SyncJson.Reader Struct Reference

An implementation of ISyncManager for reading JSON objects. Designed to be both fast and flexible, this implementation normally reads UTF8 directly into values without allocating intermediate strings, and without requiring the entire JSON file to be loaded into memory at once. SupportsReordering, SupportsDeduplication and SupportsNextField are all true, and non-strict JSON is allowed (e.g. comments are accepted but ignored, unless you turn off support in the Options.) More...

---

Source file:

• /Core/Loyc.SyncLib.SyncJson/SyncJson.Reader.cs

Inheritance diagram for SyncJson.Reader:

Remarks

An implementation of ISyncManager for reading JSON objects. Designed to be both fast and flexible, this implementation normally reads UTF8 directly into values without allocating intermediate strings, and without requiring the entire JSON file to be loaded into memory at once. SupportsReordering, SupportsDeduplication and SupportsNextField are all true, and non-strict JSON is allowed (e.g. comments are accepted but ignored, unless you turn off support in the Options.)

For best performance, your synchronizers should read the JSON data in the same order it was written. Synchronizers written in the usual way naturally work this way. Out-of-order reads are supported but are slower and, when reading large JSON files, may use more memory.

Since the JSON is always read in a single pass from an IScanner{byte}, trying to read a JSON property that doesn't exist can, in the worst case, cause the whole file to be buffered into memory. However, JSON files that are (essentially) large arrays won't have this problem, since you can only read arrays in order.

This type can read JSON files larger than 2GB, provided that an out-of- order read doesn't cause over 2GB of data to be scanned at once.

While normally you can read properties in any order, metadata properties such as $id and $ref must be located at the beginning of a JSON object in order to be detected during deserialization. (Newtonsoft.Json has the same restriction, by the way.) In addition, object IDs must always be represented by the same byte sequence, e.g. "3" and "\u0033" are not treated as the same ID even though they both represent "3" in JSON.

This is a struct rather than a class for performance reasons. Don't try to use a default(Reader); it'll throw NullReferenceException.

Public fields
SyncMode	Mode => SyncMode.Reading
bool	IsReading => true
bool	IsWriting => false
bool	SupportsReordering => true
bool	SupportsDeduplication => true
bool	NeedsIntegerIds => false
bool	IsPlainText => true
ISyncOptions	Options => _s._opt
bool	IsInsideList => _s.IsInsideList
bool	ReachedEndOfList => _s.IsInsideList ? _s.ReachedEndOfList : null
int	MinimumListLength => 0
int	Depth => _s.Depth
bool	SupportsNextField => true
FieldId	NextField => _s.NextField
bool	Begun
bool int	Length

Properties
object	CurrentObject [set]

Public Member Functions
bool int object Object	BeginSubObject (FieldId name, object?childKey, ObjectMode mode, int listLength=-1)
void	EndSubObject ()
SyncType	GetFieldType (FieldId name, SyncType expectedType=SyncType.Unknown)
string	SyncTypeTag (string?tag)
bool	Sync (FieldId name, bool savable)
sbyte	Sync (FieldId name, sbyte savable)
byte	Sync (FieldId name, byte savable)
short	Sync (FieldId name, short savable)
ushort	Sync (FieldId name, ushort savable)
int	Sync (FieldId name, int savable)
uint	Sync (FieldId name, uint savable)
long	Sync (FieldId name, long savable)
ulong	Sync (FieldId name, ulong savable)
float	Sync (FieldId name, float savable)
double	Sync (FieldId name, double savable)
decimal	Sync (FieldId name, decimal savable)
BigInteger	Sync (FieldId name, BigInteger savable)
char	Sync (FieldId name, char savable)
int	Sync (FieldId name, int savable, int bits, bool signed=true)
long	Sync (FieldId name, long savable, int bits, bool signed=true)
BigInteger	Sync (FieldId name, BigInteger savable, int bits, bool signed=true)
List	SyncListBoolImpl< Scanner, List, ListBuilder > (FieldId name, Scanner scanner, List?saving, ListBuilder builder, ObjectMode mode, int tupleLength=-1)
List	SyncListByteImpl< Scanner, List, ListBuilder > (FieldId name, Scanner scanner, List?saving, ListBuilder builder, ObjectMode mode, int tupleLength=-1)
List	SyncListCharImpl< Scanner, List, ListBuilder > (FieldId name, Scanner scanner, List?saving, ListBuilder builder, ObjectMode mode, int tupleLength=-1)
bool	Sync (FieldId name, bool?savable)
sbyte	Sync (FieldId name, sbyte?savable)
byte	Sync (FieldId name, byte?savable)
short	Sync (FieldId name, short?savable)
ushort	Sync (FieldId name, ushort?savable)
int	Sync (FieldId name, int?savable)
uint	Sync (FieldId name, uint?savable)
long	Sync (FieldId name, long?savable)
ulong	Sync (FieldId name, ulong?savable)
float	Sync (FieldId name, float?savable)
double	Sync (FieldId name, double?savable)
decimal	Sync (FieldId name, decimal?savable)
BigInteger	Sync (FieldId name, BigInteger?savable)
char	Sync (FieldId name, char?savable)
string	Sync (FieldId name, string?savable, ObjectMode mode=ObjectMode.Normal)