2011-07-20 03:50:46 +08:00
|
|
|
= How to use the QAPI code generator =
|
|
|
|
|
|
|
|
QAPI is a native C API within QEMU which provides management-level
|
|
|
|
functionality to internal/external users. For external
|
|
|
|
users/processes, this interface is made available by a JSON-based
|
|
|
|
QEMU Monitor protocol that is provided by the QMP server.
|
|
|
|
|
|
|
|
To map QMP-defined interfaces to the native C QAPI implementations,
|
|
|
|
a JSON-based schema is used to define types and function
|
|
|
|
signatures, and a set of scripts is used to generate types/signatures,
|
|
|
|
and marshaling/dispatch code. The QEMU Guest Agent also uses these
|
2011-11-14 05:24:27 +08:00
|
|
|
scripts, paired with a separate schema, to generate
|
2011-07-20 03:50:46 +08:00
|
|
|
marshaling/dispatch code for the guest agent server running in the
|
|
|
|
guest.
|
|
|
|
|
|
|
|
This document will describe how the schemas, scripts, and resulting
|
2014-09-26 23:20:33 +08:00
|
|
|
code are used.
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
|
|
|
|
== QMP/Guest agent schema ==
|
|
|
|
|
|
|
|
This file defines the types, commands, and events used by QMP. It should
|
|
|
|
fully describe the interface used by QMP.
|
|
|
|
|
|
|
|
This file is designed to be loosely based on JSON although it's technically
|
|
|
|
executable Python. While dictionaries are used, they are parsed as
|
|
|
|
OrderedDicts so that ordering is preserved.
|
|
|
|
|
|
|
|
There are two basic syntaxes used, type definitions and command definitions.
|
|
|
|
|
|
|
|
The first syntax defines a type and is represented by a dictionary. There are
|
2013-07-16 19:17:27 +08:00
|
|
|
three kinds of user-defined types that are supported: complex types,
|
|
|
|
enumeration types and union types.
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2013-07-16 19:17:27 +08:00
|
|
|
Generally speaking, types definitions should always use CamelCase for the type
|
|
|
|
names. Command names should be all lower case with words separated by a hyphen.
|
|
|
|
|
2014-05-08 02:46:15 +08:00
|
|
|
|
|
|
|
=== Includes ===
|
|
|
|
|
|
|
|
The QAPI schema definitions can be modularized using the 'include' directive:
|
|
|
|
|
|
|
|
{ 'include': 'path/to/file.json'}
|
|
|
|
|
|
|
|
The directive is evaluated recursively, and include paths are relative to the
|
2014-05-16 18:51:56 +08:00
|
|
|
file using the directive. Multiple includes of the same file are safe.
|
2014-05-08 02:46:15 +08:00
|
|
|
|
|
|
|
|
2013-07-16 19:17:27 +08:00
|
|
|
=== Complex types ===
|
|
|
|
|
|
|
|
A complex type is a dictionary containing a single key whose value is a
|
2011-07-20 03:50:46 +08:00
|
|
|
dictionary. This corresponds to a struct in C or an Object in JSON. An
|
|
|
|
example of a complex type is:
|
|
|
|
|
|
|
|
{ 'type': 'MyType',
|
2011-10-28 22:58:26 +08:00
|
|
|
'data': { 'member1': 'str', 'member2': 'int', '*member3': 'str' } }
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-05-07 09:57:41 +08:00
|
|
|
The use of '*' as a prefix to the name means the member is optional.
|
|
|
|
|
|
|
|
The default initialization value of an optional argument should not be changed
|
|
|
|
between versions of QEMU unless the new default maintains backward
|
|
|
|
compatibility to the user-visible behavior of the old default.
|
|
|
|
|
|
|
|
With proper documentation, this policy still allows some flexibility; for
|
|
|
|
example, documenting that a default of 0 picks an optimal buffer size allows
|
|
|
|
one release to declare the optimal size at 512 while another release declares
|
|
|
|
the optimal size at 4096 - the user-visible behavior is not the bytes used by
|
|
|
|
the buffer, but the fact that the buffer was optimal size.
|
|
|
|
|
|
|
|
On input structures (only mentioned in the 'data' side of a command), changing
|
|
|
|
from mandatory to optional is safe (older clients will supply the option, and
|
|
|
|
newer clients can benefit from the default); changing from optional to
|
|
|
|
mandatory is backwards incompatible (older clients may be omitting the option,
|
|
|
|
and must continue to work).
|
|
|
|
|
|
|
|
On output structures (only mentioned in the 'returns' side of a command),
|
|
|
|
changing from mandatory to optional is in general unsafe (older clients may be
|
|
|
|
expecting the field, and could crash if it is missing), although it can be done
|
|
|
|
if the only way that the optional argument will be omitted is when it is
|
|
|
|
triggered by the presence of a new input flag to the command that older clients
|
|
|
|
don't know to send. Changing from optional to mandatory is safe.
|
|
|
|
|
|
|
|
A structure that is used in both input and output of various commands
|
|
|
|
must consider the backwards compatibility constraints of both directions
|
|
|
|
of use.
|
2013-09-19 17:56:36 +08:00
|
|
|
|
|
|
|
A complex type definition can specify another complex type as its base.
|
|
|
|
In this case, the fields of the base type are included as top-level fields
|
|
|
|
of the new complex type's dictionary in the QMP wire format. An example
|
|
|
|
definition is:
|
|
|
|
|
|
|
|
{ 'type': 'BlockdevOptionsGenericFormat', 'data': { 'file': 'str' } }
|
|
|
|
{ 'type': 'BlockdevOptionsGenericCOWFormat',
|
|
|
|
'base': 'BlockdevOptionsGenericFormat',
|
|
|
|
'data': { '*backing': 'str' } }
|
|
|
|
|
|
|
|
An example BlockdevOptionsGenericCOWFormat object on the wire could use
|
|
|
|
both fields like this:
|
|
|
|
|
|
|
|
{ "file": "/some/place/my-image",
|
|
|
|
"backing": "/some/place/my-backing-file" }
|
|
|
|
|
2013-07-16 19:17:27 +08:00
|
|
|
=== Enumeration types ===
|
|
|
|
|
|
|
|
An enumeration type is a dictionary containing a single key whose value is a
|
2011-07-20 03:50:46 +08:00
|
|
|
list of strings. An example enumeration is:
|
|
|
|
|
|
|
|
{ 'enum': 'MyEnum', 'data': [ 'value1', 'value2', 'value3' ] }
|
|
|
|
|
2013-07-16 19:17:27 +08:00
|
|
|
=== Union types ===
|
|
|
|
|
|
|
|
Union types are used to let the user choose between several different data
|
|
|
|
types. A union type is defined using a dictionary as explained in the
|
|
|
|
following paragraphs.
|
|
|
|
|
|
|
|
|
|
|
|
A simple union type defines a mapping from discriminator values to data types
|
|
|
|
like in this example:
|
|
|
|
|
|
|
|
{ 'type': 'FileOptions', 'data': { 'filename': 'str' } }
|
|
|
|
{ 'type': 'Qcow2Options',
|
|
|
|
'data': { 'backing-file': 'str', 'lazy-refcounts': 'bool' } }
|
|
|
|
|
|
|
|
{ 'union': 'BlockdevOptions',
|
|
|
|
'data': { 'file': 'FileOptions',
|
|
|
|
'qcow2': 'Qcow2Options' } }
|
|
|
|
|
|
|
|
In the QMP wire format, a simple union is represented by a dictionary that
|
|
|
|
contains the 'type' field as a discriminator, and a 'data' field that is of the
|
|
|
|
specified data type corresponding to the discriminator value:
|
|
|
|
|
|
|
|
{ "type": "qcow2", "data" : { "backing-file": "/some/place/my-image",
|
|
|
|
"lazy-refcounts": true } }
|
|
|
|
|
|
|
|
|
|
|
|
A union definition can specify a complex type as its base. In this case, the
|
|
|
|
fields of the complex type are included as top-level fields of the union
|
|
|
|
dictionary in the QMP wire format. An example definition is:
|
|
|
|
|
|
|
|
{ 'type': 'BlockdevCommonOptions', 'data': { 'readonly': 'bool' } }
|
|
|
|
{ 'union': 'BlockdevOptions',
|
|
|
|
'base': 'BlockdevCommonOptions',
|
|
|
|
'data': { 'raw': 'RawOptions',
|
|
|
|
'qcow2': 'Qcow2Options' } }
|
|
|
|
|
|
|
|
And it looks like this on the wire:
|
|
|
|
|
|
|
|
{ "type": "qcow2",
|
|
|
|
"readonly": false,
|
|
|
|
"data" : { "backing-file": "/some/place/my-image",
|
|
|
|
"lazy-refcounts": true } }
|
|
|
|
|
2013-07-03 21:58:57 +08:00
|
|
|
|
|
|
|
Flat union types avoid the nesting on the wire. They are used whenever a
|
|
|
|
specific field of the base type is declared as the discriminator ('type' is
|
2014-03-05 10:44:39 +08:00
|
|
|
then no longer generated). The discriminator must be of enumeration type.
|
2013-07-03 21:58:57 +08:00
|
|
|
The above example can then be modified as follows:
|
|
|
|
|
2014-03-07 09:08:56 +08:00
|
|
|
{ 'enum': 'BlockdevDriver', 'data': [ 'raw', 'qcow2' ] }
|
2013-07-03 21:58:57 +08:00
|
|
|
{ 'type': 'BlockdevCommonOptions',
|
2014-03-07 09:08:56 +08:00
|
|
|
'data': { 'driver': 'BlockdevDriver', 'readonly': 'bool' } }
|
2013-07-03 21:58:57 +08:00
|
|
|
{ 'union': 'BlockdevOptions',
|
|
|
|
'base': 'BlockdevCommonOptions',
|
|
|
|
'discriminator': 'driver',
|
|
|
|
'data': { 'raw': 'RawOptions',
|
|
|
|
'qcow2': 'Qcow2Options' } }
|
|
|
|
|
|
|
|
Resulting in this JSON object:
|
|
|
|
|
|
|
|
{ "driver": "qcow2",
|
|
|
|
"readonly": false,
|
|
|
|
"backing-file": "/some/place/my-image",
|
|
|
|
"lazy-refcounts": true }
|
|
|
|
|
|
|
|
|
2013-07-08 22:14:21 +08:00
|
|
|
A special type of unions are anonymous unions. They don't form a dictionary in
|
|
|
|
the wire format but allow the direct use of different types in their place. As
|
|
|
|
they aren't structured, they don't have any explicit discriminator but use
|
|
|
|
the (QObject) data type of their value as an implicit discriminator. This means
|
|
|
|
that they are restricted to using only one discriminator value per QObject
|
|
|
|
type. For example, you cannot have two different complex types in an anonymous
|
|
|
|
union, or two different integer types.
|
|
|
|
|
|
|
|
Anonymous unions are declared using an empty dictionary as their discriminator.
|
|
|
|
The discriminator values never appear on the wire, they are only used in the
|
|
|
|
generated C code. Anonymous unions cannot have a base type.
|
|
|
|
|
|
|
|
{ 'union': 'BlockRef',
|
|
|
|
'discriminator': {},
|
|
|
|
'data': { 'definition': 'BlockdevOptions',
|
|
|
|
'reference': 'str' } }
|
|
|
|
|
|
|
|
This example allows using both of the following example objects:
|
|
|
|
|
|
|
|
{ "file": "my_existing_block_device_id" }
|
|
|
|
{ "file": { "driver": "file",
|
|
|
|
"readonly": false,
|
2013-10-20 00:52:33 +08:00
|
|
|
"filename": "/tmp/mydisk.qcow2" } }
|
2013-07-08 22:14:21 +08:00
|
|
|
|
|
|
|
|
2013-07-16 19:17:27 +08:00
|
|
|
=== Commands ===
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
Commands are defined by using a list containing three members. The first
|
|
|
|
member is the command name, the second member is a dictionary containing
|
|
|
|
arguments, and the third member is the return type.
|
|
|
|
|
|
|
|
An example command is:
|
|
|
|
|
|
|
|
{ 'command': 'my-command',
|
|
|
|
'data': { 'arg1': 'str', '*arg2': 'str' },
|
2011-10-28 22:58:26 +08:00
|
|
|
'returns': 'str' }
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-06-18 14:43:28 +08:00
|
|
|
=== Events ===
|
|
|
|
|
|
|
|
Events are defined with the keyword 'event'. When 'data' is also specified,
|
2014-06-25 07:33:59 +08:00
|
|
|
additional info will be included in the event. Finally there will be C API
|
|
|
|
generated in qapi-event.h; when called by QEMU code, a message with timestamp
|
|
|
|
will be emitted on the wire. If timestamp is -1, it means failure to retrieve
|
|
|
|
host time.
|
2014-06-18 14:43:28 +08:00
|
|
|
|
|
|
|
An example event is:
|
|
|
|
|
|
|
|
{ 'event': 'EVENT_C',
|
|
|
|
'data': { '*a': 'int', 'b': 'str' } }
|
|
|
|
|
|
|
|
Resulting in this JSON object:
|
|
|
|
|
|
|
|
{ "event": "EVENT_C",
|
|
|
|
"data": { "b": "test string" },
|
|
|
|
"timestamp": { "seconds": 1267020223, "microseconds": 435656 } }
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
|
2011-07-20 03:50:46 +08:00
|
|
|
== Code generation ==
|
|
|
|
|
|
|
|
Schemas are fed into 3 scripts to generate all the code/files that, paired
|
|
|
|
with the core QAPI libraries, comprise everything required to take JSON
|
|
|
|
commands read in by a QMP/guest agent server, unmarshal the arguments into
|
|
|
|
the underlying C types, call into the corresponding C function, and map the
|
|
|
|
response back to a QMP/guest agent response to be returned to the user.
|
|
|
|
|
|
|
|
As an example, we'll use the following schema, which describes a single
|
|
|
|
complex user-defined type (which will produce a C struct, along with a list
|
|
|
|
node structure that can be used to chain together a list of such types in
|
|
|
|
case we want to accept/return a list of this type with a command), and a
|
|
|
|
command which takes that type as a parameter and returns the same type:
|
|
|
|
|
2014-05-14 23:27:23 +08:00
|
|
|
$ cat example-schema.json
|
2011-07-20 03:50:46 +08:00
|
|
|
{ 'type': 'UserDefOne',
|
|
|
|
'data': { 'integer': 'int', 'string': 'str' } }
|
|
|
|
|
|
|
|
{ 'command': 'my-command',
|
|
|
|
'data': {'arg1': 'UserDefOne'},
|
|
|
|
'returns': 'UserDefOne' }
|
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
{ 'event': 'MY_EVENT' }
|
|
|
|
|
2011-07-20 03:50:46 +08:00
|
|
|
=== scripts/qapi-types.py ===
|
|
|
|
|
|
|
|
Used to generate the C types defined by a schema. The following files are
|
|
|
|
created:
|
|
|
|
|
|
|
|
$(prefix)qapi-types.h - C types corresponding to types defined in
|
|
|
|
the schema you pass in
|
|
|
|
$(prefix)qapi-types.c - Cleanup functions for the above C types
|
|
|
|
|
|
|
|
The $(prefix) is an optional parameter used as a namespace to keep the
|
|
|
|
generated code from one schema/code-generation separated from others so code
|
|
|
|
can be generated/used from multiple schemas without clobbering previously
|
|
|
|
created code.
|
|
|
|
|
|
|
|
Example:
|
|
|
|
|
2014-05-14 23:27:23 +08:00
|
|
|
$ python scripts/qapi-types.py --output-dir="qapi-generated" \
|
|
|
|
--prefix="example-" --input-file=example-schema.json
|
|
|
|
$ cat qapi-generated/example-qapi-types.c
|
2014-05-07 15:53:43 +08:00
|
|
|
[Uninteresting stuff omitted...]
|
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
void qapi_free_UserDefOneList(UserDefOneList *obj)
|
2014-05-07 15:53:43 +08:00
|
|
|
{
|
|
|
|
QapiDeallocVisitor *md;
|
|
|
|
Visitor *v;
|
|
|
|
|
|
|
|
if (!obj) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
md = qapi_dealloc_visitor_new();
|
|
|
|
v = qapi_dealloc_get_visitor(md);
|
|
|
|
visit_type_UserDefOneList(v, &obj, NULL, NULL);
|
|
|
|
qapi_dealloc_visitor_cleanup(md);
|
|
|
|
}
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
void qapi_free_UserDefOne(UserDefOne *obj)
|
2011-07-20 03:50:46 +08:00
|
|
|
{
|
|
|
|
QapiDeallocVisitor *md;
|
|
|
|
Visitor *v;
|
|
|
|
|
|
|
|
if (!obj) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
md = qapi_dealloc_visitor_new();
|
|
|
|
v = qapi_dealloc_get_visitor(md);
|
|
|
|
visit_type_UserDefOne(v, &obj, NULL, NULL);
|
|
|
|
qapi_dealloc_visitor_cleanup(md);
|
|
|
|
}
|
|
|
|
|
2014-05-14 23:27:23 +08:00
|
|
|
$ cat qapi-generated/example-qapi-types.h
|
2014-05-07 15:53:43 +08:00
|
|
|
[Uninteresting stuff omitted...]
|
|
|
|
|
|
|
|
#ifndef EXAMPLE_QAPI_TYPES_H
|
|
|
|
#define EXAMPLE_QAPI_TYPES_H
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-05-07 15:53:43 +08:00
|
|
|
[Builtin types omitted...]
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
typedef struct UserDefOne UserDefOne;
|
|
|
|
|
|
|
|
typedef struct UserDefOneList
|
|
|
|
{
|
2014-05-07 15:53:43 +08:00
|
|
|
union {
|
|
|
|
UserDefOne *value;
|
|
|
|
uint64_t padding;
|
|
|
|
};
|
2011-07-20 03:50:46 +08:00
|
|
|
struct UserDefOneList *next;
|
|
|
|
} UserDefOneList;
|
|
|
|
|
2014-05-07 15:53:43 +08:00
|
|
|
[Functions on builtin types omitted...]
|
|
|
|
|
2011-07-20 03:50:46 +08:00
|
|
|
struct UserDefOne
|
|
|
|
{
|
|
|
|
int64_t integer;
|
2014-09-26 23:20:33 +08:00
|
|
|
char *string;
|
2011-07-20 03:50:46 +08:00
|
|
|
};
|
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
void qapi_free_UserDefOneList(UserDefOneList *obj);
|
|
|
|
void qapi_free_UserDefOne(UserDefOne *obj);
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
=== scripts/qapi-visit.py ===
|
|
|
|
|
|
|
|
Used to generate the visitor functions used to walk through and convert
|
|
|
|
a QObject (as provided by QMP) to a native C data structure and
|
|
|
|
vice-versa, as well as the visitor function used to dealloc a complex
|
|
|
|
schema-defined C type.
|
|
|
|
|
|
|
|
The following files are generated:
|
|
|
|
|
|
|
|
$(prefix)qapi-visit.c: visitor function for a particular C type, used
|
|
|
|
to automagically convert QObjects into the
|
|
|
|
corresponding C type and vice-versa, as well
|
|
|
|
as for deallocating memory for an existing C
|
|
|
|
type
|
|
|
|
|
|
|
|
$(prefix)qapi-visit.h: declarations for previously mentioned visitor
|
|
|
|
functions
|
|
|
|
|
|
|
|
Example:
|
|
|
|
|
2014-05-14 23:27:23 +08:00
|
|
|
$ python scripts/qapi-visit.py --output-dir="qapi-generated"
|
|
|
|
--prefix="example-" --input-file=example-schema.json
|
|
|
|
$ cat qapi-generated/example-qapi-visit.c
|
2014-05-07 15:53:43 +08:00
|
|
|
[Uninteresting stuff omitted...]
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
static void visit_type_UserDefOne_fields(Visitor *m, UserDefOne **obj, Error **errp)
|
2014-05-07 15:53:43 +08:00
|
|
|
{
|
|
|
|
Error *err = NULL;
|
|
|
|
visit_type_int(m, &(*obj)->integer, "integer", &err);
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
if (err) {
|
|
|
|
goto out;
|
|
|
|
}
|
2014-05-07 15:53:43 +08:00
|
|
|
visit_type_str(m, &(*obj)->string, "string", &err);
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
if (err) {
|
|
|
|
goto out;
|
|
|
|
}
|
2014-05-07 15:53:43 +08:00
|
|
|
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
out:
|
2014-05-07 15:53:43 +08:00
|
|
|
error_propagate(errp, err);
|
|
|
|
}
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
void visit_type_UserDefOne(Visitor *m, UserDefOne **obj, const char *name, Error **errp)
|
2011-07-20 03:50:46 +08:00
|
|
|
{
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
Error *err = NULL;
|
|
|
|
|
|
|
|
visit_start_struct(m, (void **)obj, "UserDefOne", name, sizeof(UserDefOne), &err);
|
|
|
|
if (!err) {
|
|
|
|
if (*obj) {
|
|
|
|
visit_type_UserDefOne_fields(m, obj, errp);
|
2014-05-07 15:53:43 +08:00
|
|
|
}
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
visit_end_struct(m, &err);
|
2014-05-07 15:53:43 +08:00
|
|
|
}
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
error_propagate(errp, err);
|
2011-07-20 03:50:46 +08:00
|
|
|
}
|
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
void visit_type_UserDefOneList(Visitor *m, UserDefOneList **obj, const char *name, Error **errp)
|
2011-07-20 03:50:46 +08:00
|
|
|
{
|
2014-05-07 15:53:43 +08:00
|
|
|
Error *err = NULL;
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
GenericList *i, **prev;
|
2014-05-07 15:53:43 +08:00
|
|
|
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
visit_start_list(m, name, &err);
|
|
|
|
if (err) {
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (prev = (GenericList **)obj;
|
|
|
|
!err && (i = visit_next_list(m, prev, &err)) != NULL;
|
|
|
|
prev = &i) {
|
|
|
|
UserDefOneList *native_i = (UserDefOneList *)i;
|
|
|
|
visit_type_UserDefOne(m, &native_i->value, NULL, &err);
|
2011-07-20 03:50:46 +08:00
|
|
|
}
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
|
|
|
|
error_propagate(errp, err);
|
|
|
|
err = NULL;
|
|
|
|
visit_end_list(m, &err);
|
|
|
|
out:
|
|
|
|
error_propagate(errp, err);
|
2011-07-20 03:50:46 +08:00
|
|
|
}
|
2014-05-14 23:27:23 +08:00
|
|
|
$ python scripts/qapi-commands.py --output-dir="qapi-generated" \
|
|
|
|
--prefix="example-" --input-file=example-schema.json
|
|
|
|
$ cat qapi-generated/example-qapi-visit.h
|
2014-05-07 15:53:43 +08:00
|
|
|
[Uninteresting stuff omitted...]
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-05-07 15:53:43 +08:00
|
|
|
#ifndef EXAMPLE_QAPI_VISIT_H
|
|
|
|
#define EXAMPLE_QAPI_VISIT_H
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-05-07 15:53:43 +08:00
|
|
|
[Visitors for builtin types omitted...]
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
void visit_type_UserDefOne(Visitor *m, UserDefOne **obj, const char *name, Error **errp);
|
|
|
|
void visit_type_UserDefOneList(Visitor *m, UserDefOneList **obj, const char *name, Error **errp);
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
=== scripts/qapi-commands.py ===
|
|
|
|
|
|
|
|
Used to generate the marshaling/dispatch functions for the commands defined
|
|
|
|
in the schema. The following files are generated:
|
|
|
|
|
|
|
|
$(prefix)qmp-marshal.c: command marshal/dispatch functions for each
|
|
|
|
QMP command defined in the schema. Functions
|
|
|
|
generated by qapi-visit.py are used to
|
2011-08-29 03:45:40 +08:00
|
|
|
convert QObjects received from the wire into
|
2011-07-20 03:50:46 +08:00
|
|
|
function parameters, and uses the same
|
|
|
|
visitor functions to convert native C return
|
|
|
|
values to QObjects from transmission back
|
|
|
|
over the wire.
|
|
|
|
|
|
|
|
$(prefix)qmp-commands.h: Function prototypes for the QMP commands
|
|
|
|
specified in the schema.
|
|
|
|
|
|
|
|
Example:
|
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
$ python scripts/qapi-commands.py --output-dir="qapi-generated"
|
|
|
|
--prefix="example-" --input-file=example-schema.json
|
2014-05-14 23:27:23 +08:00
|
|
|
$ cat qapi-generated/example-qmp-marshal.c
|
2014-05-07 15:53:43 +08:00
|
|
|
[Uninteresting stuff omitted...]
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
static void qmp_marshal_output_my_command(UserDefOne *ret_in, QObject **ret_out, Error **errp)
|
2011-07-20 03:50:46 +08:00
|
|
|
{
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
Error *local_err = NULL;
|
2011-07-20 03:50:46 +08:00
|
|
|
QmpOutputVisitor *mo = qmp_output_visitor_new();
|
2014-05-07 15:53:44 +08:00
|
|
|
QapiDeallocVisitor *md;
|
2011-07-20 03:50:46 +08:00
|
|
|
Visitor *v;
|
|
|
|
|
|
|
|
v = qmp_output_get_visitor(mo);
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
visit_type_UserDefOne(v, &ret_in, "unused", &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
2014-05-07 15:53:43 +08:00
|
|
|
}
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
*ret_out = qmp_output_get_qobject(mo);
|
|
|
|
|
|
|
|
out:
|
|
|
|
error_propagate(errp, local_err);
|
2014-05-07 15:53:43 +08:00
|
|
|
qmp_output_visitor_cleanup(mo);
|
2014-05-07 15:53:44 +08:00
|
|
|
md = qapi_dealloc_visitor_new();
|
2011-07-20 03:50:46 +08:00
|
|
|
v = qapi_dealloc_get_visitor(md);
|
2014-05-07 15:53:43 +08:00
|
|
|
visit_type_UserDefOne(v, &ret_in, "unused", NULL);
|
2011-07-20 03:50:46 +08:00
|
|
|
qapi_dealloc_visitor_cleanup(md);
|
|
|
|
}
|
|
|
|
|
2014-05-07 15:53:43 +08:00
|
|
|
static void qmp_marshal_input_my_command(QDict *args, QObject **ret, Error **errp)
|
2011-07-20 03:50:46 +08:00
|
|
|
{
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
Error *local_err = NULL;
|
2014-09-26 23:20:33 +08:00
|
|
|
UserDefOne *retval = NULL;
|
2014-05-07 15:53:44 +08:00
|
|
|
QmpInputVisitor *mi = qmp_input_visitor_new_strict(QOBJECT(args));
|
2011-07-20 03:50:46 +08:00
|
|
|
QapiDeallocVisitor *md;
|
|
|
|
Visitor *v;
|
2014-09-26 23:20:33 +08:00
|
|
|
UserDefOne *arg1 = NULL;
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
v = qmp_input_get_visitor(mi);
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
visit_type_UserDefOne(v, &arg1, "arg1", &local_err);
|
|
|
|
if (local_err) {
|
2011-07-20 03:50:46 +08:00
|
|
|
goto out;
|
|
|
|
}
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
|
|
|
|
retval = qmp_my_command(arg1, &local_err);
|
|
|
|
if (local_err) {
|
|
|
|
goto out;
|
2014-05-07 15:53:43 +08:00
|
|
|
}
|
2011-07-20 03:50:46 +08:00
|
|
|
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
qmp_marshal_output_my_command(retval, ret, &local_err);
|
|
|
|
|
2011-07-20 03:50:46 +08:00
|
|
|
out:
|
qapi: Replace uncommon use of the error API by the common one
We commonly use the error API like this:
err = NULL;
foo(..., &err);
if (err) {
goto out;
}
bar(..., &err);
Every error source is checked separately. The second function is only
called when the first one succeeds. Both functions are free to pass
their argument to error_set(). Because error_set() asserts no error
has been set, this effectively means they must not be called with an
error set.
The qapi-generated code uses the error API differently:
// *errp was initialized to NULL somewhere up the call chain
frob(..., errp);
gnat(..., errp);
Errors accumulate in *errp: first error wins, subsequent errors get
dropped. To make this work, the second function does nothing when
called with an error set. Requires non-null errp, or else the second
function can't see the first one fail.
This usage has also bled into visitor tests, and two device model
object property getters rtc_get_date() and balloon_stats_get_all().
With the "accumulate" technique, you need fewer error checks in
callers, and buy that with an error check in every callee. Can be
nice.
However, mixing the two techniques is confusing. You can't use the
"accumulate" technique with functions designed for the "check
separately" technique. You can use the "check separately" technique
with functions designed for the "accumulate" technique, but then
error_set() can't catch you setting an error more than once.
Standardize on the "check separately" technique for now, because it's
overwhelmingly prevalent.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
2014-05-07 15:53:54 +08:00
|
|
|
error_propagate(errp, local_err);
|
2014-05-07 15:53:44 +08:00
|
|
|
qmp_input_visitor_cleanup(mi);
|
2011-07-20 03:50:46 +08:00
|
|
|
md = qapi_dealloc_visitor_new();
|
|
|
|
v = qapi_dealloc_get_visitor(md);
|
2014-05-07 15:53:43 +08:00
|
|
|
visit_type_UserDefOne(v, &arg1, "arg1", NULL);
|
2011-07-20 03:50:46 +08:00
|
|
|
qapi_dealloc_visitor_cleanup(md);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void qmp_init_marshal(void)
|
|
|
|
{
|
2014-05-07 15:53:43 +08:00
|
|
|
qmp_register_command("my-command", qmp_marshal_input_my_command, QCO_NO_OPTIONS);
|
2011-07-20 03:50:46 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
qapi_init(qmp_init_marshal);
|
2014-05-14 23:27:23 +08:00
|
|
|
$ cat qapi-generated/example-qmp-commands.h
|
2014-05-07 15:53:43 +08:00
|
|
|
[Uninteresting stuff omitted...]
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-05-07 15:53:43 +08:00
|
|
|
#ifndef EXAMPLE_QMP_COMMANDS_H
|
|
|
|
#define EXAMPLE_QMP_COMMANDS_H
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
#include "example-qapi-types.h"
|
2014-05-07 15:53:43 +08:00
|
|
|
#include "qapi/qmp/qdict.h"
|
|
|
|
#include "qapi/error.h"
|
2011-07-20 03:50:46 +08:00
|
|
|
|
2014-09-26 23:20:33 +08:00
|
|
|
UserDefOne *qmp_my_command(UserDefOne *arg1, Error **errp);
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
=== scripts/qapi-event.py ===
|
|
|
|
|
|
|
|
Used to generate the event-related C code defined by a schema. The
|
|
|
|
following files are created:
|
|
|
|
|
|
|
|
$(prefix)qapi-event.h - Function prototypes for each event type, plus an
|
|
|
|
enumeration of all event names
|
|
|
|
$(prefix)qapi-event.c - Implementation of functions to send an event
|
|
|
|
|
|
|
|
Example:
|
|
|
|
|
|
|
|
$ python scripts/qapi-event.py --output-dir="qapi-generated"
|
|
|
|
--prefix="example-" --input-file=example-schema.json
|
|
|
|
$ cat qapi-generated/example-qapi-event.c
|
|
|
|
[Uninteresting stuff omitted...]
|
|
|
|
|
|
|
|
void qapi_event_send_my_event(Error **errp)
|
|
|
|
{
|
|
|
|
QDict *qmp;
|
|
|
|
Error *local_err = NULL;
|
|
|
|
QMPEventFuncEmit emit;
|
|
|
|
emit = qmp_event_get_func_emit();
|
|
|
|
if (!emit) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
qmp = qmp_event_build_dict("MY_EVENT");
|
|
|
|
|
|
|
|
emit(EXAMPLE_QAPI_EVENT_MY_EVENT, qmp, &local_err);
|
|
|
|
|
|
|
|
error_propagate(errp, local_err);
|
|
|
|
QDECREF(qmp);
|
|
|
|
}
|
|
|
|
|
|
|
|
const char *EXAMPLE_QAPIEvent_lookup[] = {
|
|
|
|
"MY_EVENT",
|
|
|
|
NULL,
|
|
|
|
};
|
|
|
|
$ cat qapi-generated/example-qapi-event.h
|
|
|
|
[Uninteresting stuff omitted...]
|
|
|
|
|
|
|
|
#ifndef EXAMPLE_QAPI_EVENT_H
|
|
|
|
#define EXAMPLE_QAPI_EVENT_H
|
|
|
|
|
|
|
|
#include "qapi/error.h"
|
|
|
|
#include "qapi/qmp/qdict.h"
|
|
|
|
#include "example-qapi-types.h"
|
|
|
|
|
|
|
|
|
|
|
|
void qapi_event_send_my_event(Error **errp);
|
|
|
|
|
|
|
|
extern const char *EXAMPLE_QAPIEvent_lookup[];
|
|
|
|
typedef enum EXAMPLE_QAPIEvent
|
|
|
|
{
|
|
|
|
EXAMPLE_QAPI_EVENT_MY_EVENT = 0,
|
|
|
|
EXAMPLE_QAPI_EVENT_MAX = 1,
|
|
|
|
} EXAMPLE_QAPIEvent;
|
2011-07-20 03:50:46 +08:00
|
|
|
|
|
|
|
#endif
|