DOKK / manpages / debian 12 / odb / odb.1.en
ODB(1) General Commands Manual ODB(1)

odb - object-relational mapping (ORM) compiler for C++

odb [ options ] file [ file... ]

Given a set of C++ classes in a header file, odb generates C++ code that allows you to persist, query, and update objects of these classes in a relational database (RDBMS). The relational database that the generated code should target is specified with the required --database option (see below).

For an input file in the form name.hxx (other file extensions can be used instead of .hxx), in the single-database mode (the default), the generated C++ files by default have the following names: name-odb.hxx (header file), name-odb.ixx (inline file), and name-odb.cxx (source file). Additionally, if the --generate-schema option is specified and the sql schema format is requested (see --schema-format), the name.sql database schema file is generated. If the separate schema format is requested, the database creation code is generated into the separate name-schema.cxx file.

In the multi-database mode (see the --multi-database option below), the generated files corresponding to the common database have the same names as in the single-database mode. For other databases, the file names include the database name: name-odb-db.hxx, name-odb-db.ixx, name-odb-db.cxx, name-db.sql, and name-schema-db.cxx (where db is the database name).

Print usage information and exit.

Print version and exit.

Add dir to the beginning of the list of directories to be searched for included header files.

Define macro name with definition def. If definition is omitted, define name to be 1.

Cancel any previous definitions of macro name, either built-in or provided with the -D option.

Generate code for the db database. Valid values are mssql, mysql, oracle, pgsql, sqlite, and common (multi-database mode only).

Enable multi-database support and specify its type. Valid values for this option are static and dynamic.

In the multi-database mode, options that determine the kind (for example, --schema-format), names (for example, --odb-file-suffix), or content (for example, prologue and epilogue options) of the output files can be prefixed with the database name followed by a colon, for example, mysql:value. This restricts the value of such an option to only apply to generated files corresponding to this database.

When static multi-database support is used, specify the database that should be made the default. When dynamic multi-database support is used, common is always made the default database.

Generate query support code. Without this support you cannot use views and can only load objects via their ids.

Generate prepared query execution support code.

Omit un-prepared (once-off) query execution support code.

Generate session support code. With this option session support will be enabled by default for all the persistent classes except those for which it was explicitly disabled using the db session pragma.

Generate the database schema. The database schema contains SQL statements that create database tables necessary to store persistent classes defined in the file being compiled. Note that by applying this schema, all the existing information stored in such tables will be lost.

Depending on the database being used (--database option), the schema is generated either as a standalone SQL file or embedded into the generated C++ code. By default the SQL file is generated for the MySQL, PostgreSQL, Oracle, and Microsoft SQL Server databases and the schema is embedded into the C++ code for the SQLite database. Use the --schema-format option to alter the default schema format.

If database schema evolution support is enabled (that is, the object model version is specified), then this option also triggers the generation of database schema migration statements, again either as standalong SQL files or embedded into the generated C++ code. You can suppress the generation of schema migration statements by specifying the --suppress-migration option.

Generate only the database schema. Note that this option is only valid when generating schema as a standalone SQL file (see --schema-format for details).

Suppress the generation of database schema migration statements.

Suppress the generation of schema version table. If you specify this option then you are also expected to manually specify the database schema version and migration state at runtime using the odb::database::schema_version() function.

Specify the alternative schema version table name instead of the default schema_version. If you specify this option then you are also expected to manually specify the schema version table name at runtime using the odb::database::schema_version_table() function. The table name can be qualified.

Generate the database schema in the specified format. Pass sql as format to generate the database schema as a standalone SQL file or pass embedded to embed the schema into the generated C++ code. The separate value is similar to embedded except the schema creation code is generated into a separate C++ file (name-schema.cxx by default). This value is primarily useful if you want to place the schema creation functionality into a separate program or library. Repeat this option to generate the same database schema in multiple formats.

Omit DROP statements from the generated database schema.

Omit CREATE statements from the generated database schema.

Use name as the database schema name. Schema names are primarily used to distinguish between multiple embedded schemas in the schema catalog. They are not to be confused with database schemas (database namespaces) which are specified with the --schema option. If this option is not specified, the empty name, which is the default schema name, is used.

Use constraint checking mode m in foreign keys generated for object relationships. Valid values for this option are not_deferrable, immediate, and deferred (default). MySQL and SQL Server do not support deferrable foreign keys and for these databases such keys are generated commented out. Other foreign keys generated by the ODB compiler (such as the ones used to support containers and polymorphic hierarchies) are always generated as not deferrable.

Note also that if you use either not_deferrable or immediate mode, then the order in which you persist, update, and erase objects within a transaction becomes important.

Use ptr as the default pointer for persistent objects and views. Objects and views that do not have a pointer assigned with the db pointer pragma will use this pointer by default. The value of this option can be * which denotes the raw pointer and is the default, or qualified name of a smart pointer class template, for example, std::auto_ptr. In the latter case, the ODB compiler constructs the object or view pointer by adding a single template argument of the object or view type to the qualified name, for example std::auto_ptr<object>. The ODB runtime uses object and view pointers to return, and, in case of objects, pass and cache dynamically allocated instances of object and view types.

Except for the raw pointer and the standard smart pointers defined in the <memory> header file, you are expected to include the definition of the default pointer at the beginning of the generated header file. There are two common ways to achieve this: you can either include the necessary header in the file being compiled or you can use the --hxx-prologue option to add the necessary #include directive to the generated code.

Use type as the alternative session type instead of the default odb::session. This option can be used to specify a custom session implementation to be use by the persistent classes. Note that you will also need to include the definition of the custom session type into the generated header file. This is normally achieved with the --hxx-prologue* options.

Specify a profile that should be used during compilation. A profile is an options file. The ODB compiler first looks for a database-specific version with the name constructed by appending the -database.options suffix to name, where database is the database name as specified with the --database option. If this file is not found, then the ODB compiler looks for a database-independant version with the name constructed by appending just the .options suffix.

The profile options files are searched for in the same set of directories as C++ headers included with the #include <...> directive (built-in paths plus those specified with the -I options). The options file is first searched for in the directory itself and then in its odb/ subdirectory.

For the format of the options file refer to the --options-file option below. You can repeat this option to specify more than one profile.

Generate code for all the input files as well as for all the files that they include at once. The result is a single set of source/schema files that contain all the generated code. If more than one input file is specified together with this option, then the --input-name option must also be specified in order to provide the base name for the output files. In this case, the directory part of such a base name is used as the location of the combined file. This can be important for the #include directive resolution.

Specify a database schema (database namespace) that should be assigned to the persistent classes in the file being compiled. Database schemas are not to be confused with database schema names (schema catalog names) which are specified with the --schema-name option.

Insert symbol in places where DLL export/import control statements (__declspec(dllexport/dllimport)) are necessary. See also the --extern-symbol option below.

If symbol is defined, insert it in places where a template instantiation must be declared extern. This option is normally used together with --export-symbol when both multi-database support and queries are enabled.

Specify the C++ standard that should be used during compilation. Valid values are c++98 (default), c++11, and c++14.

Warn about hard-added data members.

Warn about hard-deleted data members and persistent classes.

Warn about both hard-added and hard-deleted data members and persistent classes.

Write the generated files to dir instead of the current directory.

Use name instead of the input file to derive the names of the generated files. If the --at-once option is specified, then the directory part of name is used as the location of the combined file. Refer to the --at-once option for details.

Read/write changelog from/to file instead of the default changelog file. The default changelog file name is derived from the input file name and it is placed into the same directory as the input file. Note that the --output-dir option does not affect the changelog file location. In other words, by default, the changelog file is treated as another input rather than output even though the ODB compiler may modify it. Use the --changelog-in and --changelog-out options to specify different input and output chaneglog files.

Read changelog from file instead of the default changelog file. If this option is specified, then you must also specify the output chanegelog file with --changelog-out.

Write changelog to file instead of the default changelog file. If this option is specified, then you must also specify the input chanegelog file with --changelog-in.

Use dir instead of the input file directory as the changelog file directory. This directory is also added to changelog files specified with the --changelog, --changelog-in, and --changelog-in options unless they are absolute paths.

Force re-initialization of the changelog even if one exists (all the existing change history will be lost). This option is primarily useful for automated testing.

Use suffix to construct the names of the generated C++ files. In the single-database mode the default value for this option is -odb. In the multi-database mode it is -odb for the files corresponding to the common database and -odb-db (where db is the database name) for other databases.

Use suffix to construct the name of the generated schema SQL file. In the single-database mode by default no suffix is used. In the multi-database mode the default value for this option is -db (where db is the database name).

Use suffix to construct the name of the generated schema C++ source file. In the single-database mode the default value for this option is -schema. In the multi-database mode it is -schema-db (where db is the database name). See the --schema-format option for details.

Use sfx to construct the name of the changelog file. In the single-database mode by default no suffix is used. In the multi-database mode the default value for this option is -db (where db is the database name).

Use suffix instead of the default .hxx to construct the name of the generated C++ header file.

Use suffix instead of the default .ixx to construct the name of the generated C++ inline file.

Use suffix instead of the default .cxx to construct the name of the generated C++ source file.

Use suffix instead of the default .sql to construct the name of the generated database schema file.

Use suffix instead of the default .xml to construct the name of the changelog file.

Insert text at the beginning of the generated C++ header file.

Insert text at the beginning of the generated C++ inline file.

Insert text at the beginning of the generated C++ source file.

Insert text at the beginning of the generated schema C++ source file.

Insert text at the beginning of the generated database schema file.

Insert text at the beginning of the generated database migration file.

Insert text after all the DROP and before any CREATE statements in the generated database schema file.

Insert text at the end of the generated C++ header file.

Insert text at the end of the generated C++ inline file.

Insert text at the end of the generated C++ source file.

Insert text at the end of the generated schema C++ source file.

Insert text at the end of the generated database schema file.

Insert text at the end of the generated database migration file.

Insert the content of file at the beginning of the generated C++ header file.

Insert the content of file at the beginning of the generated C++ inline file.

Insert the content of file at the beginning of the generated C++ source file.

Insert the content of file at the beginning of the generated schema C++ source file.

Insert the content of file at the beginning of the generated database schema file.

Insert the content of file f at the beginning of the generated database migration file.

Insert the content of file after all the DROP and before any CREATE statements in the generated database schema file.

Insert the content of file at the end of the generated C++ header file.

Insert the content of file at the end of the generated C++ inline file.

Insert the content of file at the end of the generated C++ source file.

Insert the content of file at the end of the generated schema C++ source file.

Insert the content of file at the end of the generated database schema file.

Insert the content of file f at the end of the generated database migration file.

Compile text before the input header file. This option allows you to add additional declarations, such as custom traits specializations, to the ODB compilation process.

Compile file contents before the input header file. Prologue files are compiled after all the prologue text fragments (--odb-prologue option).

Compile text after the input header file. This option allows you to add additional declarations, such as custom traits specializations, to the ODB compilation process.

Compile file contents after the input header file. Epilogue files are compiled after all the epilogue text fragments (--odb-epilogue option).

Add prefix to table names and, for databases that have global index and/or foreign key names, to those names as well. The prefix is added to both names that were specified with the db table and db index pragmas and those that were automatically derived from class and data member names. If you require a separator, such as an underscore, between the prefix and the name, then you should include it into the prefix value.

Use suffix instead of the default _i to construct index names. The suffix is only added to names that were automatically derived from data member names. If you require a separator, such as an underscore, between the name and the suffix, then you should include it into the suffix value.

Use suffix instead of the default _fk to construct foreign key names. If you require a separator, such as an underscore, between the name and the suffix, then you should include it into the suffix value.

Use suffix instead of the default _seq to construct sequence names. If you require a separator, such as an underscore, between the name and the suffix, then you should include it into the suffix value.

Convert all automatically-derived SQL names to upper or lower case. Valid values for this option are upper and lower.

Add regex to the list of regular expressions that is used to transform automatically-derived table names. See the SQL NAME TRANSFORMATIONS section below for details.

Add regex to the list of regular expressions that is used to transform automatically-derived column names. See the SQL NAME TRANSFORMATIONS section below for details.

Add regex to the list of regular expressions that is used to transform automatically-derived index names. See the SQL NAME TRANSFORMATIONS section below for details.

Add regex to the list of regular expressions that is used to transform automatically-derived foreign key names. See the SQL NAME TRANSFORMATIONS section below for details.

Add regex to the list of regular expressions that is used to transform automatically-derived sequence names. See the SQL NAME TRANSFORMATIONS section below for details.

Add regex to the list of regular expressions that is used to transform automatically-derived prepared statement names. See the SQL NAME TRANSFORMATIONS section below for details.

Add regex to the list of regular expressions that is used to transform all automatically-derived SQL names. See the SQL NAME TRANSFORMATIONS section below for details.

Trace the process of applying regular expressions specified with the SQL name --*-regex options. Use this option to find out why your regular expressions don't do what you expected them to do.

Add regex to the list of regular expressions used to transform data member names to function names when searching for a suitable accessor function. The argument to this option is a Perl-like regular expression in the form /pattern/replacement/. Any character can be used as a delimiter instead of / and the delimiter can be escaped inside pattern and replacement with a backslash (\). You can specify multiple regular expressions by repeating this option.

All the regular expressions are tried in the order specified and the first expression that produces a suitable accessor function is used. Each expression is tried twice: first with the actual member name and then with the member's public name which is obtained by removing the common member name decorations, such as leading and trailing underscores, the m_ prefix, etc. The ODB compiler also includes a number of built-in expressions for commonly used accessor names, such as get_foo, getFoo, getfoo, and just foo. The built-in expressions are tried last.

As an example, the following expression transforms data members with public names in the form foo to accessor names in the form GetFoo:

/(.+)/Get\u$1/

See also the REGEX AND SHELL QUOTING section below.

Trace the process of applying regular expressions specified with the --accessor-regex option. Use this option to find out why your regular expressions don't do what you expected them to do.

Add regex to the list of regular expressions used to transform data member names to function names when searching for a suitable modifier function. The argument to this option is a Perl-like regular expression in the form /pattern/replacement/. Any character can be used as a delimiter instead of / and the delimiter can be escaped inside pattern and replacement with a backslash (\). You can specify multiple regular expressions by repeating this option.

All the regular expressions are tried in the order specified and the first expression that produces a suitable modifier function is used. Each expression is tried twice: first with the actual member name and then with the member's public name which is obtained by removing the common member name decorations, such as leading and trailing underscores, the m_ prefix, etc. The ODB compiler also includes a number of built-in expressions for commonly used modifier names, such as set_foo, setFoo, setfoo, and just foo. The built-in expressions are tried last.

As an example, the following expression transforms data members with public names in the form foo to modifier names in the form SetFoo:

/(.+)/Set\u$1/

See also the REGEX AND SHELL QUOTING section below.

Trace the process of applying regular expressions specified with the --modifier-regex option. Use this option to find out why your regular expressions don't do what you expected them to do.

Use angle brackets (<>) instead of quotes ("") in the generated #include directives.

Add prefix to the generated #include directive paths.

Add regex to the list of regular expressions used to transform generated #include directive paths. The argument to this option is a Perl-like regular expression in the form /pattern/replacement/. Any character can be used as a delimiter instead of / and the delimiter can be escaped inside pattern and replacement with a backslash (\). You can specify multiple regular expressions by repeating this option. All the regular expressions are tried in the order specified and the first expression that matches is used.

As an example, the following expression transforms include paths in the form foo/bar-odb.h to paths in the form foo/generated/bar-odb.h:

%foo/(.+)-odb.h%foo/generated/$1-odb.h%

See also the REGEX AND SHELL QUOTING section below.

Trace the process of applying regular expressions specified with the --include-regex option. Use this option to find out why your regular expressions don't do what you expected them to do.

Add prefix to the generated header inclusion guards. The prefix is transformed to upper case and characters that are illegal in a preprocessor macro name are replaced with underscores.

Print the number of generated physical source lines of code (SLOC).

Check that the number of generated physical source lines of code (SLOC) does not exceed num.

Read additional options from file with each option appearing on a separate line optionally followed by space and an option value. Empty lines and lines starting with # are ignored. Option values can be enclosed in double (") or single (') quotes to preserve leading and trailing whitespaces as well as to specify empty values. If the value itself contains trailing or leading quotes, enclose it with an extra pair of quotes, for example '"x"'. Non-leading and non-trailing quotes are interpreted as being part of the option value.

The semantics of providing options in a file is equivalent to providing the same set of options in the same order on the command line at the point where the --options-file option is specified except that the shell escaping and quoting is not required. You can repeat this option to specify more than one options file.

Pass option to the underlying C++ compiler (g++). The option value that doesn't start with - is considered the g++ executable name.

Print the commands executed to run the stages of compilation.

Trace the compilation process.

Use engine instead of the default InnoDB in the generated database schema file. For more information on the storage engine options see the MySQL documentation. If you would like to use the database-default engine, pass default as the value for this option.

Make all columns in the generated database schema allow NULL values. This is primarily useful in schema migration since SQLite does not support dropping of columns. By making all columns NULL we can later "delete" them by setting their values to NULL. Note that this option overrides even the not_null pragma.

Do not force monotonically increasing automatically-assigned object ids. In this mode the generated database schema omits the AUTOINCREMENT keyword which results in faster object persistence but may lead to automatically-assigned ids not being in a strictly ascending order. Refer to the SQLite documentation for details.

Specify the minimum PostgreSQL server version with which the generated C++ code and schema will be used. This information is used to enable version-specific optimizations and workarounds in the generated C++ code and schema. The version must be in the major.minor form, for example, 9.1. If this option is not specified, then 7.4 or later is assumed.

Specify the minimum Oracle client library (OCI) version with which the generated C++ code will be linked. This information is used to enable version-specific optimizations and workarounds in the generated C++ code. The version must be in the major.minor form, for example, 11.2. If this option is not specified, then 10.1 or later is assumed.

Warn about SQL names that are longer than 30 characters and are therefore truncated. Note that during database schema generation (--generate-schema) ODB detects when such truncations lead to name conflicts and issues diagnostics even without this option specified.

Specify the minimum SQL Server server version with which the generated C++ code and schema will be used. This information is used to enable version-specific optimizations and workarounds in the generated C++ code and schema. The version must be in the major.minor form, for example, 9.0 (SQL Server 2005), 10.5 (2008R2), or 11.0 (2012). If this option is not specified, then 10.0 (SQL Server 2008) or later is assumed.

Specify the short data size limit. If a character, national character, or binary data type has a maximum length (in bytes) less than or equal to this limit, then it is treated as short data, otherwise it is long data. For short data ODB pre-allocates an intermediate buffer of the maximum size and binds it directly to a parameter or result column. This way the underlying API (ODBC) can read/write directly from/to this buffer. In the case of long data, the data is read/written in chunks using the SQLGetData()/SQLPutData() ODBC functions. While the long data approach reduces the amount of memory used by the application, it may require greater CPU resources. The default short data limit is 1024 bytes. When setting a custom short data limit, make sure that it is sufficiently large so that no object id in the application is treated as long data.

The ODB compiler provides a number of mechanisms for transforming automatically-derived SQL names, such as tables, columns, etc., to match a specific naming convention. At the higher level, we can add a prefix to global names (tables and, for some databases, indexes and/or foreign keys) with the --table-prefix option. Similarly, we can specify custom suffixes for automatically-derived index (--index-suffix; default is _i), foreign key (--fkey-suffix; default is _fk), and sequence (--sequence-suffix; default is _seq) names. Finally, we can also convert all the names to upper or lower case with the --sql-name-case option (valid values are upper and lower).

At the lower level we can specify a set of regular expressions to implement arbitrary transformations of the automatically-derived SQL names. If we want a particular regular expression only to apply to a specific name, for example, table or column, then we use one of the --kind-regex options, where kind can be table, column, index, fkey, sequence, or statement. On the other hand, if we want our regular expressions to apply to all SQL names, then we use the --sql-name-regex option.

The interaction between the higher and lower level transformations is as follows. Prefixes and suffixes are added first. Then the regular expression transformations are applied. Finally, if requested, the name is converted to upper or lower case. Note also that all of these transformations except for --table-prefix only apply to automatically-derived names. In other words, if a table, column, etc., name was explicitly specified with a pragma, then it is used as is, without applying any (except for the table prefix) transformations.

The value for the --*-regex options is a Perl-like regular expression in the form /pattern/replacement/. Any character can be used as a delimiter instead of / and the delimiter can be escaped inside pattern and replacement with a backslash (\). You can also specify multiple regular expressions by repeating these options.

All the regular expressions are tried in the order specified with the name-specific expressions (for example, --table-regex) tried first followed by the generic expressions (--sql-name-regex). The first expression that matches is used.

As an example, consider a regular expression that transforms a class name in the form CFoo to a table name in the form FOO:

--table-regex '/C(.+)/\U$1/'

As a more interesting example, consider the transformation of class names that follow the upper camel case convention (for example, FooBar) to table names that follow the underscore-separated, all upper case convention (for example, FOO_BAR). For this case we have to use separate expressions to handle one-word, two-word, etc., names:

--table-regex '/([A-z][a-z]+)/\U$1/'

--table-regex '/([A-z][a-z]+)([A-z][a-z]+)/\U$1_$2/'

See also the REGEX AND SHELL QUOTING section below.

When entering a regular expression argument in the shell command line it is often necessary to use quoting (enclosing the argument in " " or ' ') in order to prevent the shell from interpreting certain characters, for example, spaces as argument separators and $ as variable expansions.

Unfortunately it is hard to achieve this in a manner that is portable across POSIX shells, such as those found on GNU/Linux and UNIX, and Windows shell. For example, if you use " " for quoting you will get a wrong result with POSIX shells if your expression contains $. The standard way of dealing with this on POSIX systems is to use ' ' instead. Unfortunately, Windows shell does not remove ' ' from arguments when they are passed to applications. As a result you may have to use ' ' for POSIX and " " for Windows ($ is not treated as a special character on Windows).

Alternatively, you can save regular expression options into a file, one option per line, and use this file with the --options-file option. With this approach you don't need to worry about shell quoting.

If the input file is not valid C++, odb will issue diagnostic messages to STDERR and exit with non-zero exit code.

Send bug reports to the odb-users@codesynthesis.com mailing list.

Copyright (c) 2009-2015 Code Synthesis Tools CC.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, version 1.2; with no Invariant Sections, no Front-Cover Texts and no Back-Cover Texts. Copy of the license can be obtained from http://www.codesynthesis.com/licenses/fdl-1.3.txt

February 2015 ODB 2.4.0