Currently the attribute is not used.
However we would like to make sure when present that the array
is always zero terminated to simplify usages after demarshalling.
Signed-off-by: Frediano Ziglio <freddy77@gmail.com>
Acked-by: Victor Toso <victortoso@redhat.com>
Current propagated attributes are propagated only from member to
member type.
But for pointers you probably want to use propagated attribus to
underlying type (like an array for instance).
Signed-off-by: Frediano Ziglio <freddy77@gmail.com>
Acked-by: Victor Toso <victortoso@redhat.com>
If the size is not constant the array has to be allocated in some
way in the output and so there must be a specification for the
output (as default is write into the C structure all data).
The only exceptions are when the length is constant (in this case
a constant length array in the C structure is used) or a pointer
(in this case the pointer allocate the array).
Signed-off-by: Frediano Ziglio <freddy77@gmail.com>
Acked-by: Victor Toso <victortoso@redhat.com>
This syntax was only used in protocol 1 which has been removed
time ago.
Beside not being used it's confusing and prone to errors,
array size is specified using 2 identifiers, one reporting
bytes and the other number of items, one used for marshalling,
the other for demarshalling.
Signed-off-by: Frediano Ziglio <freddy77@gmail.com>
We want to obsolete from enumeration values, currently CELT.
The removal of check in has_attr is not an issue, as the attributes
are already checked by fix_attributes calls.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Victor Toso <victortoso@redhat.com>
Check that combination of fields for an array does not
lead to unsafe code.
check_valid method came from generate_c_declaration with
some more check and it's use in demarshaller to validate
the array if the structure is not generated.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Victor Toso <victortoso@redhat.com>
Output attributes are the attributes that specify how to store
that field in the C structure.
There can be only one output type specified.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Victor Toso <victortoso@redhat.com>
member = None is set before the if/else condition.
In the else code, when member is set it is checked
and if not-None it breaks out of the loop.
If the code is still in the loop for sure member is None.
Found by covscan.
Signed-off-by: Uri Lublin <uril@redhat.com>
Acked-by: Frediano Ziglio <fziglio@redhat.com>
Allows to specify a @declare attribute for messages and structure
that can generate the needed C structures.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
This option was used in protocol 1 to generate 64 bit pointers.
A pointer in the protocol is an offset in the current message.
This allows the possibility to have messages with pointers with more
than 4GB. This feature was removed and not used in protocol 2.
The reason this feature was correctly removed in protocol 2 is that
having 64 bit pointers in the protocol would require messages larger
than 4GB which would cause:
- huge latency as a single message would take more than 4 seconds
to be send in a 10Gb connection;
- huge memory requirements.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
Does not make sense to specify the same field to have 2
different C implementation at the same time.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
c_type() method is supposed to return the type to use for
C structure field. But the name is not a C type but a
protocol name.
Return the type name of the aliased type (for instance
uint32_t for a uint32 type).
This does not change the generated code.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
The idea in version 1 of the protocol was to extend it using the minor
version. However this was replaced by the usage of capabilities and the
minor attribute (which was not much used in version 1) was abandoned in
version 2.
This patch create a big difference in the code generated but only because
the minor version was passed between all possible functions as argument.
Note that exported functions retain the minor argument for compatibility
reasons.
The demarshaller code export directly spice_get_client_channel_parser or
spice_get_server_channel_parser functions which returns internal module
functions which parse message of specific channels.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
This attribute was used only in SPICE version 1.
The intention was use fixed size for switch type in the protocol.
However this does not bring any improvement, just increase network
bytes used.
Generated code does not change.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
This attribute was used only in SPICE version 1.
Its usage was confusing, and was replaced by the simple usage of
array size.
Generated code does not change.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
Make sure there are not 2 messages with the same value in the
same channel.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
Make sure there are not 2 messages with the same name in the
same channel.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
Code that handled client and server messages check was the same, just
changed some variable names.
Instead use a class to store same information and reuse the code.
This allows easier extension of the 2 path of code.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Lukáš Hrázký <lhrazky@redhat.com>
When testing out some experimental protocol changes, I managed to
trigger the following error:
GEN generated_client_demarshallers.c
Traceback (most recent call last):
File "../../../spice-common/spice_codegen.py", line 267, in <module>
demarshal.write_protocol_parser(writer, proto, True)
File "/home/jjongsma/work/spice/spice-common/python_modules/demarshal.py", line 1270, in write_protocol_parser
parsers[channel.value] = (channel.channel_type, write_channel_parser(writer, channel.channel_type, is_server))
File "/home/jjongsma/work/spice/spice-common/python_modules/demarshal.py", line 1163, in write_channel_parser
func = write_msg_parser(helpers, ids[i].message_type)
File "/home/jjongsma/work/spice/spice-common/python_modules/demarshal.py", line 1061, in write_msg_parser
num_pointers = message.get_num_pointers()
File "/home/jjongsma/work/spice/spice-common/python_modules/ptypes.py", line 855, in get_num_pointers
count = count + m.get_num_pointers()
File "/home/jjongsma/work/spice/spice-common/python_modules/ptypes.py", line 662, in get_num_pointers
return self.member_type.get_num_pointers()
File "/home/jjongsma/work/spice/spice-common/python_modules/ptypes.py", line 507, in get_num_pointers
if self.is_constant_length(self):
TypeError: is_constant_length() takes exactly 1 argument (2 given)
Calling a member function will implicitly pass 'self' as the first
argument, but we were also explicitly passing it as an argument
(self.is_constant_length(self)). This resulted in the above error.
Acked-by: Lukáš Hrázký <lhrazky@redhat.com>
Not supported by code so trigger an error to avoid invalid usages
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Acked-by: Christophe Fergeau <cfergeau@redhat.com>
This patch is a left-over from the fd passing commit 267391c8fd as
suggested by Frediano Ziglio during review.
Signed-off-by: Marc-André Lureau <marcandre.lureau@gmail.com>
Acked-by: Frediano Ziglio <fziglio@redhat.com>
(cherry picked from spice-protocol commit 47076559628d71c128e14e11147ce36b92677885)
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Add a new type, "unix_fd", used to describe file descriptor sharing via
socket ancillary data (these messages are local only).
The marshaller/demarshaller can't serialize this in memory (consume_fd
implementation is empty), so it is the responsability of the marshaller
user to handle sending and receiving the handles, which are appended at
the end of the message with an extra stream byte (because some Unix
requires sending at least a byte with ancillary data).
Even if there is no fd to send (or if the fd is invalid etc), the
receiver side expects an extra byte anyway.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Acked-by: Frediano Ziglio <fziglio@redhat.com>
(cherry-picked from spice-protocol commit 267391c8fd7c90c067b3e4845ff0227a2580e2e2)
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
This reverts commit 7665dcf1bb.
Also revert the related build-sys changes to fix the build.
codegen generated code depends on spice-common code (marshaller,
messages etc), it makes more sense to keep the generator along
this. Otherwise a newer protocol release will fail to build older
projects.
*.proto files are required as well, since it generates code that parent
modules depend on unconditionnaly.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
This patch changes the type of 'valid_attributes' and
'attributes_with_arguments'.
Both of them are list of different strings and are kept in sets.
This was the intention of the original code, but this patch
use a specific set([ strings ]) format, instead of { strings }.
This fixes the build for me on RHEL-6 (python-2.6.6).
Build error is:
File "/home/ulublin/git/spice/spice-common/python_modules/ptypes.py",
line 67
'end',
^
SyntaxError: invalid syntax
Verify that the attribute is known. This could help for instance to
avoid some future typo mistakes.
We also now have a list of attributes that we can comment for
documentation purpose.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
This is a new version of my previous patch that does not include six.py.
It's still kind of big, but at least it's all spice-common changes now.
There are also a few other fixes that Christophe brought to my attention.
Note that six now needs to be installed on the system (python-six on
Fedora and Debian, six on PyPI).
This *should* be enough to make spice_codegen.py work on both Python 2
and Python 3. The major changes are as follows:
* cStringIO.StringIO -> io.StringIO
* str vs. unicode updates (io.StringIO doesn't like str)
* integer division
* foo.has_key(bar) -> bar in foo
* import internal_thing -> from . import internal_thing
* removed from __future__ import with_statement
(might break Python 2.5?)
* changed some lambdas to list comprehensions (done by 2to3)
* cast some_dict.keys() to list where needed (e.g. for sorting)
* use normal type names with isinstance instead of types.WhateverType
Signed-off-by: Alexander Wauck <awauck@codeweavers.com>
The wireshark protocol dissector is a bit out-of-date. Several new channel types
and enums have been added. It would be nice if these values (and the
translation between the value and the name) could be automatically generated so
that updating the dissector was a slightly less manual process. This patch adds
a commandline switch which generates both the enums and value-name lists in the
format that wireshark expects.
Fixes the resulting enums.h from the invocation of:
./spice_codegen.py --generate-enums spice.proto spice-protocol/spice/enums.h
Right now any enum will contain None as the enum members, with this fix
it will contain the real enum members, i.e. SPICE_FOO.
We want to be able to add an @ifdef annotation to optional messages
For example, we want to compile in the smartcard messages only if
libcacard is available
With the new usbredir code we have the new concept of the abstract /
generic spicevmc channel type (which just tunnels data from a qemu chardev),
and we've the usbredir channel, which is the only current user of this.
This was reflected in the protocols enum in spice-protocol.h by a manual
edit done by me, my bad. This patch teaches spice.proto about the relation
between the abstract spicevmc channel and the usbredir channel and
modifies codegen to deal with this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Commit 9d5ef9beeca722b2ceff7d15aaa3aaaaf07ecfbf in spice-protocol
introduced a typedef manually in the generated enums.h header.
This patch adds them automatically to all enums during enums.h generation.
