NetworkManager/src/libnm-core-impl/nm-setting-serial.c

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

342 lines
11 KiB
C
Raw Permalink Normal View History

/* SPDX-License-Identifier: LGPL-2.1-or-later */
/*
* Copyright (C) 2007 - 2018 Red Hat, Inc.
* Copyright (C) 2007 - 2008 Novell, Inc.
*/
#include "libnm-core-impl/nm-default-libnm-core.h"
#include "nm-setting-serial.h"
#include "nm-setting-private.h"
/**
* SECTION:nm-setting-serial
* @short_description: Describes connection properties for devices that use
* serial communications
*
* The #NMSettingSerial object is a #NMSetting subclass that describes
* properties necessary for connections that may use serial communications,
* such as mobile broadband or analog telephone connections.
**/
/*****************************************************************************/
NM_GOBJECT_PROPERTIES_DEFINE_BASE(PROP_BAUD,
PROP_BITS,
PROP_PARITY,
PROP_STOPBITS,
PROP_SEND_DELAY, );
typedef struct {
guint64 send_delay;
guint32 baud;
guint32 bits;
guint32 stopbits;
char parity;
} NMSettingSerialPrivate;
/**
* NMSettingSerial:
*
* Serial Link Settings
*/
struct _NMSettingSerial {
libnm: embed private structure in NMSetting and avoid g_type_class_add_private() Historically, the NMSetting types were in public headers. Theoretically, that allowed users to subtype our classes. However in practice that was impossible because they lacked required access to internal functions to fully create an NMSetting class outside of libnm. And it also was not useful, because you simply cannot extend libnm by subtyping a libnm class. And supporting such a use case would be hard and limit what we can do in libnm. Having GObject structs in public headers also require that we don't change it's layout. The ABI of those structs must not change, if anybody out there was actually subclassing our GObjects. In libnm 1.34 (commit e46d484fae9e ('libnm: hide NMSetting types from public headers')) we moved the structs from headers to internal. This would have caused a compiler error if anybody was using those struct definitions. However, we still didn't change the ABI/layout so that we didn't break users who relied on it (for whatever reason). It doesn't seem there were any affected user. We waited long enough. Change internal ABI. No longer use g_type_class_add_private(). Instead, embed the private structs directly (_NM_GET_PRIVATE()) or indirectly (_NM_GET_PRIVATE_PTR()) in the object. The main benefit is for debugging in the debugger, where we can now easily find the private data. Previously that was so cumbersome to be effectively impossible. It's also the fastest possible way, since NM_SETTING_*_GET_PRIVATE() literally resolves to "&self->_priv" (plus static asserts and nm_assert() for type checking). _NM_GET_PRIVATE() also propagates constness and requires that the argument is a compatible pointer type (at compile time). Note that g_type_class_add_private() is also deprecated in glib 2.58 and replaced by G_ADD_PRIVATE(). For one, we still don't rely on 2.58. Also, G_ADD_PRIVATE() is a worse solution as it supports a usecase that we don't care for (public structs in headers). _NM_GET_PRIVATE() is still faster, works with older glib and most importantly: is better for debugging as you can find the private data from an object pointer. For NMSettingIPConfig this is rather awkward, because all direct properties require a common "klass->private_offset". This was however the case before this change. Nothing new here. And if you ever touch this and do something wrong, many unit tests will fail. It's almost impossible to get wrong, albeit it can be confusing to understand. https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1773
2023-10-24 19:05:50 +02:00
NMSetting parent;
NMSettingSerialPrivate _priv;
};
struct _NMSettingSerialClass {
NMSettingClass parent;
};
G_DEFINE_TYPE(NMSettingSerial, nm_setting_serial, NM_TYPE_SETTING)
#define NM_SETTING_SERIAL_GET_PRIVATE(o) \
libnm: embed private structure in NMSetting and avoid g_type_class_add_private() Historically, the NMSetting types were in public headers. Theoretically, that allowed users to subtype our classes. However in practice that was impossible because they lacked required access to internal functions to fully create an NMSetting class outside of libnm. And it also was not useful, because you simply cannot extend libnm by subtyping a libnm class. And supporting such a use case would be hard and limit what we can do in libnm. Having GObject structs in public headers also require that we don't change it's layout. The ABI of those structs must not change, if anybody out there was actually subclassing our GObjects. In libnm 1.34 (commit e46d484fae9e ('libnm: hide NMSetting types from public headers')) we moved the structs from headers to internal. This would have caused a compiler error if anybody was using those struct definitions. However, we still didn't change the ABI/layout so that we didn't break users who relied on it (for whatever reason). It doesn't seem there were any affected user. We waited long enough. Change internal ABI. No longer use g_type_class_add_private(). Instead, embed the private structs directly (_NM_GET_PRIVATE()) or indirectly (_NM_GET_PRIVATE_PTR()) in the object. The main benefit is for debugging in the debugger, where we can now easily find the private data. Previously that was so cumbersome to be effectively impossible. It's also the fastest possible way, since NM_SETTING_*_GET_PRIVATE() literally resolves to "&self->_priv" (plus static asserts and nm_assert() for type checking). _NM_GET_PRIVATE() also propagates constness and requires that the argument is a compatible pointer type (at compile time). Note that g_type_class_add_private() is also deprecated in glib 2.58 and replaced by G_ADD_PRIVATE(). For one, we still don't rely on 2.58. Also, G_ADD_PRIVATE() is a worse solution as it supports a usecase that we don't care for (public structs in headers). _NM_GET_PRIVATE() is still faster, works with older glib and most importantly: is better for debugging as you can find the private data from an object pointer. For NMSettingIPConfig this is rather awkward, because all direct properties require a common "klass->private_offset". This was however the case before this change. Nothing new here. And if you ever touch this and do something wrong, many unit tests will fail. It's almost impossible to get wrong, albeit it can be confusing to understand. https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1773
2023-10-24 19:05:50 +02:00
_NM_GET_PRIVATE(o, NMSettingSerial, NM_IS_SETTING_SERIAL, NMSetting)
/*****************************************************************************/
/**
* nm_setting_serial_get_baud:
* @setting: the #NMSettingSerial
*
* Returns: the #NMSettingSerial:baud property of the setting
**/
guint
nm_setting_serial_get_baud(NMSettingSerial *setting)
{
g_return_val_if_fail(NM_IS_SETTING_SERIAL(setting), 0);
return NM_SETTING_SERIAL_GET_PRIVATE(setting)->baud;
}
/**
* nm_setting_serial_get_bits:
* @setting: the #NMSettingSerial
*
* Returns: the #NMSettingSerial:bits property of the setting
**/
guint
nm_setting_serial_get_bits(NMSettingSerial *setting)
{
g_return_val_if_fail(NM_IS_SETTING_SERIAL(setting), 0);
return NM_SETTING_SERIAL_GET_PRIVATE(setting)->bits;
}
/**
* nm_setting_serial_get_parity:
* @setting: the #NMSettingSerial
*
* Returns: the #NMSettingSerial:parity property of the setting
**/
NMSettingSerialParity
nm_setting_serial_get_parity(NMSettingSerial *setting)
{
g_return_val_if_fail(NM_IS_SETTING_SERIAL(setting), 0);
return NM_SETTING_SERIAL_GET_PRIVATE(setting)->parity;
}
/**
* nm_setting_serial_get_stopbits:
* @setting: the #NMSettingSerial
*
* Returns: the #NMSettingSerial:stopbits property of the setting
**/
guint
nm_setting_serial_get_stopbits(NMSettingSerial *setting)
{
g_return_val_if_fail(NM_IS_SETTING_SERIAL(setting), 0);
return NM_SETTING_SERIAL_GET_PRIVATE(setting)->stopbits;
}
/**
* nm_setting_serial_get_send_delay:
* @setting: the #NMSettingSerial
*
* Returns: the #NMSettingSerial:send-delay property of the setting
**/
guint64
nm_setting_serial_get_send_delay(NMSettingSerial *setting)
{
g_return_val_if_fail(NM_IS_SETTING_SERIAL(setting), 0);
return NM_SETTING_SERIAL_GET_PRIVATE(setting)->send_delay;
}
/*****************************************************************************/
static GVariant *
libnm: use macros function arguments for NMSettInfoPropertType These functions tend to have many arguments. They are also quite som boilerplate to implement the hundereds of properties we have, while we want that properties have common behaviors and similarities. Instead of repeatedly spelling out the function arguments, use a macro. Advantages: - the usage of a _NM_SETT_INFO_PROP_*_FCN_ARGS macro signals that this is an implementation of a property. You can now grep for these macros to find all implementation. That was previously rather imprecise, you could only `git grep '\.to_dbus_fcn'` to find the uses, but not the implementations. As the goal is to keep properties "similar", there is a desire to reduce the number of similar implementations and to find them. - changing the arguments now no longer will require you to go through all implementations. At least not, if you merely add an argument that has a reasonable default behavior and does not require explicit handling by most implementation. - it's convenient to be able to patch the argument list to let the compiler help to reason about something. For example, the "connection_dict" argument to from_dbus_fcn() is usually unused. If you'd like to find who uses it, rename the parameter, and review the (few) compiler errors. - it does save 573 LOC of boilerplate with no actual logic or useful information. I argue, that this simplifies the code and review, by increasing the relative amount of actually meaningful code. Disadvantages: - the user no longer directly sees the argument list. They would need cscope/ctags or an IDE to jump to the macro definition and conveniently see all arguments. Also use _nm_nil, so that clang-format interprets this as a function parameter list. Otherwise, it formats the function differently.
2021-07-26 23:45:31 +02:00
parity_to_dbus_fcn(_NM_SETT_INFO_PROP_TO_DBUS_FCN_ARGS _nm_nil)
{
switch (nm_setting_serial_get_parity(NM_SETTING_SERIAL(setting))) {
case NM_SETTING_SERIAL_PARITY_EVEN:
return g_variant_new_byte('E');
case NM_SETTING_SERIAL_PARITY_ODD:
return g_variant_new_byte('o');
case NM_SETTING_SERIAL_PARITY_NONE:
/* the default, serializes to NULL. */
return NULL;
default:
return NULL;
}
}
static void
libnm: use macros function arguments for NMSettInfoPropertType These functions tend to have many arguments. They are also quite som boilerplate to implement the hundereds of properties we have, while we want that properties have common behaviors and similarities. Instead of repeatedly spelling out the function arguments, use a macro. Advantages: - the usage of a _NM_SETT_INFO_PROP_*_FCN_ARGS macro signals that this is an implementation of a property. You can now grep for these macros to find all implementation. That was previously rather imprecise, you could only `git grep '\.to_dbus_fcn'` to find the uses, but not the implementations. As the goal is to keep properties "similar", there is a desire to reduce the number of similar implementations and to find them. - changing the arguments now no longer will require you to go through all implementations. At least not, if you merely add an argument that has a reasonable default behavior and does not require explicit handling by most implementation. - it's convenient to be able to patch the argument list to let the compiler help to reason about something. For example, the "connection_dict" argument to from_dbus_fcn() is usually unused. If you'd like to find who uses it, rename the parameter, and review the (few) compiler errors. - it does save 573 LOC of boilerplate with no actual logic or useful information. I argue, that this simplifies the code and review, by increasing the relative amount of actually meaningful code. Disadvantages: - the user no longer directly sees the argument list. They would need cscope/ctags or an IDE to jump to the macro definition and conveniently see all arguments. Also use _nm_nil, so that clang-format interprets this as a function parameter list. Otherwise, it formats the function differently.
2021-07-26 23:45:31 +02:00
parity_from_dbus(_NM_SETT_INFO_PROP_FROM_DBUS_GPROP_FCN_ARGS _nm_nil)
{
switch (g_variant_get_byte(from)) {
case 'E':
g_value_set_enum(to, NM_SETTING_SERIAL_PARITY_EVEN);
break;
case 'o':
g_value_set_enum(to, NM_SETTING_SERIAL_PARITY_ODD);
break;
case 'n':
default:
g_value_set_enum(to, NM_SETTING_SERIAL_PARITY_NONE);
break;
}
}
/*****************************************************************************/
static void
get_property(GObject *object, guint prop_id, GValue *value, GParamSpec *pspec)
{
NMSettingSerial *setting = NM_SETTING_SERIAL(object);
switch (prop_id) {
case PROP_PARITY:
g_value_set_enum(value, nm_setting_serial_get_parity(setting));
break;
default:
_nm_setting_property_get_property_direct(object, prop_id, value, pspec);
break;
}
}
static void
set_property(GObject *object, guint prop_id, const GValue *value, GParamSpec *pspec)
{
NMSettingSerialPrivate *priv = NM_SETTING_SERIAL_GET_PRIVATE(object);
switch (prop_id) {
case PROP_PARITY:
priv->parity = g_value_get_enum(value);
break;
default:
_nm_setting_property_set_property_direct(object, prop_id, value, pspec);
break;
}
}
/*****************************************************************************/
static void
nm_setting_serial_init(NMSettingSerial *self)
{
NMSettingSerialPrivate *priv = NM_SETTING_SERIAL_GET_PRIVATE(self);
nm_assert(priv->parity == NM_SETTING_SERIAL_PARITY_NONE);
}
/**
* nm_setting_serial_new:
*
* Creates a new #NMSettingSerial object with default values.
*
* Returns: (transfer full): the new empty #NMSettingSerial object
**/
NMSetting *
nm_setting_serial_new(void)
{
return g_object_new(NM_TYPE_SETTING_SERIAL, NULL);
}
static void
nm_setting_serial_class_init(NMSettingSerialClass *klass)
{
GObjectClass *object_class = G_OBJECT_CLASS(klass);
NMSettingClass *setting_class = NM_SETTING_CLASS(klass);
GArray *properties_override = _nm_sett_info_property_override_create_array();
object_class->get_property = get_property;
object_class->set_property = set_property;
/**
* NMSettingSerial:baud:
*
* Speed to use for communication over the serial port. Note that this
* value usually has no effect for mobile broadband modems as they generally
* ignore speed settings and use the highest available speed.
**/
_nm_setting_property_define_direct_uint32(properties_override,
obj_properties,
NM_SETTING_SERIAL_BAUD,
PROP_BAUD,
0,
G_MAXUINT32,
57600,
NM_SETTING_PARAM_NONE,
NMSettingSerialPrivate,
baud);
/**
* NMSettingSerial:bits:
*
* Byte-width of the serial communication. The 8 in "8n1" for example.
**/
_nm_setting_property_define_direct_uint32(properties_override,
obj_properties,
NM_SETTING_SERIAL_BITS,
PROP_BITS,
5,
8,
8,
NM_SETTING_PARAM_NONE,
NMSettingSerialPrivate,
bits);
/**
* NMSettingSerial:parity:
*
* Parity setting of the serial port.
**/
/* ---keyfile---
* property: parity
* format: 'e', 'o', or 'n'
* description: The connection parity; even, odd, or none. Note that older
* versions of NetworkManager stored this as an integer: 69 ('E') for even,
* 111 ('o') for odd, or 110 ('n') for none.
* example: parity=n
* ---end---
*/
/* ---dbus---
* property: parity
* format: byte
* description: The connection parity: 69 (ASCII 'E') for even parity,
* 111 (ASCII 'o') for odd, 110 (ASCII 'n') for none.
* ---end---
*/
obj_properties[PROP_PARITY] = g_param_spec_enum(NM_SETTING_SERIAL_PARITY,
"",
"",
NM_TYPE_SETTING_SERIAL_PARITY,
NM_SETTING_SERIAL_PARITY_NONE,
G_PARAM_READWRITE | G_PARAM_STATIC_STRINGS);
_nm_properties_override_gobj(
properties_override,
obj_properties[PROP_PARITY],
NM_SETT_INFO_PROPERT_TYPE_DBUS(G_VARIANT_TYPE_BYTE,
.compare_fcn = _nm_setting_property_compare_fcn_default,
.to_dbus_fcn = parity_to_dbus_fcn,
.typdata_from_dbus.gprop_fcn = parity_from_dbus,
.from_dbus_fcn = _nm_setting_property_from_dbus_fcn_gprop,
.from_dbus_is_full = TRUE));
/**
* NMSettingSerial:stopbits:
*
* Number of stop bits for communication on the serial port. Either 1 or 2.
* The 1 in "8n1" for example.
**/
_nm_setting_property_define_direct_uint32(properties_override,
obj_properties,
NM_SETTING_SERIAL_STOPBITS,
PROP_STOPBITS,
1,
2,
1,
NM_SETTING_PARAM_NONE,
NMSettingSerialPrivate,
stopbits);
/**
* NMSettingSerial:send-delay:
*
* Time to delay between each byte sent to the modem, in microseconds.
**/
_nm_setting_property_define_direct_uint64(properties_override,
obj_properties,
NM_SETTING_SERIAL_SEND_DELAY,
PROP_SEND_DELAY,
0,
G_MAXUINT64,
0,
NM_SETTING_PARAM_NONE,
NMSettingSerialPrivate,
send_delay);
g_object_class_install_properties(object_class, _PROPERTY_ENUMS_LAST, obj_properties);
_nm_setting_class_commit(setting_class,
NM_META_SETTING_TYPE_SERIAL,
NULL,
properties_override,
libnm: embed private structure in NMSetting and avoid g_type_class_add_private() Historically, the NMSetting types were in public headers. Theoretically, that allowed users to subtype our classes. However in practice that was impossible because they lacked required access to internal functions to fully create an NMSetting class outside of libnm. And it also was not useful, because you simply cannot extend libnm by subtyping a libnm class. And supporting such a use case would be hard and limit what we can do in libnm. Having GObject structs in public headers also require that we don't change it's layout. The ABI of those structs must not change, if anybody out there was actually subclassing our GObjects. In libnm 1.34 (commit e46d484fae9e ('libnm: hide NMSetting types from public headers')) we moved the structs from headers to internal. This would have caused a compiler error if anybody was using those struct definitions. However, we still didn't change the ABI/layout so that we didn't break users who relied on it (for whatever reason). It doesn't seem there were any affected user. We waited long enough. Change internal ABI. No longer use g_type_class_add_private(). Instead, embed the private structs directly (_NM_GET_PRIVATE()) or indirectly (_NM_GET_PRIVATE_PTR()) in the object. The main benefit is for debugging in the debugger, where we can now easily find the private data. Previously that was so cumbersome to be effectively impossible. It's also the fastest possible way, since NM_SETTING_*_GET_PRIVATE() literally resolves to "&self->_priv" (plus static asserts and nm_assert() for type checking). _NM_GET_PRIVATE() also propagates constness and requires that the argument is a compatible pointer type (at compile time). Note that g_type_class_add_private() is also deprecated in glib 2.58 and replaced by G_ADD_PRIVATE(). For one, we still don't rely on 2.58. Also, G_ADD_PRIVATE() is a worse solution as it supports a usecase that we don't care for (public structs in headers). _NM_GET_PRIVATE() is still faster, works with older glib and most importantly: is better for debugging as you can find the private data from an object pointer. For NMSettingIPConfig this is rather awkward, because all direct properties require a common "klass->private_offset". This was however the case before this change. Nothing new here. And if you ever touch this and do something wrong, many unit tests will fail. It's almost impossible to get wrong, albeit it can be confusing to understand. https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1773
2023-10-24 19:05:50 +02:00
G_STRUCT_OFFSET(NMSettingSerial, _priv));
}