com.amazon.ion.impl.lite.IonStructLite Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of ion-java Show documentation
Show all versions of ion-java Show documentation
A Java implementation of the Amazon Ion data notation.
/*
* Copyright 2007-2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License").
* You may not use this file except in compliance with the License.
* A copy of the License is located at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* or in the "license" file accompanying this file. This file is distributed
* on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either
* express or implied. See the License for the specific language governing
* permissions and limitations under the License.
*/
package com.amazon.ion.impl.lite;
import com.amazon.ion.ContainedValueException;
import com.amazon.ion.IonStruct;
import com.amazon.ion.IonType;
import com.amazon.ion.IonValue;
import com.amazon.ion.SymbolToken;
import com.amazon.ion.ValueFactory;
import com.amazon.ion.ValueVisitor;
import com.amazon.ion.impl._Private_CurriedValueFactory;
import com.amazon.ion.UnknownSymbolException;
import java.io.PrintWriter;
import java.util.Arrays;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Iterator;
import java.util.Map;
import java.util.Map.Entry;
import java.util.Set;
final class IonStructLite
extends IonContainerLite
implements IonStruct
{
private static final int HASH_SIGNATURE =
IonType.STRUCT.toString().hashCode();
// TODO amazon-ion/ion-java/issues/41: add support for _isOrdered
IonStructLite(ContainerlessContext context, boolean isNull)
{
super(context, isNull);
}
private IonStructLite(IonStructLite existing, IonContext context)
{
super(existing, context);
// The field map provides optimized lookups when the struct is larger than a few fields. Rather than being
// eagerly copied during clone, it is created on-demand. This has been shown to improve performance when cloning
// structs, even in cases where the field map is eventually needed.
this._field_map = null;
this._field_map_duplicate_count = existing._field_map_duplicate_count;
this.hasNullFieldName = existing.hasNullFieldName;
}
private Map _field_map;
private boolean hasNullFieldName = false;
public int _field_map_duplicate_count;
@Override
public IonStructLite clone()
{
return (IonStructLite) deepClone(false);
}
@Override
IonValueLite shallowClone(IonContext context) {
return new IonStructLite(this, context);
}
@Override
protected void transitionToLargeSize(int size)
{
if (_field_map != null) return;
build_field_map();
}
private void build_field_map()
{
int size = (_children == null) ? 0 : _children.length;
_field_map = new HashMap<>((int) Math.ceil(size / 0.75f), 0.75f); // avoids unconditional growth
_field_map_duplicate_count = 0;
int count = get_child_count();
for (int ii=0; ii newFieldIdx) {
newFieldIdx = idx;
}
}
_field_map.put(fieldName, newFieldIdx);
}
private void remove_field(String fieldName, int lowest_idx, int copies)
{
if (_field_map == null) {
return;
}
_field_map.remove(fieldName);
_field_map_duplicate_count -= (copies - 1);
}
private void remove_field_from_field_map(String fieldName, int idx)
{
Integer field_idx = _field_map.get(fieldName);
assert(field_idx != null);
if (field_idx != idx) {
// if the map has a different index, this must
// be a duplicate, and this copy isn't in the map
assert(_field_map_duplicate_count > 0);
_field_map_duplicate_count--;
}
else if (_field_map_duplicate_count > 0) {
// if we have any duplicates we have to check
// every time since we don't track which field
// is duplicated - so any dup can be expensive
int ii = find_last_duplicate(fieldName, idx);
if (ii == -1) {
// this is the last copy of this key
_field_map.remove(fieldName);
}
else {
// replaces this fields (the one being
// removed) array idx in the map with
// the preceding duplicates index
_field_map.put(fieldName, ii);
_field_map_duplicate_count--;
}
}
else {
// since there are not dup's we can just update
// the map by removing this fieldname
_field_map.remove(fieldName);
}
}
private void patch_map_elements_helper(int removed_idx)
{
if (_field_map == null) {
return;
}
if (removed_idx >= get_child_count()) {
// if this was the at the end of the list
// there's nothing to change
return;
}
for (int ii=removed_idx; ii> it = _field_map.entrySet().iterator();
out.print(" map: [");
boolean first = true;
while (it.hasNext()) {
Entry e = it.next();
if (!first) {
out.print(",");
}
out.print(e.getKey()+":"+e.getValue());
first = false;
}
out.println("]");
}
@Override
public String validate()
{
if (_field_map == null) {
return null;
}
String error = "";
Iterator> it = _field_map.entrySet().iterator();
while (it.hasNext()) {
Entry e = it.next();
int idx = e.getValue();
IonValueLite v = (idx >= 0 && idx < get_child_count()) ? get_child(idx) : null;
if (v == null || idx != v._elementid() || (e.getKey().equals(v.getFieldName()) == false)) {
error += "map entry ["+e+"] doesn't match list value ["+v+"]\n";
}
}
return (error == "") ? null : error;
}
private int find_last_duplicate(String fieldName, int existing_idx)
{
for (int ii=existing_idx; ii>0; ) {
ii--;
IonValueLite field = get_child(ii);
if (fieldName.equals(field.getFieldName())) {
return ii;
}
}
assert(there_is_only_one(fieldName, existing_idx));
return -1;
}
private boolean there_is_only_one(String fieldName, int existing_idx)
{
int count = 0;
for (int ii=0; ii fields =
new HashSet(Arrays.asList(fieldNames));
if (keep && fields.contains(null))
{
throw new NullPointerException("Can't retain an unknown field name");
}
clone = getSystem().newEmptyStruct();
clone._children = new IonValueLite[_children.length];
clone._child_count = 0;
for (int i = 0; i < _child_count; i++) {
IonValueLite value = _children[i];
if (fields.contains(value._fieldName) == keep)
{
IonValueLite clonedChild = (IonValueLite) value.clone();
// This ensures that we don't copy an unknown field name.
clonedChild._fieldName = value.getFieldName();
clonedChild._elementid(clone._child_count);
clonedChild._context = clone;
if (!clone._isSymbolIdPresent() && clonedChild._isSymbolIdPresent())
{
clone.cascadeSIDPresentToContextRoot();
}
clone._children[clone._child_count++] = clonedChild;
}
}
}
if (_annotations != null && _annotations.length > 0) {
clone._annotations = _annotations.clone();
clone.checkAnnotationsForSids();
}
return clone;
}
@Override
public IonType getType()
{
return IonType.STRUCT;
}
public boolean containsKey(Object fieldName)
{
String name = (String) fieldName;
return (null != get(name));
}
public boolean containsValue(Object value)
{
IonValue v = (IonValue) value;
return (v.getContainer() == this);
}
public IonValue get(String fieldName)
{
int field_idx = find_field_helper(fieldName);
IonValue field;
if (field_idx < 0) {
if(hasNullFieldName) throw new UnknownSymbolException("Unable to determine whether the field exists because the struct contains field names with unknown text.");
field = null;
} else {
field = get_child(field_idx);
}
return field;
}
private boolean fieldMapIsActive(int proposedSize) {
if (_field_map != null) return true;
if (proposedSize <= STRUCT_INITIAL_SIZE) return false;
build_field_map();
return true;
}
private int find_field_helper(String fieldName)
{
validateFieldName(fieldName);
if (isNullValue()) {
// nothing to see here, move along
}
else if (fieldMapIsActive(_child_count)) {
Integer idx = _field_map.get(fieldName);
if (idx != null) {
return idx;
}
}
else {
int ii, size = get_child_count();
for (ii=0; ii m)
{
// TODO this is very inefficient
for (Entry entry : m.entrySet())
{
put(entry.getKey(), entry.getValue());
}
}
/**
* put is "make this value the one and only value
* associated with this fieldName". The side effect
* is that if there were multiple fields with this
* name when put is complete there will only be the
* one value in the collection.
*/
public void put(String fieldName, IonValue value)
{
checkForLock();
validateFieldName(fieldName);
if (value != null) validateNewChild(value);
int lowestRemovedIndex = _child_count;
boolean any_removed = false;
// first we remove the any existing fields
// associated with fieldName (which may be none)
if (_field_map_duplicate_count == 0 && fieldMapIsActive(_child_count))
{
// we have a map and no duplicates so the index
// (aka map) is all we need to find the only
// value associated with fieldName, if there is one
Integer idx = _field_map.get(fieldName);
if (idx != null) {
lowestRemovedIndex = idx;
remove_field_from_field_map(fieldName, lowestRemovedIndex);
remove_child(lowestRemovedIndex);
any_removed = true;
}
}
else {
// either we don't have a map (index) or there
// are duplicates in both cases we have to
// scan the child list directly.
// Walk backwards to minimize array movement
// as we remove fields as we encounter them.
int copies_removed = 0;
for (int ii = get_child_count(); ii > 0; )
{
ii--;
IonValueLite child = get_child(ii);
if (fieldName.equals(child._fieldName))
{
// done by remove_child: child.detachFromContainer();
remove_child(ii);
lowestRemovedIndex = ii;
copies_removed++;
any_removed = true;
}
}
if (any_removed) {
remove_field(fieldName, lowestRemovedIndex, copies_removed);
}
}
if (any_removed) {
patch_map_elements_helper(lowestRemovedIndex);
patch_elements_helper(lowestRemovedIndex);
}
// once we've removed any existing copy we now add,
// this (delete + add == put) turns out be be the
// right choice since:
// 1 - because of possible duplicates we can't
// guarantee the idx is stable
// 2 - we have to maintain the hash and that
// really means we end up with the delete
// anyway
// strictly speaking this approach, while simpler,
// is more expensive when we don't have a has and
// the value already exists, and it's not at the
// end of the field list anyway.
if (value != null) {
add(fieldName, value);
}
}
@Override
void beforeIteratorRemove(IonValueLite value, int idx) {
if (_field_map != null) {
remove_field_from_field_map(value.getFieldName(), idx);
}
}
@Override
void afterIteratorRemove(IonValueLite value, int idx) {
if (_field_map != null) {
patch_map_elements_helper(idx);
}
}
public IonValue remove(String fieldName)
{
checkForLock();
IonValue field = get(fieldName);
if (field == null) {
return null;
}
int idx = ((IonValueLite)field)._elementid();
// update the hash map first we don't want
// the child list changed until we've done
// this since the map update expects the
// index value of the remove field to be
// correct and unchanged.
if (_field_map != null) {
remove_field_from_field_map(fieldName, idx);
}
super.remove(field);
if (_field_map != null) {
patch_map_elements_helper(idx);
}
return field;
}
@Override
public boolean remove(IonValue element)
{
if (element == null) {
throw new NullPointerException();
}
checkForLock();
if (element.getContainer() != this) {
return false;
}
IonValueLite concrete = (IonValueLite) element;
int idx = concrete._elementid();
// update the hash map first we don't want
// the child list changed until we've done
// this since the map update expects the
// index value of the remove field to be
// correct and unchanged.
if (_field_map != null) {
remove_field_from_field_map(concrete.getFieldName(), idx);
}
super.remove(concrete);
if (_field_map != null) {
patch_map_elements_helper(idx);
}
return true;
}
public boolean removeAll(String... fieldNames)
{
boolean removedAny = false;
checkForLock();
int size = get_child_count();
for (int ii=size; ii>0; ) {
ii--;
IonValue field = get_child(ii);
if (isListedField(field, fieldNames)) {
field.removeFromContainer();
removedAny = true;
}
}
return removedAny;
}
public boolean retainAll(String... fieldNames)
{
checkForLock();
boolean removedAny = false;
int size = get_child_count();
for (int ii=size; ii>0; ) {
ii--;
IonValue field = get_child(ii);
if (! isListedField(field, fieldNames))
{
field.removeFromContainer();
removedAny = true;
}
}
return removedAny;
}
/**
*
* @param field must not be null. It is not required to have a field name.
* @param fields must not be null, and must not contain and nulls.
* @return true if {@code field.getFieldName()} is in {@code fields}.
*/
private static boolean isListedField(IonValue field, String[] fields)
{
String fieldName = field.getFieldName();
for (String key : fields)
{
if (key.equals(fieldName)) return true;
}
return false;
}
/**
* Ensures that a given field name is valid. Used as a helper for
* methods that have that precondition.
*
* @throws NullPointerException if the fieldName
* is null.
*/
private static void validateFieldName(String fieldName)
{
if (fieldName == null)
{
throw new NullPointerException("fieldName is null");
}
}
@Override
public void accept(ValueVisitor visitor) throws Exception
{
visitor.visit(this);
}
}