descriptor.proto

// Protocol Buffers - Google's data interchange format
// Copyright 2008 Google Inc.  All rights reserved.
// http://code.google.com/p/protobuf/
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
//     * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
//     * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following disclaimer
// in the documentation and/or other materials provided with the
// distribution.
//     * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

// Author: kenton@google.com (Kenton Varda)
//  Based on original Protocol Buffers design by
//  Sanjay Ghemawat, Jeff Dean, and others.
//
// The messages in this file describe the definitions found in .proto files.
// A valid .proto file can be translated directly to a FileDescriptorProto
// without any other information (e.g. without reading its imports).



package google.protobuf;
option java_package = "com.google.protobuf";
option java_outer_classname = "DescriptorProtos";

// descriptor.proto must be optimized for speed because reflection-based
// algorithms don't work during bootstrapping.
option optimize_for = SPEED;

// The protocol compiler can output a FileDescriptorSet containing the .proto
// files it parses.
message FileDescriptorSet {
  repeated FileDescriptorProto file = 1;
}

// Describes a complete .proto file.
message FileDescriptorProto {
  optional string name = 1;       // file name, relative to root of source tree
  optional string package = 2;    // e.g. "foo", "foo.bar", etc.

  // Names of files imported by this file.
  repeated string dependency = 3;

  // All top-level definitions in this file.
  repeated DescriptorProto message_type = 4;
  repeated EnumDescriptorProto enum_type = 5;
  repeated ServiceDescriptorProto service = 6;
  repeated FieldDescriptorProto extension = 7;

  optional FileOptions options = 8;

  // This field contains optional information about the original source code.
  // You may safely remove this entire field whithout harming runtime
  // functionality of the descriptors -- the information is needed only by
  // development tools.
  optional SourceCodeInfo source_code_info = 9;
}

// Describes a message type.
message DescriptorProto {
  optional string name = 1;

  repeated FieldDescriptorProto field = 2;
  repeated FieldDescriptorProto extension = 6;

  repeated DescriptorProto nested_type = 3;
  repeated EnumDescriptorProto enum_type = 4;

  message ExtensionRange {
    optional int32 start = 1;
    optional int32 end = 2;
  }
  repeated ExtensionRange extension_range = 5;

  optional MessageOptions options = 7;
}

// Describes a field within a message.
message FieldDescriptorProto {
  enum Type {
    // 0 is reserved for errors.
    // Order is weird for historical reasons.
    TYPE_DOUBLE         = 1;
    TYPE_FLOAT          = 2;
    TYPE_INT64          = 3;   // Not ZigZag encoded.  Negative numbers
                               // take 10 bytes.  Use TYPE_SINT64 if negative
                               // values are likely.
    TYPE_UINT64         = 4;
    TYPE_INT32          = 5;   // Not ZigZag encoded.  Negative numbers
                               // take 10 bytes.  Use TYPE_SINT32 if negative
                               // values are likely.
    TYPE_FIXED64        = 6;
    TYPE_FIXED32        = 7;
    TYPE_BOOL           = 8;
    TYPE_STRING         = 9;
    TYPE_GROUP          = 10;  // Tag-delimited aggregate.
    TYPE_MESSAGE        = 11;  // Length-delimited aggregate.

    // New in version 2.
    TYPE_BYTES          = 12;
    TYPE_UINT32         = 13;
    TYPE_ENUM           = 14;
    TYPE_SFIXED32       = 15;
    TYPE_SFIXED64       = 16;
    TYPE_SINT32         = 17;  // Uses ZigZag encoding.
    TYPE_SINT64         = 18;  // Uses ZigZag encoding.
  };

  enum Label {
    // 0 is reserved for errors
    LABEL_OPTIONAL      = 1;
    LABEL_REQUIRED      = 2;
    LABEL_REPEATED      = 3;
    // TODO(sanjay): Should we add LABEL_MAP?
  };

  optional string name = 1;
  optional int32 number = 3;
  optional Label label = 4;

  // If type_name is set, this need not be set.  If both this and type_name
  // are set, this must be either TYPE_ENUM or TYPE_MESSAGE.
  optional Type type = 5;

  // For message and enum types, this is the name of the type.  If the name
  // starts with a '.', it is fully-qualified.  Otherwise, C++-like scoping
  // rules are used to find the type (i.e. first the nested types within this
  // message are searched, then within the parent, on up to the root
  // namespace).
  optional string type_name = 6;

  // For extensions, this is the name of the type being extended.  It is
  // resolved in the same manner as type_name.
  optional string extendee = 2;

  // For numeric types, contains the original text representation of the value.
  // For booleans, "true" or "false".
  // For strings, contains the default text contents (not escaped in any way).
  // For bytes, contains the C escaped value.  All bytes >= 128 are escaped.
  // TODO(kenton):  Base-64 encode?
  optional string default_value = 7;

  optional FieldOptions options = 8;
}

// Describes an enum type.
message EnumDescriptorProto {
  optional string name = 1;

  repeated EnumValueDescriptorProto value = 2;

  optional EnumOptions options = 3;
}

// Describes a value within an enum.
message EnumValueDescriptorProto {
  optional string name = 1;
  optional int32 number = 2;

  optional EnumValueOptions options = 3;
}

// Describes a service.
message ServiceDescriptorProto {
  optional string name = 1;
  repeated MethodDescriptorProto method = 2;

  optional ServiceOptions options = 3;
}

// Describes a method of a service.
message MethodDescriptorProto {
  optional string name = 1;

  // Input and output type names.  These are resolved in the same way as
  // FieldDescriptorProto.type_name, but must refer to a message type.
  optional string input_type = 2;
  optional string output_type = 3;

  optional MethodOptions options = 4;
}

// ===================================================================
// Options

// Each of the definitions above may have "options" attached.  These are
// just annotations which may cause code to be generated slightly differently
// or may contain hints for code that manipulates protocol messages.
//
// Clients may define custom options as extensions of the *Options messages.
// These extensions may not yet be known at parsing time, so the parser cannot
// store the values in them.  Instead it stores them in a field in the *Options
// message called uninterpreted_option. This field must have the same name
// across all *Options messages. We then use this field to populate the
// extensions when we build a descriptor, at which point all protos have been
// parsed and so all extensions are known.
//
// Extension numbers for custom options may be chosen as follows:
// * For options which will only be used within a single application or
//   organization, or for experimental options, use field numbers 50000
//   through 99999.  It is up to you to ensure that you do not use the
//   same number for multiple options.
// * For options which will be published and used publicly by multiple
//   independent entities, e-mail kenton@google.com to reserve extension
//   numbers.  Simply tell me how many you need and I'll send you back a
//   set of numbers to use -- there's no need to explain how you intend to
//   use them.  If this turns out to be popular, a web service will be set up
//   to automatically assign option numbers.


message FileOptions {

  // Sets the Java package where classes generated from this .proto will be
  // placed.  By default, the proto package is used, but this is often
  // inappropriate because proto packages do not normally start with backwards
  // domain names.
  optional string java_package = 1;


  // If set, all the classes from the .proto file are wrapped in a single
  // outer class with the given name.  This applies to both Proto1
  // (equivalent to the old "--one_java_file" option) and Proto2 (where
  // a .proto always translates to a single class, but you may want to
  // explicitly choose the class name).
  optional string java_outer_classname = 8;

  // If set true, then the Java code generator will generate a separate .java
  // file for each top-level message, enum, and service defined in the .proto
  // file.  Thus, these types will *not* be nested inside the outer class
  // named by java_outer_classname.  However, the outer class will still be
  // generated to contain the file's getDescriptor() method as well as any
  // top-level extensions defined in the file.
  optional bool java_multiple_files = 10 [default=false];

  // If set true, then the Java code generator will generate equals() and
  // hashCode() methods for all messages defined in the .proto file. This is
  // purely a speed optimization, as the AbstractMessage base class includes
  // reflection-based implementations of these methods.
  optional bool java_generate_equals_and_hash = 20 [default=false];

  // Generated classes can be optimized for speed or code size.
  enum OptimizeMode {
    SPEED = 1;        // Generate complete code for parsing, serialization,
                      // etc.
    CODE_SIZE = 2;    // Use ReflectionOps to implement these methods.
    LITE_RUNTIME = 3; // Generate code using MessageLite and the lite runtime.
  }
  optional OptimizeMode optimize_for = 9 [default=SPEED];




  // Should generic services be generated in each language?  "Generic" services
  // are not specific to any particular RPC system.  They are generated by the
  // main code generators in each language (without additional plugins).
  // Generic services were the only kind of service generation supported by
  // early versions of proto2.
  //
  // Generic services are now considered deprecated in favor of using plugins
  // that generate code specific to your particular RPC system.  Therefore,
  // these default to false.  Old code which depends on generic services should
  // explicitly set them to true.
  optional bool cc_generic_services = 16 [default=false];
  optional bool java_generic_services = 17 [default=false];
  optional bool py_generic_services = 18 [default=false];

  // The parser stores options it doesn't recognize here. See above.
  repeated UninterpretedOption uninterpreted_option = 999;

  // Clients can define custom options in extensions of this message. See above.
  extensions 1000 to max;
}

message MessageOptions {
  // Set true to use the old proto1 MessageSet wire format for extensions.
  // This is provided for backwards-compatibility with the MessageSet wire
  // format.  You should not use this for any other reason:  It's less
  // efficient, has fewer features, and is more complicated.
  //
  // The message must be defined exactly as follows:
  //   message Foo {
  //     option message_set_wire_format = true;
  //     extensions 4 to max;
  //   }
  // Note that the message cannot have any defined fields; MessageSets only
  // have extensions.
  //
  // All extensions of your type must be singular messages; e.g. they cannot
  // be int32s, enums, or repeated messages.
  //
  // Because this is an option, the above two restrictions are not enforced by
  // the protocol compiler.
  optional bool message_set_wire_format = 1 [default=false];

  // Disables the generation of the standard "descriptor()" accessor, which can
  // conflict with a field of the same name.  This is meant to make migration
  // from proto1 easier; new code should avoid fields named "descriptor".
  optional bool no_standard_descriptor_accessor = 2 [default=false];

  // The parser stores options it doesn't recognize here. See above.
  repeated UninterpretedOption uninterpreted_option = 999;

  // Clients can define custom options in extensions of this message. See above.
  extensions 1000 to max;
}

message FieldOptions {
  // The ctype option instructs the C++ code generator to use a different
  // representation of the field than it normally would.  See the specific
  // options below.  This option is not yet implemented in the open source
  // release -- sorry, we'll try to include it in a future version!
  optional CType ctype = 1 [default = STRING];
  enum CType {
    // Default mode.
    STRING = 0;

    CORD = 1;

    STRING_PIECE = 2;
  }
  // The packed option can be enabled for repeated primitive fields to enable
  // a more efficient representation on the wire. Rather than repeatedly
  // writing the tag and type for each element, the entire array is encoded as
  // a single length-delimited blob.
  optional bool packed = 2;


  // Is this field deprecated?
  // Depending on the target platform, this can emit Deprecated annotations
  // for accessors, or it will be completely ignored; in the very least, this
  // is a formalization for deprecating fields.
  optional bool deprecated = 3 [default=false];

  // EXPERIMENTAL.  DO NOT USE.
  // For "map" fields, the name of the field in the enclosed type that
  // is the key for this map.  For example, suppose we have:
  //   message Item {
  //     required string name = 1;
  //     required string value = 2;
  //   }
  //   message Config {
  //     repeated Item items = 1 [experimental_map_key="name"];
  //   }
  // In this situation, the map key for Item will be set to "name".
  // TODO: Fully-implement this, then remove the "experimental_" prefix.
  optional string experimental_map_key = 9;

  // The parser stores options it doesn't recognize here. See above.
  repeated UninterpretedOption uninterpreted_option = 999;

  // Clients can define custom options in extensions of this message. See above.
  extensions 1000 to max;
}

message EnumOptions {

  // The parser stores options it doesn't recognize here. See above.
  repeated UninterpretedOption uninterpreted_option = 999;

  // Clients can define custom options in extensions of this message. See above.
  extensions 1000 to max;
}

message EnumValueOptions {
  // The parser stores options it doesn't recognize here. See above.
  repeated UninterpretedOption uninterpreted_option = 999;

  // Clients can define custom options in extensions of this message. See above.
  extensions 1000 to max;
}

message ServiceOptions {

  // Note:  Field numbers 1 through 32 are reserved for Google's internal RPC
  //   framework.  We apologize for hoarding these numbers to ourselves, but
  //   we were already using them long before we decided to release Protocol
  //   Buffers.

  // The parser stores options it doesn't recognize here. See above.
  repeated UninterpretedOption uninterpreted_option = 999;

  // Clients can define custom options in extensions of this message. See above.
  extensions 1000 to max;
}

message MethodOptions {

  // Note:  Field numbers 1 through 32 are reserved for Google's internal RPC
  //   framework.  We apologize for hoarding these numbers to ourselves, but
  //   we were already using them long before we decided to release Protocol
  //   Buffers.

  // The parser stores options it doesn't recognize here. See above.
  repeated UninterpretedOption uninterpreted_option = 999;

  // Clients can define custom options in extensions of this message. See above.
  extensions 1000 to max;
}

// A message representing a option the parser does not recognize. This only
// appears in options protos created by the compiler::Parser class.
// DescriptorPool resolves these when building Descriptor objects. Therefore,
// options protos in descriptor objects (e.g. returned by Descriptor::options(),
// or produced by Descriptor::CopyTo()) will never have UninterpretedOptions
// in them.
message UninterpretedOption {
  // The name of the uninterpreted option.  Each string represents a segment in
  // a dot-separated name.  is_extension is true iff a segment represents an
  // extension (denoted with parentheses in options specs in .proto files).
  // E.g.,{ ["foo", false], ["bar.baz", true], ["qux", false] } represents
  // "foo.(bar.baz).qux".
  message NamePart {
    required string name_part = 1;
    required bool is_extension = 2;
  }
  repeated NamePart name = 2;

  // The value of the uninterpreted option, in whatever type the tokenizer
  // identified it as during parsing. Exactly one of these should be set.
  optional string identifier_value = 3;
  optional uint64 positive_int_value = 4;
  optional int64 negative_int_value = 5;
  optional double double_value = 6;
  optional bytes string_value = 7;
  optional string aggregate_value = 8;
}

// ===================================================================
// Optional source code info

// Encapsulates information about the original source file from which a
// FileDescriptorProto was generated.
message SourceCodeInfo {
  // A Location identifies a piece of source code in a .proto file which
  // corresponds to a particular definition.  This information is intended
  // to be useful to IDEs, code indexers, documentation generators, and similar
  // tools.
  //
  // For example, say we have a file like:
  //   message Foo {
  //     optional string foo = 1;
  //   }
  // Let's look at just the field definition:
  //   optional string foo = 1;
  //   ^       ^^     ^^  ^  ^^^
  //   a       bc     de  f  ghi
  // We have the following locations:
  //   span   path               represents
  //   [a,i)  [ 4, 0, 2, 0 ]     The whole field definition.
  //   [a,b)  [ 4, 0, 2, 0, 4 ]  The label (optional).
  //   [c,d)  [ 4, 0, 2, 0, 5 ]  The type (string).
  //   [e,f)  [ 4, 0, 2, 0, 1 ]  The name (foo).
  //   [g,h)  [ 4, 0, 2, 0, 3 ]  The number (1).
  //
  // Notes:
  // - A location may refer to a repeated field itself (i.e. not to any
  //   particular index within it).  This is used whenever a set of elements are
  //   logically enclosed in a single code segment.  For example, an entire
  //   extend block (possibly containing multiple extension definitions) will
  //   have an outer location whose path refers to the "extensions" repeated
  //   field without an index.
  // - Multiple locations may have the same path.  This happens when a single
  //   logical declaration is spread out across multiple places.  The most
  //   obvious example is the "extend" block again -- there may be multiple
  //   extend blocks in the same scope, each of which will have the same path.
  // - A location's span is not always a subset of its parent's span.  For
  //   example, the "extendee" of an extension declaration appears at the
  //   beginning of the "extend" block and is shared by all extensions within
  //   the block.
  // - Just because a location's span is a subset of some other location's span
  //   does not mean that it is a descendent.  For example, a "group" defines
  //   both a type and a field in a single declaration.  Thus, the locations
  //   corresponding to the type and field and their components will overlap.
  // - Code which tries to interpret locations should probably be designed to
  //   ignore those that it doesn't understand, as more types of locations could
  //   be recorded in the future.
  repeated Location location = 1;
  message Location {
    // Identifies which part of the FileDescriptorProto was defined at this
    // location.
    //
    // Each element is a field number or an index.  They form a path from
    // the root FileDescriptorProto to the place where the definition.  For
    // example, this path:
    //   [ 4, 3, 2, 7, 1 ]
    // refers to:
    //   file.message_type(3)  // 4, 3
    //       .field(7)         // 2, 7
    //       .name()           // 1
    // This is because FileDescriptorProto.message_type has field number 4:
    //   repeated DescriptorProto message_type = 4;
    // and DescriptorProto.field has field number 2:
    //   repeated FieldDescriptorProto field = 2;
    // and FieldDescriptorProto.name has field number 1:
    //   optional string name = 1;
    //
    // Thus, the above path gives the location of a field name.  If we removed
    // the last element:
    //   [ 4, 3, 2, 7 ]
    // this path refers to the whole field declaration (from the beginning
    // of the label to the terminating semicolon).
    repeated int32 path = 1 [packed=true];

    // Always has exactly three or four elements: start line, start column,
    // end line (optional, otherwise assumed same as start line), end column.
    // These are packed into a single field for efficiency.  Note that line
    // and column numbers are zero-based -- typically you will want to add
    // 1 to each before displaying to a user.
    repeated int32 span = 2 [packed=true];

    // TODO(kenton):  Record comments appearing before and after the
    // declaration.
  }
}