> addLabels (UnicodeSet additions)

Add more index characters (aside from what are in the locale)

Parameters
additions UnicodeSet: additional characters to add to the index, such as A-Z.

Returns
AlphabeticIndex<V> this, for chaining

addLabels

Added in API level 24
public AlphabeticIndex<V> addLabels (Locale... additions)

Add more index characters (aside from what are in the locale)

Parameters
additions Locale: additional characters to add to the index, such as those in Swedish.

Returns
AlphabeticIndex<V> this, for chaining

addRecord

Added in API level 24
public AlphabeticIndex<V> addRecord (CharSequence name, 
                V data)

Add a record (name and data) to the index. The name will be used to sort the items into buckets, and to sort within the bucket. Two records may have the same name. When they do, the sort order is according to the order added: the first added comes first.

Parameters
name CharSequence: Name, such as a name

data V: Data, such as an address or link

Returns
AlphabeticIndex<V> this, for chaining

buildImmutableIndex

Added in API level 24
public ImmutableIndex<V> buildImmutableIndex ()

Builds an immutable, thread-safe version of this instance, without data records.

Returns
ImmutableIndex<V> an immutable index instance

clearRecords

Added in API level 24
public AlphabeticIndex<V> clearRecords ()

Clear the index.

Returns
AlphabeticIndex<V> this, for chaining

getBucketCount

Added in API level 24
public int getBucketCount ()

Return the number of buckets in the index. This will be the same as the number of labels, plus buckets for the underflow, overflow, and inflow(s).

Returns
int number of buckets

getBucketIndex

Added in API level 24
public int getBucketIndex (CharSequence name)

Get the bucket number for the given name. This routine permits callers to implement their own bucket handling mechanisms, including client-server handling. For example, when a new name is created on the client, it can ask the server for the bucket for that name, and the sortkey (using getCollator). Once the client has that information, it can put the name into the right bucket, and sort it within that bucket, without having access to the index or collator.

Note that the bucket number (and sort key) are only valid for the settings of the current AlphabeticIndex; if those are changed, then the bucket number and sort key must be regenerated.

Parameters
name CharSequence: Name, such as a name

Returns
int the bucket index for the name

getBucketLabels

Added in API level 24
public List<String> getBucketLabels ()

Get the labels.

Returns
List<String> The list of bucket labels, after processing.

getCollator

Added in API level 24
public RuleBasedCollator getCollator ()

Get a clone of the collator used internally. Note that for performance reasons, the clone is only done once, and then stored. The next time it is accessed, the same instance is returned.

Don't use this method across threads if you are changing the settings on the collator, at least not without synchronizing.

Returns
RuleBasedCollator a clone of the collator used internally

getInflowLabel

Added in API level 24
public String getInflowLabel ()

Get the default label used for abbreviated buckets between other labels. For example, consider the labels for Latin and Greek are used: X Y Z ... Α Β Γ.

Returns
String inflow label

getMaxLabelCount

Added in API level 24
public int getMaxLabelCount ()

Get the limit on the number of labels in the index. The number of buckets can be slightly larger: see getBucketCount().

Returns
int maxLabelCount maximum number of labels.

getOverflowLabel

Added in API level 24
public String getOverflowLabel ()

Get the default label used in the IndexCharacters' locale for overflow, eg the first item in: ... A B C

Returns
String overflow label

getRecordCount

Added in API level 24
public int getRecordCount ()

Return the number of records in the index: that is, the total number of distinct <name,data> pairs added with addRecord(...), over all the buckets.

Returns
int total number of records in buckets

getUnderflowLabel

Added in API level 24
public String getUnderflowLabel ()

Get the default label used in the IndexCharacters' locale for underflow, eg the last item in: X Y Z ...

Returns
String underflow label

iterator

Added in API level 24
public Iterator<Bucket<V>> iterator ()

Return an iterator over the buckets.

Returns
Iterator<Bucket<V>> iterator over buckets.

setInflowLabel

Added in API level 24
public AlphabeticIndex<V> setInflowLabel (String inflowLabel)

Set the inflowLabel label

Parameters
inflowLabel String: see class description

Returns
AlphabeticIndex<V> this, for chaining

setMaxLabelCount

Added in API level 24
public AlphabeticIndex<V> setMaxLabelCount (int maxLabelCount)

Set a limit on the number of labels in the index. The number of buckets can be slightly larger: see getBucketCount().

Parameters
maxLabelCount int: Set the maximum number of labels. Currently, if the number is exceeded, then every nth item is removed to bring the count down. A more sophisticated mechanism may be available in the future.

Returns
AlphabeticIndex<V> this, for chaining

setOverflowLabel

Added in API level 24
public AlphabeticIndex<V> setOverflowLabel (String overflowLabel)

Set the overflow label

Parameters
overflowLabel String: see class description

Returns
AlphabeticIndex<V> this, for chaining

setUnderflowLabel

Added in API level 24
public AlphabeticIndex<V> setUnderflowLabel (String underflowLabel)

Set the underflowLabel label

Parameters
underflowLabel String: see class description

Returns
AlphabeticIndex<V> this, for chaining