Package google.golang.org/appengine/datastore (v1.6.8)

Package datastore provides a client for App Engine's datastore service.

Basic Operations

Entities are the unit of storage and are associated with a key. A key consists of an optional parent key, a string application ID, a string kind (also known as an entity type), and either a StringID or an IntID. A StringID is also known as an entity name or key name.

It is valid to create a key with a zero StringID and a zero IntID; this is called an incomplete key, and does not refer to any saved entity. Putting an entity into the datastore under an incomplete key will cause a unique key to be generated for that entity, with a non-zero IntID.

An entity's contents are a mapping from case-sensitive field names to values. Valid value types are:

• signed integers (int, int8, int16, int32 and int64),
• bool,
• string,
• float32 and float64,
• []byte (up to 1 megabyte in length),
• any type whose underlying type is one of the above predeclared types,
• ByteString,
• *Key,
• time.Time (stored with microsecond precision),
• appengine.BlobKey,
• appengine.GeoPoint,
• structs whose fields are all valid value types,
• slices of any of the above.

Slices of structs are valid, as are structs that contain slices. However, if one struct contains another, then at most one of those can be repeated. This disqualifies recursively defined struct types: any struct T that (directly or indirectly) contains a []T.

The Get and Put functions load and save an entity's contents. An entity's contents are typically represented by a struct pointer.

Example code:

typeEntitystruct{
Valuestring
}
funchandle(whttp.ResponseWriter,r*http.Request){
ctx:=appengine.NewContext(r)
k:=datastore.NewKey(ctx,"Entity","stringID",0,nil)
e:=new(Entity)
iferr:=datastore.Get(ctx,k,e);err!=nil{
http.Error(w,err.Error(),500)
return
}
old:=e.Value
e.Value=r.URL.Path
if_,err:=datastore.Put(ctx,k,e);err!=nil{
http.Error(w,err.Error(),500)
return
}
w.Header().Set("Content-Type","text/plain; charset=utf-8")
fmt.Fprintf(w,"old=%q\nnew=%q\n",old,e.Value)
}

GetMulti, PutMulti and DeleteMulti are batch versions of the Get, Put and Delete functions. They take a []*Key instead of a *Key, and may return an appengine.MultiError when encountering partial failure.

Properties

An entity's contents can be represented by a variety of types. These are typically struct pointers, but can also be any type that implements the PropertyLoadSaver interface. If using a struct pointer, you do not have to explicitly implement the PropertyLoadSaver interface; the datastore will automatically convert via reflection. If a struct pointer does implement that interface then those methods will be used in preference to the default behavior for struct pointers. Struct pointers are more strongly typed and are easier to use; PropertyLoadSavers are more flexible.

The actual types passed do not have to match between Get and Put calls or even across different calls to datastore. It is valid to put a *PropertyList and get that same entity as a *myStruct, or put a *myStruct0 and get a *myStruct1. Conceptually, any entity is saved as a sequence of properties, and is loaded into the destination value on a property-by-property basis. When loading into a struct pointer, an entity that cannot be completely represented (such as a missing field) will result in an ErrFieldMismatch error but it is up to the caller whether this error is fatal, recoverable or ignorable.

By default, for struct pointers, all properties are potentially indexed, and the property name is the same as the field name (and hence must start with an upper case letter).

Fields may have a datastore:"name,options" tag. The tag name is the property name, which must be one or more valid Go identifiers joined by ".", but may start with a lower case letter. An empty tag name means to just use the field name. A "-" tag name means that the datastore will ignore that field.

The only valid options are "omitempty" and "noindex".

If the options include "omitempty" and the value of the field is empty, then the field will be omitted on Save. The empty values are false, 0, any nil interface value, and any array, slice, map, or string of length zero. Struct field values will never be empty.

If options include "noindex" then the field will not be indexed. All fields are indexed by default. Strings or byte slices longer than 1500 bytes cannot be indexed; fields used to store long strings and byte slices must be tagged with "noindex" or they will cause Put operations to fail.

To use multiple options together, separate them by a comma. The order does not matter.

If the options is "" then the comma may be omitted.

Example code:

// A and B are renamed to a and b.
// A, C and J are not indexed.
// D's tag is equivalent to having no tag at all (E).
// I is ignored entirely by the datastore.
// J has tag information for both the datastore and json packages.
typeTaggedStructstruct{
Aint`datastore:"a,noindex"`
Bint`datastore:"b"`
Cint`datastore:",noindex"`
Dint`datastore:""`
Eint
Iint`datastore:"-"`
Jint`datastore:",noindex"json:"j"`
}

Structured Properties

If the struct pointed to contains other structs, then the nested or embedded structs are flattened. For example, given these definitions:

typeInner1struct{
Wint32
Xstring
}
typeInner2struct{
Yfloat64
}
typeInner3struct{
Zbool
}
typeOuterstruct{
Aint16
I[]Inner1
JInner2
Inner3
}

then an Outer's properties would be equivalent to those of:

typeOuterEquivalentstruct{
Aint16
IDotW[]int32`datastore:"I.W"`
IDotX[]string`datastore:"I.X"`
JDotYfloat64`datastore:"J.Y"`
Zbool
}

If Outer's embedded Inner3 field was tagged as datastore:"Foo" then the equivalent field would instead be: FooDotZ bool datastore:"Foo.Z".

If an outer struct is tagged "noindex" then all of its implicit flattened fields are effectively "noindex".

The PropertyLoadSaver Interface

An entity's contents can also be represented by any type that implements the PropertyLoadSaver interface. This type may be a struct pointer, but it does not have to be. The datastore package will call Load when getting the entity's contents, and Save when putting the entity's contents. Possible uses include deriving non-stored fields, verifying fields, or indexing a field only if its value is positive.

Example code:

typeCustomPropsExamplestruct{
I,Jint
//Sumisnotstored,butshouldalwaysbeequaltoI+J.
Sumint`datastore:"-"`
}
func(x*CustomPropsExample)Load(ps[]datastore.Property)error{
//LoadIandJasusual.
iferr:=datastore.LoadStruct(x,ps);err!=nil{
returnerr
}
//DerivetheSumfield.
x.Sum=x.I+x.J
returnnil
}
func(x*CustomPropsExample)Save()([]datastore.Property,error){
//ValidatetheSumfield.
ifx.Sum!=x.I+x.J{
returnnil,errors.New("CustomPropsExample has inconsistent sum")
}
//SaveIandJasusual.Thecodebelowisequivalenttocalling
//"return datastore.SaveStruct(x)",butisdonemanuallyfor
//demonstrationpurposes.
return[]datastore.Property{
{
Name:"I",
Value:int64(x.I),
},
{
Name:"J",
Value:int64(x.J),
},
},nil
}

The *PropertyList type implements PropertyLoadSaver, and can therefore hold an arbitrary entity's contents.

Queries

Queries retrieve entities based on their properties or key's ancestry. Running a query yields an iterator of results: either keys or (key, entity) pairs. Queries are re-usable and it is safe to call Query.Run from concurrent goroutines. Iterators are not safe for concurrent use.

Queries are immutable, and are either created by calling NewQuery, or derived from an existing query by calling a method like Filter or Order that returns a new query value. A query is typically constructed by calling NewQuery followed by a chain of zero or more such methods. These methods are:

• Ancestor and Filter constrain the entities returned by running a query.
• Order affects the order in which they are returned.
• Project constrains the fields returned.
• Distinct de-duplicates projected entities.
• KeysOnly makes the iterator return only keys, not (key, entity) pairs.
• Start, End, Offset and Limit define which sub-sequence of matching entities to return. Start and End take cursors, Offset and Limit take integers. Start and Offset affect the first result, End and Limit affect the last result. If both Start and Offset are set, then the offset is relative to Start. If both End and Limit are set, then the earliest constraint wins. Limit is relative to Start+Offset, not relative to End. As a special case, a negative limit means unlimited.

Example code:

typeWidgetstruct{
Descriptionstring
Priceint
}
funchandle(whttp.ResponseWriter,r*http.Request){
ctx:=appengine.NewContext(r)
q:=datastore.NewQuery("Widget").
Filter("Price <",1000).
Order("-Price")
b:=new(bytes.Buffer)
fort:=q.Run(ctx);;{
varxWidget
key,err:=t.Next(&x)
iferr==datastore.Done{
break
}
iferr!=nil{
serveError(ctx,w,err)
return
}
fmt.Fprintf(b,"Key=%v\nWidget=%#v\n\n",key,x)
}
w.Header().Set("Content-Type","text/plain; charset=utf-8")
io.Copy(w,b)
}

Transactions

RunInTransaction runs a function in a transaction.

Example code:

typeCounterstruct{
Countint
}
funcinc(ctxcontext.Context,key*datastore.Key)(int,error){
varxCounter
iferr:=datastore.Get(ctx,key,&x);err!=nil && err!=datastore.ErrNoSuchEntity{
return0,err
}
x.Count++
if_,err:=datastore.Put(ctx,key,&x);err!=nil{
return0,err
}
returnx.Count,nil
}
funchandle(whttp.ResponseWriter,r*http.Request){
ctx:=appengine.NewContext(r)
varcountint
err:=datastore.RunInTransaction(ctx,func(ctxcontext.Context)error{
varerr1error
count,err1=inc(ctx,datastore.NewKey(ctx,"Counter","singleton",0,nil))
returnerr1
},nil)
iferr!=nil{
serveError(ctx,w,err)
return
}
w.Header().Set("Content-Type","text/plain; charset=utf-8")
fmt.Fprintf(w,"Count=%d",count)
}

Metadata

The datastore package provides access to some of App Engine's datastore metadata. This metadata includes information about the entity groups, namespaces, entity kinds, and properties in the datastore, as well as the property representations for each property.

Example code:

funchandle(whttp.ResponseWriter,r*http.Request){
//Printallthekindsinthedatastore,withalltheindexed
//properties(andtheirrepresentations)foreach.
ctx:=appengine.NewContext(r)
kinds,err:=datastore.Kinds(ctx)
iferr!=nil{
serveError(ctx,w,err)
return
}
w.Header().Set("Content-Type","text/plain; charset=utf-8")
for_,kind:=rangekinds{
fmt.Fprintf(w,"%s:\n",kind)
props,err:=datastore.KindProperties(ctx,kind)
iferr!=nil{
fmt.Fprintln(w,"\t(unable to retrieve properties)")
continue
}
forp,rep:=rangeprops{
fmt.Fprintf(w,"\t-%s (%s)\n",p,strings.Join(rep,", "))
}
}
}

Done is returned when a query iteration has completed.

ErrConcurrentTransaction is returned when a transaction is rolled back due to a conflict with a concurrent transaction.

AllocateIDRange allocates a range of IDs with specific endpoints. The range is inclusive at both the low and high end. Once these IDs have been allocated, you can manually assign them to newly created entities.

The Datastore's automatic ID allocator never assigns a key that has already been allocated (either through automatic ID allocation or through an explicit AllocateIDs call). As a result, entities written to the given key range will never be overwritten. However, writing entities with manually assigned keys in this range may overwrite existing entities (or new entities written by a separate request), depending on the error returned.

Use this only if you have an existing numeric ID range that you want to reserve (for example, bulk loading entities that already have IDs). If you don't care about which IDs you receive, use AllocateIDs instead.

AllocateIDRange returns nil if the range is successfully allocated. If one or more entities with an ID in the given range already exist, it returns a KeyRangeCollisionError. If the Datastore has already cached IDs in this range (e.g. from a previous call to AllocateIDRange), it returns a KeyRangeContentionError. Errors of other types indicate problems with arguments or an error returned directly from the Datastore.

AllocateIDs returns a range of n integer IDs with the given kind and parent combination. kind cannot be empty; parent may be nil. The IDs in the range returned will not be used by the datastore's automatic ID sequence generator and may be used with NewKey without conflict.

The range is inclusive at the low end and exclusive at the high end. In other words, valid intIDs x satisfy low <= x && x < high.

If no error is returned, low + n == high.

Delete deletes the entity for the given key.

DeleteMulti is a batch version of Delete.

EnableKeyConversion enables encoded key compatibility with the Cloud Datastore client library (cloud.google.com/go/datastore). Encoded keys generated by the Cloud Datastore client library will be decoded into App Engine datastore keys.

The context provided must be an App Engine context if running in App Engine first generation runtime. This can be called in the /_ah/start handler. It is safe to call multiple times, and is cheap to call, so can also be inserted as middleware.

Enabling key compatibility does not affect the encoding format used by Key.Encode, it only expands the type of keys that are able to be decoded with DecodeKey.

Get loads the entity stored for k into dst, which must be a struct pointer or implement PropertyLoadSaver. If there is no such entity for the key, Get returns ErrNoSuchEntity.

The values of dst's unmatched struct fields are not modified, and matching slice-typed fields are not reset before appending to them. In particular, it is recommended to pass a pointer to a zero valued struct on each Get call.

ErrFieldMismatch is returned when a field is to be loaded into a different type than the one it was stored from, or when a field is missing or unexported in the destination struct. ErrFieldMismatch is only returned if dst is a struct pointer.

GetMulti is a batch version of Get.

dst must be a []S, []*S, []I or []P, for some struct type S, some interface type I, or some non-interface non-pointer type P such that P or *P implements PropertyLoadSaver. If an []I, each element must be a valid dst for Get: it must be a struct pointer or implement PropertyLoadSaver.

As a special case, PropertyList is an invalid type for dst, even though a PropertyList is a slice of structs. It is treated as invalid to avoid being mistakenly passed when []PropertyList was intended.

KindProperties returns all the indexed properties for the given kind. The properties are returned as a map of property names to a slice of the representation types. The representation types for the supported Go property types are:

"INT64": signed integers and time.Time
"DOUBLE": float32 and float64
"BOOLEAN": bool
"STRING": string, []byte and ByteString
"POINT": appengine.GeoPoint
"REFERENCE": *Key
"USER": (not used in the Go runtime)

Kinds returns the names of all the kinds in the current namespace.

LoadStruct loads the properties from p to dst. dst must be a struct pointer.

Namespaces returns all the datastore namespaces.

RunInTransaction runs f in a transaction. It calls f with a transaction context tc that f should use for all App Engine operations.

If f returns nil, RunInTransaction attempts to commit the transaction, returning nil if it succeeds. If the commit fails due to a conflicting transaction, RunInTransaction retries f, each time with a new transaction context. It gives up and returns ErrConcurrentTransaction after three failed attempts. The number of attempts can be configured by specifying TransactionOptions.Attempts.

If f returns non-nil, then any datastore changes will not be applied and RunInTransaction returns that same error. The function f is not retried.

Note that when f returns, the transaction is not yet committed. Calling code must be careful not to assume that any of f's changes have been committed until RunInTransaction returns nil.

Since f may be called multiple times, f should usually be idempotent. datastore.Get is not idempotent when unmarshaling slice fields.

Nested transactions are not supported; c may not be a transaction context.

ByteString is a short byte slice (up to 1500 bytes) that can be indexed.

Cursor is an iterator's position. It can be converted to and from an opaque string. A cursor can be used from different HTTP requests, but only with a query with the same kind, ancestor, filter and order constraints.

DecodeCursor decodes a cursor from its base-64 string representation.

String returns a base-64 string representation of a cursor.

An Entity is the value type for a nested struct. This type is only used for a Property's Value.

ErrFieldMismatch is returned when a field is to be loaded into a different type than the one it was stored from, or when a field is missing or unexported in the destination struct. StructType is the type of the struct pointed to by the destination argument passed to Get or to Iterator.Next.

Iterator is the result of running a query.

Cursor returns a cursor for the iterator's current location.

Next returns the key of the next result. When there are no more results, Done is returned as the error.

If the query is not keys only and dst is non-nil, it also loads the entity stored for that key into the struct pointer or PropertyLoadSaver dst, with the same semantics and possible errors as for the Get function.

Key represents the datastore key for a stored entity, and is immutable.

DecodeKey decodes a key from the opaque representation returned by Encode.

NewIncompleteKey creates a new incomplete key. kind cannot be empty.

NewKey creates a new key. kind cannot be empty. Either one or both of stringID and intID must be zero. If both are zero, the key returned is incomplete. parent must either be a complete key or nil.

Put saves the entity src into the datastore with key k. src must be a struct pointer or implement PropertyLoadSaver; if a struct pointer then any unexported fields of that struct will be skipped. If k is an incomplete key, the returned key will be a unique key generated by the datastore.

PutMulti is a batch version of Put.

src must satisfy the same conditions as the dst argument to GetMulti.

AppID returns the key's application ID.

Encode returns an opaque representation of the key suitable for use in HTML and URLs. This is compatible with the Python and Java runtimes.

Equal returns whether two keys are equal.

Incomplete returns whether the key does not refer to a stored entity. In particular, whether the key has a zero StringID and a zero IntID.

IntID returns the key's integer ID, which may be 0.

Kind returns the key's kind (also known as entity type).

Namespace returns the key's namespace.

Parent returns the key's parent key, which may be nil.

String returns a string representation of the key.

StringID returns the key's string ID (also known as an entity name or key name), which may be "".

Property is a name/value pair plus some metadata. A datastore entity's contents are loaded and saved as a sequence of Properties. An entity can have multiple Properties with the same name, provided that p.Multiple is true on all of that entity's Properties with that name.

SaveStruct returns the properties from src as a slice of Properties. src must be a struct pointer.

PropertyList converts a []Property to implement PropertyLoadSaver.

Load loads all of the provided properties into l. It does not first reset *l to an empty slice.

Save saves all of l's properties as a slice or Properties.

PropertyLoadSaver can be converted from and to a slice of Properties.

Query represents a datastore query.

NewQuery creates a new Query for a specific entity kind.

An empty kind means to return all entities, including entities created and managed by other App Engine features, and is called a kindless query. Kindless queries cannot include filters or sort orders on property values.

Ancestor returns a derivative query with an ancestor filter. The ancestor should not be nil.

BatchSize returns a derivative query to fetch the supplied number of results at once. This value should be greater than zero, and equal to or less than the Limit.

Count returns the number of results for the query.

The running time and number of API calls made by Count scale linearly with the sum of the query's offset and limit. Unless the result count is expected to be small, it is best to specify a limit; otherwise Count will continue until it finishes counting or the provided context expires.

Distinct returns a derivative query that yields de-duplicated entities with respect to the set of projected fields. It is only used for projection queries. Distinct cannot be used with DistinctOn.

DistinctOn returns a derivative query that yields de-duplicated entities with respect to the set of the specified fields. It is only used for projection queries. The field list should be a subset of the projected field list. DistinctOn cannot be used with Distinct.

End returns a derivative query with the given end point.

EventualConsistency returns a derivative query that returns eventually consistent results. It only has an effect on ancestor queries.

Filter returns a derivative query with a field-based filter. The filterStr argument must be a field name followed by optional space, followed by an operator, one of ">", "<", ">=", "<=", or "=". Fields are compared against the provided value using the operator. Multiple filters are AND'ed together.

GetAll runs the query in the given context and returns all keys that match that query, as well as appending the values to dst.

dst must have type []S or *[]S or *[]P, for some struct type S or some non- interface, non-pointer type P such that P or *P implements PropertyLoadSaver.

As a special case, *PropertyList is an invalid type for dst, even though a PropertyList is a slice of structs. It is treated as invalid to avoid being mistakenly passed when *[]PropertyList was intended.

The keys returned by GetAll will be in a 1-1 correspondence with the entities added to dst.

If q is a "keys-only" query, GetAll ignores dst and only returns the keys.

The running time and number of API calls made by GetAll scale linearly with the sum of the query's offset and limit. Unless the result count is expected to be small, it is best to specify a limit; otherwise GetAll will continue until it finishes collecting results or the provided context expires.

KeysOnly returns a derivative query that yields only keys, not keys and entities. It cannot be used with projection queries.

Limit returns a derivative query that has a limit on the number of results returned. A negative value means unlimited.

Offset returns a derivative query that has an offset of how many keys to skip over before returning results. A negative value is invalid.

Order returns a derivative query with a field-based sort order. Orders are applied in the order they are added. The default order is ascending; to sort in descending order prefix the fieldName with a minus sign (-).

Project returns a derivative query that yields only the given fields. It cannot be used with KeysOnly.

Run runs the query in the given context.

Start returns a derivative query with the given start point.

TransactionOptions are the options for running a transaction.