1FLATC(1) User Manuals FLATC(1)
2
6 flatc - flatbuffers compiler
7
9 flatc [generator-options] [-o PATH] [-I=path] [-S] files... [--
10 files...]
11 The files are read and parsed in order, and can contain either schemas
13 or data (see below). Data files are processed according to the defini‐
14 tions of the most recent schema specified. -- indicates that the fol‐
15 lowing files are binary files in FlatBuffer format conforming to the
16 schema indicated before it. Depending on the flags passed, additional
17 files may be generated for each file processed.
18
20 For any schema input files, one or more generators can be specified:
21 --cpp, -c
23 Generate a C++ header for all definitions in this file (as file‐
24 name_generated.h).
25 --java, -j
27 Generate Java code.
28 --csharp, -n
30 Generate C# code.
31 --go, -g
33 Generate Go code.
34 --python, -p
36 Generate Python code.
37 --js, -s
39 Generate JavaScript code.
40 --ts
42 Generate TypeScript code.
43 --php
45 Generate PHP code.
46 --grpc
48 Generate RPC stub code for GRPC.
49 --dart
51 Generate Dart code.
52 --lua
54 Generate Lua code.
55 --lobster
57 Generate Lobster code.
58 --rust, -r
60 Generate Rust code.
61 For any data input files:
63 --binary, -b
65 If data is contained in this file, generate a filename.bin contain‐
66 ing the binary flatbuffer (or a different extension if one is specified
67 in the schema).
68 --json, -t
70 If data is contained in this file, generate a filename.json repre‐
71 senting the data in the flatbuffer.
72 Additional options:
74 -o PATH
76 Output all generated files to PATH (either absolute, or relative to
77 the current directory). If omitted, PATH will be the current directory.
78 PATH should end in your systems path separator, e.g. / or \.
79 -I PATH
81 when encountering include statements, attempt to load the files from
82 this path. Paths will be tried in the order given, and if all fail (or
83 none are specified) it will try to load relative to the path of the
84 schema file being parsed.
85 -M
87 Print make rules for generated files.
88 --strict-json
90 Require & generate strict JSON (field names are enclosed in quotes,
91 no trailing commas in tables/vectors). By default, no quotes are
92 required/generated, and trailing commas are allowed.
93 --defaults-json
95 Output fields whose value is equal to the default value when writing
96 JSON text.
97 --no-prefix
99 Don't prefix enum values in generated C++ by their enum type.
100 --scoped-enums
102 Use C++11 style scoped and strongly typed enums in generated C++.
103 This also implies --no-prefix.
104 --gen-includes
106 (deprecated), this is the default behavior. If the original behavior
107 is required (no include statements) use --no-includes.
108 --no-includes
110 Don't generate include statements for included schemas the generated
111 file depends on (C++).
112 --gen-mutable
114 Generate additional non-const accessors for mutating FlatBuffers
115 in-place.
116 --gen-object-api
118 Generate an additional object-based API. This API is more convenient
119 for object construction and mutation than the base API, at the cost of
120 efficiency (object allocation). Recommended only to be used if other
121 options are insufficient.
122 --gen-compare
124 Generate operator== for object-based API types.
125 --gen-onefile
127 Generate single output file (useful for C#)
128 --gen-all
130 Generate not just code for the current schema files, but for all
131 files it includes as well. If the language uses a single file for out‐
132 put (by default the case for C++ and JS), all code will end up in this
133 one file.
134 --no-js-exports
136 Removes Node.js style export lines (useful for JS)
137 --goog-js-export
139 Uses goog.exportsSymbol and goog.exportsProperty instead of Node.js
140 style exporting. Needed for compatibility with the Google closure com‐
141 piler (useful for JS).
142 --es6-js-export
144 Generates ECMAScript v6 style export definitions instead of Node.js
145 style exporting. Useful when integrating flatbuffers with modern
146 Javascript projects.
147 --raw-binary
149 Allow binaries without a file_indentifier to be read. This may crash
150 flatc given a mismatched schema.
151 --proto
153 Expect input files to be .proto files (protocol buffers). Output the
154 corresponding .fbs file. Currently supports: package, message, enum,
155 nested declarations, import (use -I for paths), extend, oneof, group.
156 Does not support, but will skip without error: option, service, exten‐
157 sions, and most everything else.
158 --schema
160 Serialize schemas instead of JSON (use with -b). This will output a
161 binary version of the specified schema that itself corresponds to the
162 reflection/reflection.fbs schema. Loading this binary file is the basis
163 for reflection functionality.
164 --bfbs-comments
166 Add doc comments to the binary schema files.
167 --conform FILE
169 Specify a schema the following schemas should be an evolution of.
170 Gives errors if not. Useful to check if schema modifications don't
171 break schema evolution rules.
172 --include-prefix PATH
174 Prefix this path to any generated include statements.
175 --keep-prefix
177 Keep original prefix of schema include statement.
178 --reflect-types
180 Add minimal type reflection to code generation.
181 --reflect-names
183 Add minimal type/name reflection.
184 --root-type T
186 Select or override the default root_type.
187 --force-defaults
189 Emit default values in binary output from JSON.
190 --force-empty
192 When serializing from object API representation, force strings and
193 vectors to empty rather than null.
194 NOTE: short-form options for generators are deprecated, use the long
196 form whenever possible.
197
199 flatbuffers(7), Official documentation ⟨http://google.github.io/flat‐
200 buffers⟩
201Linux APRIL 2018 FLATC(1)