freemarker.template.SimpleHash Maven / Gradle / Ivy
/*
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License 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 freemarker.template;
import java.io.Serializable;
import java.util.ConcurrentModificationException;
import java.util.HashMap;
import java.util.Iterator;
import java.util.Map;
import java.util.SortedMap;
import java.util.TreeMap;
import freemarker.core._DelayedJQuote;
import freemarker.core._TemplateModelException;
import freemarker.ext.beans.BeansWrapper;
/**
* A simple implementation of the {@link TemplateHashModelEx} interface, using its own underlying {@link Map} or
* {@link SortedMap} for storing the hash entries. If you are wrapping an already existing {@link Map}, you should
* certainly use {@link DefaultMapAdapter} instead (see comparison below).
*
*
* This class is thread-safe if you don't call modifying methods (like {@link #put(String, Object)},
* {@link #remove(String)}, etc.) after you have made the object available for multiple threads (assuming you have
* published it safely to the other threads; see JSR-133 Java Memory Model). These methods aren't called by FreeMarker,
* so it's usually not a concern.
*
*
* {@link SimpleHash} VS {@link DefaultMapAdapter} - Which to use when?
*
*
* For a {@link Map} that exists regardless of FreeMarker, only you need to access it from templates,
* {@link DefaultMapAdapter} should be the default choice, as it reflects the exact behavior of the underlying
* {@link Map} (no surprises), can be unwrapped to the originally wrapped object (important when passing it to Java
* methods from the template), and has more predictable performance (no spikes).
*
*
* For a hash that's made specifically to be used from templates, creating an empty {@link SimpleHash} then filling it
* with {@link SimpleHash#put(String, Object)} is usually the way to go, as the resulting hash is significantly faster
* to read from templates than a {@link DefaultMapAdapter} (though it's somewhat slower to read from a plain Java method
* to which it had to be passed adapted to a {@link Map}).
*
*
* It also matters if for how many times will the same {@link Map} entry be read from the template(s) later, on
* average. If, on average, you read each entry for more than 4 times, {@link SimpleHash} will be most certainly faster,
* but if for 2 times or less (and especially if not at all) then {@link DefaultMapAdapter} will be faster. Before
* choosing based on performance though, pay attention to the behavioral differences; {@link SimpleHash} will
* shallow-copy the original {@link Map} at construction time, so key order will be lost in some cases, and it won't
* reflect {@link Map} content changes after the {@link SimpleHash} construction, also {@link SimpleHash} can't be
* unwrapped to the original {@link Map} instance.
*
* @see DefaultMapAdapter
* @see TemplateHashModelEx
*/
public class SimpleHash extends WrappingTemplateModel implements TemplateHashModelEx2, Serializable {
private final Map map;
private boolean putFailed;
private Map unwrappedMap;
/**
* Constructs an empty hash that uses the default wrapper set in
* {@link WrappingTemplateModel#setDefaultObjectWrapper(ObjectWrapper)}.
*
* @deprecated Use {@link #SimpleHash(ObjectWrapper)}
*/
@Deprecated
public SimpleHash() {
this((ObjectWrapper) null);
}
/**
* Creates a new simple hash with the copy of the underlying map and the
* default wrapper set in
* {@link WrappingTemplateModel#setDefaultObjectWrapper(ObjectWrapper)}.
* @param map The Map to use for the key/value pairs. It makes a copy for
* internal use. If the map implements the {@link SortedMap} interface, the
* internal copy will be a {@link TreeMap}, otherwise it will be a
* {@link HashMap}.
*
* @deprecated Use {@link #SimpleHash(Map, ObjectWrapper)}
*/
@Deprecated
public SimpleHash(Map map) {
this(map, null);
}
/**
* Creates an empty simple hash using the specified object wrapper.
* @param wrapper The object wrapper to use to wrap objects into
* {@link TemplateModel} instances. If null, the default wrapper set in
* {@link WrappingTemplateModel#setDefaultObjectWrapper(ObjectWrapper)} is
* used.
*/
public SimpleHash(ObjectWrapper wrapper) {
super(wrapper);
map = new HashMap();
}
/**
* Creates an instance that will use the specified {@link Map} directly as its backing store; beware, the
* {@link Map} will be possibly modified by {@link SimpleHash}, even if you only read the {@link SimpleHash}.
* That's because when a value is read, it's replaced with the corresponding {@link TemplateModel}.
*
*
The goal of this constructor is to allow you to control technical aspects, like the initial capacity, and
* ordering of the underlying {@link Map} implementation. The iteration order of the {@link SimpleHash} will be
* the same as of the underlying {@link Map}.
*
* @param directMap The map that the instance will use directly as backing storage; possibly will be modified,
* even when you only read the {@link SimpleHash}! Must allow any kind of object as value,
* including {@code null}! Supporting {@code null} keys is not needed.
*
* @param overloadDistinction To avoid ambiguity with other overloads; the value is unused.
*
* @since 2.3.30
*/
public SimpleHash(Map directMap, ObjectWrapper wrapper, int overloadDistinction) {
super(wrapper);
this.map = directMap;
}
/**
* Creates a new hash by shallow-coping (possibly cloning) the underlying map; in many applications you should use
* {@link DefaultMapAdapter} instead.
*
* @param map
* The Map to use for the key/value pairs. It makes a copy for internal use. If the map implements the
* {@link SortedMap} interface, the internal copy will be a {@link TreeMap}, otherwise it will be a
* @param wrapper
* The object wrapper to use to wrap contained objects into {@link TemplateModel} instances. Using
* {@code null} is deprecated but allowed, in which case the deprecated default wrapper set in
* {@link WrappingTemplateModel#setDefaultObjectWrapper(ObjectWrapper)} is used.
*/
public SimpleHash(Map map, ObjectWrapper wrapper) {
super(wrapper);
Map mapCopy;
try {
mapCopy = copyMap(map);
} catch (ConcurrentModificationException cme) {
//This will occur extremely rarely.
//If it does, we just wait 5 ms and try again. If
// the ConcurrentModificationException
// is thrown again, we just let it bubble up this time.
// TODO: Maybe we should log here.
try {
Thread.sleep(5);
} catch (InterruptedException ie) {
}
synchronized (map) {
mapCopy = copyMap(map);
}
}
this.map = mapCopy;
}
protected Map copyMap(Map map) {
if (map instanceof HashMap) {
return (Map) ((HashMap) map).clone();
}
if (map instanceof SortedMap) {
if (map instanceof TreeMap) {
return (Map) ((TreeMap) map).clone();
} else {
return new TreeMap((SortedMap) map);
}
}
return new HashMap(map);
}
/**
* Adds a key-value entry to this hash.
*
* @param key
* The name by which the object is identified in the template.
* @param value
* The value to which the name will be associated. This will only be wrapped to {@link TemplateModel}
* lazily when it's first read.
*/
public void put(String key, Object value) {
map.put(key, value);
unwrappedMap = null;
}
/**
* Puts a boolean in the map
*
* @param key the name by which the resulting TemplateModel
* is identified in the template.
* @param b the boolean to store.
*/
public void put(String key, boolean b) {
put(key, b ? TemplateBooleanModel.TRUE : TemplateBooleanModel.FALSE);
}
@Override
public TemplateModel get(String key) throws TemplateModelException {
Object result;
try {
result = map.get(key);
} catch (ClassCastException e) {
throw new _TemplateModelException(e,
"ClassCastException while getting Map entry with String key ",
new _DelayedJQuote(key));
} catch (NullPointerException e) {
throw new _TemplateModelException(e,
"NullPointerException while getting Map entry with String key ",
new _DelayedJQuote(key));
}
// The key to use for putting -- it's the key that already exists in
// the map (either key or charKey below). This way, we'll never put a
// new key in the map, avoiding spurious ConcurrentModificationException
// from another thread iterating over the map, see bug #1939742 in
// SourceForge tracker.
Object putKey = null;
if (result == null) {
// Check for Character key if this is a single-character string.
// In SortedMap-s, however, we can't do that safely, as it can cause ClassCastException.
if (key.length() == 1 && !(map instanceof SortedMap)) {
Character charKey = Character.valueOf(key.charAt(0));
try {
result = map.get(charKey);
if (result != null || map.containsKey(charKey)) {
putKey = charKey;
}
} catch (ClassCastException e) {
throw new _TemplateModelException(e,
"ClassCastException while getting Map entry with Character key ",
new _DelayedJQuote(key));
} catch (NullPointerException e) {
throw new _TemplateModelException(e,
"NullPointerException while getting Map entry with Character key ",
new _DelayedJQuote(key));
}
}
if (putKey == null) {
if (!map.containsKey(key)) {
return null;
} else {
putKey = key;
}
}
} else {
putKey = key;
}
if (result instanceof TemplateModel) {
return (TemplateModel) result;
}
TemplateModel tm = wrap(result);
if (!putFailed) {
try {
map.put(putKey, tm);
} catch (Exception e) {
// If it's immutable or something, we just keep going.
putFailed = true;
}
}
return tm;
}
/**
* Tells if the map contains a key or not, regardless if the associated value is {@code null} or not.
* @since 2.3.20
*/
public boolean containsKey(String key) {
return map.containsKey(key);
}
/**
* Removes the given key from the underlying map.
*
* @param key the key to be removed
*/
public void remove(String key) {
map.remove(key);
}
/**
* Adds all the key/value entries in the map
* @param m the map with the entries to add, the keys are assumed to be strings.
*/
public void putAll(Map m) {
for (Iterator it = m.entrySet().iterator(); it.hasNext(); ) {
Map.Entry entry = (Map.Entry) it.next();
this.put((String) entry.getKey(), entry.getValue());
}
}
/**
* Note that this method creates and returns a deep-copy of the underlying hash used
* internally. This could be a gotcha for some people
* at some point who want to alter something in the data model,
* but we should maintain our immutability semantics (at least using default SimpleXXX wrappers)
* for the data model. It will recursively unwrap the stuff in the underlying container.
*/
public Map toMap() throws TemplateModelException {
if (unwrappedMap == null) {
Class mapClass = this.map.getClass();
Map m = null;
try {
m = (Map) mapClass.newInstance();
} catch (Exception e) {
throw new TemplateModelException("Error instantiating map of type " + mapClass.getName() + "\n" + e.getMessage());
}
// Create a copy to maintain immutability semantics and
// Do nested unwrapping of elements if necessary.
BeansWrapper bw = BeansWrapper.getDefaultInstance();
for (Iterator it = map.entrySet().iterator(); it.hasNext(); ) {
Map.Entry entry = (Map.Entry) it.next();
Object key = entry.getKey();
Object value = entry.getValue();
if (value instanceof TemplateModel) {
value = bw.unwrap((TemplateModel) value);
}
m.put(key, value);
}
unwrappedMap = m;
}
return unwrappedMap;
}
/**
* Returns the {@code toString()} of the underlying {@link Map}.
*/
@Override
public String toString() {
return map.toString();
}
@Override
public int size() {
return map.size();
}
@Override
public boolean isEmpty() {
return map == null || map.isEmpty();
}
@Override
public TemplateCollectionModel keys() {
return new SimpleCollection(map.keySet(), getObjectWrapper());
}
@Override
public TemplateCollectionModel values() {
return new SimpleCollection(map.values(), getObjectWrapper());
}
@Override
public KeyValuePairIterator keyValuePairIterator() {
return new MapKeyValuePairIterator(map, getObjectWrapper());
}
public SimpleHash synchronizedWrapper() {
return new SynchronizedHash();
}
private class SynchronizedHash extends SimpleHash {
@Override
public boolean isEmpty() {
synchronized (SimpleHash.this) {
return SimpleHash.this.isEmpty();
}
}
@Override
public void put(String key, Object obj) {
synchronized (SimpleHash.this) {
SimpleHash.this.put(key, obj);
}
}
@Override
public TemplateModel get(String key) throws TemplateModelException {
synchronized (SimpleHash.this) {
return SimpleHash.this.get(key);
}
}
@Override
public void remove(String key) {
synchronized (SimpleHash.this) {
SimpleHash.this.remove(key);
}
}
@Override
public int size() {
synchronized (SimpleHash.this) {
return SimpleHash.this.size();
}
}
@Override
public TemplateCollectionModel keys() {
synchronized (SimpleHash.this) {
return SimpleHash.this.keys();
}
}
@Override
public TemplateCollectionModel values() {
synchronized (SimpleHash.this) {
return SimpleHash.this.values();
}
}
@Override
public KeyValuePairIterator keyValuePairIterator() {
synchronized (SimpleHash.this) {
return SimpleHash.this.keyValuePairIterator();
}
}
@Override
public Map toMap() throws TemplateModelException {
synchronized (SimpleHash.this) {
return SimpleHash.this.toMap();
}
}
}
}