Commit | Line | Data |
---|---|---|
1f7d590d MW |
1 | %%% -*-latex-*- |
2 | %%% | |
3 | %%% Conceptual background | |
4 | %%% | |
5 | %%% (c) 2015 Straylight/Edgeware | |
6 | %%% | |
7 | ||
8 | %%%----- Licensing notice --------------------------------------------------- | |
9 | %%% | |
e0808c47 | 10 | %%% This file is part of the Sensible Object Design, an object system for C. |
1f7d590d MW |
11 | %%% |
12 | %%% SOD is free software; you can redistribute it and/or modify | |
13 | %%% it under the terms of the GNU General Public License as published by | |
14 | %%% the Free Software Foundation; either version 2 of the License, or | |
15 | %%% (at your option) any later version. | |
16 | %%% | |
17 | %%% SOD is distributed in the hope that it will be useful, | |
18 | %%% but WITHOUT ANY WARRANTY; without even the implied warranty of | |
19 | %%% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
20 | %%% GNU General Public License for more details. | |
21 | %%% | |
22 | %%% You should have received a copy of the GNU General Public License | |
23 | %%% along with SOD; if not, write to the Free Software Foundation, | |
24 | %%% Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. | |
25 | ||
3cc520db | 26 | \chapter{Concepts} \label{ch:concepts} |
1f7d590d | 27 | |
3cc520db | 28 | %%%-------------------------------------------------------------------------- |
3cc520db MW |
29 | \section{Modules} \label{sec:concepts.modules} |
30 | ||
31 | A \emph{module} is the top-level syntactic unit of input to the Sod | |
32 | translator. As described above, given an input module, the translator | |
33 | generates C source and header files. | |
34 | ||
35 | A module can \emph{import} other modules. This makes the type names and | |
36 | classes defined in those other modules available to class definitions in the | |
37 | importing module. Sod's module system is intentionally very simple. There | |
38 | are no private declarations or attempts to hide things. | |
39 | ||
40 | As well as importing existing modules, a module can include a number of | |
41 | different kinds of \emph{items}: | |
42 | \begin{itemize} | |
43 | \item \emph{class definitions} describe new classes, possibly in terms of | |
44 | existing classes; | |
45 | \item \emph{type name declarations} introduce new type names to Sod's | |
46 | parser;\footnote{% | |
47 | This is unfortunately necessary because C syntax, upon which Sod's input | |
48 | language is based for obvious reasons, needs to treat type names | |
49 | differently from other kinds of identifiers.} % | |
50 | and | |
51 | \item \emph{code fragments} contain literal C code to be dropped into an | |
52 | appropriate place in an output file. | |
53 | \end{itemize} | |
54 | Each kind of item, and, indeed, a module as a whole, can have a collection of | |
55 | \emph{properties} associated with it. A property has a \emph{name} and a | |
56 | \emph{value}. Properties are an open-ended way of attaching additional | |
57 | information to module items, so extensions can make use of them without | |
58 | having to implement additional syntax. | |
59 | ||
60 | %%%-------------------------------------------------------------------------- | |
61 | \section{Classes, instances, and slots} \label{sec:concepts.classes} | |
62 | ||
63 | For the most part, Sod takes a fairly traditional view of what it means to be | |
64 | an object system. | |
65 | ||
66 | An \emph{object} maintains \emph{state} and exhibits \emph{behaviour}. An | |
67 | object's state is maintained in named \emph{slots}, each of which can store a | |
68 | C value of an appropriate (scalar or aggregate) type. An object's behaviour | |
69 | is stimulated by sending it \emph{messages}. A message has a name, and may | |
70 | carry a number of arguments, which are C values; sending a message may result | |
71 | in the state of receiving object (or other objects) being changed, and a C | |
72 | value being returned to the sender. | |
73 | ||
74 | Every object is a (direct) instance of some \emph{class}. The class | |
75 | determines which slots its instances have, which messages its instances can | |
76 | be sent, and which methods are invoked when those messages are received. The | |
77 | Sod translator's main job is to read class definitions and convert them into | |
78 | appropriate C declarations, tables, and functions. An object cannot | |
79 | (usually) change its direct class, and the direct class of an object is not | |
80 | affected by, for example, the static type of a pointer to it. | |
81 | ||
0a2d4b68 | 82 | |
3cc520db MW |
83 | \subsection{Superclasses and inheritance} |
84 | \label{sec:concepts.classes.inherit} | |
85 | ||
86 | \subsubsection{Class relationships} | |
87 | Each class has zero or more \emph{direct superclasses}. | |
88 | ||
89 | A class with no direct superclasses is called a \emph{root class}. The Sod | |
90 | runtime library includes a root class named @|SodObject|; making new root | |
91 | classes is somewhat tricky, and won't be discussed further here. | |
92 | ||
93 | Classes can have more than one direct superclass, i.e., Sod supports | |
94 | \emph{multiple inheritance}. A Sod class definition for a class~$C$ lists | |
95 | the direct superclasses of $C$ in a particular order. This order is called | |
96 | the \emph{local precedence order} of $C$, and the list which consists of $C$ | |
97 | follows by $C$'s direct superclasses in local precedence order is called the | |
98 | $C$'s \emph{local precedence list}. | |
99 | ||
100 | The multiple inheritance in Sod works similarly to multiple inheritance in | |
101 | Lisp-like languages, such as Common Lisp, EuLisp, Dylan, and Python, which is | |
102 | very different from how multiple inheritance works in \Cplusplus.\footnote{% | |
103 | The latter can be summarized as `badly'. By default in \Cplusplus, an | |
104 | instance receives an additional copy of superclass's state for each path | |
105 | through the class graph from the instance's direct class to that | |
106 | superclass, though this behaviour can be overridden by declaring | |
107 | superclasses to be @|virtual|. Also, \Cplusplus\ offers only trivial | |
108 | method combination (\xref{sec:concepts.methods}), leaving programmers to | |
109 | deal with delegation manually and (usually) statically.} % | |
110 | ||
111 | If $C$ is a class, then the \emph{superclasses} of $C$ are | |
112 | \begin{itemize} | |
113 | \item $C$ itself, and | |
114 | \item the superclasses of each of $C$'s direct superclasses. | |
115 | \end{itemize} | |
116 | The \emph{proper superclasses} of a class $C$ are the superclasses of $C$ | |
117 | except for $C$ itself. If a class $B$ is a (direct, proper) superclass of | |
118 | $C$, then $C$ is a \emph{(direct, proper) subclass} of $B$. If $C$ is a root | |
119 | class then the only superclass of $C$ is $C$ itself, and $C$ has no proper | |
120 | superclasses. | |
121 | ||
122 | If an object is a direct instance of class~$C$ then the object is also an | |
123 | (indirect) instance of every superclass of $C$. | |
124 | ||
054e8f8f MW |
125 | If $C$ has a proper superclass $B$, then $B$ must not have $C$ as a direct |
126 | superclass. In different terms, if we construct a graph, whose vertices are | |
127 | classes, and draw an edge from each class to each of its direct superclasses, | |
128 | then this graph must be acyclic. In yet other terms, the `is a superclass | |
129 | of' relation is a partial order on classes. | |
3cc520db MW |
130 | |
131 | \subsubsection{The class precedence list} | |
132 | This partial order is not quite sufficient for our purposes. For each class | |
133 | $C$, we shall need to extend it into a total order on $C$'s superclasses. | |
134 | This calculation is called \emph{superclass linearization}, and the result is | |
135 | a \emph{class precedence list}, which lists each of $C$'s superclasses | |
136 | exactly once. If a superclass $B$ precedes (resp.\ follows) some other | |
137 | superclass $A$ in $C$'s class precedence list, then we say that $B$ is a more | |
138 | (resp.\ less) \emph{specific} superclass of $C$ than $A$ is. | |
139 | ||
140 | The superclass linearization algorithm isn't fixed, and extensions to the | |
141 | translator can introduce new linearizations for special effects, but the | |
142 | following properties are expected to hold. | |
143 | \begin{itemize} | |
144 | \item The first class in $C$'s class precedence list is $C$ itself; i.e., | |
145 | $C$ is always its own most specific superclass. | |
146 | \item If $A$ and $B$ are both superclasses of $C$, and $A$ is a proper | |
147 | superclass of $B$ then $A$ appears after $B$ in $C$'s class precedence | |
148 | list, i.e., $B$ is a more specific superclass of $C$ than $A$ is. | |
149 | \end{itemize} | |
150 | The default linearization algorithm used in Sod is the \emph{C3} algorithm, | |
151 | which has a number of good properties described in~\cite{FIXME:C3}. | |
152 | It works as follows. | |
153 | \begin{itemize} | |
154 | \item A \emph{merge} of some number of input lists is a single list | |
155 | containing each item that is in any of the input lists exactly once, and no | |
156 | other items; if an item $x$ appears before an item $y$ in any input list, | |
157 | then $x$ also appears before $y$ in the merge. If a collection of lists | |
158 | have no merge then they are said to be \emph{inconsistent}. | |
159 | \item The class precedence list of a class $C$ is a merge of the local | |
160 | precedence list of $C$ together with the class precedence lists of each of | |
161 | $C$'s direct superclasses. | |
162 | \item If there are no such merges, then the definition of $C$ is invalid. | |
163 | \item Suppose that there are multiple candidate merges. Consider the | |
164 | earliest position in these candidate merges at which they disagree. The | |
165 | \emph{candidate classes} at this position are the classes appearing at this | |
166 | position in the candidate merges. Each candidate class must be a | |
781a8fbd | 167 | superclass of distinct direct superclasses of $C$, since otherwise the |
3cc520db MW |
168 | candidates would be ordered by their common subclass's class precedence |
169 | list. The class precedence list contains, at this position, that candidate | |
170 | class whose subclass appears earliest in $C$'s local precedence order. | |
171 | \end{itemize} | |
172 | ||
173 | \subsubsection{Class links and chains} | |
174 | The definition for a class $C$ may distinguish one of its proper superclasses | |
175 | as being the \emph{link superclass} for class $C$. Not every class need have | |
176 | a link superclass, and the link superclass of a class $C$, if it exists, need | |
177 | not be a direct superclass of $C$. | |
178 | ||
179 | Superclass links must obey the following rule: if $C$ is a class, then there | |
756e9293 MW |
180 | must be no three distinct superclasses $X$, $Y$ and~$Z$ of $C$ such that $Z$ |
181 | is the link superclass of both $X$ and $Y$. As a consequence of this rule, | |
182 | the superclasses of $C$ can be partitioned into linear \emph{chains}, such | |
183 | that superclasses $A$ and $B$ are in the same chain if and only if one can | |
184 | trace a path from $A$ to $B$ by following superclass links, or \emph{vice | |
185 | versa}. | |
3cc520db MW |
186 | |
187 | Since a class links only to one of its proper superclasses, the classes in a | |
188 | chain are naturally ordered from most- to least-specific. The least specific | |
189 | class in a chain is called the \emph{chain head}; the most specific class is | |
190 | the \emph{chain tail}. Chains are often named after their chain head | |
191 | classes. | |
192 | ||
193 | \subsection{Names} | |
194 | \label{sec:concepts.classes.names} | |
195 | ||
196 | Classes have a number of other attributes: | |
197 | \begin{itemize} | |
198 | \item A \emph{name}, which is a C identifier. Class names must be globally | |
199 | unique. The class name is used in the names of a number of associated | |
200 | definitions, to be described later. | |
201 | \item A \emph{nickname}, which is also a C identifier. Unlike names, | |
202 | nicknames are not required to be globally unique. If $C$ is any class, | |
203 | then all the superclasses of $C$ must have distinct nicknames. | |
204 | \end{itemize} | |
205 | ||
0a2d4b68 | 206 | |
3cc520db MW |
207 | \subsection{Slots} \label{sec:concepts.classes.slots} |
208 | ||
209 | Each class defines a number of \emph{slots}. Much like a structure member, a | |
210 | slot has a \emph{name}, which is a C identifier, and a \emph{type}. Unlike | |
211 | many other object systems, different superclasses of a class $C$ can define | |
212 | slots with the same name without ambiguity, since slot references are always | |
213 | qualified by the defining class's nickname. | |
214 | ||
215 | \subsubsection{Slot initializers} | |
216 | As well as defining slot names and types, a class can also associate an | |
217 | \emph{initial value} with each slot defined by itself or one of its | |
98da9322 | 218 | subclasses. A class $C$ provides an \emph{initialization message} (see |
d24d47f5 | 219 | \xref{sec:concepts.lifecycle.birth}, and \xref{sec:structures.root.sodclass}) |
98da9322 MW |
220 | whose methods set the slots of a \emph{direct} instance of the class to the |
221 | correct initial values. If several of $C$'s superclasses define initializers | |
222 | for the same slot then the initializer from the most specific such class is | |
223 | used. If none of $C$'s superclasses define an initializer for some slot then | |
224 | that slot will be left uninitialized. | |
3cc520db MW |
225 | |
226 | The initializer for a slot with scalar type may be any C expression. The | |
227 | initializer for a slot with aggregate type must contain only constant | |
228 | expressions if the generated code is expected to be processed by a | |
229 | implementation of C89. Initializers will be evaluated once each time an | |
230 | instance is initialized. | |
231 | ||
27ec3825 MW |
232 | Slots are initialized in reverse-precedence order of their defining classes; |
233 | i.e., slots defined by a less specific superclass are initialized earlier | |
234 | than slots defined by a more specific superclass. Slots defined by the same | |
235 | class are initialized in the order in which they appear in the class | |
236 | definition. | |
237 | ||
238 | The initializer for a slot may refer to other slots in the same object, via | |
239 | the @|me| pointer: in an initializer for a slot defined by a class $C$, @|me| | |
240 | has type `pointer to $C$'. (Note that the type of @|me| depends only on the | |
241 | class which defined the slot, not the class which defined the initializer.) | |
242 | ||
0a2d4b68 | 243 | |
3cc520db MW |
244 | \subsection{C language integration} \label{sec:concepts.classes.c} |
245 | ||
246 | For each class~$C$, the Sod translator defines a C type, the \emph{class | |
247 | type}, with the same name. This is the usual type used when considering an | |
248 | object as an instance of class~$C$. No entire object will normally have a | |
249 | class type,\footnote{% | |
250 | In general, a class type only captures the structure of one of the | |
251 | superclass chains of an instance. A full instance layout contains multiple | |
252 | chains. See \xref{sec:structures.layout} for the full details.} % | |
253 | so access to instances is almost always via pointers. | |
254 | ||
255 | \subsubsection{Access to slots} | |
256 | The class type for a class~$C$ is actually a structure. It contains one | |
257 | member for each class in $C$'s superclass chain, named with that class's | |
258 | nickname. Each of these members is also a structure, containing the | |
259 | corresponding class's slots, one member per slot. There's nothing special | |
260 | about these slot members: C code can access them in the usual way. | |
261 | ||
262 | For example, if @|MyClass| has the nickname @|mine|, and defines a slot @|x| | |
263 | of type @|int|, then the simple function | |
264 | \begin{prog} | |
c18d6aba | 265 | int get_x(MyClass *m) \{ return (m@->mine.x); \} |
3cc520db MW |
266 | \end{prog} |
267 | will extract the value of @|x| from an instance of @|MyClass|. | |
268 | ||
269 | All of this means that there's no such thing as `private' or `protected' | |
270 | slots. If you want to hide implementation details, the best approach is to | |
271 | stash them in a dynamically allocated private structure, and leave a pointer | |
272 | to it in a slot. (This will also help preserve binary compatibility, because | |
273 | the private structure can grow more members as needed. See | |
e4ea29d8 | 274 | \xref{sec:fixme.compatibility} for more details.) |
3cc520db | 275 | |
caa6f4b9 MW |
276 | \subsubsection{Vtables} |
277 | ||
278 | ||
3cc520db MW |
279 | \subsubsection{Class objects} |
280 | In Sod's object system, classes are objects too. Therefore classes are | |
281 | themselves instances; the class of a class is called a \emph{metaclass}. The | |
282 | consequences of this are explored in \xref{sec:concepts.metaclasses}. The | |
283 | \emph{class object} has the same name as the class, suffixed with | |
284 | `@|__class|'\footnote{% | |
285 | This is not quite true. @|$C$__class| is actually a macro. See | |
286 | \xref{sec:structures.layout.additional} for the gory details.} % | |
287 | and its type is usually @|SodClass|; @|SodClass|'s nickname is @|cls|. | |
288 | ||
289 | A class object's slots contain or point to useful information, tables and | |
290 | functions for working with that class's instances. (The @|SodClass| class | |
054e8f8f MW |
291 | doesn't define any messages, so it doesn't have any methods other than for |
292 | the @|SodObject| lifecycle messages @|init| and @|teardown|; see | |
293 | \xref{sec:concepts.lifecycle}. In Sod, a class slot containing a function | |
294 | pointer is not at all the same thing as a method.) | |
3cc520db | 295 | |
3cc520db | 296 | \subsubsection{Conversions} |
e4ea29d8 MW |
297 | Suppose one has a value of type pointer-to-class-type for some class~$C$, and |
298 | wants to convert it to a pointer-to-class-type for some other class~$B$. | |
3cc520db MW |
299 | There are three main cases to distinguish. |
300 | \begin{itemize} | |
301 | \item If $B$ is a superclass of~$C$, in the same chain, then the conversion | |
302 | is an \emph{in-chain upcast}. The conversion can be performed using the | |
303 | appropriate generated upcast macro (see below), or by simply casting the | |
304 | pointer, using C's usual cast operator (or the \Cplusplus\ @|static_cast<>| | |
305 | operator). | |
306 | \item If $B$ is a superclass of~$C$, in a different chain, then the | |
307 | conversion is a \emph{cross-chain upcast}. The conversion is more than a | |
308 | simple type change: the pointer value must be adjusted. If the direct | |
309 | class of the instance in question is not known, the conversion will require | |
310 | a lookup at runtime to find the appropriate offset by which to adjust the | |
311 | pointer. The conversion can be performed using the appropriate generated | |
312 | upcast macro (see below); the general case is handled by the macro | |
58f9b400 | 313 | \descref{SOD_XCHAIN}{mac}. |
e4ea29d8 | 314 | \item If $B$ is a subclass of~$C$ then the conversion is a \emph{downcast}; |
3cc520db MW |
315 | otherwise the conversion is a~\emph{cross-cast}. In either case, the |
316 | conversion can fail: the object in question might not be an instance of~$B$ | |
e4ea29d8 | 317 | after all. The macro \descref{SOD_CONVERT}{mac} and the function |
58f9b400 | 318 | \descref{sod_convert}{fun} perform general conversions. They return a null |
054e8f8f | 319 | pointer if the conversion fails. (These are therefore your analogue to the |
e4ea29d8 | 320 | \Cplusplus\ @|dynamic_cast<>| operator.) |
3cc520db MW |
321 | \end{itemize} |
322 | The Sod translator generates macros for performing both in-chain and | |
323 | cross-chain upcasts. For each class~$C$, and each proper superclass~$B$ | |
324 | of~$C$, a macro is defined: given an argument of type pointer to class type | |
325 | of~$C$, it returns a pointer to the same instance, only with type pointer to | |
326 | class type of~$B$, adjusted as necessary in the case of a cross-chain | |
327 | conversion. The macro is named by concatenating | |
328 | \begin{itemize} | |
329 | \item the name of class~$C$, in upper case, | |
330 | \item the characters `@|__CONV_|', and | |
331 | \item the nickname of class~$B$, in upper case; | |
332 | \end{itemize} | |
333 | e.g., if $C$ is named @|MyClass|, and $B$'s name is @|SuperClass| with | |
334 | nickname @|super|, then the macro @|MYCLASS__CONV_SUPER| converts a | |
335 | @|MyClass~*| to a @|SuperClass~*|. See | |
336 | \xref{sec:structures.layout.additional} for the formal description. | |
337 | ||
338 | %%%-------------------------------------------------------------------------- | |
9e91c8e7 MW |
339 | \section{Keyword arguments} \label{sec:concepts.keywords} |
340 | ||
341 | In standard C, the actual arguments provided to a function are matched up | |
342 | with the formal arguments given in the function definition according to their | |
343 | ordering in a list. Unless the (rather cumbersome) machinery for dealing | |
344 | with variable-length argument tails (@|<stdarg.h>|) is used, exactly the | |
345 | correct number of arguments must be supplied, and in the correct order. | |
346 | ||
347 | A \emph{keyword argument} is matched by its distinctive \emph{name}, rather | |
348 | than by its position in a list. Keyword arguments may be \emph{omitted}, | |
349 | causing some default behaviour by the function. A function can detect | |
350 | whether a particular keyword argument was supplied: so the default behaviour | |
351 | need not be the same as that caused by any specific value of the argument. | |
352 | ||
353 | Keyword arguments can be provided in three ways. | |
354 | \begin{enumerate} | |
355 | \item Directly, as a variable-length argument tail, consisting (for the most | |
356 | part) of alternating keyword names, as pointers to null-terminated strings, | |
357 | and argument values, and terminated by a null pointer. This is somewhat | |
358 | error-prone, and the support library defines some macros which help ensure | |
359 | that keyword argument lists are well formed. | |
360 | \item Indirectly, through a @|va_list| object capturing a variable-length | |
361 | argument tail passed to some other function. Such indirect argument tails | |
362 | have the same structure as the direct argument tails described above. | |
363 | Because @|va_list| objects are hard to copy, the keyword-argument support | |
364 | library consistently passes @|va_list| objects \emph{by reference} | |
365 | throughout its programming interface. | |
366 | \item Indirectly, through a vector of @|struct kwval| objects, each of which | |
367 | contains a keyword name, as a pointer to a null-terminated string, and the | |
368 | \emph{address} of a corresponding argument value. (This indirection is | |
369 | necessary so that the items in the vector can be of uniform size.) | |
370 | Argument vectors are rather inconvenient to use, but are the only practical | |
371 | way in which a caller can decide at runtime which arguments to include in a | |
372 | call, which is useful when writing wrapper functions. | |
373 | \end{enumerate} | |
374 | ||
375 | Keyword arguments are provided as a general feature for C functions. | |
43073476 | 376 | However, Sod has special support for messages which accept keyword arguments |
8ec911fa | 377 | (\xref{sec:concepts.methods.keywords}); and they play an essential rôle in |
a142609c | 378 | the instance construction protocol (\xref{sec:concepts.lifecycle.birth}). |
9e91c8e7 MW |
379 | |
380 | %%%-------------------------------------------------------------------------- | |
3cc520db MW |
381 | \section{Messages and methods} \label{sec:concepts.methods} |
382 | ||
383 | Objects can be sent \emph{messages}. A message has a \emph{name}, and | |
384 | carries a number of \emph{arguments}. When an object is sent a message, a | |
385 | function, determined by the receiving object's class, is invoked, passing it | |
386 | the receiver and the message arguments. This function is called the | |
387 | class's \emph{effective method} for the message. The effective method can do | |
388 | anything a C function can do, including reading or updating program state or | |
389 | object slots, sending more messages, calling other functions, issuing system | |
390 | calls, or performing I/O; if it finishes, it may return a value, which is | |
391 | returned in turn to the message sender. | |
392 | ||
393 | The set of messages an object can receive, characterized by their names, | |
394 | argument types, and return type, is determined by the object's class. Each | |
395 | class can define new messages, which can be received by any instance of that | |
396 | class. The messages defined by a single class must have distinct names: | |
397 | there is no `function overloading'. As with slots | |
398 | (\xref{sec:concepts.classes.slots}), messages defined by distinct classes are | |
399 | always distinct, even if they have the same names: references to messages are | |
400 | always qualified by the defining class's name or nickname. | |
401 | ||
402 | Messages may take any number of arguments, of any non-array value type. | |
403 | Since message sends are effectively function calls, arguments of array type | |
404 | are implicitly converted to values of the corresponding pointer type. While | |
405 | message definitions may ascribe an array type to an argument, the formal | |
406 | argument will have pointer type, as is usual for C functions. A message may | |
407 | accept a variable-length argument suffix, denoted @|\dots|. | |
408 | ||
409 | A class definition may include \emph{direct methods} for messages defined by | |
410 | it or any of its superclasses. | |
411 | ||
412 | Like messages, direct methods define argument lists and return types, but | |
8ec911fa | 413 | they may also have a \emph{body}, and a \emph{rôle}. |
3cc520db MW |
414 | |
415 | A direct method need not have the same argument list or return type as its | |
416 | message. The acceptable argument lists and return types for a method depend | |
417 | on the message, in particular its method combination | |
8ec911fa | 418 | (\xref{sec:concepts.methods.combination}), and the method's rôle. |
3cc520db MW |
419 | |
420 | A direct method body is a block of C code, and the Sod translator usually | |
421 | defines, for each direct method, a function with external linkage, whose body | |
422 | contains a copy of the direct method body. Within the body of a direct | |
423 | method defined for a class $C$, the variable @|me|, of type pointer to class | |
424 | type of $C$, refers to the receiving object. | |
425 | ||
0a2d4b68 | 426 | |
3cc520db MW |
427 | \subsection{Effective methods and method combinations} |
428 | \label{sec:concepts.methods.combination} | |
429 | ||
430 | For each message a direct instance of a class might receive, there is a set | |
431 | of \emph{applicable methods}, which are exactly the direct methods defined on | |
432 | the object's class and its superclasses. These direct methods are combined | |
433 | together to form the \emph{effective method} for that particular class and | |
434 | message. Direct methods can be combined into an effective method in | |
435 | different ways, according to the \emph{method combination} specified by the | |
8ec911fa MW |
436 | message. The method combination determines which direct method rôles are |
437 | acceptable, and, for each rôle, the appropriate argument lists and return | |
3cc520db MW |
438 | types. |
439 | ||
440 | One direct method, $M$, is said to be more (resp.\ less) \emph{specific} than | |
441 | another, $N$, with respect to a receiving class~$C$, if the class defining | |
442 | $M$ is a more (resp.\ less) specific superclass of~$C$ than the class | |
443 | defining $N$. | |
444 | ||
43073476 | 445 | \subsubsection{The standard method combination} |
3cc520db MW |
446 | The default method combination is called the \emph{standard method |
447 | combination}; other method combinations are useful occasionally for special | |
8ec911fa | 448 | effects. The standard method combination accepts four direct method rôles, |
9761db0d | 449 | called `primary' (the default), @|before|, @|after|, and @|around|. |
3cc520db MW |
450 | |
451 | All direct methods subject to the standard method combination must have | |
452 | argument lists which \emph{match} the message's argument list: | |
453 | \begin{itemize} | |
454 | \item the method's arguments must have the same types as the message, though | |
455 | the arguments may have different names; and | |
456 | \item if the message accepts a variable-length argument suffix then the | |
457 | direct method must instead have a final argument of type @|va_list|. | |
458 | \end{itemize} | |
b1254eb6 MW |
459 | Primary and @|around| methods must have the same return type as the message; |
460 | @|before| and @|after| methods must return @|void| regardless of the | |
461 | message's return type. | |
3cc520db MW |
462 | |
463 | If there are no applicable primary methods then no effective method is | |
464 | constructed: the vtables contain null pointers in place of pointers to method | |
465 | entry functions. | |
466 | ||
f1aa19a8 MW |
467 | \begin{figure} |
468 | \begin{tikzpicture} | |
469 | [>=stealth, thick, | |
470 | order/.append style={color=green!70!black}, | |
471 | code/.append style={font=\sffamily}, | |
472 | action/.append style={font=\itshape}, | |
473 | method/.append style={rectangle, draw=black, thin, fill=blue!30, | |
474 | text height=\ht\strutbox, text depth=\dp\strutbox, | |
475 | minimum width=40mm}] | |
476 | ||
477 | \def\delgstack#1#2#3{ | |
478 | \node (#10) [method, #2] {#3}; | |
479 | \node (#11) [method, above=6mm of #10] {#3}; | |
480 | \draw [->] ($(#10.north)!.5!(#10.north west) + (0mm, 1mm)$) -- | |
481 | ++(0mm, 4mm) | |
482 | node [code, left=4pt, midway] {next_method}; | |
483 | \draw [<-] ($(#10.north)!.5!(#10.north east) + (0mm, 1mm)$) -- | |
484 | ++(0mm, 4mm) | |
485 | node [action, right=4pt, midway] {return}; | |
486 | \draw [->] ($(#11.north)!.5!(#11.north west) + (0mm, 1mm)$) -- | |
487 | ++(0mm, 4mm) | |
488 | node [code, left=4pt, midway] {next_method} | |
489 | node (ld) [above] {$\smash\vdots\mathstrut$}; | |
490 | \draw [<-] ($(#11.north)!.5!(#11.north east) + (0mm, 1mm)$) -- | |
491 | ++(0mm, 4mm) | |
492 | node [action, right=4pt, midway] {return} | |
493 | node (rd) [above] {$\smash\vdots\mathstrut$}; | |
494 | \draw [->] ($(ld.north) + (0mm, 1mm)$) -- ++(0mm, 4mm) | |
495 | node [code, left=4pt, midway] {next_method}; | |
496 | \draw [<-] ($(rd.north) + (0mm, 1mm)$) -- ++(0mm, 4mm) | |
497 | node [action, right=4pt, midway] {return}; | |
498 | \node (p) at ($(ld.north)!.5!(rd.north)$) {}; | |
499 | \node (#1n) [method, above=5mm of p] {#3}; | |
500 | \draw [->, order] ($(#10.south east) + (4mm, 1mm)$) -- | |
501 | ($(#1n.north east) + (4mm, -1mm)$) | |
502 | node [midway, right, align=left] | |
503 | {Most to \\ least \\ specific};} | |
504 | ||
dc20d91f | 505 | \delgstack{a}{}{@|around| method} |
f1aa19a8 MW |
506 | \draw [<-] ($(a0.south)!.5!(a0.south west) - (0mm, 1mm)$) -- |
507 | ++(0mm, -4mm); | |
508 | \draw [->] ($(a0.south)!.5!(a0.south east) - (0mm, 1mm)$) -- | |
509 | ++(0mm, -4mm) | |
510 | node [action, right=4pt, midway] {return}; | |
511 | ||
512 | \draw [->] ($(an.north)!.6!(an.north west) + (0mm, 1mm)$) -- | |
513 | ++(-8mm, 8mm) | |
514 | node [code, midway, left=3mm] {next_method} | |
515 | node (b0) [method, above left = 1mm + 4mm and -6mm - 4mm] {}; | |
516 | \node (b1) [method] at ($(b0) - (2mm, 2mm)$) {}; | |
dc20d91f | 517 | \node (bn) [method] at ($(b1) - (2mm, 2mm)$) {@|before| method}; |
f1aa19a8 MW |
518 | \draw [->, order] ($(bn.west) - (6mm, 0mm)$) -- ++(12mm, 12mm) |
519 | node [midway, above left, align=center] {Most to \\ least \\ specific}; | |
520 | \draw [->] ($(b0.north east) + (-10mm, 1mm)$) -- ++(8mm, 8mm) | |
521 | node (p) {}; | |
522 | ||
523 | \delgstack{m}{above right=1mm and 0mm of an.west |- p}{Primary method} | |
524 | \draw [->] ($(mn.north)!.5!(mn.north west) + (0mm, 1mm)$) -- ++(0mm, 4mm) | |
525 | node [code, left=4pt, midway] {next_method} | |
526 | node [above right = 0mm and -8mm] | |
527 | {$\vcenter{\hbox{\Huge\textcolor{red}{!}}} | |
528 | \vcenter{\hbox{\begin{tabular}[c]{l} | |
529 | \textsf{next_method} \\ | |
530 | pointer is null | |
531 | \end{tabular}}}$}; | |
532 | ||
533 | \draw [->, color=blue, dotted] | |
534 | ($(m0.south)!.2!(m0.south east) - (0mm, 1mm)$) -- | |
535 | ($(an.north)!.2!(an.north east) + (0mm, 1mm)$) | |
536 | node [midway, sloped, below] {Return value}; | |
537 | ||
538 | \draw [<-] ($(an.north)!.6!(an.north east) + (0mm, 1mm)$) -- | |
539 | ++(8mm, 8mm) | |
540 | node [action, midway, right=3mm] {return} | |
541 | node (f0) [method, above right = 1mm and -6mm] {}; | |
542 | \node (f1) [method] at ($(f0) + (-2mm, 2mm)$) {}; | |
dc20d91f | 543 | \node (fn) [method] at ($(f1) + (-2mm, 2mm)$) {@|after| method}; |
f1aa19a8 MW |
544 | \draw [<-, order] ($(f0.east) + (6mm, 0mm)$) -- ++(-12mm, 12mm) |
545 | node [midway, above right, align=center] | |
546 | {Least to \\ most \\ specific}; | |
547 | \draw [<-] ($(fn.north west) + (6mm, 1mm)$) -- ++(-8mm, 8mm); | |
548 | ||
549 | \end{tikzpicture} | |
550 | ||
551 | \caption{The standard method combination} | |
552 | \label{fig:concepts.methods.stdmeth} | |
553 | \end{figure} | |
554 | ||
3cc520db | 555 | The effective method for a message with standard method combination works as |
f1aa19a8 | 556 | follows (see also~\xref{fig:concepts.methods.stdmeth}). |
3cc520db MW |
557 | \begin{enumerate} |
558 | ||
8ec911fa | 559 | \item If any applicable methods have the @|around| rôle, then the most |
3cc520db MW |
560 | specific such method, with respect to the class of the receiving object, is |
561 | invoked. | |
562 | ||
b1254eb6 | 563 | Within the body of an @|around| method, the variable @|next_method| is |
3cc520db MW |
564 | defined, having pointer-to-function type. The method may call this |
565 | function, as described below, any number of times. | |
566 | ||
b1254eb6 MW |
567 | If there any remaining @|around| methods, then @|next_method| invokes the |
568 | next most specific such method, returning whichever value that method | |
dc20d91f MW |
569 | returns; otherwise the behaviour of @|next_method| is to invoke the |
570 | @|before| methods (if any), followed by the most specific primary method, | |
b0563651 | 571 | followed by the @|after| methods (if any), and to return whichever value |
dc20d91f MW |
572 | was returned by the most specific primary method, as described in the |
573 | following items. That is, the behaviour of the least specific @|around| | |
574 | method's @|next_method| function is exactly the behaviour that the | |
575 | effective method would have if there were no @|around| methods. Note that | |
576 | if the least-specific @|around| method calls its @|next_method| more than | |
577 | once then the whole sequence of @|before|, primary, and @|after| methods | |
578 | occurs multiple times. | |
3cc520db | 579 | |
b1254eb6 MW |
580 | The value returned by the most specific @|around| method is the value |
581 | returned by the effective method. | |
3cc520db | 582 | |
8ec911fa | 583 | \item If any applicable methods have the @|before| rôle, then they are all |
3cc520db MW |
584 | invoked, starting with the most specific. |
585 | ||
586 | \item The most specific applicable primary method is invoked. | |
587 | ||
588 | Within the body of a primary method, the variable @|next_method| is | |
589 | defined, having pointer-to-function type. If there are no remaining less | |
590 | specific primary methods, then @|next_method| is a null pointer. | |
591 | Otherwise, the method may call the @|next_method| function any number of | |
592 | times. | |
593 | ||
594 | The behaviour of the @|next_method| function, if it is not null, is to | |
595 | invoke the next most specific applicable primary method, and to return | |
596 | whichever value that method returns. | |
597 | ||
b1254eb6 MW |
598 | If there are no applicable @|around| methods, then the value returned by |
599 | the most specific primary method is the value returned by the effective | |
600 | method; otherwise the value returned by the most specific primary method is | |
601 | returned to the least specific @|around| method, which called it via its | |
602 | own @|next_method| function. | |
3cc520db | 603 | |
8ec911fa | 604 | \item If any applicable methods have the @|after| rôle, then they are all |
3cc520db | 605 | invoked, starting with the \emph{least} specific. (Hence, the most |
b1254eb6 | 606 | specific @|after| method is invoked with the most `afterness'.) |
3cc520db MW |
607 | |
608 | \end{enumerate} | |
609 | ||
b1254eb6 MW |
610 | A typical use for @|around| methods is to allow a base class to set up the |
611 | dynamic environment appropriately for the primary methods of its subclasses, | |
756e9293 | 612 | e.g., by claiming a lock, and releasing it afterwards. |
3cc520db | 613 | |
9761db0d | 614 | The @|next_method| function provided to methods with the primary and |
8ec911fa | 615 | @|around| rôles accepts the same arguments, and returns the same type, as the |
3cc520db MW |
616 | message, except that one or two additional arguments are inserted at the |
617 | front of the argument list. The first additional argument is always the | |
618 | receiving object, @|me|. If the message accepts a variable argument suffix, | |
619 | then the second addition argument is a @|va_list|; otherwise there is no | |
620 | second additional argument; otherwise, In the former case, a variable | |
621 | @|sod__master_ap| of type @|va_list| is defined, containing a separate copy | |
622 | of the argument pointer (so the method body can process the variable argument | |
623 | suffix itself, and still pass a fresh copy on to the next method). | |
624 | ||
8ec911fa | 625 | A method with the primary or @|around| rôle may use the convenience macro |
3cc520db MW |
626 | @|CALL_NEXT_METHOD|, which takes no arguments itself, and simply calls |
627 | @|next_method| with appropriate arguments: the receiver @|me| pointer, the | |
628 | argument pointer @|sod__master_ap| (if applicable), and the method's | |
629 | arguments. If the method body has overwritten its formal arguments, then | |
630 | @|CALL_NEXT_METHOD| will pass along the updated values, rather than the | |
631 | original ones. | |
632 | ||
781a8fbd MW |
633 | A primary or @|around| method which invokes its @|next_method| function is |
634 | said to \emph{extend} the message behaviour; a method which does not invoke | |
635 | its @|next_method| is said to \emph{override} the behaviour. Note that a | |
636 | method may make a decision to override or extend at runtime. | |
637 | ||
43073476 | 638 | \subsubsection{Aggregating method combinations} |
3cc520db MW |
639 | A number of other method combinations are provided. They are called |
640 | `aggregating' method combinations because, instead of invoking just the most | |
641 | specific primary method, as the standard method combination does, they invoke | |
642 | the applicable primary methods in turn and aggregate the return values from | |
643 | each. | |
644 | ||
8ec911fa | 645 | The aggregating method combinations accept the same four rôles as the |
b1254eb6 MW |
646 | standard method combination, and @|around|, @|before|, and @|after| methods |
647 | work in the same way. | |
3cc520db MW |
648 | |
649 | The aggregating method combinations provided are as follows. | |
650 | \begin{description} \let\makelabel\code | |
651 | \item[progn] The message must return @|void|. The applicable primary methods | |
652 | are simply invoked in turn, most specific first. | |
653 | \item[sum] The message must return a numeric type.\footnote{% | |
654 | The Sod translator does not check this, since it doesn't have enough | |
655 | insight into @|typedef| names.} % | |
656 | The applicable primary methods are invoked in turn, and their return values | |
657 | added up. The final result is the sum of the individual values. | |
658 | \item[product] The message must return a numeric type. The applicable | |
659 | primary methods are invoked in turn, and their return values multiplied | |
660 | together. The final result is the product of the individual values. | |
661 | \item[min] The message must return a scalar type. The applicable primary | |
662 | methods are invoked in turn. The final result is the smallest of the | |
663 | individual values. | |
664 | \item[max] The message must return a scalar type. The applicable primary | |
665 | methods are invoked in turn. The final result is the largest of the | |
666 | individual values. | |
665a0455 MW |
667 | \item[and] The message must return a scalar type. The applicable primary |
668 | methods are invoked in turn. If any method returns zero then the final | |
669 | result is zero and no further methods are invoked. If all of the | |
670 | applicable primary methods return nonzero, then the final result is the | |
671 | result of the last primary method. | |
672 | \item[or] The message must return a scalar type. The applicable primary | |
673 | methods are invoked in turn. If any method returns nonzero then the final | |
674 | result is that nonzero value and no further methods are invoked. If all of | |
675 | the applicable primary methods return zero, then the final result is zero. | |
3cc520db MW |
676 | \end{description} |
677 | ||
678 | There is also a @|custom| aggregating method combination, which is described | |
679 | in \xref{sec:fixme.custom-aggregating-method-combination}. | |
680 | ||
43073476 | 681 | |
caa6f4b9 MW |
682 | \subsection{Sending messages in C} \label{sec:concepts.methods.c} |
683 | ||
684 | Each instance is associated with its direct class [FIXME] | |
685 | ||
686 | The effective methods for each class are determined at translation time, by | |
687 | the Sod translator. For each effective method, one or more \emph{method | |
688 | entry functions} are constructed. A method entry function has three | |
689 | responsibilities. | |
690 | \begin{itemize} | |
691 | \item It converts the receiver pointer to the correct type. Method entry | |
692 | functions can perform these conversions extremely efficiently: there are | |
693 | separate method entries for each chain of each class which can receive a | |
694 | message, so method entry functions are in the privileged situation of | |
695 | knowing the \emph{exact} class of the receiving object. | |
696 | \item If the message accepts a variable-length argument tail, then two method | |
697 | entry functions are created for each chain of each class: one receives a | |
698 | variable-length argument tail, as intended, and captures it in a @|va_list| | |
699 | object; the other accepts an argument of type @|va_list| in place of the | |
700 | variable-length tail and arranges for it to be passed along to the direct | |
701 | methods. | |
702 | \item It invokes the effective method with the appropriate arguments. There | |
703 | might or might not be an actual function corresponding to the effective | |
704 | method itself: the translator may instead open-code the effective method's | |
705 | behaviour into each method entry function; and the machinery for handling | |
706 | `delegation chains', such as is used for @|around| methods and primary | |
707 | methods in the standard method combination, is necessarily scattered among | |
708 | a number of small functions. | |
709 | \end{itemize} | |
710 | ||
711 | ||
43073476 MW |
712 | \subsection{Messages with keyword arguments} |
713 | \label{sec:concepts.methods.keywords} | |
714 | ||
715 | A message or a direct method may declare that it accepts keyword arguments. | |
716 | A message which accepts keyword arguments is called a \emph{keyword message}; | |
717 | a direct method which accepts keyword arguments is called a \emph{keyword | |
718 | method}. | |
719 | ||
720 | While method combinations may set their own rules, usually keyword methods | |
721 | can only be defined on keyword messages, and all methods defined on a keyword | |
722 | message must be keyword methods. The direct methods defined on a keyword | |
723 | message may differ in the keywords they accept, both from each other, and | |
724 | from the message. If two superclasses of some common class both define | |
725 | keyword methods on the same message, and the methods both accept a keyword | |
726 | argument with the same name, then these two keyword arguments must also have | |
727 | the same type. Different applicable methods may declare keyword arguments | |
728 | with the same name but different defaults; see below. | |
729 | ||
730 | The keyword arguments acceptable in a message sent to an object are the | |
731 | keywords listed in the message definition, together with all of the keywords | |
732 | accepted by any applicable method. There is no easy way to determine at | |
733 | runtime whether a particular keyword is acceptable in a message to a given | |
734 | instance. | |
735 | ||
736 | At runtime, a direct method which accepts one or more keyword arguments | |
737 | receives an additional argument named @|suppliedp|. This argument is a small | |
738 | structure. For each keyword argument named $k$ accepted by the direct | |
739 | method, @|suppliedp| contains a one-bit-wide bitfield member of type | |
740 | @|unsigned|, also named $k$. If a keyword argument named $k$ was passed in | |
741 | the message, then @|suppliedp.$k$| is one, and $k$ contains the argument | |
742 | value; otherwise @|suppliedp.$k$| is zero, and $k$ contains the default value | |
743 | from the direct method definition if there was one, or an unspecified value | |
744 | otherwise. | |
745 | ||
3cc520db | 746 | %%%-------------------------------------------------------------------------- |
d24d47f5 MW |
747 | \section{The object lifecycle} \label{sec:concepts.lifecycle} |
748 | ||
749 | \subsection{Creation} \label{sec:concepts.lifecycle.birth} | |
750 | ||
751 | Construction of a new instance of a class involves three steps. | |
752 | \begin{enumerate} | |
753 | \item \emph{Allocation} arranges for there to be storage space for the | |
754 | instance's slots and associated metadata. | |
755 | \item \emph{Imprinting} fills in the instance's metadata, associating the | |
756 | instance with its class. | |
757 | \item \emph{Initialization} stores appropriate initial values in the | |
758 | instance's slots, and maybe links it into any external data structures as | |
759 | necessary. | |
760 | \end{enumerate} | |
761 | The \descref{SOD_DECL}[macro]{mac} handles constructing instances with | |
a42893dd MW |
762 | automatic storage duration (`on the stack'). Similarly, the |
763 | \descref{SOD_MAKE}[macro]{mac} and the \descref{sod_make}{fun} and | |
764 | \descref{sod_makev}{fun} functions construct instances allocated from the | |
765 | standard @|malloc| heap. Programmers can add support for other allocation | |
766 | strategies by using the \descref{SOD_INIT}[macro]{mac} and the | |
767 | \descref{sod_init}{fun} and \descref{sod_initv}{fun} functions, which package | |
768 | up imprinting and initialization. | |
d24d47f5 MW |
769 | |
770 | \subsubsection{Allocation} | |
771 | Instances of most classes (specifically including those classes defined by | |
772 | Sod itself) can be held in any storage of sufficient size. The in-memory | |
773 | layout of an instance of some class~$C$ is described by the type @|struct | |
774 | $C$__ilayout|, and if the relevant class is known at compile time then the | |
775 | best way to discover the layout size is with the @|sizeof| operator. Failing | |
776 | that, the size required to hold an instance of $C$ is available in a slot in | |
777 | $C$'s class object, as @|$C$__class@->cls.initsz|. | |
778 | ||
779 | It is not in general sufficient to declare, or otherwise allocate, an object | |
780 | of the class type $C$. The class type only describes a single chain of the | |
781 | object's layout. It is nearly always an error to use the class type as if it | |
782 | is a \emph{complete type}, e.g., to declare objects or arrays of the class | |
783 | type, or to enquire about its size or alignment requirements. | |
784 | ||
785 | Instance layouts may be declared as objects with automatic storage duration | |
786 | (colloquially, `allocated on the stack') or allocated dynamically, e.g., | |
787 | using @|malloc|. They may be included as members of structures or unions, or | |
788 | elements of arrays. Sod's runtime system doesn't retain addresses of | |
789 | instances, so, for example, Sod doesn't make using fancy allocators which | |
790 | sometimes move objects around in memory any more difficult than it needs to | |
791 | be. | |
792 | ||
793 | There isn't any way to discover the alignment required for a particular | |
794 | class's instances at runtime; it's best to be conservative and assume that | |
795 | the platform's strictest alignment requirement applies. | |
796 | ||
797 | The following simple function correctly allocates and returns space for an | |
798 | instance of a class given a pointer to its class object @<cls>. | |
799 | \begin{prog} | |
020b9e2b | 800 | void *allocate_instance(const SodClass *cls) \\ \ind |
d24d47f5 MW |
801 | \{ return malloc(cls@->cls.initsz); \} |
802 | \end{prog} | |
803 | ||
804 | \subsubsection{Imprinting} | |
805 | Once storage has been allocated, it must be \emph{imprinted} before it can be | |
806 | used as an instance of a class, e.g., before any messages can be sent to it. | |
807 | ||
808 | Imprinting an instance stores some metadata about its direct class in the | |
809 | instance structure, so that the rest of the program (and Sod's runtime | |
810 | library) can tell what sort of object it is, and how to use it.\footnote{% | |
811 | Specifically, imprinting an instance's storage involves storing the | |
812 | appropriate vtable pointers in the right places in it.} % | |
813 | A class object's @|imprint| slot points to a function which will correctly | |
814 | imprint storage for one of that class's instances. | |
815 | ||
816 | Once an instance's storage has been imprinted, it is technically possible to | |
817 | send messages to the instance; however the instance's slots are still | |
756e9293 MW |
818 | uninitialized at this point, so the applicable methods are unlikely to do |
819 | much of any use unless they've been written specifically for the purpose. | |
d24d47f5 MW |
820 | |
821 | The following simple function imprints storage at address @<p> as an instance | |
822 | of a class, given a pointer to its class object @<cls>. | |
823 | \begin{prog} | |
020b9e2b | 824 | void imprint_instance(const SodClass *cls, void *p) \\ \ind |
d24d47f5 MW |
825 | \{ cls@->cls.imprint(p); \} |
826 | \end{prog} | |
827 | ||
828 | \subsubsection{Initialization} | |
829 | The final step for constructing a new instance is to \emph{initialize} it, to | |
830 | establish the necessary invariants for the instance itself and the | |
831 | environment in which it operates. | |
832 | ||
833 | Details of initialization are necessarily class-specific, but typically it | |
834 | involves setting the instance's slots to appropriate values, and possibly | |
d1b394fa MW |
835 | linking it into some larger data structure to keep track of it. It is |
836 | possible for initialization methods to attempt to allocate resources, but | |
837 | this must be done carefully: there is currently no way to report an error | |
838 | from object initialization, so the object must be marked as incompletely | |
839 | initialized, and left in a state where it will be safe to tear down later. | |
d24d47f5 | 840 | |
a142609c MW |
841 | Initialization is performed by sending the imprinted instance an @|init| |
842 | message, defined by the @|SodObject| class. This message uses a nonstandard | |
843 | method combination which works like the standard combination, except that the | |
844 | \emph{default behaviour}, if there is no overriding method, is to initialize | |
b2983f35 MW |
845 | the instance's slots, as described below, and to invoke each superclass's |
846 | initialization fragments. This default behaviour may be invoked multiple | |
847 | times if some method calls on its @|next_method| more than once, unless some | |
848 | other method takes steps to prevent this. | |
a142609c | 849 | |
27ec3825 MW |
850 | Slots are initialized in a well-defined order. |
851 | \begin{itemize} | |
054e8f8f MW |
852 | \item Slots defined by a more specific superclass are initialized after slots |
853 | defined by a less specific superclass. | |
27ec3825 MW |
854 | \item Slots defined by the same class are initialized in the order in which |
855 | their definitions appear. | |
856 | \end{itemize} | |
857 | ||
a42893dd MW |
858 | A class can define \emph{initialization fragments}: pieces of literal code to |
859 | be executed to set up a new instance. Each superclass's initialization | |
860 | fragments are executed with @|me| bound to an instance pointer of the | |
861 | appropriate superclass type, immediately after that superclass's slots (if | |
862 | any) have been initialized; therefore, fragments defined by a more specific | |
13cb243a | 863 | superclass are executed after fragments defined by a less specific |
a42893dd MW |
864 | superclass. A class may define more than one initialization fragment: the |
865 | fragments are executed in the order in which they appear in the class | |
866 | definition. It is possible for an initialization fragment to use @|return| | |
867 | or @|goto| for special control-flow effects, but this is not likely to be a | |
868 | good idea. | |
869 | ||
b2983f35 MW |
870 | The @|init| message accepts keyword arguments |
871 | (\xref{sec:concepts.methods.keywords}). The set of acceptable keywords is | |
872 | determined by the applicable methods as usual, but also by the | |
873 | \emph{initargs} defined by the receiving instance's class and its | |
874 | superclasses, which are made available to slot initializers and | |
875 | initialization fragments. | |
876 | ||
877 | There are two kinds of initarg definitions. \emph{User initargs} are defined | |
878 | by an explicit @|initarg| item appearing in a class definition: the item | |
879 | defines a name, type, and (optionally) a default value for the initarg. | |
880 | \emph{Slot initargs} are defined by attaching an @|initarg| property to a | |
756e9293 MW |
881 | slot or slot initializer item: the property's value determines the initarg's |
882 | name, while the type is taken from the underlying slot type; slot initargs do | |
883 | not have default values. Both kinds define a \emph{direct initarg} for the | |
b2983f35 MW |
884 | containing class. |
885 | ||
886 | Initargs are inherited. The \emph{applicable} direct initargs for an @|init| | |
887 | effective method are those defined by the receiving object's class, and all | |
888 | of its superclasses. Applicable direct initargs with the same name are | |
889 | merged to form \emph{effective initargs}. An error is reported if two | |
890 | applicable direct initargs have the same name but different types. The | |
891 | default value of an effective initarg is taken from the most specific | |
892 | applicable direct initarg which specifies a defalt value; if no applicable | |
893 | direct initarg specifies a default value then the effective initarg has no | |
894 | default. | |
895 | ||
896 | All initarg values are made available at runtime to user code -- | |
897 | initialization fragments and slot initializer expressions -- through local | |
898 | variables and a @|suppliedp| structure, as in a direct method | |
899 | (\xref{sec:concepts.methods.keywords}). Furthermore, slot initarg | |
900 | definitions influence the initialization of slots. | |
901 | ||
902 | The process for deciding how to initialize a particular slot works as | |
903 | follows. | |
904 | \begin{enumerate} | |
905 | \item If there are any slot initargs defined on the slot, or any of its slot | |
906 | initializers, \emph{and} the sender supplied a value for one or more of the | |
907 | corresponding effective initargs, then the value of the most specific slot | |
908 | initarg is stored in the slot. | |
909 | \item Otherwise, if there are any slot initializers defined which include an | |
910 | initializer expression, then the initializer expression from the most | |
911 | specific such slot initializer is evaluated and its value stored in the | |
912 | slot. | |
913 | \item Otherwise, the slot is left uninitialized. | |
914 | \end{enumerate} | |
915 | Note that the default values (if any) of effective initargs do \emph{not} | |
916 | affect this procedure. | |
d24d47f5 | 917 | |
d24d47f5 MW |
918 | |
919 | \subsection{Destruction} | |
920 | \label{sec:concepts.lifecycle.death} | |
921 | ||
922 | Destruction of an instance, when it is no longer required, consists of two | |
923 | steps. | |
924 | \begin{enumerate} | |
925 | \item \emph{Teardown} releases any resources held by the instance and | |
926 | disentangles it from any external data structures. | |
927 | \item \emph{Deallocation} releases the memory used to store the instance so | |
928 | that it can be reused. | |
929 | \end{enumerate} | |
a42893dd MW |
930 | Teardown alone, for objects which require special deallocation, or for which |
931 | deallocation occurs automatically (e.g., instances with automatic storage | |
932 | duration, or instances whose storage will be garbage-collected), is performed | |
933 | using the \descref{sod_teardown}[function]{fun}. Destruction of instances | |
934 | allocated from the standard @|malloc| heap is done using the | |
935 | \descref{sod_destroy}[function]{fun}. | |
d24d47f5 MW |
936 | |
937 | \subsubsection{Teardown} | |
7646dc4c MW |
938 | Details of teardown are necessarily class-specific, but typically it |
939 | involves releasing resources held by the instance, and disentangling it from | |
940 | any data structures it might be linked into. | |
a42893dd MW |
941 | |
942 | Teardown is performed by sending the instance the @|teardown| message, | |
943 | defined by the @|SodObject| class. The message returns an integer, used as a | |
944 | boolean flag. If the message returns zero, then the instance's storage | |
945 | should be deallocated. If the message returns nonzero, then it is safe for | |
946 | the caller to forget about instance, but should not deallocate its storage. | |
947 | This is \emph{not} an error return: if some teardown method fails then the | |
948 | program may be in an inconsistent state and should not continue. | |
d24d47f5 | 949 | |
a42893dd MW |
950 | This simple protocol can be used, for example, to implement a reference |
951 | counting system, as follows. | |
d24d47f5 | 952 | \begin{prog} |
020b9e2b | 953 | [nick = ref] \\ |
d7451ac3 | 954 | class ReferenceCountedObject: SodObject \{ \\ \ind |
020b9e2b MW |
955 | unsigned nref = 1; \\- |
956 | void inc() \{ me@->ref.nref++; \} \\- | |
957 | [role = around] \\ | |
958 | int obj.teardown() \\ | |
959 | \{ \\ \ind | |
960 | if (--\,--me@->ref.nref) return (1); \\ | |
961 | else return (CALL_NEXT_METHOD); \-\\ | |
962 | \} \-\\ | |
d24d47f5 MW |
963 | \} |
964 | \end{prog} | |
965 | ||
fa7e2d72 MW |
966 | The @|teardown| message uses a nonstandard method combination which works |
967 | like the standard combination, except that the \emph{default behaviour}, if | |
968 | there is no overriding method, is to execute the superclass's teardown | |
969 | fragments, and to return zero. This default behaviour may be invoked | |
970 | multiple times if some method calls on its @|next_method| more than once, | |
971 | unless some other method takes steps to prevent this. | |
a42893dd MW |
972 | |
973 | A class can define \emph{teardown fragments}: pieces of literal code to be | |
974 | executed to shut down an instance. Each superclass's teardown fragments are | |
975 | executed with @|me| bound to an instance pointer of the appropriate | |
976 | superclass type; fragments defined by a more specific superclass are executed | |
13cb243a | 977 | before fragments defined by a less specific superclass. A class may define |
a42893dd MW |
978 | more than one teardown fragment: the fragments are executed in the order in |
979 | which they appear in the class definition. It is possible for an | |
980 | initialization fragment to use @|return| or @|goto| for special control-flow | |
981 | effects, but this is not likely to be a good idea. Similarly, it's probably | |
982 | a better idea to use an @|around| method to influence the return value than | |
983 | to write an explicit @|return| statement in a teardown fragment. | |
984 | ||
d24d47f5 MW |
985 | \subsubsection{Deallocation} |
986 | The details of instance deallocation are obviously specific to the allocation | |
987 | strategy used by the instance, and this is often orthogonal from the object's | |
988 | class. | |
989 | ||
990 | The code which makes the decision to destroy an object may often not be aware | |
991 | of the object's direct class. Low-level details of deallocation often | |
992 | require the proper base address of the instance's storage, which can be | |
993 | determined using the \descref{SOD_INSTBASE}[macro]{mac}. | |
994 | ||
d24d47f5 | 995 | %%%-------------------------------------------------------------------------- |
3cc520db | 996 | \section{Metaclasses} \label{sec:concepts.metaclasses} |
1f7d590d | 997 | |
caa6f4b9 MW |
998 | %%%-------------------------------------------------------------------------- |
999 | \section{Compatibility considerations} \label{sec:concepts.compatibility} | |
1000 | ||
1001 | Sod doesn't make source-level compatibility especially difficult. As long as | |
1002 | classes, slots, and messages don't change names or dissappear, and slots and | |
1003 | messages retain their approximate types, everything will be fine. | |
1004 | ||
1005 | Binary compatibility is much more difficult. Unfortunately, Sod classes have | |
1006 | rather fragile binary interfaces.\footnote{% | |
1007 | Research suggestion: investigate alternative instance and vtable layouts | |
1008 | which improve binary compatibility, probably at the expense of instance | |
1009 | compactness, and efficiency of slot access and message sending. There may | |
1010 | be interesting trade-offs to be made.} % | |
1011 | ||
1012 | If instances are allocated [FIXME] | |
1013 | ||
1f7d590d MW |
1014 | %%%----- That's all, folks -------------------------------------------------- |
1015 | ||
1016 | %%% Local variables: | |
1017 | %%% mode: LaTeX | |
1018 | %%% TeX-master: "sod.tex" | |
1019 | %%% TeX-PDF-mode: t | |
1020 | %%% End: |