source: doc/papers/general/Paper.tex @ d52a55b

aaron-thesisarm-ehcleanup-dtorsdeferred_resndemanglerjacob/cs343-translationjenkins-sandboxnew-astnew-ast-unique-exprnew-envno_listpersistent-indexerwith_gc
Last change on this file since d52a55b was d52a55b, checked in by Peter A. Buhr <pabuhr@…>, 4 years ago

description of CFA translator

  • Property mode set to 100644
File size: 158.0 KB
Line 
1\documentclass[AMA,STIX1COL]{WileyNJD-v2}
2
3\articletype{RESEARCH ARTICLE}%
4
5\received{26 April 2016}
6\revised{6 June 2016}
7\accepted{6 June 2016}
8
9\raggedbottom
10
11%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12
13% Latex packages used in the document.
14
15\usepackage{epic,eepic}
16\usepackage{xspace}
17\usepackage{comment}
18\usepackage{upquote}                                            % switch curled `'" to straight
19\usepackage{listings}                                           % format program code
20\captionsetup{justification=raggedright,singlelinecheck=false}
21%\usepackage{enumitem}
22%\setlist[itemize]{topsep=3pt,itemsep=2pt,parsep=0pt}% global
23%\usepackage{rotating}
24
25\hypersetup{breaklinks=true}
26\definecolor{ForestGreen}{cmyk}{1, 0, 0.99995, 0}
27
28\usepackage[pagewise]{lineno}
29\renewcommand{\linenumberfont}{\scriptsize\sffamily}
30
31\lefthyphenmin=3                                                        % hyphen only after 4 characters
32\righthyphenmin=3
33
34%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
35
36% Names used in the document.
37
38\newcommand{\CFAIcon}{\textsf{C}\raisebox{\depth}{\rotatebox{180}{\textsf{A}}}\xspace} % Cforall symbolic name
39\newcommand{\CFA}{\protect\CFAIcon}             % safe for section/caption
40\newcommand{\CFL}{\textrm{Cforall}\xspace}      % Cforall symbolic name
41\newcommand{\Celeven}{\textrm{C11}\xspace}      % C11 symbolic name
42\newcommand{\CC}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}\xspace} % C++ symbolic name
43\newcommand{\CCeleven}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}11\xspace} % C++11 symbolic name
44\newcommand{\CCfourteen}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}14\xspace} % C++14 symbolic name
45\newcommand{\CCseventeen}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}17\xspace} % C++17 symbolic name
46\newcommand{\CCtwenty}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}20\xspace} % C++20 symbolic name
47\newcommand{\CCV}{\rm C\kern-.1em\hbox{+\kern-.25em+}obj\xspace} % C++ virtual symbolic name
48\newcommand{\Csharp}{C\raisebox{-0.7ex}{\Large$^\sharp$}\xspace} % C# symbolic name
49
50%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
51
52\newcommand{\Textbf}[2][red]{{\color{#1}{\textbf{#2}}}}
53%\newcommand{\TODO}[1]{\textbf{TODO}: {\itshape #1}} % TODO included
54\newcommand{\TODO}[1]{} % TODO elided
55
56% Default underscore is too low and wide. Cannot use lstlisting "literate" as replacing underscore
57% removes it as a variable-name character so keywords in variables are highlighted. MUST APPEAR
58% AFTER HYPERREF.
59%\DeclareTextCommandDefault{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.1ex}}}
60\renewcommand{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.075ex}}}
61
62\makeatletter
63% parindent is relative, i.e., toggled on/off in environments like itemize, so store the value for
64% use rather than use \parident directly.
65\newlength{\parindentlnth}
66\setlength{\parindentlnth}{\parindent}
67
68\newcommand{\LstBasicStyle}[1]{{\lst@basicstyle{\lst@basicstyle{#1}}}}
69\newcommand{\LstKeywordStyle}[1]{{\lst@basicstyle{\lst@keywordstyle{#1}}}}
70\newcommand{\LstCommentStyle}[1]{{\lst@basicstyle{\lst@commentstyle{#1}}}}
71
72\newlength{\gcolumnposn}                                        % temporary hack because lstlisting does not handle tabs correctly
73\newlength{\columnposn}
74\setlength{\gcolumnposn}{3.5in}
75\setlength{\columnposn}{\gcolumnposn}
76
77\newcommand{\C}[2][\@empty]{\ifx#1\@empty\else\global\setlength{\columnposn}{#1}\global\columnposn=\columnposn\fi\hfill\makebox[\textwidth-\columnposn][l]{\lst@basicstyle{\LstCommentStyle{#2}}}}
78\newcommand{\CRT}{\global\columnposn=\gcolumnposn}
79
80% Denote newterms in particular font and index them without particular font and in lowercase, e.g., \newterm{abc}.
81% The option parameter provides an index term different from the new term, e.g., \newterm[\texttt{abc}]{abc}
82% The star version does not lowercase the index information, e.g., \newterm*{IBM}.
83\newcommand{\newtermFontInline}{\emph}
84\newcommand{\newterm}{\@ifstar\@snewterm\@newterm}
85\newcommand{\@newterm}[2][\@empty]{\lowercase{\def\temp{#2}}{\newtermFontInline{#2}}\ifx#1\@empty\index{\temp}\else\index{#1@{\protect#2}}\fi}
86\newcommand{\@snewterm}[2][\@empty]{{\newtermFontInline{#2}}\ifx#1\@empty\index{#2}\else\index{#1@{\protect#2}}\fi}
87
88% Latin abbreviation
89\newcommand{\abbrevFont}{\textit}                       % set empty for no italics
90\@ifundefined{eg}{
91\newcommand{\EG}{\abbrevFont{e}\abbrevFont{g}}
92\newcommand*{\eg}{%
93        \@ifnextchar{,}{\EG}%
94                {\@ifnextchar{:}{\EG}%
95                        {\EG,\xspace}}%
96}}{}%
97\@ifundefined{ie}{
98\newcommand{\IE}{\abbrevFont{i}\abbrevFont{e}}
99\newcommand*{\ie}{%
100        \@ifnextchar{,}{\IE}%
101                {\@ifnextchar{:}{\IE}%
102                        {\IE,\xspace}}%
103}}{}%
104\@ifundefined{etc}{
105\newcommand{\ETC}{\abbrevFont{etc}}
106\newcommand*{\etc}{%
107        \@ifnextchar{.}{\ETC}%
108        {\ETC.\xspace}%
109}}{}%
110\@ifundefined{etal}{
111\newcommand{\ETAL}{\abbrevFont{et}~\abbrevFont{al}}
112\newcommand*{\etal}{%
113        \@ifnextchar{.}{\protect\ETAL}%
114                {\protect\ETAL.\xspace}%
115}}{}%
116\@ifundefined{viz}{
117\newcommand{\VIZ}{\abbrevFont{viz}}
118\newcommand*{\viz}{%
119        \@ifnextchar{.}{\VIZ}%
120                {\VIZ.\xspace}%
121}}{}%
122\makeatother
123
124\newenvironment{cquote}{%
125        \list{}{\lstset{resetmargins=true,aboveskip=0pt,belowskip=0pt}\topsep=3pt\parsep=0pt\leftmargin=\parindentlnth\rightmargin\leftmargin}%
126        \item\relax
127}{%
128        \endlist
129}% cquote
130
131% CFA programming language, based on ANSI C (with some gcc additions)
132\lstdefinelanguage{CFA}[ANSI]{C}{
133        morekeywords={
134                _Alignas, _Alignof, __alignof, __alignof__, asm, __asm, __asm__, __attribute, __attribute__,
135                auto, _Bool, catch, catchResume, choose, _Complex, __complex, __complex__, __const, __const__,
136                coroutine, disable, dtype, enable, exception, __extension__, fallthrough, fallthru, finally,
137                __float80, float80, __float128, float128, forall, ftype, _Generic, _Imaginary, __imag, __imag__,
138                inline, __inline, __inline__, __int128, int128, __label__, monitor, mutex, _Noreturn, one_t, or,
139                otype, restrict, __restrict, __restrict__, __signed, __signed__, _Static_assert, thread,
140                _Thread_local, throw, throwResume, timeout, trait, try, ttype, typeof, __typeof, __typeof__,
141                virtual, __volatile, __volatile__, waitfor, when, with, zero_t},
142        moredirectives={defined,include_next}%
143}
144
145\lstset{
146language=CFA,
147columns=fullflexible,
148basicstyle=\linespread{0.9}\sf,                                                 % reduce line spacing and use sanserif font
149stringstyle=\tt,                                                                                % use typewriter font
150tabsize=5,                                                                                              % N space tabbing
151xleftmargin=\parindentlnth,                                                             % indent code to paragraph indentation
152%mathescape=true,                                                                               % LaTeX math escape in CFA code $...$
153escapechar=\$,                                                                                  % LaTeX escape in CFA code
154keepspaces=true,                                                                                %
155showstringspaces=false,                                                                 % do not show spaces with cup
156showlines=true,                                                                                 % show blank lines at end of code
157aboveskip=4pt,                                                                                  % spacing above/below code block
158belowskip=3pt,
159% replace/adjust listing characters that look bad in sanserif
160literate={-}{\makebox[1ex][c]{\raisebox{0.4ex}{\rule{0.8ex}{0.1ex}}}}1 {^}{\raisebox{0.6ex}{$\scriptstyle\land\,$}}1
161        {~}{\raisebox{0.3ex}{$\scriptstyle\sim\,$}}1 % {`}{\ttfamily\upshape\hspace*{-0.1ex}`}1
162        {<-}{$\leftarrow$}2 {=>}{$\Rightarrow$}2 {->}{\makebox[1ex][c]{\raisebox{0.4ex}{\rule{0.8ex}{0.075ex}}}\kern-0.2ex{\textgreater}}2,
163moredelim=**[is][\color{red}]{`}{`},
164}% lstset
165
166\lstnewenvironment{cfa}[1][]
167{\lstset{#1}}
168{}
169\lstnewenvironment{C++}[1][]                            % use C++ style
170{\lstset{language=C++,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
171{}
172
173% inline code @...@
174\lstMakeShortInline@%
175
176
177\title{\texorpdfstring{\protect\CFA : Adding Modern Programming Language Features to C}{Cforall : Adding Modern Programming Language Features to C}}
178
179\author[1]{Aaron Moss}
180\author[1]{Robert Schluntz}
181\author[1]{Peter A. Buhr*}
182\authormark{MOSS \textsc{et al}}
183
184\address[1]{\orgdiv{Cheriton School of Computer Science}, \orgname{University of Waterloo}, \orgaddress{\state{Waterloo, ON}, \country{Canada}}}
185
186\corres{*Peter A. Buhr, Cheriton School of Computer Science, University of Waterloo, 200 University Avenue West, Waterloo, ON, N2L 3G1, Canada. \email{pabuhr{\char`\@}uwaterloo.ca}}
187
188\fundingInfo{Natural Sciences and Engineering Research Council of Canada}
189
190\abstract[Summary]{
191The C programming language is a foundational technology for modern computing with millions of lines of code implementing everything from commercial operating-systems to hobby projects.
192This installation base and the programmers producing it represent a massive software-engineering investment spanning decades and likely to continue for decades more.
193Nevertheless, C, first standardized over thirty years ago, lacks many features that make programming in more modern languages safer and more productive.
194The goal of the \CFA project is to create an extension of C that provides modern safety and productivity features while still ensuring strong backwards compatibility with C and its programmers.
195Prior projects have attempted similar goals but failed to honour C programming-style; for instance, adding object-oriented or functional programming with garbage collection is a non-starter for many C developers.
196Specifically, \CFA is designed to have an orthogonal feature-set based closely on the C programming paradigm, so that \CFA features can be added \emph{incrementally} to existing C code-bases, and C programmers can learn \CFA extensions on an as-needed basis, preserving investment in existing code and programmers.
197This paper presents a quick tour of \CFA features showing how their design avoids shortcomings of similar features in C and other C-like languages.
198Finally, experimental results are presented to validate several of the new features.
199}%
200
201\keywords{generic types, tuple types, variadic types, polymorphic functions, C, Cforall}
202
203
204\begin{document}
205\linenumbers                                            % comment out to turn off line numbering
206
207\maketitle
208
209
210\section{Introduction}
211The C programming language is a foundational technology for modern computing with millions of lines of code implementing everything from commercial operating-systems to hobby projects.
212This installation base and the programmers producing it represent a massive software-engineering investment spanning decades and likely to continue for decades more.
213The TIOBE~\cite{TIOBE} ranks the top 5 most \emph{popular} programming languages as: Java 15\%, \Textbf{C 12\%}, \Textbf{\CC 5.5\%}, Python 5\%, \Csharp 4.5\% = 42\%, where the next 50 languages are less than 4\% each with a long tail.
214The top 3 rankings over the past 30 years are:
215\begin{center}
216\setlength{\tabcolsep}{10pt}
217\lstDeleteShortInline@%
218\begin{tabular}{@{}rccccccc@{}}
219                & 2018  & 2013  & 2008  & 2003  & 1998  & 1993  & 1988  \\ \hline
220Java    & 1             & 2             & 1             & 1             & 18    & -             & -             \\
221\Textbf{C}& \Textbf{2} & \Textbf{1} & \Textbf{2} & \Textbf{2} & \Textbf{1} & \Textbf{1} & \Textbf{1} \\
222\CC             & 3             & 4             & 3             & 3             & 2             & 2             & 5             \\
223\end{tabular}
224\lstMakeShortInline@%
225\end{center}
226Love it or hate it, C is extremely popular, highly used, and one of the few systems languages.
227In many cases, \CC is often used solely as a better C.
228Nevertheless, C, first standardized over thirty years ago, lacks many features that make programming in more modern languages safer and more productive.
229
230\CFA (pronounced ``C-for-all'', and written \CFA or Cforall) is an evolutionary extension of the C programming language that adds modern language-features to C, while maintaining both source and runtime compatibility with C and a familiar programming model for programmers.
231The four key design goals for \CFA~\cite{Bilson03} are:
232(1) The behaviour of standard C code must remain the same when translated by a \CFA compiler as when translated by a C compiler;
233(2) Standard C code must be as fast and as small when translated by a \CFA compiler as when translated by a C compiler;
234(3) \CFA code must be at least as portable as standard C code;
235(4) Extensions introduced by \CFA must be translated in the most efficient way possible.
236These goals ensure existing C code-bases can be converted to \CFA incrementally with minimal effort, and C programmers can productively generate \CFA code without training beyond the features being used.
237\CC is used similarly, but has the disadvantages of multiple legacy design-choices that cannot be updated and active divergence of the language model from C, requiring significant effort and training to incrementally add \CC to a C-based project.
238
239All languages features discussed in this paper are working, except some advanced exception-handling features.
240Not discussed in this paper are the integrated concurrency-constructs and user-level threading-library~\cite{Delisle18}.
241\CFA is an \emph{open-source} project implemented as an source-to-source translator from \CFA to the gcc-dialect of C~\cite{GCCExtensions}, allowing it to leverage the portability and code optimizations provided by gcc, meeting goals (1)--(3).
242Ultimately, a compiler is necessary for advanced features and optimal performance.
243% @plg2[9]% cd cfa-cc/src; cloc ArgTweak CodeGen CodeTools Common Concurrency ControlStruct Designators GenPoly InitTweak MakeLibCfa.cc MakeLibCfa.h Parser ResolvExpr SymTab SynTree Tuples driver prelude main.cc
244% -------------------------------------------------------------------------------
245% Language                     files          blank        comment           code
246% -------------------------------------------------------------------------------
247% C++                            108           5420           5232          34961
248% C/C++ Header                    86           2379           2450           8464
249% Teamcenter def                   2            115             65           1387
250% make                             5            168             87           1052
251% C                               20            109            403            488
252% awk                              1             12             26            121
253% sed                              1              0              0              6
254% -------------------------------------------------------------------------------
255% SUM:                           223           8203           8263          46479
256% -------------------------------------------------------------------------------
257The \CFA translator is 200+ files and 46,000+ lines of code written in C/\CC.
258Starting with a translator versus a compiler makes it easier and faster to generate and debug C object-code rather than intermediate, assembler or machine code.
259The translator design is based on the \emph{visitor pattern}, allowing multiple passes over the abstract code-tree, which works well for incrementally adding new feature through additional visitor passes.
260At the heart of the translator is the type resolver, which handles the polymorphic routine/type overload-resolution.
261% @plg2[8]% cd cfa-cc/src; cloc libcfa
262% -------------------------------------------------------------------------------
263% Language                     files          blank        comment           code
264% -------------------------------------------------------------------------------
265% C                               35           1256           1240           9116
266% C/C++ Header                    54            358           1106           1198
267% make                             2            201            325           1167
268% C++                              3             18             17            124
269% Assembly                         3             56             97            111
270% Bourne Shell                     2              2              0             25
271% awk                              1              4              0             22
272% -------------------------------------------------------------------------------
273% SUM:                           100           1895           2785          11763
274% -------------------------------------------------------------------------------
275The \CFA runtime system is 100+ files and 11,000+ lines of code, written in \CFA.
276Currently, the \CFA runtime is the largest \emph{user} of \CFA providing a vehicle to test the language features and implementation.
277% @plg2[6]% cd cfa-cc/src; cloc tests examples benchmark
278% -------------------------------------------------------------------------------
279% Language                     files          blank        comment           code
280% -------------------------------------------------------------------------------
281% C                              237          12260           2869          23286
282% make                             8            464            245           2838
283% C/C++ Header                    22            225            175            785
284% Python                           5            131             93            420
285% C++                             10             48              5            201
286% Lua                              2             31              4            126
287% Java                             4              5              0             80
288% Go                               2             11              9             40
289% -------------------------------------------------------------------------------
290% SUM:                           290          13175           3400          27776
291% -------------------------------------------------------------------------------
292The \CFA tests are 290+ files and 27,000+ lines of code.
293The tests illustrate syntactic and semantic features in \CFA, plus a growing number of runtime benchmarks.
294The tests check for correctness and are used for daily regression testing of commits (3800+).
295
296Finally, it is impossible to describe a programming language without usages before definitions.
297Therefore, syntax and semantics appear before explanations;
298hence, patience is necessary until details are presented.
299
300
301\section{Polymorphic Functions}
302
303\CFA introduces both ad-hoc and parametric polymorphism to C, with a design originally formalized by Ditchfield~\cite{Ditchfield92}, and first implemented by Bilson~\cite{Bilson03}.
304Shortcomings are identified in existing approaches to generic and variadic data types in C-like languages and how these shortcomings are avoided in \CFA.
305Specifically, the solution is both reusable and type-checked, as well as conforming to the design goals of \CFA with ergonomic use of existing C abstractions.
306The new constructs are empirically compared with C and \CC approaches via performance experiments in Section~\ref{sec:eval}.
307
308
309\subsection{Name Overloading}
310\label{s:NameOverloading}
311
312\begin{quote}
313There are only two hard things in Computer Science: cache invalidation and \emph{naming things} -- Phil Karlton
314\end{quote}
315\vspace{-10pt}
316C already has a limited form of ad-hoc polymorphism in the form of its basic arithmetic operators, which apply to a variety of different types using identical syntax.
317\CFA extends the built-in operator overloading by allowing users to define overloads for any function, not just operators, and even any variable;
318Section~\ref{sec:libraries} includes a number of examples of how this overloading simplifies \CFA programming relative to C.
319Code generation for these overloaded functions and variables is implemented by the usual approach of mangling the identifier names to include a representation of their type, while \CFA decides which overload to apply based on the same ``usual arithmetic conversions'' used in C to disambiguate operator overloads.
320As an example:
321\begin{cfa}
322int max = 2147483647;                                           $\C[4in]{// (1)}$
323double max = 1.7976931348623157E+308;           $\C{// (2)}$
324int max( int a, int b ) { return a < b ? b : a; }  $\C{// (3)}$
325double max( double a, double b ) { return a < b ? b : a; }  $\C{// (4)}\CRT$
326max( 7, -max );                                         $\C[2.75in]{// uses (3) and (1), by matching int from constant 7}$
327max( max, 3.14 );                                       $\C{// uses (4) and (2), by matching double from constant 3.14}$
328max( max, -max );                                       $\C{// ERROR: ambiguous}$
329int m = max( max, -max );                       $\C{// uses (3) and (1) twice, by matching return type}\CRT$
330\end{cfa}
331
332\CFA maximizes the ability to reuse names to aggressively address the naming problem.
333In some cases, hundreds of names can be reduced to tens, resulting in a significant cognitive reduction.
334In the above, the name @max@ has a consistent meaning, and a programmer only needs to remember the single concept: maximum.
335To prevent significant ambiguities, \CFA uses the return type in selecting overloads, \eg in the assignment to @m@, the compiler use @m@'s type to unambiguously select the most appropriate call to function @max@ (as does Ada).
336As is shown later, there are a number of situations where \CFA takes advantage of available type information to disambiguate, where other programming languages generate ambiguities.
337
338\Celeven added @_Generic@ expressions, which is used in preprocessor macros to provide a form of ad-hoc polymorphism;
339however, this polymorphism is both functionally and ergonomically inferior to \CFA name overloading.
340The macro wrapping the generic expression imposes some limitations;
341\eg, it cannot implement the example above, because the variables @max@ are ambiguous with the functions @max@.
342Ergonomic limitations of @_Generic@ include the necessity to put a fixed list of supported types in a single place and manually dispatch to appropriate overloads, as well as possible namespace pollution from the dispatch functions, which must all have distinct names.
343For backwards compatibility, \CFA supports @_Generic@ expressions, but it is an unnecessary mechanism. \TODO{actually implement that}
344
345% http://fanf.livejournal.com/144696.html
346% http://www.robertgamble.net/2012/01/c11-generic-selections.html
347% https://abissell.com/2014/01/16/c11s-_generic-keyword-macro-applications-and-performance-impacts/
348
349
350\subsection{\texorpdfstring{\protect\lstinline{forall} Functions}{forall Functions}}
351\label{sec:poly-fns}
352
353The signature feature of \CFA is parametric-polymorphic functions~\cite{forceone:impl,Cormack90,Duggan96} with functions generalized using a @forall@ clause (giving the language its name):
354\begin{cfa}
355`forall( otype T )` T identity( T val ) { return val; }
356int forty_two = identity( 42 );         $\C{// T is bound to int, forty\_two == 42}$
357\end{cfa}
358This @identity@ function can be applied to any complete \newterm{object type} (or @otype@).
359The type variable @T@ is transformed into a set of additional implicit parameters encoding sufficient information about @T@ to create and return a variable of that type.
360The \CFA implementation passes the size and alignment of the type represented by an @otype@ parameter, as well as an assignment operator, constructor, copy constructor and destructor.
361If this extra information is not needed, \eg for a pointer, the type parameter can be declared as a \newterm{data type} (or @dtype@).
362
363In \CFA, the polymorphic runtime-cost is spread over each polymorphic call, because more arguments are passed to polymorphic functions;
364the experiments in Section~\ref{sec:eval} show this overhead is similar to \CC virtual-function calls.
365A design advantage is that, unlike \CC template-functions, \CFA polymorphic-functions are compatible with C \emph{separate compilation}, preventing compilation and code bloat.
366
367Since bare polymorphic-types provide a restricted set of available operations, \CFA provides a \newterm{type assertion}~\cite[pp.~37-44]{Alphard} mechanism to provide further type information, where type assertions may be variable or function declarations that depend on a polymorphic type-variable.
368For example, the function @twice@ can be defined using the \CFA syntax for operator overloading:
369\begin{cfa}
370forall( otype T `| { T ?+?(T, T); }` ) T twice( T x ) { return x `+` x; }  $\C{// ? denotes operands}$
371int val = twice( twice( 3.7 ) );
372\end{cfa}
373which works for any type @T@ with a matching addition operator.
374The polymorphism is achieved by creating a wrapper function for calling @+@ with @T@ bound to @double@, then passing this function to the first call of @twice@.
375There is now the option of using the same @twice@ and converting the result to @int@ on assignment, or creating another @twice@ with type parameter @T@ bound to @int@ because \CFA uses the return type~\cite{Cormack81,Baker82,Ada} in its type analysis.
376The first approach has a late conversion from @double@ to @int@ on the final assignment, while the second has an eager conversion to @int@.
377\CFA minimizes the number of conversions and their potential to lose information, so it selects the first approach, which corresponds with C-programmer intuition.
378
379Crucial to the design of a new programming language are the libraries to access thousands of external software features.
380Like \CC, \CFA inherits a massive compatible library-base, where other programming languages must rewrite or provide fragile inter-language communication with C.
381A simple example is leveraging the existing type-unsafe (@void *@) C @bsearch@ to binary search a sorted float array:
382\begin{cfa}
383void * bsearch( const void * key, const void * base, size_t nmemb, size_t size,
384                                int (* compar)( const void *, const void * ));
385int comp( const void * t1, const void * t2 ) {
386         return *(double *)t1 < *(double *)t2 ? -1 : *(double *)t2 < *(double *)t1 ? 1 : 0;
387}
388double key = 5.0, vals[10] = { /* 10 sorted float values */ };
389double * val = (double *)bsearch( &key, vals, 10, sizeof(vals[0]), comp ); $\C{// search sorted array}$
390\end{cfa}
391which can be augmented simply with generalized, type-safe, \CFA-overloaded wrappers:
392\begin{cfa}
393forall( otype T | { int ?<?( T, T ); } ) T * bsearch( T key, const T * arr, size_t size ) {
394        int comp( const void * t1, const void * t2 ) { /* as above with double changed to T */ }
395        return (T *)bsearch( &key, arr, size, sizeof(T), comp );
396}
397forall( otype T | { int ?<?( T, T ); } ) unsigned int bsearch( T key, const T * arr, size_t size ) {
398        T * result = bsearch( key, arr, size ); $\C{// call first version}$
399        return result ? result - arr : size; $\C{// pointer subtraction includes sizeof(T)}$
400}
401double * val = bsearch( 5.0, vals, 10 ); $\C{// selection based on return type}$
402int posn = bsearch( 5.0, vals, 10 );
403\end{cfa}
404The nested function @comp@ provides the hidden interface from typed \CFA to untyped (@void *@) C, plus the cast of the result.
405Providing a hidden @comp@ function in \CC is awkward as lambdas do not use C calling-conventions and template declarations cannot appear at block scope.
406As well, an alternate kind of return is made available: position versus pointer to found element.
407\CC's type-system cannot disambiguate between the two versions of @bsearch@ because it does not use the return type in overload resolution, nor can \CC separately compile a template @bsearch@.
408
409\CFA has replacement libraries condensing hundreds of existing C functions into tens of \CFA overloaded functions, all without rewriting the actual computations (see Section~\ref{sec:libraries}).
410For example, it is possible to write a type-safe \CFA wrapper @malloc@ based on the C @malloc@, where the return type supplies the type/size of the allocation, which is impossible in most type systems.
411\begin{cfa}
412forall( dtype T | sized(T) ) T * malloc( void ) { return (T *)malloc( sizeof(T) ); }
413// select type and size from left-hand side
414int * ip = malloc();  double * dp = malloc();  struct S {...} * sp = malloc();
415\end{cfa}
416
417Call-site inferencing and nested functions provide a localized form of inheritance.
418For example, the \CFA @qsort@ only sorts in ascending order using @<@.
419However, it is trivial to locally change this behaviour:
420\begin{cfa}
421forall( otype T | { int ?<?( T, T ); } ) void qsort( const T * arr, size_t size ) { /* use C qsort */ }
422{
423        int ?<?( double x, double y ) { return x `>` y; } $\C{// locally override behaviour}$
424        qsort( vals, size );                                    $\C{// descending sort}$
425}
426\end{cfa}
427The local version of @?<?@ performs @?>?@ overriding the built-in @?<?@ so it is passed to @qsort@.
428Hence, programmers can easily form local environments, adding and modifying appropriate functions, to maximize reuse of other existing functions and types.
429
430To reducing duplication, it is possible to distribute a group of @forall@ (and storage-class qualifiers) over functions/types, so each block declaration is prefixed by the group (see example in Appendix~\ref{s:CforallStack}).
431\begin{cfa}
432forall( otype `T` ) {                                                   $\C{// distribution block, add forall qualifier to declarations}$
433        struct stack { stack_node(`T`) * head; };       $\C{// generic type}$
434        inline {                                                                        $\C{// nested distribution block, add forall/inline to declarations}$
435                void push( stack(`T`) & s, `T` value ) ...      $\C{// generic operations}$
436        }
437}
438\end{cfa}
439
440
441\vspace*{-2pt}
442\subsection{Traits}
443
444\CFA provides \newterm{traits} to name a group of type assertions, where the trait name allows specifying the same set of assertions in multiple locations, preventing repetition mistakes at each function declaration:
445
446\begin{cquote}
447\lstDeleteShortInline@%
448\begin{tabular}{@{}l@{\hspace{\parindentlnth}}|@{\hspace{\parindentlnth}}l@{}}
449\begin{cfa}
450trait `sumable`( otype T ) {
451        void `?{}`( T &, zero_t ); // 0 literal constructor
452        T ?+?( T, T );                   // assortment of additions
453        T `?+=?`( T &, T );
454        T ++?( T & );
455        T ?++( T & );
456};
457\end{cfa}
458&
459\begin{cfa}
460forall( otype T `| sumable( T )` ) // use trait
461T sum( T a[$\,$], size_t size ) {
462        `T` total = { `0` };  // initialize by 0 constructor
463        for ( size_t i = 0; i < size; i += 1 )
464                total `+=` a[i]; // select appropriate +
465        return total;
466}
467\end{cfa}
468\end{tabular}
469\lstMakeShortInline@%
470\end{cquote}
471
472In fact, the set of @summable@ trait operators is incomplete, as it is missing assignment for type @T@, but @otype@ is syntactic sugar for the following implicit trait:
473\begin{cfa}
474trait otype( dtype T | sized(T) ) {  // sized is a pseudo-trait for types with known size and alignment
475        void ?{}( T & );                                                $\C{// default constructor}$
476        void ?{}( T &, T );                                             $\C{// copy constructor}$
477        void ?=?( T &, T );                                             $\C{// assignment operator}$
478        void ^?{}( T & );                                               $\C{// destructor}$
479};
480\end{cfa}
481Given the information provided for an @otype@, variables of polymorphic type can be treated as if they were a complete type: stack-allocatable, default or copy-initialized, assigned, and deleted.
482
483In summation, the \CFA type-system uses \newterm{nominal typing} for concrete types, matching with the C type-system, and \newterm{structural typing} for polymorphic types.
484Hence, trait names play no part in type equivalence;
485the names are simply macros for a list of polymorphic assertions, which are expanded at usage sites.
486Nevertheless, trait names form a logical subtype-hierarchy with @dtype@ at the top, where traits often contain overlapping assertions, \eg operator @+@.
487Traits are used like interfaces in Java or abstract base-classes in \CC, but without the nominal inheritance-relationships.
488Instead, each polymorphic function (or generic type) defines the structural type needed for its execution (polymorphic type-key), and this key is fulfilled at each call site from the lexical environment, which is similar to Go~\cite{Go} interfaces.
489Hence, new lexical scopes and nested functions are used extensively to create local subtypes, as in the @qsort@ example, without having to manage a nominal-inheritance hierarchy.
490(Nominal inheritance can be approximated with traits using marker variables or functions, as is done in Go.)
491
492% Nominal inheritance can be simulated with traits using marker variables or functions:
493% \begin{cfa}
494% trait nominal(otype T) {
495%     T is_nominal;
496% };
497% int is_nominal;                                                               $\C{// int now satisfies the nominal trait}$
498% \end{cfa}
499%
500% Traits, however, are significantly more powerful than nominal-inheritance interfaces; most notably, traits may be used to declare a relationship \emph{among} multiple types, a property that may be difficult or impossible to represent in nominal-inheritance type systems:
501% \begin{cfa}
502% trait pointer_like(otype Ptr, otype El) {
503%     lvalue El *?(Ptr);                                                $\C{// Ptr can be dereferenced into a modifiable value of type El}$
504% }
505% struct list {
506%     int value;
507%     list * next;                                                              $\C{// may omit "struct" on type names as in \CC}$
508% };
509% typedef list * list_iterator;
510%
511% lvalue int *?( list_iterator it ) { return it->value; }
512% \end{cfa}
513% In the example above, @(list_iterator, int)@ satisfies @pointer_like@ by the user-defined dereference function, and @(list_iterator, list)@ also satisfies @pointer_like@ by the built-in dereference operator for pointers. Given a declaration @list_iterator it@, @*it@ can be either an @int@ or a @list@, with the meaning disambiguated by context (\eg @int x = *it;@ interprets @*it@ as an @int@, while @(*it).value = 42;@ interprets @*it@ as a @list@).
514% While a nominal-inheritance system with associated types could model one of those two relationships by making @El@ an associated type of @Ptr@ in the @pointer_like@ implementation, few such systems could model both relationships simultaneously.
515
516
517\vspace*{-2pt}
518\section{Generic Types}
519
520A significant shortcoming of standard C is the lack of reusable type-safe abstractions for generic data structures and algorithms.
521Broadly speaking, there are three approaches to implement abstract data-structures in C.
522One approach is to write bespoke data-structures for each context in which they are needed.
523While this approach is flexible and supports integration with the C type-checker and tooling, it is also tedious and error-prone, especially for more complex data structures.
524A second approach is to use @void *@-based polymorphism, \eg the C standard-library functions @bsearch@ and @qsort@, which allow reuse of code with common functionality.
525However, basing all polymorphism on @void *@ eliminates the type-checker's ability to ensure that argument types are properly matched, often requiring a number of extra function parameters, pointer indirection, and dynamic allocation that is not otherwise needed.
526A third approach to generic code is to use preprocessor macros, which does allow the generated code to be both generic and type-checked, but errors may be difficult to interpret.
527Furthermore, writing and using preprocessor macros is unnatural and inflexible.
528
529\CC, Java, and other languages use \newterm{generic types} to produce type-safe abstract data-types.
530\CFA generic types integrate efficiently and naturally with the existing polymorphic functions, while retaining backwards compatibility with C and providing separate compilation.
531However, for known concrete parameters, the generic-type definition can be inlined, like \CC templates.
532
533A generic type can be declared by placing a @forall@ specifier on a @struct@ or @union@ declaration, and instantiated using a parenthesized list of types after the type name:
534\begin{cquote}
535\lstDeleteShortInline@%
536\begin{tabular}{@{}l|@{\hspace{2\parindentlnth}}l@{}}
537\begin{cfa}
538forall( otype R, otype S ) struct pair {
539        R first;        S second;
540};
541`forall( otype T )` // dynamic
542T value( pair(const char *, T) p ) { return p.second; }
543`forall( dtype F, otype T )` // dtype-static (concrete)
544T value( pair(F *, T * ) p) { return *p.second; }
545\end{cfa}
546&
547\begin{cfa}
548pair(const char *, int) p = {"magic", 42}; // concrete
549int i = value( p );
550pair(void *, int *) q = { 0, &p.second }; // concrete
551i = value( q );
552double d = 1.0;
553pair(double *, double *) r = { &d, &d }; // concrete
554d = value( r );
555\end{cfa}
556\end{tabular}
557\lstMakeShortInline@%
558\end{cquote}
559
560\CFA classifies generic types as either \newterm{concrete} or \newterm{dynamic}.
561Concrete types have a fixed memory layout regardless of type parameters, while dynamic types vary in memory layout depending on their type parameters.
562A \newterm{dtype-static} type has polymorphic parameters but is still concrete.
563Polymorphic pointers are an example of dtype-static types, \eg @forall(dtype T) T *@ is a polymorphic type, but for any @T@, @T *@  is a fixed-sized pointer, and therefore, can be represented by a @void *@ in code generation.
564
565\CFA generic types also allow checked argument-constraints.
566For example, the following declaration of a sorted set-type ensures the set key supports equality and relational comparison:
567\begin{cfa}
568forall( otype Key | { _Bool ?==?(Key, Key); _Bool ?<?(Key, Key); } ) struct sorted_set;
569\end{cfa}
570
571
572\subsection{Concrete Generic-Types}
573
574The \CFA translator template-expands concrete generic-types into new structure types, affording maximal inlining.
575To enable inter-operation among equivalent instantiations of a generic type, the translator saves the set of instantiations currently in scope and reuses the generated structure declarations where appropriate.
576A function declaration that accepts or returns a concrete generic-type produces a declaration for the instantiated structure in the same scope, which all callers may reuse.
577For example, the concrete instantiation for @pair( const char *, int )@ is:
578\begin{cfa}
579struct _pair_conc0 {
580        const char * first;
581        int second;
582};
583\end{cfa}
584
585A concrete generic-type with dtype-static parameters is also expanded to a structure type, but this type is used for all matching instantiations.
586In the above example, the @pair( F *, T * )@ parameter to @value@ is such a type; its expansion is below and it is used as the type of the variables @q@ and @r@ as well, with casts for member access where appropriate:
587\begin{cfa}
588struct _pair_conc1 {
589        void * first;
590        void * second;
591};
592\end{cfa}
593
594
595\subsection{Dynamic Generic-Types}
596
597Though \CFA implements concrete generic-types efficiently, it also has a fully general system for dynamic generic types.
598As mentioned in Section~\ref{sec:poly-fns}, @otype@ function parameters (in fact all @sized@ polymorphic parameters) come with implicit size and alignment parameters provided by the caller.
599Dynamic generic-types also have an \newterm{offset array} containing structure-member offsets.
600A dynamic generic-@union@ needs no such offset array, as all members are at offset 0, but size and alignment are still necessary.
601Access to members of a dynamic structure is provided at runtime via base-displacement addressing with the structure pointer and the member offset (similar to the @offsetof@ macro), moving a compile-time offset calculation to runtime.
602
603The offset arrays are statically generated where possible.
604If a dynamic generic-type is declared to be passed or returned by value from a polymorphic function, the translator can safely assume the generic type is complete (\ie has a known layout) at any call-site, and the offset array is passed from the caller;
605if the generic type is concrete at the call site, the elements of this offset array can even be statically generated using the C @offsetof@ macro.
606As an example, the body of the second @value@ function is implemented as:
607\begin{cfa}
608_assign_T( _retval, p + _offsetof_pair[1] ); $\C{// return *p.second}$
609\end{cfa}
610@_assign_T@ is passed in as an implicit parameter from @otype T@, and takes two @T *@ (@void *@ in the generated code), a destination and a source; @_retval@ is the pointer to a caller-allocated buffer for the return value, the usual \CFA method to handle dynamically-sized return types.
611@_offsetof_pair@ is the offset array passed into @value@; this array is generated at the call site as:
612\begin{cfa}
613size_t _offsetof_pair[] = { offsetof( _pair_conc0, first ), offsetof( _pair_conc0, second ) }
614\end{cfa}
615
616In some cases the offset arrays cannot be statically generated.
617For instance, modularity is generally provided in C by including an opaque forward-declaration of a structure and associated accessor and mutator functions in a header file, with the actual implementations in a separately-compiled @.c@ file.
618\CFA supports this pattern for generic types, but the caller does not know the actual layout or size of the dynamic generic-type, and only holds it by a pointer.
619The \CFA translator automatically generates \newterm{layout functions} for cases where the size, alignment, and offset array of a generic struct cannot be passed into a function from that function's caller.
620These layout functions take as arguments pointers to size and alignment variables and a caller-allocated array of member offsets, as well as the size and alignment of all @sized@ parameters to the generic structure (un@sized@ parameters are forbidden from being used in a context that affects layout).
621Results of these layout functions are cached so that they are only computed once per type per function. %, as in the example below for @pair@.
622Layout functions also allow generic types to be used in a function definition without reflecting them in the function signature.
623For instance, a function that strips duplicate values from an unsorted @vector(T)@ likely has a pointer to the vector as its only explicit parameter, but uses some sort of @set(T)@ internally to test for duplicate values.
624This function could acquire the layout for @set(T)@ by calling its layout function with the layout of @T@ implicitly passed into the function.
625
626Whether a type is concrete, dtype-static, or dynamic is decided solely on the @forall@'s type parameters.
627This design allows opaque forward declarations of generic types, \eg @forall(otype T)@ @struct Box@ -- like in C, all uses of @Box(T)@ can be separately compiled, and callers from other translation units know the proper calling conventions to use.
628If the definition of a structure type is included in deciding whether a generic type is dynamic or concrete, some further types may be recognized as dtype-static (\eg @forall(otype T)@ @struct unique_ptr { T * p }@ does not depend on @T@ for its layout, but the existence of an @otype@ parameter means that it \emph{could}.), but preserving separate compilation (and the associated C compatibility) in the existing design is judged to be an appropriate trade-off.
629
630
631\subsection{Applications}
632\label{sec:generic-apps}
633
634The reuse of dtype-static structure instantiations enables useful programming patterns at zero runtime cost.
635The most important such pattern is using @forall(dtype T) T *@ as a type-checked replacement for @void *@, \eg creating a lexicographic comparison for pairs of pointers used by @bsearch@ or @qsort@:
636\begin{cfa}
637forall( dtype T ) int lexcmp( pair( T *, T * ) * a, pair( T *, T * ) * b, int (* cmp)( T *, T * ) ) {
638        return cmp( a->first, b->first ) ? : cmp( a->second, b->second );
639}
640\end{cfa}
641Since @pair( T *, T * )@ is a concrete type, there are no implicit parameters passed to @lexcmp@, so the generated code is identical to a function written in standard C using @void *@, yet the \CFA version is type-checked to ensure the fields of both pairs and the arguments to the comparison function match in type.
642
643Another useful pattern enabled by reused dtype-static type instantiations is zero-cost \newterm{tag-structures}.
644Sometimes information is only used for type-checking and can be omitted at runtime, \eg:
645\begin{cquote}
646\lstDeleteShortInline@%
647\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
648\begin{cfa}
649forall( dtype Unit ) struct scalar { unsigned long value; };
650struct metres {};
651struct litres {};
652forall( dtype U ) scalar(U) ?+?( scalar(U) a, scalar(U) b ) {
653        return (scalar(U)){ a.value + b.value };
654}
655\end{cfa}
656&
657\begin{cfa}
658scalar(metres) half_marathon = { 21_098 };
659scalar(litres) pool = { 2_500_000 };
660scalar(metres) marathon = half_marathon +
661                                                        half_marathon;
662scalar(litres) two_pools = pool + pool;
663`marathon + pool;`      // compilation ERROR
664\end{cfa}
665\end{tabular}
666\lstMakeShortInline@%
667\end{cquote}
668@scalar@ is a dtype-static type, so all uses have a single structure definition, containing @unsigned long@, and can share the same implementations of common functions like @?+?@.
669These implementations may even be separately compiled, unlike \CC template functions.
670However, the \CFA type-checker ensures matching types are used by all calls to @?+?@, preventing nonsensical computations like adding a length to a volume.
671
672
673\section{Tuples}
674\label{sec:tuples}
675
676In many languages, functions can return at most one value;
677however, many operations have multiple outcomes, some exceptional.
678Consider C's @div@ and @remquo@ functions, which return the quotient and remainder for a division of integer and float values, respectively.
679\begin{cfa}
680typedef struct { int quo, rem; } div_t;         $\C{// from include stdlib.h}$
681div_t div( int num, int den );
682double remquo( double num, double den, int * quo );
683div_t qr = div( 13, 5 );                                        $\C{// return quotient/remainder aggregate}$
684int q;
685double r = remquo( 13.5, 5.2, &q );                     $\C{// return remainder, alias quotient}$
686\end{cfa}
687@div@ aggregates the quotient/remainder in a structure, while @remquo@ aliases a parameter to an argument.
688Both approaches are awkward.
689Alternatively, a programming language can directly support returning multiple values, \eg in \CFA:
690\begin{cfa}
691[ int, int ] div( int num, int den );           $\C{// return two integers}$
692[ double, double ] div( double num, double den ); $\C{// return two doubles}$
693int q, r;                                                                       $\C{// overloaded variable names}$
694double q, r;
695[ q, r ] = div( 13, 5 );                                        $\C{// select appropriate div and q, r}$
696[ q, r ] = div( 13.5, 5.2 );                            $\C{// assign into tuple}$
697\end{cfa}
698This approach is straightforward to understand and use;
699therefore, why do few programming languages support this obvious feature or provide it awkwardly?
700To answer, there are complex consequences that cascade through multiple aspects of the language, especially the type-system.
701This section show these consequences and how \CFA handles them.
702
703
704\subsection{Tuple Expressions}
705
706The addition of multiple-return-value functions (MRVF) are \emph{useless} without a syntax for accepting multiple values at the call-site.
707The simplest mechanism for capturing the return values is variable assignment, allowing the values to be retrieved directly.
708As such, \CFA allows assigning multiple values from a function into multiple variables, using a square-bracketed list of lvalue expressions (as above), called a \newterm{tuple}.
709
710However, functions also use \newterm{composition} (nested calls), with the direct consequence that MRVFs must also support composition to be orthogonal with single-returning-value functions (SRVF), \eg:
711\begin{cfa}
712printf( "%d %d\n", div( 13, 5 ) );                      $\C{// return values seperated into arguments}$
713\end{cfa}
714Here, the values returned by @div@ are composed with the call to @printf@ by flattening the tuple into separate arguments.
715However, the \CFA type-system must support significantly more complex composition:
716\begin{cfa}
717[ int, int ] foo$\(_1\)$( int );                        $\C{// overloaded foo functions}$
718[ double ] foo$\(_2\)$( int );
719void bar( int, double, double );
720`bar`( foo( 3 ), foo( 3 ) );
721\end{cfa}
722The type-resolver only has the tuple return-types to resolve the call to @bar@ as the @foo@ parameters are identical, which involves unifying the possible @foo@ functions with @bar@'s parameter list.
723No combination of @foo@s are an exact match with @bar@'s parameters, so the resolver applies C conversions.
724The minimal cost is @bar( foo@$_1$@( 3 ), foo@$_2$@( 3 ) )@, giving (@int@, {\color{ForestGreen}@int@}, @double@) to (@int@, {\color{ForestGreen}@double@}, @double@) with one {\color{ForestGreen}safe} (widening) conversion from @int@ to @double@ versus ({\color{red}@double@}, {\color{ForestGreen}@int@}, {\color{ForestGreen}@int@}) to ({\color{red}@int@}, {\color{ForestGreen}@double@}, {\color{ForestGreen}@double@}) with one {\color{red}unsafe} (narrowing) conversion from @double@ to @int@ and two safe conversions.
725
726
727\subsection{Tuple Variables}
728
729An important observation from function composition is that new variable names are not required to initialize parameters from an MRVF.
730\CFA also allows declaration of tuple variables that can be initialized from an MRVF, since it can be awkward to declare multiple variables of different types, \eg:
731\begin{cfa}
732[ int, int ] qr = div( 13, 5 );                         $\C{// tuple-variable declaration and initialization}$
733[ double, double ] qr = div( 13.5, 5.2 );
734\end{cfa}
735where the tuple variable-name serves the same purpose as the parameter name(s).
736Tuple variables can be composed of any types, except for array types, since array sizes are generally unknown in C.
737
738One way to access the tuple-variable components is with assignment or composition:
739\begin{cfa}
740[ q, r ] = qr;                                                          $\C{// access tuple-variable components}$
741printf( "%d %d\n", qr );
742\end{cfa}
743\CFA also supports \newterm{tuple indexing} to access single components of a tuple expression:
744\begin{cfa}
745[int, int] * p = &qr;                                           $\C{// tuple pointer}$
746int rem = qr`.1`;                                                       $\C{// access remainder}$
747int quo = div( 13, 5 )`.0`;                                     $\C{// access quotient}$
748p`->0` = 5;                                                                     $\C{// change quotient}$
749bar( qr`.1`, qr );                                                      $\C{// pass remainder and quotient/remainder}$
750rem = [div( 13, 5 ), 42]`.0.1`;                         $\C{// access 2nd component of 1st component}$
751\end{cfa}
752
753
754\subsection{Flattening and Restructuring}
755
756In function call contexts, tuples support implicit flattening and restructuring conversions.
757Tuple flattening recursively expands a tuple into the list of its basic components.
758Tuple structuring packages a list of expressions into a value of tuple type, \eg:
759\begin{cfa}
760int f( int, int );
761[int] g( [int, int] );
762[int] h( int, [int, int] );
763[int, int] x;
764int y;
765f( x );                                                                         $\C{// flatten}$
766g( y, 10 );                                                                     $\C{// structure}$
767h( x, y );                                                                      $\C{// flatten and structure}$
768\end{cfa}
769In the call to @f@, @x@ is implicitly flattened so the components of @x@ are passed as the two arguments.
770In the call to @g@, the values @y@ and @10@ are structured into a single argument of type @[int, int]@ to match the parameter type of @g@.
771Finally, in the call to @h@, @x@ is flattened to yield an argument list of length 3, of which the first component of @x@ is passed as the first parameter of @h@, and the second component of @x@ and @y@ are structured into the second argument of type @[int, int]@.
772The flexible structure of tuples permits a simple and expressive function call syntax to work seamlessly with both SRVF and MRVF, and with any number of arguments of arbitrarily complex structure.
773
774
775\subsection{Tuple Assignment}
776
777An assignment where the left side is a tuple type is called \newterm{tuple assignment}.
778There are two kinds of tuple assignment depending on whether the right side of the assignment operator has a tuple type or a non-tuple type, called \newterm{multiple} and \newterm{mass assignment}, respectively.
779\begin{cfa}
780int x = 10;
781double y = 3.5;
782[int, double] z;
783z = [x, y];                                                                     $\C{// multiple assignment}$
784[x, y] = z;                                                                     $\C{// multiple assignment}$
785z = 10;                                                                         $\C{// mass assignment}$
786[y, x] = 3.14;                                                          $\C{// mass assignment}$
787\end{cfa}
788Both kinds of tuple assignment have parallel semantics, so that each value on the left and right side is evaluated before any assignments occur.
789As a result, it is possible to swap the values in two variables without explicitly creating any temporary variables or calling a function, \eg, @[x, y] = [y, x]@.
790This semantics means mass assignment differs from C cascading assignment (\eg @a = b = c@) in that conversions are applied in each individual assignment, which prevents data loss from the chain of conversions that can happen during a cascading assignment.
791For example, @[y, x] = 3.14@ performs the assignments @y = 3.14@ and @x = 3.14@, yielding @y == 3.14@ and @x == 3@;
792whereas, C cascading assignment @y = x = 3.14@ performs the assignments @x = 3.14@ and @y = x@, yielding @3@ in @y@ and @x@.
793Finally, tuple assignment is an expression where the result type is the type of the left-hand side of the assignment, just like all other assignment expressions in C.
794This example shows mass, multiple, and cascading assignment used in one expression:
795\begin{cfa}
796[void] f( [int, int] );
797f( [x, y] = z = 1.5 );                                          $\C{// assignments in parameter list}$
798\end{cfa}
799
800
801\subsection{Member Access}
802
803It is also possible to access multiple fields from a single expression using a \newterm{member-access}.
804The result is a single tuple-valued expression whose type is the tuple of the types of the members, \eg:
805\begin{cfa}
806struct S { int x; double y; char * z; } s;
807s.[x, y, z] = 0;
808\end{cfa}
809Here, the mass assignment sets all members of @s@ to zero.
810Since tuple-index expressions are a form of member-access expression, it is possible to use tuple-index expressions in conjunction with member-tuple expressions to manually restructure a tuple (\eg rearrange, drop, and duplicate components).
811\begin{cfa}
812[int, int, long, double] x;
813void f( double, long );
814x.[0, 1] = x.[1, 0];                                            $\C{// rearrange: [x.0, x.1] = [x.1, x.0]}$
815f( x.[0, 3] );                                                          $\C{// drop: f(x.0, x.3)}$
816[int, int, int] y = x.[2, 0, 2];                        $\C{// duplicate: [y.0, y.1, y.2] = [x.2, x.0.x.2]}$
817\end{cfa}
818It is also possible for a member access to contain other member accesses, \eg:
819\begin{cfa}
820struct A { double i; int j; };
821struct B { int * k; short l; };
822struct C { int x; A y; B z; } v;
823v.[x, y.[i, j], z.k];                                           $\C{// [v.x, [v.y.i, v.y.j], v.z.k]}$
824\end{cfa}
825
826
827\begin{comment}
828\subsection{Casting}
829
830In C, the cast operator is used to explicitly convert between types.
831In \CFA, the cast operator has a secondary use as type ascription.
832That is, a cast can be used to select the type of an expression when it is ambiguous, as in the call to an overloaded function:
833\begin{cfa}
834int f();     // (1)
835double f()// (2)
836
837f();       // ambiguous - (1),(2) both equally viable
838(int)f()// choose (2)
839\end{cfa}
840
841Since casting is a fundamental operation in \CFA, casts should be given a meaningful interpretation in the context of tuples.
842Taking a look at standard C provides some guidance with respect to the way casts should work with tuples:
843\begin{cfa}
844int f();
845void g();
846
847(void)f()// (1)
848(int)g()// (2)
849\end{cfa}
850In C, (1) is a valid cast, which calls @f@ and discards its result.
851On the other hand, (2) is invalid, because @g@ does not produce a result, so requesting an @int@ to materialize from nothing is nonsensical.
852Generalizing these principles, any cast wherein the number of components increases as a result of the cast is invalid, while casts that have the same or fewer number of components may be valid.
853
854Formally, a cast to tuple type is valid when $T_n \leq S_m$, where $T_n$ is the number of components in the target type and $S_m$ is the number of components in the source type, and for each $i$ in $[0, n)$, $S_i$ can be cast to $T_i$.
855Excess elements ($S_j$ for all $j$ in $[n, m)$) are evaluated, but their values are discarded so that they are not included in the result expression.
856This approach follows naturally from the way that a cast to @void@ works in C.
857
858For example, in
859\begin{cfa}
860[int, int, int] f();
861[int, [int, int], int] g();
862
863([int, double])f();           $\C{// (1)}$
864([int, int, int])g();         $\C{// (2)}$
865([void, [int, int]])g();      $\C{// (3)}$
866([int, int, int, int])g();    $\C{// (4)}$
867([int, [int, int, int]])g()$\C{// (5)}$
868\end{cfa}
869
870(1) discards the last element of the return value and converts the second element to @double@.
871Since @int@ is effectively a 1-element tuple, (2) discards the second component of the second element of the return value of @g@.
872If @g@ is free of side effects, this expression is equivalent to @[(int)(g().0), (int)(g().1.0), (int)(g().2)]@.
873Since @void@ is effectively a 0-element tuple, (3) discards the first and third return values, which is effectively equivalent to @[(int)(g().1.0), (int)(g().1.1)]@).
874
875Note that a cast is not a function call in \CFA, so flattening and structuring conversions do not occur for cast expressions\footnote{User-defined conversions have been considered, but for compatibility with C and the existing use of casts as type ascription, any future design for such conversions requires more precise matching of types than allowed for function arguments and parameters.}.
876As such, (4) is invalid because the cast target type contains 4 components, while the source type contains only 3.
877Similarly, (5) is invalid because the cast @([int, int, int])(g().1)@ is invalid.
878That is, it is invalid to cast @[int, int]@ to @[int, int, int]@.
879\end{comment}
880
881
882\subsection{Polymorphism}
883
884Tuples also integrate with \CFA polymorphism as a kind of generic type.
885Due to the implicit flattening and structuring conversions involved in argument passing, @otype@ and @dtype@ parameters are restricted to matching only with non-tuple types, \eg:
886\begin{cfa}
887forall( otype T, dtype U ) void f( T x, U * y );
888f( [5, "hello"] );
889\end{cfa}
890where @[5, "hello"]@ is flattened, giving argument list @5, "hello"@, and @T@ binds to @int@ and @U@ binds to @const char@.
891Tuples, however, may contain polymorphic components.
892For example, a plus operator can be written to sum two triples.
893\begin{cfa}
894forall( otype T | { T ?+?( T, T ); } ) [T, T, T] ?+?( [T, T, T] x, [T, T, T] y ) {
895        return [x.0 + y.0, x.1 + y.1, x.2 + y.2];
896}
897[int, int, int] x;
898int i1, i2, i3;
899[i1, i2, i3] = x + ([10, 20, 30]);
900\end{cfa}
901
902Flattening and restructuring conversions are also applied to tuple types in polymorphic type assertions.
903\begin{cfa}
904[int] f( [int, double], double );
905forall( otype T, otype U | { T f( T, U, U ); } ) void g( T, U );
906g( 5, 10.21 );
907\end{cfa}
908Hence, function parameter and return lists are flattened for the purposes of type unification allowing the example to pass expression resolution.
909This relaxation is possible by extending the thunk scheme described by Bilson~\cite{Bilson03}.
910% Whenever a candidate's parameter structure does not exactly match the formal parameter's structure, a thunk is generated to specialize calls to the actual function:
911% \begin{cfa}
912% int _thunk( int _p0, double _p1, double _p2 ) { return f( [_p0, _p1], _p2 ); }
913% \end{cfa}
914% so the thunk provides flattening and structuring conversions to inferred functions, improving the compatibility of tuples and polymorphism.
915% These thunks are generated locally using gcc nested-functions, rather hoisting them to the external scope, so they can easily access local state.
916
917
918\subsection{Variadic Tuples}
919\label{sec:variadic-tuples}
920
921To define variadic functions, \CFA adds a new kind of type parameter, @ttype@ (tuple type).
922Matching against a @ttype@ parameter consumes all remaining argument components and packages them into a tuple, binding to the resulting tuple of types.
923In a given parameter list, there must be at most one @ttype@ parameter that occurs last, which matches normal variadic semantics, with a strong feeling of similarity to \CCeleven variadic templates.
924As such, @ttype@ variables are also called \newterm{argument packs}.
925
926Like variadic templates, @ttype@ polymorphic functions are primarily manipulated via recursion.
927Since nothing is known about a parameter pack by default, assertion parameters are key to doing anything meaningful.
928Unlike variadic templates, @ttype@ polymorphic functions can be separately compiled.
929For example, a generalized @sum@ function:
930\begin{cfa}
931int sum$\(_0\)$() { return 0; }
932forall( ttype Params | { int sum( Params ); } ) int sum$\(_1\)$( int x, Params rest ) {
933        return x + sum( rest );
934}
935sum( 10, 20, 30 );
936\end{cfa}
937Since @sum@\(_0\) does not accept any arguments, it is not a valid candidate function for the call @sum(10, 20, 30)@.
938In order to call @sum@\(_1\), @10@ is matched with @x@, and the argument resolution moves on to the argument pack @rest@, which consumes the remainder of the argument list and @Params@ is bound to @[20, 30]@.
939The process continues until @Params@ is bound to @[]@, requiring an assertion @int sum()@, which matches @sum@\(_0\) and terminates the recursion.
940Effectively, this algorithm traces as @sum(10, 20, 30)@ $\rightarrow$ @10 + sum(20, 30)@ $\rightarrow$ @10 + (20 + sum(30))@ $\rightarrow$ @10 + (20 + (30 + sum()))@ $\rightarrow$ @10 + (20 + (30 + 0))@.
941
942It is reasonable to take the @sum@ function a step further to enforce a minimum number of arguments:
943\begin{cfa}
944int sum( int x, int y ) { return x + y; }
945forall( ttype Params | { int sum( int, Params ); } ) int sum( int x, int y, Params rest ) {
946        return sum( x + y, rest );
947}
948\end{cfa}
949One more step permits the summation of any summable type with all arguments of the same type:
950\begin{cfa}
951trait summable( otype T ) {
952        T ?+?( T, T );
953};
954forall( otype R | summable( R ) ) R sum( R x, R y ) {
955        return x + y;
956}
957forall( otype R, ttype Params | summable(R) | { R sum(R, Params); } ) R sum(R x, R y, Params rest) {
958        return sum( x + y, rest );
959}
960\end{cfa}
961Unlike C variadic functions, it is unnecessary to hard code the number and expected types.
962Furthermore, this code is extendable for any user-defined type with a @?+?@ operator.
963Summing arbitrary heterogeneous lists is possible with similar code by adding the appropriate type variables and addition operators.
964
965It is also possible to write a type-safe variadic print function to replace @printf@:
966\begin{cfa}
967struct S { int x, y; };
968forall( otype T, ttype Params | { void print(T); void print(Params); } ) void print(T arg, Params rest) {
969        print(arg);  print(rest);
970}
971void print( const char * x ) { printf( "%s", x ); }
972void print( int x ) { printf( "%d", x ); }
973void print( S s ) { print( "{ ", s.x, ",", s.y, " }" ); }
974print( "s = ", (S){ 1, 2 }, "\n" );
975\end{cfa}
976This example showcases a variadic-template-like decomposition of the provided argument list.
977The individual @print@ functions allow printing a single element of a type.
978The polymorphic @print@ allows printing any list of types, where as each individual type has a @print@ function.
979The individual print functions can be used to build up more complicated @print@ functions, such as @S@, which cannot be done with @printf@ in C.
980This mechanism is used to seamlessly print tuples in the \CFA I/O library (see Section~\ref{s:IOLibrary}).
981
982Finally, it is possible to use @ttype@ polymorphism to provide arbitrary argument forwarding functions.
983For example, it is possible to write @new@ as a library function:
984\begin{cfa}
985forall( otype R, otype S ) void ?{}( pair(R, S) *, R, S );
986forall( dtype T, ttype Params | sized(T) | { void ?{}( T *, Params ); } ) T * new( Params p ) {
987        return ((T *)malloc()){ p };                    $\C{// construct into result of malloc}$
988}
989pair( int, char ) * x = new( 42, '!' );
990\end{cfa}
991The @new@ function provides the combination of type-safe @malloc@ with a \CFA constructor call, making it impossible to forget constructing dynamically allocated objects.
992This function provides the type-safety of @new@ in \CC, without the need to specify the allocated type again, thanks to return-type inference.
993
994
995\subsection{Implementation}
996
997Tuples are implemented in the \CFA translator via a transformation into \newterm{generic types}.
998For each $N$, the first time an $N$-tuple is seen in a scope a generic type with $N$ type parameters is generated, \eg:
999\begin{cfa}
1000[int, int] f() {
1001        [double, double] x;
1002        [int, double, int] y;
1003}
1004\end{cfa}
1005is transformed into:
1006\begin{cfa}
1007forall( dtype T0, dtype T1 | sized(T0) | sized(T1) ) struct _tuple2 {
1008        T0 field_0;                                                             $\C{// generated before the first 2-tuple}$
1009        T1 field_1;
1010};
1011_tuple2(int, int) f() {
1012        _tuple2(double, double) x;
1013        forall( dtype T0, dtype T1, dtype T2 | sized(T0) | sized(T1) | sized(T2) ) struct _tuple3 {
1014                T0 field_0;                                                     $\C{// generated before the first 3-tuple}$
1015                T1 field_1;
1016                T2 field_2;
1017        };
1018        _tuple3(int, double, int) y;
1019}
1020\end{cfa}
1021{\sloppy
1022Tuple expressions are then simply converted directly into compound literals, \eg @[5, 'x', 1.24]@ becomes @(_tuple3(int, char, double)){ 5, 'x', 1.24 }@.
1023\par}%
1024
1025\begin{comment}
1026Since tuples are essentially structures, tuple indexing expressions are just field accesses:
1027\begin{cfa}
1028void f(int, [double, char]);
1029[int, double] x;
1030
1031x.0+x.1;
1032printf("%d %g\n", x);
1033f(x, 'z');
1034\end{cfa}
1035Is transformed into:
1036\begin{cfa}
1037void f(int, _tuple2(double, char));
1038_tuple2(int, double) x;
1039
1040x.field_0+x.field_1;
1041printf("%d %g\n", x.field_0, x.field_1);
1042f(x.field_0, (_tuple2){ x.field_1, 'z' });
1043\end{cfa}
1044Note that due to flattening, @x@ used in the argument position is converted into the list of its fields.
1045In the call to @f@, the second and third argument components are structured into a tuple argument.
1046Similarly, tuple member expressions are recursively expanded into a list of member access expressions.
1047
1048Expressions that may contain side effects are made into \newterm{unique expressions} before being expanded by the flattening conversion.
1049Each unique expression is assigned an identifier and is guaranteed to be executed exactly once:
1050\begin{cfa}
1051void g(int, double);
1052[int, double] h();
1053g(h());
1054\end{cfa}
1055Internally, this expression is converted to two variables and an expression:
1056\begin{cfa}
1057void g(int, double);
1058[int, double] h();
1059
1060_Bool _unq0_finished_ = 0;
1061[int, double] _unq0;
1062g(
1063        (_unq0_finished_ ? _unq0 : (_unq0 = f(), _unq0_finished_ = 1, _unq0)).0,
1064        (_unq0_finished_ ? _unq0 : (_unq0 = f(), _unq0_finished_ = 1, _unq0)).1,
1065);
1066\end{cfa}
1067Since argument evaluation order is not specified by the C programming language, this scheme is built to work regardless of evaluation order.
1068The first time a unique expression is executed, the actual expression is evaluated and the accompanying boolean is set to true.
1069Every subsequent evaluation of the unique expression then results in an access to the stored result of the actual expression.
1070Tuple member expressions also take advantage of unique expressions in the case of possible impurity.
1071
1072Currently, the \CFA translator has a very broad, imprecise definition of impurity, where any function call is assumed to be impure.
1073This notion could be made more precise for certain intrinsic, auto-generated, and builtin functions, and could analyze function bodies when they are available to recursively detect impurity, to eliminate some unique expressions.
1074
1075The various kinds of tuple assignment, constructors, and destructors generate GNU C statement expressions.
1076A variable is generated to store the value produced by a statement expression, since its fields may need to be constructed with a non-trivial constructor and it may need to be referred to multiple time, \eg in a unique expression.
1077The use of statement expressions allows the translator to arbitrarily generate additional temporary variables as needed, but binds the implementation to a non-standard extension of the C language.
1078However, there are other places where the \CFA translator makes use of GNU C extensions, such as its use of nested functions, so this restriction is not new.
1079\end{comment}
1080
1081
1082\section{Control Structures}
1083
1084\CFA identifies inconsistent, problematic, and missing control structures in C, and extends, modifies, and adds control structures to increase functionality and safety.
1085
1086
1087\subsection{\texorpdfstring{\protect\lstinline{if} Statement}{if Statement}}
1088
1089The @if@ expression allows declarations, similar to @for@ declaration expression:
1090\begin{cfa}
1091if ( int x = f() ) ...                                          $\C{// x != 0}$
1092if ( int x = f(), y = g() ) ...                         $\C{// x != 0 \&\& y != 0}$
1093if ( int x = f(), y = g(); `x < y` ) ...        $\C{// relational expression}$
1094\end{cfa}
1095Unless a relational expression is specified, each variable is compared not equal to 0, which is the standard semantics for the @if@ expression, and the results are combined using the logical @&&@ operator.\footnote{\CC only provides a single declaration always compared not equal to 0.}
1096The scope of the declaration(s) is local to the @if@ statement but exist within both the ``then'' and ``else'' clauses.
1097
1098
1099\subsection{\texorpdfstring{\protect\lstinline{switch} Statement}{switch Statement}}
1100
1101There are a number of deficiencies with the C @switch@ statements: enumerating @case@ lists, placement of @case@ clauses, scope of the switch body, and fall through between case clauses.
1102
1103C has no shorthand for specifying a list of case values, whether the list is non-contiguous or contiguous\footnote{C provides this mechanism via fall through.}.
1104\CFA provides a shorthand for a non-contiguous list:
1105\begin{cquote}
1106\lstDeleteShortInline@%
1107\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1108\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
1109\begin{cfa}
1110case 2, 10, 34, 42:
1111\end{cfa}
1112&
1113\begin{cfa}
1114case 2: case 10: case 34: case 42:
1115\end{cfa}
1116\end{tabular}
1117\lstMakeShortInline@%
1118\end{cquote}
1119for a contiguous list:\footnote{gcc has the same mechanism but awkward syntax, \lstinline@2 ...42@, as a space is required after a number, otherwise the first period is a decimal point.}
1120\begin{cquote}
1121\lstDeleteShortInline@%
1122\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1123\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
1124\begin{cfa}
1125case 2~42:
1126\end{cfa}
1127&
1128\begin{cfa}
1129case 2: case 3: ... case 41: case 42:
1130\end{cfa}
1131\end{tabular}
1132\lstMakeShortInline@%
1133\end{cquote}
1134and a combination:
1135\begin{cfa}
1136case -12~-4, -1~5, 14~21, 34~42:
1137\end{cfa}
1138
1139C allows placement of @case@ clauses \emph{within} statements nested in the @switch@ body (called Duff's device~\cite{Duff83});
1140\begin{cfa}
1141switch ( i ) {
1142  case 0:
1143        for ( int i = 0; i < 10; i += 1 ) {
1144                ...
1145  `case 1:`             // no initialization of loop index
1146                ...
1147        }
1148}
1149\end{cfa}
1150\CFA precludes this form of transfer \emph{into} a control structure because it causes undefined behaviour, especially with respect to missed initialization, and provides very limited functionality.
1151
1152C allows placement of declaration within the @switch@ body and unreachable code at the start, resulting in undefined behaviour:
1153\begin{cfa}
1154switch ( x ) {
1155        `int y = 1;`                                                    $\C{// unreachable initialization}$
1156        `x = 7;`                                                                $\C{// unreachable code without label/branch}$
1157  case 0:
1158        ...
1159        `int z = 0;`                                                    $\C{// unreachable initialization, cannot appear after case}$
1160        z = 2;
1161  case 1:
1162        `x = z;`                                                                $\C{// without fall through, z is undefined}$
1163}
1164\end{cfa}
1165\CFA allows the declaration of local variables, \eg @y@, at the start of the @switch@ with scope across the entire @switch@ body, \ie all @case@ clauses.
1166\CFA disallows the declaration of local variable, \eg @z@, directly within the @switch@ body, because a declaration cannot occur immediately after a @case@ since a label can only be attached to a statement, and the use of @z@ is undefined in @case 1@ as neither storage allocation nor initialization may have occurred.
1167
1168C @switch@ provides multiple entry points into the statement body, but once an entry point is selected, control continues across \emph{all} @case@ clauses until the end of the @switch@ body, called \newterm{fall through};
1169@case@ clauses are made disjoint by the @break@ statement.
1170While fall through \emph{is} a useful form of control flow, it does not match well with programmer intuition, resulting in errors from missing @break@ statements.
1171For backwards compatibility, \CFA provides a \emph{new} control structure, @choose@, which mimics @switch@, but reverses the meaning of fall through (see Figure~\ref{f:ChooseSwitchStatements}), similar to Go.
1172
1173\begin{figure}
1174\centering
1175\lstDeleteShortInline@%
1176\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1177\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
1178\begin{cfa}
1179`choose` ( day ) {
1180  case Mon~Thu:  // program
1181
1182  case Fri:  // program
1183        wallet += pay;
1184        `fallthrough;`
1185  case Sat:  // party
1186        wallet -= party;
1187
1188  case Sun:  // rest
1189
1190  default:  // error
1191}
1192\end{cfa}
1193&
1194\begin{cfa}
1195switch ( day ) {
1196  case Mon: case Tue: case Wed: case Thu:  // program
1197        `break;`
1198  case Fri:  // program
1199        wallet += pay;
1200
1201  case Sat:  // party
1202        wallet -= party;
1203        `break;`
1204  case Sun:  // rest
1205        `break;`
1206  default:  // error
1207}
1208\end{cfa}
1209\end{tabular}
1210\lstMakeShortInline@%
1211\caption{\lstinline|choose| versus \lstinline|switch| Statements}
1212\label{f:ChooseSwitchStatements}
1213\end{figure}
1214
1215Finally, Figure~\ref{f:FallthroughStatement} shows @fallthrough@ may appear in contexts other than terminating a @case@ clause, and have an explicit transfer label allowing separate cases but common final-code for a set of cases.
1216The target label must be below the @fallthrough@ and may not be nested in a control structure, \ie @fallthrough@ cannot form a loop, and the target label must be at the same or higher level as the containing @case@ clause and located at the same level as a @case@ clause;
1217the target label may be case @default@, but only associated with the current @switch@/@choose@ statement.
1218
1219\begin{figure}
1220\centering
1221\lstDeleteShortInline@%
1222\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1223\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{non-terminator}}  & \multicolumn{1}{c}{\textbf{target label}}     \\
1224\begin{cfa}
1225choose ( ... ) {
1226  case 3:
1227        if ( ... ) {
1228                ... `fallthrough;`  // goto case 4
1229        } else {
1230                ...
1231        }
1232        // implicit break
1233  case 4:
1234\end{cfa}
1235&
1236\begin{cfa}
1237choose ( ... ) {
1238  case 3:
1239        ... `fallthrough common;`
1240  case 4:
1241        ... `fallthrough common;`
1242  `common`: // below fallthrough at same level as case clauses
1243        ...      // common code for cases 3 and 4
1244        // implicit break
1245  case 4:
1246\end{cfa}
1247\end{tabular}
1248\lstMakeShortInline@%
1249\caption{\lstinline|fallthrough| Statement}
1250\label{f:FallthroughStatement}
1251\end{figure}
1252
1253
1254\subsection{\texorpdfstring{Labelled \protect\lstinline{continue} / \protect\lstinline{break}}{Labelled continue / break}}
1255
1256While C provides @continue@ and @break@ statements for altering control flow, both are restricted to one level of nesting for a particular control structure.
1257Unfortunately, this restriction forces programmers to use @goto@ to achieve the equivalent control-flow for more than one level of nesting.
1258To prevent having to switch to the @goto@, \CFA extends the @continue@ and @break@ with a target label to support static multi-level exit~\cite{Buhr85}, as in Java.
1259For both @continue@ and @break@, the target label must be directly associated with a @for@, @while@ or @do@ statement;
1260for @break@, the target label can also be associated with a @switch@, @if@ or compound (@{}@) statement.
1261Figure~\ref{f:MultiLevelExit} shows @continue@ and @break@ indicating the specific control structure, and the corresponding C program using only @goto@ and labels.
1262The innermost loop has 7 exit points, which cause continuation or termination of one or more of the 7 nested control-structures.
1263
1264\begin{figure}
1265\lstDeleteShortInline@%
1266\begin{tabular}{@{\hspace{\parindentlnth}}l@{\hspace{\parindentlnth}}l@{\hspace{\parindentlnth}}l@{}}
1267\multicolumn{1}{@{\hspace{\parindentlnth}}c@{\hspace{\parindentlnth}}}{\textbf{\CFA}}   & \multicolumn{1}{@{\hspace{\parindentlnth}}c}{\textbf{C}}      \\
1268\begin{cfa}
1269`LC:` {
1270        ... $declarations$ ...
1271        `LS:` switch ( ... ) {
1272          case 3:
1273                `LIF:` if ( ... ) {
1274                        `LF:` for ( ... ) {
1275                                ... break `LC`; ...
1276                                ... break `LS`; ...
1277                                ... break `LIF`; ...
1278                                ... continue `LF;` ...
1279                                ... break `LF`; ...
1280                        } // for
1281                } else {
1282                        ... break `LIF`; ...
1283                } // if
1284        } // switch
1285} // compound
1286\end{cfa}
1287&
1288\begin{cfa}
1289{
1290        ... $declarations$ ...
1291        switch ( ... ) {
1292          case 3:
1293                if ( ... ) {
1294                        for ( ... ) {
1295                                ... goto `LC`; ...
1296                                ... goto `LS`; ...
1297                                ... goto `LIF`; ...
1298                                ... goto `LFC`; ...
1299                                ... goto `LFB`; ...
1300                          `LFC:` ; } `LFB:` ;
1301                } else {
1302                        ... goto `LIF`; ...
1303                } `LIF:` ;
1304        } `LS:` ;
1305} `LC:` ;
1306\end{cfa}
1307&
1308\begin{cfa}
1309
1310
1311
1312
1313
1314
1315
1316// terminate compound
1317// terminate switch
1318// terminate if
1319// continue loop
1320// terminate loop
1321
1322
1323
1324// terminate if
1325
1326\end{cfa}
1327\end{tabular}
1328\lstMakeShortInline@%
1329\caption{Multi-level Exit}
1330\label{f:MultiLevelExit}
1331\end{figure}
1332
1333With respect to safety, both labelled @continue@ and @break@ are a @goto@ restricted in the following ways:
1334\begin{itemize}
1335\item
1336They cannot create a loop, which means only the looping constructs cause looping.
1337This restriction means all situations resulting in repeated execution are clearly delineated.
1338\item
1339They cannot branch into a control structure.
1340This restriction prevents missing declarations and/or initializations at the start of a control structure resulting in undefined behaviour.
1341\end{itemize}
1342The advantage of the labelled @continue@/@break@ is allowing static multi-level exits without having to use the @goto@ statement, and tying control flow to the target control structure rather than an arbitrary point in a program.
1343Furthermore, the location of the label at the \emph{beginning} of the target control structure informs the reader (eye candy) that complex control-flow is occurring in the body of the control structure.
1344With @goto@, the label is at the end of the control structure, which fails to convey this important clue early enough to the reader.
1345Finally, using an explicit target for the transfer instead of an implicit target allows new constructs to be added or removed without affecting existing constructs.
1346Otherwise, the implicit targets of the current @continue@ and @break@, \ie the closest enclosing loop or @switch@, change as certain constructs are added or removed.
1347
1348
1349\subsection{Exception Handling}
1350
1351The following framework for \CFA exception handling is in place, excluding some runtime type-information and virtual functions.
1352\CFA provides two forms of exception handling: \newterm{fix-up} and \newterm{recovery} (see Figure~\ref{f:CFAExceptionHandling})~\cite{Buhr92b,Buhr00a}.
1353Both mechanisms provide dynamic call to a handler using dynamic name-lookup, where fix-up has dynamic return and recovery has static return from the handler.
1354\CFA restricts exception types to those defined by aggregate type @exception@.
1355The form of the raise dictates the set of handlers examined during propagation: \newterm{resumption propagation} (@resume@) only examines resumption handlers (@catchResume@); \newterm{terminating propagation} (@throw@) only examines termination handlers (@catch@).
1356If @resume@ or @throw@ have no exception type, it is a reresume/rethrow, meaning the currently exception continues propagation.
1357If there is no current exception, the reresume/rethrow results in a runtime error.
1358
1359\begin{figure}
1360\begin{cquote}
1361\lstDeleteShortInline@%
1362\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1363\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{Resumption}}      & \multicolumn{1}{c}{\textbf{Termination}}      \\
1364\begin{cfa}
1365`exception R { int fix; };`
1366void f() {
1367        R r;
1368        ... `resume( r );` ...
1369        ... r.fix // control returns here after handler
1370}
1371`try` {
1372        ... f(); ...
1373} `catchResume( R r )` {
1374        ... r.fix = ...; // return correction to raise
1375} // dynamic return to _Resume
1376\end{cfa}
1377&
1378\begin{cfa}
1379`exception T {};`
1380void f() {
1381
1382        ... `throw( T{} );` ...
1383        // control does NOT return here after handler
1384}
1385`try` {
1386        ... f(); ...
1387} `catch( T t )` {
1388        ... // recover and continue
1389} // static return to next statement
1390\end{cfa}
1391\end{tabular}
1392\lstMakeShortInline@%
1393\end{cquote}
1394\caption{\CFA Exception Handling}
1395\label{f:CFAExceptionHandling}
1396\end{figure}
1397
1398The set of exception types in a list of catch clause may include both a resumption and termination handler:
1399\begin{cfa}
1400try {
1401        ... resume( `R{}` ); ...
1402} catchResume( `R` r ) { ... throw( R{} ); ... } $\C{\color{red}// H1}$
1403   catch( `R` r ) { ... }                                       $\C{\color{red}// H2}$
1404
1405\end{cfa}
1406The resumption propagation raises @R@ and the stack is not unwound;
1407the exception is caught by the @catchResume@ clause and handler H1 is invoked.
1408The termination propagation in handler H1 raises @R@ and the stack is unwound;
1409the exception is caught by the @catch@ clause and handler H2 is invoked.
1410The termination handler is available because the resumption propagation did not unwind the stack.
1411
1412An additional feature is conditional matching in a catch clause:
1413\begin{cfa}
1414try {
1415        ... write( `datafile`, ... ); ...               $\C{// may throw IOError}$
1416        ... write( `logfile`, ... ); ...
1417} catch ( IOError err; `err.file == datafile` ) { ... } $\C{// handle datafile error}$
1418   catch ( IOError err; `err.file == logfile` ) { ... } $\C{// handle logfile error}$
1419   catch ( IOError err ) { ... }                        $\C{// handler error from other files}$
1420\end{cfa}
1421where the throw inserts the failing file-handle into the I/O exception.
1422Conditional catch cannot be trivially mimicked by other mechanisms because once an exception is caught, handler clauses in that @try@ statement are no longer eligible..
1423
1424The resumption raise can specify an alternate stack on which to raise an exception, called a \newterm{nonlocal raise}:
1425\begin{cfa}
1426resume( $\emph{exception-type}$, $\emph{alternate-stack}$ )
1427resume( $\emph{alternate-stack}$ )
1428\end{cfa}
1429These overloads of @resume@ raise the specified exception or the currently propagating exception (reresume) at another \CFA coroutine or task~\cite{Delisle18}.
1430Nonlocal raise is restricted to resumption to provide the exception handler the greatest flexibility because processing the exception does not unwind its stack, allowing it to continue after the handler returns.
1431
1432To facilitate nonlocal raise, \CFA provides dynamic enabling and disabling of nonlocal exception-propagation.
1433The constructs for controlling propagation of nonlocal exceptions are the @enable@ and the @disable@ blocks:
1434\begin{cquote}
1435\lstDeleteShortInline@%
1436\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1437\begin{cfa}
1438enable $\emph{exception-type-list}$ {
1439        // allow non-local raise
1440}
1441\end{cfa}
1442&
1443\begin{cfa}
1444disable $\emph{exception-type-list}$ {
1445        // disallow non-local raise
1446}
1447\end{cfa}
1448\end{tabular}
1449\lstMakeShortInline@%
1450\end{cquote}
1451The arguments for @enable@/@disable@ specify the exception types allowed to be propagated or postponed, respectively.
1452Specifying no exception type is shorthand for specifying all exception types.
1453Both @enable@ and @disable@ blocks can be nested, turning propagation on/off on entry, and on exit, the specified exception types are restored to their prior state.
1454Coroutines and tasks start with non-local exceptions disabled, allowing handlers to be put in place, before non-local exceptions are explicitly enabled.
1455\begin{cfa}
1456void main( mytask & t ) {                                       $\C{// thread starts here}$
1457        // non-local exceptions disabled
1458        try {                                                                   $\C{// establish handles for non-local exceptions}$
1459                enable {                                                        $\C{// allow non-local exception delivery}$
1460                        // task body
1461                }
1462        // appropriate catchResume/catch handlers
1463        }
1464}
1465\end{cfa}
1466
1467Finally, \CFA provides a Java like  @finally@ clause after the catch clauses:
1468\begin{cfa}
1469try {
1470        ... f(); ...
1471// catchResume or catch clauses
1472} `finally` {
1473        // house keeping
1474}
1475\end{cfa}
1476The finally clause is always executed, i.e., if the try block ends normally or if an exception is raised.
1477If an exception is raised and caught, the handler is run before the finally clause.
1478Like a destructor (see Section~\ref{s:ConstructorsDestructors}), a finally clause can raise an exception but not if there is an exception being propagated.
1479Mimicking the @finally@ clause with mechanisms like RAII is non-trivially when there are multiple types and local accesses.
1480
1481
1482\subsection{\texorpdfstring{\protect\lstinline{with} Statement}{with Statement}}
1483\label{s:WithStatement}
1484
1485Heterogeneous data is often aggregated into a structure/union.
1486To reduce syntactic noise, \CFA provides a @with@ statement (see Pascal~\cite[\S~4.F]{Pascal}) to elide aggregate field-qualification by opening a scope containing the field identifiers.
1487\begin{cquote}
1488\vspace*{-\baselineskip}%???
1489\lstDeleteShortInline@%
1490\begin{cfa}
1491struct S { char c; int i; double d; };
1492struct T { double m, n; };
1493// multiple aggregate parameters
1494\end{cfa}
1495\begin{tabular}{@{}l@{\hspace{\parindentlnth}}|@{\hspace{\parindentlnth}}l@{}}
1496\begin{cfa}
1497void f( S & s, T & t ) {
1498        `s.`c; `s.`i; `s.`d;
1499        `t.`m; `t.`n;
1500}
1501\end{cfa}
1502&
1503\begin{cfa}
1504void f( S & s, T & t ) `with ( s, t )` {
1505        c; i; d;                // no qualification
1506        m; n;
1507}
1508\end{cfa}
1509\end{tabular}
1510\lstMakeShortInline@%
1511\end{cquote}
1512Object-oriented programming languages only provide implicit qualification for the receiver.
1513
1514In detail, the @with@ statement has the form:
1515\begin{cfa}
1516$\emph{with-statement}$:
1517        'with' '(' $\emph{expression-list}$ ')' $\emph{compound-statement}$
1518\end{cfa}
1519and may appear as the body of a function or nested within a function body.
1520Each expression in the expression-list provides a type and object.
1521The type must be an aggregate type.
1522(Enumerations are already opened.)
1523The object is the implicit qualifier for the open structure-fields.
1524
1525All expressions in the expression list are open in parallel within the compound statement, which is different from Pascal, which nests the openings from left to right.
1526The difference between parallel and nesting occurs for fields with the same name and type:
1527\begin{cfa}
1528struct S { int `i`; int j; double m; } s, w;
1529struct T { int `i`; int k; int m; } t, w;
1530with ( s, t ) {
1531        j + k;                                                                  $\C{// unambiguous, s.j + t.k}$
1532        m = 5.0;                                                                $\C{// unambiguous, t.m = 5.0}$
1533        m = 1;                                                                  $\C{// unambiguous, s.m = 1}$
1534        int a = m;                                                              $\C{// unambiguous, a = s.i }$
1535        double b = m;                                                   $\C{// unambiguous, b = t.m}$
1536        int c = s.i + t.i;                                              $\C{// unambiguous, qualification}$
1537        (double)m;                                                              $\C{// unambiguous, cast}$
1538}
1539\end{cfa}
1540For parallel semantics, both @s.i@ and @t.i@ are visible, so @i@ is ambiguous without qualification;
1541for nested semantics, @t.i@ hides @s.i@, so @i@ implies @t.i@.
1542\CFA's ability to overload variables means fields with the same name but different types are automatically disambiguated, eliminating most qualification when opening multiple aggregates.
1543Qualification or a cast is used to disambiguate.
1544
1545There is an interesting problem between parameters and the function-body @with@, \eg:
1546\begin{cfa}
1547void ?{}( S & s, int i ) with ( s ) {           $\C{// constructor}$
1548        `s.i = i;`  j = 3;  m = 5.5;                    $\C{// initialize fields}$
1549}
1550\end{cfa}
1551Here, the assignment @s.i = i@ means @s.i = s.i@, which is meaningless, and there is no mechanism to qualify the parameter @i@, making the assignment impossible using the function-body @with@.
1552To solve this problem, parameters are treated like an initialized aggregate:
1553\begin{cfa}
1554struct Params {
1555        S & s;
1556        int i;
1557} params;
1558\end{cfa}
1559and implicitly opened \emph{after} a function-body open, to give them higher priority:
1560\begin{cfa}
1561void ?{}( S & s, int `i` ) with ( s ) `with( $\emph{\color{red}params}$ )` {
1562        s.i = `i`; j = 3; m = 5.5;
1563}
1564\end{cfa}
1565Finally, a cast may be used to disambiguate among overload variables in a @with@ expression:
1566\begin{cfa}
1567with ( w ) { ... }                                                      $\C{// ambiguous, same name and no context}$
1568with ( (S)w ) { ... }                                           $\C{// unambiguous, cast}$
1569\end{cfa}
1570and @with@ expressions may be complex expressions with type reference (see Section~\ref{s:References}) to aggregate:
1571\begin{cfa}
1572struct S { int i, j; } sv;
1573with ( sv ) {                                                           $\C{// implicit reference}$
1574        S & sr = sv;
1575        with ( sr ) {                                                   $\C{// explicit reference}$
1576                S * sp = &sv;
1577                with ( *sp ) {                                          $\C{// computed reference}$
1578                        i = 3; j = 4;                                   $\C{\color{red}// sp--{\textgreater}i, sp--{\textgreater}j}$
1579                }
1580                i = 2; j = 3;                                           $\C{\color{red}// sr.i, sr.j}$
1581        }
1582        i = 1; j = 2;                                                   $\C{\color{red}// sv.i, sv.j}$
1583}
1584\end{cfa}
1585
1586Collectively, these control-structure enhancements reduce programmer burden and increase readability and safety.
1587
1588
1589\section{Declarations}
1590
1591Declarations in C have weaknesses and omissions.
1592\CFA attempts to correct and add to C declarations, while ensuring \CFA subjectively ``feels like'' C.
1593An important part of this subjective feel is maintaining C's syntax and procedural paradigm, as opposed to functional and object-oriented approaches in other systems languages such as \CC and Rust.
1594Maintaining the C approach means that C coding-patterns remain not only useable but idiomatic in \CFA, reducing the mental burden of retraining C programmers and switching between C and \CFA development.
1595Nevertheless, some features from other approaches are undeniably convenient;
1596\CFA attempts to adapt these features to the C paradigm.
1597
1598
1599\subsection{Alternative Declaration Syntax}
1600
1601C declaration syntax is notoriously confusing and error prone.
1602For example, many C programmers are confused by a declaration as simple as:
1603\begin{cquote}
1604\lstDeleteShortInline@%
1605\begin{tabular}{@{}ll@{}}
1606\begin{cfa}
1607int * x[5]
1608\end{cfa}
1609&
1610\raisebox{-0.75\totalheight}{\input{Cdecl}}
1611\end{tabular}
1612\lstMakeShortInline@%
1613\end{cquote}
1614Is this an array of 5 pointers to integers or a pointer to an array of 5 integers?
1615If there is any doubt, it implies productivity and safety issues even for basic programs.
1616Another example of confusion results from the fact that a function name and its parameters are embedded within the return type, mimicking the way the return value is used at the function's call site.
1617For example, a function returning a pointer to an array of integers is defined and used in the following way:
1618\begin{cfa}
1619int `(*`f`())[`5`]` {...};                                      $\C{// definition}$
1620 ... `(*`f`())[`3`]` += 1;                                      $\C{// usage}$
1621\end{cfa}
1622Essentially, the return type is wrapped around the function name in successive layers (like an onion).
1623While attempting to make the two contexts consistent is a laudable goal, it has not worked out in practice.
1624
1625\CFA provides its own type, variable and function declarations, using a different syntax~\cite[pp.~856--859]{Buhr94a}.
1626The new declarations place qualifiers to the left of the base type, while C declarations place qualifiers to the right.
1627The qualifiers have the same meaning but are ordered left to right to specify a variable's type.
1628\begin{cquote}
1629\lstDeleteShortInline@%
1630\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{\hspace{2\parindentlnth}}l@{}}
1631\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
1632\begin{cfa}
1633`[5] *` int x1;
1634`* [5]` int x2;
1635`[* [5] int]` f( int p );
1636\end{cfa}
1637&
1638\begin{cfa}
1639int `*` x1 `[5]`;
1640int `(*`x2`)[5]`;
1641`int (*`f( int p )`)[5]`;
1642\end{cfa}
1643&
1644\begin{cfa}
1645// array of 5 pointers to int
1646// pointer to array of 5 int
1647// function returning pointer to array of 5 int and taking int
1648\end{cfa}
1649\end{tabular}
1650\lstMakeShortInline@%
1651\end{cquote}
1652The only exception is bit field specification, which always appear to the right of the base type.
1653% Specifically, the character @*@ is used to indicate a pointer, square brackets @[@\,@]@ are used to represent an array or function return value, and parentheses @()@ are used to indicate a function parameter.
1654However, unlike C, \CFA type declaration tokens are distributed across all variables in the declaration list.
1655For instance, variables @x@ and @y@ of type pointer to integer are defined in \CFA as follows:
1656\begin{cquote}
1657\lstDeleteShortInline@%
1658\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1659\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
1660\begin{cfa}
1661`*` int x, y;
1662int y;
1663\end{cfa}
1664&
1665\begin{cfa}
1666int `*`x, `*`y;
1667
1668\end{cfa}
1669\end{tabular}
1670\lstMakeShortInline@%
1671\end{cquote}
1672The downside of the \CFA semantics is the need to separate regular and pointer declarations.
1673
1674\begin{comment}
1675Other examples are:
1676\begin{cquote}
1677\lstDeleteShortInline@%
1678\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{\hspace{2\parindentlnth}}l@{}}
1679\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{C}}     \\
1680\begin{cfa}
1681[ 5 ] int z;
1682[ 5 ] * char w;
1683* [ 5 ] double v;
1684struct s {
1685        int f0:3;
1686        * int f1;
1687        [ 5 ] * int f2;
1688};
1689\end{cfa}
1690&
1691\begin{cfa}
1692int z[ 5 ];
1693char * w[ 5 ];
1694double (* v)[ 5 ];
1695struct s {
1696        int f0:3;
1697        int * f1;
1698        int * f2[ 5 ]
1699};
1700\end{cfa}
1701&
1702\begin{cfa}
1703// array of 5 integers
1704// array of 5 pointers to char
1705// pointer to array of 5 doubles
1706
1707// common bit field syntax
1708
1709
1710
1711\end{cfa}
1712\end{tabular}
1713\lstMakeShortInline@%
1714\end{cquote}
1715\end{comment}
1716
1717All specifiers (@extern@, @static@, \etc) and qualifiers (@const@, @volatile@, \etc) are used in the normal way with the new declarations and also appear left to right, \eg:
1718\begin{cquote}
1719\lstDeleteShortInline@%
1720\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{\hspace{2\parindentlnth}}l@{}}
1721\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{C}}     \\
1722\begin{cfa}
1723extern const * const int x;
1724static const * [5] const int y;
1725\end{cfa}
1726&
1727\begin{cfa}
1728int extern const * const x;
1729static const int (* const y)[5]
1730\end{cfa}
1731&
1732\begin{cfa}
1733// external const pointer to const int
1734// internal const pointer to array of 5 const int
1735\end{cfa}
1736\end{tabular}
1737\lstMakeShortInline@%
1738\end{cquote}
1739Specifiers must appear at the start of a \CFA function declaration\footnote{\label{StorageClassSpecifier}
1740The placement of a storage-class specifier other than at the beginning of the declaration specifiers in a declaration is an obsolescent feature.~\cite[\S~6.11.5(1)]{C11}}.
1741
1742The new declaration syntax can be used in other contexts where types are required, \eg casts and the pseudo-function @sizeof@:
1743\begin{cquote}
1744\lstDeleteShortInline@%
1745\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1746\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
1747\begin{cfa}
1748y = (* int)x;
1749i = sizeof([ 5 ] * int);
1750\end{cfa}
1751&
1752\begin{cfa}
1753y = (int *)x;
1754i = sizeof(int * [ 5 ]);
1755\end{cfa}
1756\end{tabular}
1757\lstMakeShortInline@%
1758\end{cquote}
1759
1760The syntax of the new function-prototype declaration follows directly from the new function-definition syntax;
1761as well, parameter names are optional, \eg:
1762\begin{cfa}
1763[ int x ] f ( /* void */ );             $\C[2.5in]{// returning int with no parameters}$
1764[ int x ] f (...);                              $\C{// returning int with unknown parameters}$
1765[ * int ] g ( int y );                  $\C{// returning pointer to int with int parameter}$
1766[ void ] h ( int, char );               $\C{// returning no result with int and char parameters}$
1767[ * int, int ] j ( int );               $\C{// returning pointer to int and int with int parameter}$
1768\end{cfa}
1769This syntax allows a prototype declaration to be created by cutting and pasting source text from the function-definition header (or vice versa).
1770Like C, it is possible to declare multiple function-prototypes in a single declaration, where the return type is distributed across \emph{all} function names in the declaration list, \eg:
1771\begin{cquote}
1772\lstDeleteShortInline@%
1773\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
1774\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
1775\begin{cfa}
1776[double] foo(), foo( int ), foo( double ) {...}
1777\end{cfa}
1778&
1779\begin{cfa}
1780double foo1( void ), foo2( int ), foo3( double );
1781\end{cfa}
1782\end{tabular}
1783\lstMakeShortInline@%
1784\end{cquote}
1785where \CFA allows the last function in the list to define its body.
1786
1787The syntax for pointers to \CFA functions specifies the pointer name on the right, \eg:
1788\begin{cfa}
1789* [ int x ] () fp;                              $\C{// pointer to function returning int with no parameters}$
1790* [ * int ] ( int y ) gp;               $\C{// pointer to function returning pointer to int with int parameter}$
1791* [ ] ( int, char ) hp;                 $\C{// pointer to function returning no result with int and char parameters}$
1792* [ * int, int ] ( int ) jp;    $\C{// pointer to function returning pointer to int and int with int parameter}$
1793\end{cfa}
1794Note, a function name cannot be specified:
1795\begin{cfa}
1796* [ int x ] f () fp;                    $\C{// function name "f" is disallowed}\CRT$
1797\end{cfa}
1798
1799Finally, new \CFA declarations may appear together with C declarations in the same program block, but cannot be mixed within a specific declaration.
1800Therefore, a programmer has the option of either continuing to use traditional C declarations or take advantage of the new style.
1801Clearly, both styles need to be supported for some time due to existing C-style header-files, particularly for UNIX-like systems.
1802
1803
1804\subsection{References}
1805\label{s:References}
1806
1807All variables in C have an \newterm{address}, a \newterm{value}, and a \newterm{type};
1808at the position in the program's memory denoted by the address, there exists a sequence of bits (the value), with the length and semantic meaning of this bit sequence defined by the type.
1809The C type-system does not always track the relationship between a value and its address;
1810a value that does not have a corresponding address is called a \newterm{rvalue} (for ``right-hand value''), while a value that does have an address is called a \newterm{lvalue} (for ``left-hand value'').
1811For example, in @int x; x = 42;@ the variable expression @x@ on the left-hand-side of the assignment is a lvalue, while the constant expression @42@ on the right-hand-side of the assignment is a rvalue.
1812Despite the nomenclature of ``left-hand'' and ``right-hand'', an expression's classification as lvalue or rvalue is entirely dependent on whether it has an address or not; in imperative programming, the address of a value is used for both reading and writing (mutating) a value, and as such, lvalues can be converted to rvalues and read from, but rvalues cannot be mutated because they lack a location to store the updated value.
1813
1814Within a lexical scope, lvalue expressions have an \newterm{address interpretation} for writing a value or a \newterm{value interpretation} to read a value.
1815For example, in @x = y@, @x@ has an address interpretation, while @y@ has a value interpretation.
1816While this duality of interpretation is useful, C lacks a direct mechanism to pass lvalues between contexts, instead relying on \newterm{pointer types} to serve a similar purpose.
1817In C, for any type @T@ there is a pointer type @T *@, the value of which is the address of a value of type @T@.
1818A pointer rvalue can be explicitly \newterm{dereferenced} to the pointed-to lvalue with the dereference operator @*?@, while the rvalue representing the address of a lvalue can be obtained with the address-of operator @&?@.
1819
1820\begin{cfa}
1821int x = 1, y = 2, * p1, * p2, ** p3;
1822p1 = &x;                                                                        $\C{// p1 points to x}$
1823p2 = &y;                                                                        $\C{// p2 points to y}$
1824p3 = &p1;                                                                       $\C{// p3 points to p1}$
1825*p2 = ((*p1 + *p2) * (**p3 - *p1)) / (**p3 - 15);
1826\end{cfa}
1827
1828Unfortunately, the dereference and address-of operators introduce a great deal of syntactic noise when dealing with pointed-to values rather than pointers, as well as the potential for subtle bugs because of pointer arithmetic.
1829For both brevity and clarity, it is desirable for the compiler to figure out how to elide the dereference operators in a complex expression such as the assignment to @*p2@ above.
1830However, since C defines a number of forms of \newterm{pointer arithmetic}, two similar expressions involving pointers to arithmetic types (\eg @*p1 + x@ and @p1 + x@) may each have well-defined but distinct semantics, introducing the possibility that a programmer may write one when they mean the other, and precluding any simple algorithm for elision of dereference operators.
1831To solve these problems, \CFA introduces reference types @T &@;
1832a @T &@ has exactly the same value as a @T *@, but where the @T *@ takes the address interpretation by default, a @T &@ takes the value interpretation by default, as below:
1833
1834\begin{cfa}
1835int x = 1, y = 2, & r1, & r2, && r3;
1836&r1 = &x;                                                                       $\C{// r1 points to x}$
1837&r2 = &y;                                                                       $\C{// r2 points to y}$
1838&&r3 = &&r1;                                                            $\C{// r3 points to r2}$
1839r2 = ((r1 + r2) * (r3 - r1)) / (r3 - 15);       $\C{// implicit dereferencing}$
1840\end{cfa}
1841
1842Except for auto-dereferencing by the compiler, this reference example is exactly the same as the previous pointer example.
1843Hence, a reference behaves like a variable name -- an lvalue expression which is interpreted as a value -- but also has the type system track the address of that value.
1844One way to conceptualize a reference is via a rewrite rule, where the compiler inserts a dereference operator before the reference variable for each reference qualifier in the reference variable declaration, so the previous example implicitly acts like:
1845
1846\begin{cfa}
1847`*`r2 = ((`*`r1 + `*`r2) * (`**`r3 - `*`r1)) / (`**`r3 - 15);
1848\end{cfa}
1849
1850References in \CFA are similar to those in \CC, with important improvements, which can be seen in the example above.
1851Firstly, \CFA does not forbid references to references.
1852This provides a much more orthogonal design for library implementors, obviating the need for workarounds such as @std::reference_wrapper@.
1853Secondly, \CFA references are rebindable, whereas \CC references have a fixed address.
1854\newsavebox{\LstBox}
1855\begin{lrbox}{\LstBox}
1856\lstset{basicstyle=\footnotesize\linespread{0.9}\sf}
1857\begin{cfa}
1858int & r = *new( int );
1859...                                                                                     $\C{// non-null reference}$
1860delete &r;                                                                      $\C{// unmanaged (programmer) memory-management}$
1861r += 1;                                                                         $\C{// undefined reference}$
1862\end{cfa}
1863\end{lrbox}
1864Rebinding allows \CFA references to be default-initialized (\eg to a null pointer\footnote{
1865While effort has been made into non-null reference checking in \CC and Java, the exercise seems moot for any non-managed languages (C/\CC), given that it only handles one of many different error situations:
1866\begin{cquote}
1867\usebox{\LstBox}
1868\end{cquote}
1869}%
1870) and point to different addresses throughout their lifetime, like pointers.
1871Rebinding is accomplished by extending the existing syntax and semantics of the address-of operator in C.
1872
1873In C, the address of a lvalue is always a rvalue, as in general that address is not stored anywhere in memory, and does not itself have an address.
1874In \CFA, the address of a @T &@ is a lvalue @T *@, as the address of the underlying @T@ is stored in the reference, and can thus be mutated there.
1875The result of this rule is that any reference can be rebound using the existing pointer assignment semantics by assigning a compatible pointer into the address of the reference, \eg @&r1 = &x;@ above.
1876This rebinding occurs to an arbitrary depth of reference nesting;
1877loosely speaking, nested address-of operators produce a nested lvalue pointer up to the depth of the reference.
1878These explicit address-of operators can be thought of as ``cancelling out'' the implicit dereference operators, \eg @(&`*`)r1 = &x@ or @(&(&`*`)`*`)r3 = &(&`*`)r1@ or even @(&`*`)r2 = (&`*`)`*`r3@ for @&r2 = &r3@.
1879More precisely:
1880\begin{itemize}
1881\item
1882if @R@ is an rvalue of type {@T &@$_1 \cdots$@ &@$_r$} where $r \ge 1$ references (@&@ symbols) than @&R@ has type {@T `*`&@$_{\color{red}2} \cdots$@ &@$_{\color{red}r}$}, \\ \ie @T@ pointer with $r-1$ references (@&@ symbols).
1883       
1884\item
1885if @L@ is an lvalue of type {@T &@$_1 \cdots$@ &@$_l$} where $l \ge 0$ references (@&@ symbols) then @&L@ has type {@T `*`&@$_{\color{red}1} \cdots$@ &@$_{\color{red}l}$}, \\ \ie @T@ pointer with $l$ references (@&@ symbols).
1886\end{itemize}
1887Since pointers and references share the same internal representation, code using either is equally performant; in fact the \CFA compiler converts references to pointers internally, and the choice between them is made solely on convenience, \eg many pointer or value accesses.
1888
1889By analogy to pointers, \CFA references also allow cv-qualifiers such as @const@:
1890\begin{cfa}
1891const int cx = 5;                                                       $\C{// cannot change cx}$
1892const int & cr = cx;                                            $\C{// cannot change cr's referred value}$
1893&cr = &cx;                                                                      $\C{// rebinding cr allowed}$
1894cr = 7;                                                                         $\C{// ERROR, cannot change cr}$
1895int & const rc = x;                                                     $\C{// must be initialized, like in \CC}$
1896&rc = &x;                                                                       $\C{// ERROR, cannot rebind rc}$
1897rc = 7;                                                                         $\C{// x now equal to 7}$
1898\end{cfa}
1899Given that a reference is meant to represent a lvalue, \CFA provides some syntactic shortcuts when initializing references.
1900There are three initialization contexts in \CFA: declaration initialization, argument/parameter binding, and return/temporary binding.
1901In each of these contexts, the address-of operator on the target lvalue is elided.
1902The syntactic motivation is clearest when considering overloaded operator-assignment, \eg @int ?+=?(int &, int)@; given @int x, y@, the expected call syntax is @x += y@, not @&x += y@.
1903
1904More generally, this initialization of references from lvalues rather than pointers is an instance of a ``lvalue-to-reference'' conversion rather than an elision of the address-of operator;
1905this conversion is used in any context in \CFA where an implicit conversion is allowed.
1906Similarly, use of a the value pointed to by a reference in an rvalue context can be thought of as a ``reference-to-rvalue'' conversion, and \CFA also includes a qualifier-adding ``reference-to-reference'' conversion, analogous to the @T *@ to @const T *@ conversion in standard C.
1907The final reference conversion included in \CFA is ``rvalue-to-reference'' conversion, implemented by means of an implicit temporary.
1908When an rvalue is used to initialize a reference, it is instead used to initialize a hidden temporary value with the same lexical scope as the reference, and the reference is initialized to the address of this temporary.
1909\begin{cfa}
1910struct S { double x, y; };
1911int x, y;
1912void f( int & i, int & j, S & s, int v[] );
1913f( 3, x + y, (S){ 1.0, 7.0 }, (int [3]){ 1, 2, 3 } ); $\C{// pass rvalue to lvalue \(\Rightarrow\) implicit temporary}$
1914\end{cfa}
1915This allows complex values to be succinctly and efficiently passed to functions, without the syntactic overhead of explicit definition of a temporary variable or the runtime cost of pass-by-value.
1916\CC allows a similar binding, but only for @const@ references; the more general semantics of \CFA are an attempt to avoid the \newterm{const hell} problem, in which addition of a @const@ qualifier to one reference requires a cascading chain of added qualifiers.
1917
1918
1919\subsection{Type Nesting}
1920
1921Nested types provide a mechanism to organize associated types and refactor a subset of fields into a named aggregate (\eg sub-aggregates @name@, @address@, @department@, within aggregate @employe@).
1922Java nested types are dynamic (apply to objects), \CC are static (apply to the \lstinline[language=C++]@class@), and C hoists (refactors) nested types into the enclosing scope, meaning there is no need for type qualification.
1923Since \CFA in not object-oriented, adopting dynamic scoping does not make sense;
1924instead \CFA adopts \CC static nesting, using the field-selection operator ``@.@'' for type qualification, as does Java, rather than the \CC type-selection operator ``@::@'' (see Figure~\ref{f:TypeNestingQualification}).
1925\begin{figure}
1926\centering
1927\lstDeleteShortInline@%
1928\begin{tabular}{@{}l@{\hspace{3em}}l|l@{}}
1929\multicolumn{1}{c@{\hspace{3em}}}{\textbf{C Type Nesting}}      & \multicolumn{1}{c|}{\textbf{C Implicit Hoisting}}     & \multicolumn{1}{c}{\textbf{\CFA}}     \\
1930\hline
1931\begin{cfa}
1932struct S {
1933        enum C { R, G, B };
1934        struct T {
1935                union U { int i, j; };
1936                enum C c;
1937                short int i, j;
1938        };
1939        struct T t;
1940} s;
1941
1942int rtn() {
1943        s.t.c = R;
1944        struct T t = { R, 1, 2 };
1945        enum C c;
1946        union U u;
1947}
1948\end{cfa}
1949&
1950\begin{cfa}
1951enum C { R, G, B };
1952union U { int i, j; };
1953struct T {
1954        enum C c;
1955        short int i, j;
1956};
1957struct S {
1958        struct T t;
1959} s;
1960       
1961
1962
1963
1964
1965
1966
1967\end{cfa}
1968&
1969\begin{cfa}
1970struct S {
1971        enum C { R, G, B };
1972        struct T {
1973                union U { int i, j; };
1974                enum C c;
1975                short int i, j;
1976        };
1977        struct T t;
1978} s;
1979
1980int rtn() {
1981        s.t.c = `S.`R;  // type qualification
1982        struct `S.`T t = { `S.`R, 1, 2 };
1983        enum `S.`C c;
1984        union `S.T.`U u;
1985}
1986\end{cfa}
1987\end{tabular}
1988\lstMakeShortInline@%
1989\caption{Type Nesting / Qualification}
1990\label{f:TypeNestingQualification}
1991\end{figure}
1992In the C left example, types @C@, @U@ and @T@ are implicitly hoisted outside of type @S@ into the containing block scope.
1993In the \CFA right example, the types are not hoisted and accessible.
1994
1995
1996\subsection{Constructors and Destructors}
1997\label{s:ConstructorsDestructors}
1998
1999One of the strengths (and weaknesses) of C is memory-management control, allowing resource release to be precisely specified versus unknown release with garbage-collected memory-management.
2000However, this manual approach is verbose, and it is useful to manage resources other than memory (\eg file handles) using the same mechanism as memory.
2001\CC addresses these issues using Resource Aquisition Is Initialization (RAII), implemented by means of \newterm{constructor} and \newterm{destructor} functions;
2002\CFA adopts constructors and destructors (and @finally@) to facilitate RAII.
2003While constructors and destructors are a common feature of object-oriented programming-languages, they are an independent capability allowing \CFA to adopt them while retaining a procedural paradigm.
2004Specifically, \CFA constructors and destructors are denoted by name and first parameter-type versus name and nesting in an aggregate type.
2005Constructor calls seamlessly integrate with existing C initialization syntax, providing a simple and familiar syntax to C programmers and allowing constructor calls to be inserted into legacy C code with minimal code changes.
2006
2007In \CFA, a constructor is named @?{}@ and a destructor is named @^?{}@\footnote{%
2008The symbol \lstinline+^+ is used for the destructor name because it was the last binary operator that could be used in a unary context.}.
2009The name @{}@ comes from the syntax for the initializer: @struct S { int i, j; } s = `{` 2, 3 `}`@.
2010Like other \CFA operators, these names represent the syntax used to call the constructor or destructor, \eg @?{}(x, ...)@ or @^{}(x, ...)@.
2011The constructor and destructor have return type @void@, and the first parameter is a reference to the object type to be constructed or destructed.
2012While the first parameter is informally called the @this@ parameter, as in object-oriented languages, any variable name may be used.
2013Both constructors and destructors allow additional parametes after the @this@ parameter for specifying values for initialization/de-initialization\footnote{
2014Destruction parameters are useful for specifying storage-management actions, such as de-initialize but not deallocate.}.
2015\begin{cfa}
2016struct VLA { int len, * data; };
2017void ?{}( VLA & vla ) with ( vla ) { len = 10;  data = alloc( len ); }  $\C{// default constructor}$
2018void ^?{}( VLA & vla ) with ( vla ) { free( data ); } $\C{// destructor}$
2019{
2020        VLA x;                                                                  $\C{// implicit:  ?\{\}( x );}$
2021}                                                                                       $\C{// implicit:  ?\^{}\{\}( x );}$
2022\end{cfa}
2023@VLA@ is a \newterm{managed type}\footnote{
2024A managed type affects the runtime environment versus a self-contained type.}: a type requiring a non-trivial constructor or destructor, or with a field of a managed type.
2025A managed type is implicitly constructed at allocation and destructed at deallocation to ensure proper interaction with runtime resources, in this case, the @data@ array in the heap.
2026For details of the code-generation placement of implicit constructor and destructor calls among complex executable statements see~\cite[\S~2.2]{Schluntz17}.
2027
2028\CFA also provides syntax for \newterm{initialization} and \newterm{copy}:
2029\begin{cfa}
2030void ?{}( VLA & vla, int size, char fill ) with ( vla ) {  $\C{// initialization}$
2031        len = size;  data = alloc( len, fill );
2032}
2033void ?{}( VLA & vla, VLA other ) {                      $\C{// copy, shallow}$
2034        vla.len = other.len;  vla.data = other.data;
2035}
2036\end{cfa}
2037(Note, the example is purposely simplified using shallow-copy semantics.)
2038An initialization constructor-call has the same syntax as a C initializer, except the initialization values are passed as arguments to a matching constructor (number and type of paremeters).
2039\begin{cfa}
2040VLA va = `{` 20, 0 `}`,  * arr = alloc()`{` 5, 0 `}`;
2041\end{cfa}
2042Note, the use of a \newterm{constructor expression} to initialize the storage from the dynamic storage-allocation.
2043Like \CC, the copy constructor has two parameters, the second of which is a value parameter with the same type as the first parameter;
2044appropriate care is taken to not recursively call the copy constructor when initializing the second parameter.
2045
2046\CFA constructors may be explicitly call, like Java, and destructors may be explicitly called, like \CC.
2047Explicit calls to constructors double as a \CC-style \emph{placement syntax}, useful for construction of member fields in user-defined constructors and reuse of existing storage allocations.
2048While existing call syntax works for explicit calls to constructors and destructors, \CFA also provides a more concise \newterm{operator syntax} for both:
2049\begin{cfa}
2050{
2051        VLA  x,            y = { 20, 0x01 },     z = y; $\C{// z points to y}$
2052        //      ?{}( x );   ?{}( y, 20, 0x01 );   ?{}( z, y );
2053        ^x{};                                                                   $\C{// deallocate x}$
2054        x{};                                                                    $\C{// reallocate x}$
2055        z{ 5, 0xff };                                                   $\C{// reallocate z, not pointing to y}$
2056        ^y{};                                                                   $\C{// deallocate y}$
2057        y{ x };                                                                 $\C{// reallocate y, points to x}$
2058        x{};                                                                    $\C{// reallocate x, not pointing to y}$
2059        // ^?{}(z);  ^?{}(y);  ^?{}(x);
2060}
2061\end{cfa}
2062
2063To provide a uniform type interface for @otype@ polymorphism, the \CFA compiler automatically generates a default constructor, copy constructor, assignment operator, and destructor for all types.
2064These default functions can be overridden by user-generated versions.
2065For compatibility with the standard behaviour of C, the default constructor and destructor for all basic, pointer, and reference types do nothing, while the copy constructor and assignment operator are bitwise copies;
2066if default zero-initialization is desired, the default constructors can be overridden.
2067For user-generated types, the four functions are also automatically generated.
2068@enum@ types are handled the same as their underlying integral type, and unions are also bitwise copied and no-op initialized and destructed.
2069For compatibility with C, a copy constructor from the first union member type is also defined.
2070For @struct@ types, each of the four functions are implicitly defined to call their corresponding functions on each member of the struct.
2071To better simulate the behaviour of C initializers, a set of \newterm{field constructors} is also generated for structures.
2072A constructor is generated for each non-empty prefix of a structure's member-list to copy-construct the members passed as parameters and default-construct the remaining members.
2073To allow users to limit the set of constructors available for a type, when a user declares any constructor or destructor, the corresponding generated function and all field constructors for that type are hidden from expression resolution;
2074similarly, the generated default constructor is hidden upon declaration of any constructor.
2075These semantics closely mirror the rule for implicit declaration of constructors in \CC\cite[p.~186]{ANSI98:C++}.
2076
2077In some circumstance programmers may not wish to have implicit constructor and destructor generation and calls.
2078In these cases, \CFA provides the initialization syntax \lstinline|S x `@=` {}|, and the object becomes unmanaged, so implicit constructor and destructor calls are not generated.
2079Any C initializer can be the right-hand side of an \lstinline|@=| initializer, \eg \lstinline|VLA a @= { 0, 0x0 }|, with the usual C initialization semantics.
2080The same syntax can be used in a compound literal, \eg \lstinline|a = VLA`@`{ 0, 0x0 }|, to create a C-style literal.
2081The point of \lstinline|@=| is to provide a migration path from legacy C code to \CFA, by providing a mechanism to incrementally convert to implicit initialization.
2082
2083
2084% \subsection{Default Parameters}
2085
2086
2087\section{Literals}
2088
2089C already includes limited polymorphism for literals -- @0@ can be either an integer or a pointer literal, depending on context, while the syntactic forms of literals of the various integer and float types are very similar, differing from each other only in suffix.
2090In keeping with the general \CFA approach of adding features while respecting the ``C-style'' of doing things, C's polymorphic constants and typed literal syntax are extended to interoperate with user-defined types, while maintaining a backwards-compatible semantics.
2091
2092A simple example is allowing the underscore, as in Ada, to separate prefixes, digits, and suffixes in all \CFA constants, \eg @0x`_`1.ffff`_`ffff`_`p`_`128`_`l@, where the underscore is also the standard separator in C identifiers.
2093\CC uses a single quote as a separator but it is restricted among digits, precluding its use in the literal prefix or suffix, \eg @0x1.ffff@@`'@@ffffp128l@, and causes problems with most IDEs, which must be extended to deal with this alternate use of the single quote.
2094
2095
2096\begin{comment}
2097\subsection{Integral Suffixes}
2098
2099New integral suffixes @hh@ (half of half of @int@) for @char@, @h@ (half of @int@) for @short@, and @z@ for @size_t@, and length suffixes for 8, 16, 32, 64, and 128 bit integers.
2100%Additional integral suffixes are added to cover all the integral types and lengths.
2101\begin{cquote}
2102\lstDeleteShortInline@%
2103\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{\hspace{2\parindentlnth}}l@{}}
2104\begin{cfa}
210520_`hh`     // signed char
210621_`hh`u   // unsigned char
210722_`h`       // signed short int
210823_u`h`     // unsigned short int
210924`z`         // size_t
2110\end{cfa}
2111&
2112\begin{cfa}
211320_`L8`      // int8_t
211421_u`l8`     // uint8_t
211522_`l16`     // int16_t
211623_u`l16`   // uint16_t
211724_`l32`     // int32_t
2118\end{cfa}
2119&
2120\begin{cfa}
212125_u`l32`      // uint32_t
212226_`l64`        // int64_t
212327_`l64`u      // uint64_t
212426_`L128`     // int128
212527_`L128`u   // unsigned int128
2126\end{cfa}
2127\end{tabular}
2128\lstMakeShortInline@%
2129\end{cquote}
2130\end{comment}
2131
2132
2133\subsection{0/1}
2134
2135In C, @0@ has the special property that it is the only ``false'' value;
2136from the standard, any value that compares equal to @0@ is false, while any value that compares unequal to @0@ is true.
2137As such, an expression @x@ in any boolean context (such as the condition of an @if@ or @while@ statement, or the arguments to @&&@, @||@, or @?:@\,) can be rewritten as @x != 0@ without changing its semantics.
2138Operator overloading in \CFA provides a natural means to implement this truth-value comparison for arbitrary types, but the C type system is not precise enough to distinguish an equality comparison with @0@ from an equality comparison with an arbitrary integer or pointer.
2139To provide this precision, \CFA introduces a new type @zero_t@ as the type of literal @0@ (somewhat analagous to @nullptr_t@ and @nullptr@ in \CCeleven);
2140@zero_t@ can only take the value @0@, but has implicit conversions to the integer and pointer types so that C code involving @0@ continues to work.
2141With this addition, \CFA rewrites @if (x)@ and similar expressions to @if ( (x) != 0 )@ or the appropriate analogue, and any type @T@ is ``truthy'' by defining an operator overload @int ?!=?( T, zero_t )@.
2142\CC makes types truthy by adding a conversion to @bool@;
2143prior to the addition of explicit cast operators in \CCeleven, this approach had the pitfall of making truthy types transitively convertable to any numeric type;
2144\CFA avoids this issue.
2145
2146Similarly, \CFA also has a special type for @1@, @one_t@;
2147like @zero_t@, @one_t@ has built-in implicit conversions to the various integral types so that @1@ maintains its expected semantics in legacy code for operations @++@ and @--@.
2148The addition of @one_t@ allows generic algorithms to handle the unit value uniformly for types where it is meaningful.
2149\TODO{Make this sentence true}
2150In particular, polymorphic functions in the \CFA prelude define @++x@ and @x++@ in terms of @x += 1@, allowing users to idiomatically define all forms of increment for a type @T@ by defining the single function @T & ?+=(T &, one_t)@;
2151analogous overloads for the decrement operators are present as well.
2152
2153
2154\subsection{User Literals}
2155
2156For readability, it is useful to associate units to scale literals, \eg weight (stone, pound, kilogram) or time (seconds, minutes, hours).
2157The left of Figure~\ref{f:UserLiteral} shows the \CFA alternative call-syntax (postfix: literal argument before function name), using the backquote, to convert basic literals into user literals.
2158The backquote is a small character, making the unit (function name) predominate.
2159For examples, the multi-precision integer-type in Section~\ref{s:MultiPrecisionIntegers} has user literals:
2160{\lstset{language=CFA,moredelim=**[is][\color{red}]{|}{|},deletedelim=**[is][]{`}{`}}
2161\begin{cfa}
2162y = 9223372036854775807L|`mp| * 18446744073709551615UL|`mp|;
2163y = "12345678901234567890123456789"|`mp| + "12345678901234567890123456789"|`mp|;
2164\end{cfa}
2165Because \CFA uses a standard function, all types and literals are applicable, as well as overloading and conversions, where @?`@ denotes a postfix-function name and @`@ denotes a postfix-function call.
2166}%
2167\begin{cquote}
2168\lstset{language=CFA,moredelim=**[is][\color{red}]{|}{|},deletedelim=**[is][]{`}{`}}
2169\lstDeleteShortInline@%
2170\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{\hspace{2\parindentlnth}}l@{\hspace{2\parindentlnth}}l@{}}
2171\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{postfix function}}        & \multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{constant}}      & \multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{variable/expression}}   & \multicolumn{1}{c}{\textbf{postfix pointer}}  \\
2172\begin{cfa}
2173int |?`h|( int s );
2174int |?`h|( double s );
2175int |?`m|( char c );
2176int |?`m|( const char * s );
2177int |?`t|( int a, int b, int c );
2178\end{cfa}
2179&
2180\begin{cfa}
21810 |`h|;
21823.5|`h|;
2183'1'|`m|;
2184"123" "456"|`m|;
2185[1,2,3]|`t|;
2186\end{cfa}
2187&
2188\begin{cfa}
2189int i = 7;
2190i|`h|;
2191(i + 3)|`h|;
2192(i + 3.5)|`h|;
2193
2194\end{cfa}
2195&
2196\begin{cfa}
2197int (* |?`p|)( int i );
2198|?`p| = |?`h|;
21993|`p|;
2200i|`p|;
2201(i + 3)|`p|;
2202\end{cfa}
2203\end{tabular}
2204\lstMakeShortInline@%
2205\end{cquote}
2206
2207The right of Figure~\ref{f:UserLiteral} shows the equivalent \CC version using the underscore for the call-syntax.
2208However, \CC restricts the types, \eg @unsigned long long int@ and @long double@ to represent integral and floating literals.
2209After which, user literals must match (no conversions);
2210hence, it is necessary to overload the unit with all appropriate types.
2211
2212\begin{figure}
2213\centering
2214\lstset{language=CFA,moredelim=**[is][\color{red}]{|}{|},deletedelim=**[is][]{`}{`}}
2215\lstDeleteShortInline@%
2216\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2217\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{\CC}}      \\
2218\begin{cfa}
2219struct W {
2220        double stones;
2221};
2222void ?{}( W & w ) { w.stones = 0; }
2223void ?{}( W & w, double w ) { w.stones = w; }
2224W ?+?( W l, W r ) {
2225        return (W){ l.stones + r.stones };
2226}
2227W |?`st|(double w) { return (W){ w }; }
2228W |?`lb|(double w) { return (W){ w/14.0 }; }
2229W |?`kg|(double w) { return (W){ w*0.16 }; }
2230
2231
2232
2233int main() {
2234        W w, heavy = { 20 };
2235        w = 155|`lb|;
2236        w = 0b_1111|`st|;
2237        w = 0_233|`lb|;
2238        w = 0x_9b_u|`kg|;
2239        w = 5.5|`st| + 8|`kg| + 25.01|`lb| + heavy;
2240}
2241\end{cfa}
2242&
2243\begin{cfa}
2244struct W {
2245        double stones;
2246        W() { stones = 0.0; }
2247        W( double w ) { stones = w; }
2248};
2249W operator+( W l, W r ) {
2250        return W( l.stones + r.stones );
2251}
2252W |operator""_st|(unsigned long long int w) {return W(w); }
2253W |operator""_lb|(unsigned long long int w) {return W(w/14.0); }
2254W |operator""_kg|(unsigned long long int w) {return W(w*0.16); }
2255W |operator""_st|(long double w ) { return W( w ); }
2256W |operator""_lb|(long double w ) { return W( w / 14.0 ); }
2257W |operator""_kg|(long double w ) { return W( w * 0.16 ); }
2258int main() {
2259        W w, heavy = { 20 };
2260        w = 155|_lb|;
2261        w = 0b1111|_lb|;       // error, binary unsupported
2262        w = 0${\color{red}\LstBasicStyle{'}}$233|_lb|;          // quote separator
2263        w = 0x9b|_kg|;
2264        w = 5.5d|_st| + 8|_kg| + 25.01|_lb| + heavy;
2265}
2266\end{cfa}
2267\end{tabular}
2268\lstMakeShortInline@%
2269\caption{User Literal}
2270\label{f:UserLiteral}
2271\end{figure}
2272
2273
2274\section{Libraries}
2275\label{sec:libraries}
2276
2277As stated in Section~\ref{sec:poly-fns}, \CFA inherits a large corpus of library code, where other programming languages must rewrite or provide fragile inter-language communication with C.
2278\CFA has replacement libraries condensing hundreds of existing C names into tens of \CFA overloaded names, all without rewriting the actual computations.
2279In many cases, the interface is an inline wrapper providing overloading during compilation but zero cost at runtime.
2280The following sections give a glimpse of the interface reduction to many C libraries.
2281In many cases, @signed@/@unsigned@ @char@, @short@, and @_Complex@ functions are available (but not shown) to ensure expression computations remain in a single type, as conversions can distort results.
2282
2283
2284\subsection{Limits}
2285
2286C library @limits.h@ provides lower and upper bound constants for the basic types.
2287\CFA name overloading is used to condense these typed constants, \eg:
2288\begin{cquote}
2289\lstDeleteShortInline@%
2290\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2291\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{Definition}}      & \multicolumn{1}{c}{\textbf{Usage}}    \\
2292\begin{cfa}
2293const short int `MIN` = -32768;
2294const int `MIN` = -2147483648;
2295const long int `MIN` = -9223372036854775808L;
2296\end{cfa}
2297&
2298\begin{cfa}
2299short int si = `MIN`;
2300int i = `MIN`;
2301long int li = `MIN`;
2302\end{cfa}
2303\end{tabular}
2304\lstMakeShortInline@%
2305\end{cquote}
2306The result is a significant reduction in names to access typed constants, \eg:
2307\begin{cquote}
2308\lstDeleteShortInline@%
2309\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2310\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
2311\begin{cfa}
2312MIN
2313
2314MAX
2315
2316PI
2317E
2318\end{cfa}
2319&
2320\begin{cfa}
2321SCHAR_MIN, CHAR_MIN, SHRT_MIN, INT_MIN, LONG_MIN,
2322        LLONG_MIN, FLT_MIN, DBL_MIN, LDBL_MIN
2323SCHAR_MAX, UCHAR_MAX, SHRT_MAX, INT_MAX, LONG_MAX,
2324        LLONG_MAX, FLT_MAX, DBL_MAX, LDBL_MAX
2325M_PI, M_PIl
2326M_E, M_El
2327\end{cfa}
2328\end{tabular}
2329\lstMakeShortInline@%
2330\end{cquote}
2331
2332
2333\subsection{Math}
2334
2335C library @math.h@ provides many mathematical functions.
2336\CFA function overloading is used to condense these mathematical functions, \eg:
2337\begin{cquote}
2338\lstDeleteShortInline@%
2339\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2340\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{Definition}}      & \multicolumn{1}{c}{\textbf{Usage}}    \\
2341\begin{cfa}
2342float `log`( float x );
2343double `log`( double );
2344double _Complex `log`( double _Complex x );
2345\end{cfa}
2346&
2347\begin{cfa}
2348float f = `log`( 3.5 );
2349double d = `log`( 3.5 );
2350double _Complex dc = `log`( 3.5+0.5I );
2351\end{cfa}
2352\end{tabular}
2353\lstMakeShortInline@%
2354\end{cquote}
2355The result is a significant reduction in names to access math functions, \eg:
2356\begin{cquote}
2357\lstDeleteShortInline@%
2358\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2359\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
2360\begin{cfa}
2361log
2362sqrt
2363sin
2364\end{cfa}
2365&
2366\begin{cfa}
2367logf, log, logl, clogf, clog, clogl
2368sqrtf, sqrt, sqrtl, csqrtf, csqrt, csqrtl
2369sinf, sin, sinl, csinf, csin, csinl
2370\end{cfa}
2371\end{tabular}
2372\lstMakeShortInline@%
2373\end{cquote}
2374While \Celeven has type-generic math~\cite[\S~7.25]{C11} in @tgmath.h@ to provide a similar mechanism, these macros are limited, matching a function name with a single set of floating type(s).
2375For example, it is impossible to overload @atan@ for both one and two arguments;
2376instead the names @atan@ and @atan2@ are required (see Section~\ref{s:NameOverloading}).
2377The key observation is that only a restricted set of type-generic macros are provided for a limited set of function names, which do not generalize across the type system, as in \CFA.
2378
2379
2380\subsection{Standard}
2381
2382C library @stdlib.h@ provides many general functions.
2383\CFA function overloading is used to condense these utility functions, \eg:
2384\begin{cquote}
2385\lstDeleteShortInline@%
2386\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2387\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{Definition}}      & \multicolumn{1}{c}{\textbf{Usage}}    \\
2388\begin{cfa}
2389unsigned int `abs`( int );
2390double `abs`( double );
2391double abs( double _Complex );
2392\end{cfa}
2393&
2394\begin{cfa}
2395unsigned int i = `abs`( -1 );
2396double d = `abs`( -1.5 );
2397double d = `abs`( -1.5+0.5I );
2398\end{cfa}
2399\end{tabular}
2400\lstMakeShortInline@%
2401\end{cquote}
2402The result is a significant reduction in names to access utility functions, \eg:
2403\begin{cquote}
2404\lstDeleteShortInline@%
2405\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2406\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
2407\begin{cfa}
2408abs
2409strto
2410random
2411\end{cfa}
2412&
2413\begin{cfa}
2414abs, labs, llabs, fabsf, fabs, fabsl, cabsf, cabs, cabsl
2415strtol, strtoul, strtoll, strtoull, strtof, strtod, strtold
2416srand48, mrand48, lrand48, drand48
2417\end{cfa}
2418\end{tabular}
2419\lstMakeShortInline@%
2420\end{cquote}
2421In additon, there are polymorphic functions, like @min@ and @max@, that work on any type with operators @?<?@ or @?>?@.
2422
2423The following shows one example where \CFA \emph{extends} an existing standard C interface to reduce complexity and provide safety.
2424C/\Celeven provide a number of complex and overlapping storage-management operation to support the following capabilities:
2425\begin{description}%[topsep=3pt,itemsep=2pt,parsep=0pt]
2426\item[fill]
2427an allocation with a specified character.
2428\item[resize]
2429an existing allocation to decreased or increased its size.
2430In either case, new storage may or may not be allocated and, if there is a new allocation, as much data from the existing allocation is copied.
2431For an increase in storage size, new storage after the copied data may be filled.
2432\item[align]
2433an allocation on a specified memory boundary, \eg, an address multiple of 64 or 128 for cache-line purposes.
2434\item[array]
2435allocation with a specified number of elements.
2436An array may be filled, resized, or aligned.
2437\end{description}
2438Table~\ref{t:StorageManagementOperations} shows the capabilities provided by C/\Celeven allocation-functions and how all the capabilities can be combined into two \CFA functions.
2439\CFA storage-management functions extend the C equivalents by overloading, providing shallow type-safety, and removing the need to specify the base allocation-size.
2440Figure~\ref{f:StorageAllocation} contrasts \CFA and C storage-allocation performing the same operations with the same type safety.
2441
2442\begin{table}
2443\centering
2444\lstDeleteShortInline@%
2445\lstMakeShortInline~%
2446\begin{tabular}{@{}r|r|l|l|l|l@{}}
2447\multicolumn{1}{c}{}&           & \multicolumn{1}{c|}{fill}     & resize        & align & array \\
2448\hline
2449C               & ~malloc~                      & no                    & no            & no            & no    \\
2450                & ~calloc~                      & yes (0 only)  & no            & no            & yes   \\
2451                & ~realloc~                     & no/copy               & yes           & no            & no    \\
2452                & ~memalign~            & no                    & no            & yes           & no    \\
2453                & ~posix_memalign~      & no                    & no            & yes           & no    \\
2454\hline
2455C11             & ~aligned_alloc~       & no                    & no            & yes           & no    \\
2456\hline
2457\CFA    & ~alloc~                       & yes/copy              & no/yes        & no            & yes   \\
2458                & ~align_alloc~         & yes                   & no            & yes           & yes   \\
2459\end{tabular}
2460\lstDeleteShortInline~%
2461\lstMakeShortInline@%
2462\caption{Storage-Management Operations}
2463\label{t:StorageManagementOperations}
2464\end{table}
2465
2466\begin{figure}
2467\centering
2468\begin{cquote}
2469\begin{cfa}[aboveskip=0pt]
2470size_t  dim = 10;                                                       $\C{// array dimension}$
2471char fill = '\xff';                                                     $\C{// initialization fill value}$
2472int * ip;
2473\end{cfa}
2474\lstDeleteShortInline@%
2475\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2476\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{C}}        \\
2477\begin{cfa}
2478ip = alloc();
2479ip = alloc( fill );
2480ip = alloc( dim );
2481ip = alloc( dim, fill );
2482ip = alloc( ip, 2 * dim );
2483ip = alloc( ip, 4 * dim, fill );
2484
2485ip = align_alloc( 16 );
2486ip = align_alloc( 16, fill );
2487ip = align_alloc( 16, dim );
2488ip = align_alloc( 16, dim, fill );
2489\end{cfa}
2490&
2491\begin{cfa}
2492ip = (int *)malloc( sizeof( int ) );
2493ip = (int *)malloc( sizeof( int ) ); memset( ip, fill, sizeof( int ) );
2494ip = (int *)malloc( dim * sizeof( int ) );
2495ip = (int *)malloc( sizeof( int ) ); memset( ip, fill, dim * sizeof( int ) );
2496ip = (int *)realloc( ip, 2 * dim * sizeof( int ) );
2497ip = (int *)realloc( ip, 4 * dim * sizeof( int ) );
2498                        memset( ip, fill, 4 * dim * sizeof( int ) );
2499ip = memalign( 16, sizeof( int ) );
2500ip = memalign( 16, sizeof( int ) ); memset( ip, fill, sizeof( int ) );
2501ip = memalign( 16, dim * sizeof( int ) );
2502ip = memalign( 16, dim * sizeof( int ) ); memset( ip, fill, dim * sizeof( int ) );
2503\end{cfa}
2504\end{tabular}
2505\lstMakeShortInline@%
2506\end{cquote}
2507\caption{\CFA versus C Storage-Allocation}
2508\label{f:StorageAllocation}
2509\end{figure}
2510
2511Variadic @new@ (see Section~\ref{sec:variadic-tuples}) cannot support the same overloading because extra parameters are for initialization.
2512Hence, there are @new@ and @anew@ functions for single and array variables, and the fill value is the arguments to the constructor, \eg:
2513\begin{cfa}
2514struct S { int i, j; };
2515void ?{}( S & s, int i, int j ) { s.i = i; s.j = j; }
2516S * s = new( 2, 3 );                                            $\C{// allocate storage and run constructor}$
2517S * as = anew( dim, 2, 3 );                                     $\C{// each array element initialized to 2, 3}$
2518\end{cfa}
2519Note, \CC can only initialization array elements via the default constructor.
2520
2521Finally, the \CFA memory-allocator has \newterm{sticky properties} for dynamic storage: fill and alignment are remembered with an object's storage in the heap.
2522When a @realloc@ is performed, the sticky properties are respected, so that new storage is correctly aligned and initialized with the fill character.
2523
2524
2525\subsection{I/O}
2526\label{s:IOLibrary}
2527
2528The goal of \CFA I/O is to simplify the common cases, while fully supporting polymorphism and user defined types in a consistent way.
2529The approach combines ideas from \CC and Python.
2530The \CFA header file for the I/O library is @fstream@.
2531
2532The common case is printing out a sequence of variables separated by whitespace.
2533\begin{cquote}
2534\lstDeleteShortInline@%
2535\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
2536\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{c}{\textbf{\CC}}      \\
2537\begin{cfa}
2538int x = 1, y = 2, z = 3;
2539sout | x `|` y `|` z | endl;
2540\end{cfa}
2541&
2542\begin{cfa}
2543
2544cout << x `<< " "` << y `<< " "` << z << endl;
2545\end{cfa}
2546\\
2547\begin{cfa}[showspaces=true,aboveskip=0pt,belowskip=0pt]
25481` `2` `3
2549\end{cfa}
2550&
2551\begin{cfa}[showspaces=true,aboveskip=0pt,belowskip=0pt]
25521 2 3
2553\end{cfa}
2554\end{tabular}
2555\lstMakeShortInline@%
2556\end{cquote}
2557The \CFA form has half the characters of the \CC form, and is similar to Python I/O with respect to implicit separators.
2558Similar simplification occurs for tuple I/O, which prints all tuple values separated by ``\lstinline[showspaces=true]@, @''.
2559\begin{cfa}
2560[int, [ int, int ] ] t1 = [ 1, [ 2, 3 ] ], t2 = [ 4, [ 5, 6 ] ];
2561sout | t1 | t2 | endl;                                  $\C{// print tuples}$
2562\end{cfa}
2563\begin{cfa}[showspaces=true,aboveskip=0pt]
25641`, `2`, `3 4`, `5`, `6
2565\end{cfa}
2566Finally, \CFA uses the logical-or operator for I/O as it is the lowest-priority overloadable operator, other than assignment.
2567Therefore, fewer output expressions require parenthesis.
2568\begin{cquote}
2569\lstDeleteShortInline@%
2570\begin{tabular}{@{}ll@{}}
2571\textbf{\CFA:}
2572&
2573\begin{cfa}
2574sout | x * 3 | y + 1 | z << 2 | x == y | (x | y) | (x || y) | (x > z ? 1 : 2) | endl;
2575\end{cfa}
2576\\
2577\textbf{\CC:}
2578&
2579\begin{cfa}
2580cout << x * 3 << y + 1 << `(`z << 2`)` << `(`x == y`)` << (x | y) << (x || y) << (x > z ? 1 : 2) << endl;
2581\end{cfa}
2582\\
2583&
2584\begin{cfa}[showspaces=true,aboveskip=0pt]
25853 3 12 0 3 1 2
2586\end{cfa}
2587\end{tabular}
2588\lstMakeShortInline@%
2589\end{cquote}
2590There is a weak similarity between the \CFA logical-or operator and the Shell pipe-operator for moving data, where data flows in the correct direction for input but the opposite direction for output.
2591
2592The implicit separator character (space/blank) is a separator not a terminator.
2593The rules for implicitly adding the separator are:
2594\begin{itemize}
2595\item
2596A separator does not appear at the start or end of a line.
2597\item
2598A separator does not appear before or after a character literal or variable.
2599\item
2600A separator does not appear before or after a null (empty) C string, which is a local mechanism to disable insertion of the separator character.
2601\item
2602A separator does not appear before a C string starting with the characters: \lstinline[mathescape=off,basicstyle=\tt]@([{=$@
2603\item
2604A separator does not appear after a C string ending with the characters: \lstinline[basicstyle=\tt]@,.;!?)]}%@
2605\item
2606{\lstset{language=CFA,deletedelim=**[is][]{`}{`}}
2607A separator does not appear before or after a C string beginning/ending with the quote or whitespace characters: \lstinline[basicstyle=\tt,showspaces=true]@`'": \t\v\f\r\n@
2608}%
2609\end{itemize}
2610There are functions to set and get the separator string, and manipulators to toggle separation on and off in the middle of output.
2611
2612
2613\subsection{Multi-precision Integers}
2614\label{s:MultiPrecisionIntegers}
2615
2616\CFA has an interface to the GMP multi-precision signed-integers~\cite{GMP}, similar to the \CC interface provided by GMP.
2617The \CFA interface wraps GMP functions into operator functions to make programming with multi-precision integers identical to using fixed-sized integers.
2618The \CFA type name for multi-precision signed-integers is @Int@ and the header file is @gmp@.
2619Figure~\ref{f:GMPInterface} shows a multi-precision factorial-program contrasting the GMP interface in \CFA and C.
2620
2621\begin{figure}
2622\centering
2623\lstDeleteShortInline@%
2624\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}@{\hspace{2\parindentlnth}}l@{}}
2625\multicolumn{1}{c@{\hspace{2\parindentlnth}}}{\textbf{\CFA}}    & \multicolumn{1}{@{\hspace{2\parindentlnth}}c}{\textbf{C}}     \\
2626\begin{cfa}
2627#include <gmp>
2628int main( void ) {
2629        sout | "Factorial Numbers" | endl;
2630        Int fact = 1;
2631        sout | 0 | fact | endl;
2632        for ( unsigned int i = 1; i <= 40; i += 1 ) {
2633                fact *= i;
2634                sout | i | fact | endl;
2635        }
2636}
2637\end{cfa}
2638&
2639\begin{cfa}
2640#include <gmp.h>
2641int main( void ) {
2642        `gmp_printf`( "Factorial Numbers\n" );
2643        `mpz_t` fact;  `mpz_init_set_ui`( fact, 1 );
2644        `gmp_printf`( "%d %Zd\n", 0, fact );
2645        for ( unsigned int i = 1; i <= 40; i += 1 ) {
2646                `mpz_mul_ui`( fact, fact, i );
2647                `gmp_printf`( "%d %Zd\n", i, fact );
2648        }
2649}
2650\end{cfa}
2651\end{tabular}
2652\lstMakeShortInline@%
2653\caption{GMP Interface \CFA versus C}
2654\label{f:GMPInterface}
2655\end{figure}
2656
2657
2658\section{Evaluation}
2659\label{sec:eval}
2660
2661Though \CFA provides significant added functionality over C, these features have a low runtime penalty.
2662In fact, \CFA's features for generic programming can enable faster runtime execution than idiomatic @void *@-based C code.
2663This claim is demonstrated through a set of generic-code-based micro-benchmarks in C, \CFA, and \CC (see stack implementations in Appendix~\ref{sec:BenchmarkStackImplementations}).
2664Since all these languages share a subset essentially comprising standard C, maximal-performance benchmarks should show little runtime variance, differing only in length and clarity of source code.
2665A more illustrative comparison measures the costs of idiomatic usage of each language's features.
2666Figure~\ref{fig:BenchmarkTest} shows the \CFA benchmark tests for a generic stack based on a singly linked-list.
2667The benchmark test is similar for the other languages.
2668The experiment uses element types @int@ and @pair(short, char)@, and pushes $N=40M$ elements on a generic stack, copies the stack, clears one of the stacks, and finds the maximum value in the other stack.
2669
2670\begin{figure}
2671\begin{cfa}[xleftmargin=3\parindentlnth,aboveskip=0pt,belowskip=0pt]
2672int main() {
2673        int max = 0, val = 42;
2674        stack( int ) si, ti;
2675
2676        REPEAT_TIMED( "push_int", N, push( si, val ); )
2677        TIMED( "copy_int", ti{ si }; )
2678        TIMED( "clear_int", clear( si ); )
2679        REPEAT_TIMED( "pop_int", N, int x = pop( ti ); if ( x > max ) max = x; )
2680
2681        pair( short, char ) max = { 0h, '\0' }, val = { 42h, 'a' };
2682        stack( pair( short, char ) ) sp, tp;
2683
2684        REPEAT_TIMED( "push_pair", N, push( sp, val ); )
2685        TIMED( "copy_pair", tp{ sp }; )
2686        TIMED( "clear_pair", clear( sp ); )
2687        REPEAT_TIMED( "pop_pair", N, pair(short, char) x = pop( tp ); if ( x > max ) max = x; )
2688}
2689\end{cfa}
2690\caption{\protect\CFA Benchmark Test}
2691\label{fig:BenchmarkTest}
2692\end{figure}
2693
2694The structure of each benchmark implemented is: C with @void *@-based polymorphism, \CFA with the presented features, \CC with templates, and \CC using only class inheritance for polymorphism, called \CCV.
2695The \CCV variant illustrates an alternative object-oriented idiom where all objects inherit from a base @object@ class, mimicking a Java-like interface;
2696hence runtime checks are necessary to safely down-cast objects.
2697The most notable difference among the implementations is in memory layout of generic types: \CFA and \CC inline the stack and pair elements into corresponding list and pair nodes, while C and \CCV lack such a capability and instead must store generic objects via pointers to separately-allocated objects.
2698Note that the C benchmark uses unchecked casts as there is no runtime mechanism to perform such checks, while \CFA and \CC provide type-safety statically.
2699
2700Figure~\ref{fig:eval} and Table~\ref{tab:eval} show the results of running the benchmark in Figure~\ref{fig:BenchmarkTest} and its C, \CC, and \CCV equivalents.
2701The graph plots the median of 5 consecutive runs of each program, with an initial warm-up run omitted.
2702All code is compiled at \texttt{-O2} by gcc or g++ 6.4.0, with all \CC code compiled as \CCfourteen.
2703The benchmarks are run on an Ubuntu 16.04 workstation with 16 GB of RAM and a 6-core AMD FX-6300 CPU with 3.5 GHz maximum clock frequency.
2704
2705\begin{figure}
2706\centering
2707\input{timing}
2708\caption{Benchmark Timing Results (smaller is better)}
2709\label{fig:eval}
2710\end{figure}
2711
2712\begin{table}
2713\centering
2714\caption{Properties of benchmark code}
2715\label{tab:eval}
2716\newcommand{\CT}[1]{\multicolumn{1}{c}{#1}}
2717\begin{tabular}{rrrrr}
2718                                                                        & \CT{C}        & \CT{\CFA}     & \CT{\CC}      & \CT{\CCV}             \\ \hline
2719maximum memory usage (MB)                       & 10,001        & 2,502         & 2,503         & 11,253                \\
2720source code size (lines)                        & 201           & 191           & 125           & 294                   \\
2721redundant type annotations (lines)      & 27            & 0                     & 2                     & 16                    \\
2722binary size (KB)                                        & 14            & 257           & 14            & 37                    \\
2723\end{tabular}
2724\end{table}
2725
2726The C and \CCV variants are generally the slowest with the largest memory footprint, because of their less-efficient memory layout and the pointer-indirection necessary to implement generic types;
2727this inefficiency is exacerbated by the second level of generic types in the pair benchmarks.
2728By contrast, the \CFA and \CC variants run in roughly equivalent time for both the integer and pair of @short@ and @char@ because the storage layout is equivalent, with the inlined libraries (\ie no separate compilation) and greater maturity of the \CC compiler contributing to its lead.
2729\CCV is slower than C largely due to the cost of runtime type-checking of down-casts (implemented with @dynamic_cast@);
2730The outlier in the graph for \CFA, pop @pair@, results from the complexity of the generated-C polymorphic code.
2731The gcc compiler is unable to optimize some dead code and condense nested calls; a compiler designed for \CFA could easily perform these optimizations.
2732Finally, the binary size for \CFA is larger because of static linking with the \CFA libraries.
2733
2734\CFA is also competitive in terms of source code size, measured as a proxy for programmer effort. The line counts in Table~\ref{tab:eval} include implementations of @pair@ and @stack@ types for all four languages for purposes of direct comparison, though it should be noted that \CFA and \CC have pre-written data structures in their standard libraries that programmers would generally use instead. Use of these standard library types has minimal impact on the performance benchmarks, but shrinks the \CFA and \CC benchmarks to 39 and 42 lines, respectively.
2735The difference between the \CFA and \CC line counts is primarily declaration duplication to implement separate compilation; a header-only \CFA library would be similar in length to the \CC version.
2736On the other hand, C does not have a generic collections-library in its standard distribution, resulting in frequent reimplementation of such collection types by C programmers.
2737\CCV does not use the \CC standard template library by construction, and in fact includes the definition of @object@ and wrapper classes for @char@, @short@, and @int@ in its line count, which inflates this count somewhat, as an actual object-oriented language would include these in the standard library;
2738with their omission, the \CCV line count is similar to C.
2739We justify the given line count by noting that many object-oriented languages do not allow implementing new interfaces on library types without subclassing or wrapper types, which may be similarly verbose.
2740
2741Line-count is a fairly rough measure of code complexity;
2742another important factor is how much type information the programmer must specify manually, especially where that information is not compiler-checked.
2743Such unchecked type information produces a heavier documentation burden and increased potential for runtime bugs, and is much less common in \CFA than C, with its manually specified function pointer arguments and format codes, or \CCV, with its extensive use of untype-checked downcasts, \eg @object@ to @integer@ when popping a stack.
2744To quantify this manual typing, the ``redundant type annotations'' line in Table~\ref{tab:eval} counts the number of lines on which the type of a known variable is respecified, either as a format specifier, explicit downcast, type-specific function, or by name in a @sizeof@, struct literal, or @new@ expression.
2745The \CC benchmark uses two redundant type annotations to create a new stack nodes, while the C and \CCV benchmarks have several such annotations spread throughout their code.
2746The \CFA benchmark is able to eliminate all redundant type annotations through use of the polymorphic @alloc@ function discussed in Section~\ref{sec:libraries}.
2747
2748
2749\section{Related Work}
2750
2751
2752\subsection{Polymorphism}
2753
2754\CC provides three disjoint polymorphic extensions to C: overloading, inheritance, and templates.
2755The overloading is restricted because resolution does not use the return type, inheritance requires learning object-oriented programming and coping with a restricted nominal-inheritance hierarchy, templates cannot be separately compiled resulting in compilation/code bloat and poor error messages, and determining how these mechanisms interact and which to use is confusing.
2756In contrast, \CFA has a single facility for polymorphic code supporting type-safe separate-compilation of polymorphic functions and generic (opaque) types, which uniformly leverage the C procedural paradigm.
2757The key mechanism to support separate compilation is \CFA's \emph{explicit} use of assumed type properties.
2758Until \CC concepts~\cite{C++Concepts} are standardized (anticipated for \CCtwenty), \CC provides no way to specify the requirements of a generic function beyond compilation errors during template expansion;
2759furthermore, \CC concepts are restricted to template polymorphism.
2760
2761Cyclone~\cite{Grossman06} also provides capabilities for polymorphic functions and existential types, similar to \CFA's @forall@ functions and generic types.
2762Cyclone existential types can include function pointers in a construct similar to a virtual function-table, but these pointers must be explicitly initialized at some point in the code, a tedious and potentially error-prone process.
2763Furthermore, Cyclone's polymorphic functions and types are restricted to abstraction over types with the same layout and calling convention as @void *@, \ie only pointer types and @int@.
2764In \CFA terms, all Cyclone polymorphism must be dtype-static.
2765While the Cyclone design provides the efficiency benefits discussed in Section~\ref{sec:generic-apps} for dtype-static polymorphism, it is more restrictive than \CFA's general model.
2766Smith and Volpano~\cite{Smith98} present Polymorphic C, an ML dialect with polymorphic functions, C-like syntax, and pointer types; it lacks many of C's features, however, most notably structure types, and so is not a practical C replacement.
2767
2768Objective-C~\cite{obj-c-book} is an industrially successful extension to C.
2769However, Objective-C is a radical departure from C, using an object-oriented model with message-passing.
2770Objective-C did not support type-checked generics until recently \cite{xcode7}, historically using less-efficient runtime checking of object types.
2771The GObject~\cite{GObject} framework also adds object-oriented programming with runtime type-checking and reference-counting garbage-collection to C;
2772these features are more intrusive additions than those provided by \CFA, in addition to the runtime overhead of reference-counting.
2773Vala~\cite{Vala} compiles to GObject-based C, adding the burden of learning a separate language syntax to the aforementioned demerits of GObject as a modernization path for existing C code-bases.
2774Java~\cite{Java8} included generic types in Java~5, which are type-checked at compilation and type-erased at runtime, similar to \CFA's.
2775However, in Java, each object carries its own table of method pointers, while \CFA passes the method pointers separately to maintain a C-compatible layout.
2776Java is also a garbage-collected, object-oriented language, with the associated resource usage and C-interoperability burdens.
2777
2778D~\cite{D}, Go, and Rust~\cite{Rust} are modern, compiled languages with abstraction features similar to \CFA traits, \emph{interfaces} in D and Go and \emph{traits} in Rust.
2779However, each language represents a significant departure from C in terms of language model, and none has the same level of compatibility with C as \CFA.
2780D and Go are garbage-collected languages, imposing the associated runtime overhead.
2781The necessity of accounting for data transfer between managed runtimes and the unmanaged C runtime complicates foreign-function interfaces to C.
2782Furthermore, while generic types and functions are available in Go, they are limited to a small fixed set provided by the compiler, with no language facility to define more.
2783D restricts garbage collection to its own heap by default, while Rust is not garbage-collected, and thus has a lighter-weight runtime more interoperable with C.
2784Rust also possesses much more powerful abstraction capabilities for writing generic code than Go.
2785On the other hand, Rust's borrow-checker provides strong safety guarantees but is complex and difficult to learn and imposes a distinctly idiomatic programming style.
2786\CFA, with its more modest safety features, allows direct ports of C code while maintaining the idiomatic style of the original source.
2787
2788
2789\subsection{Tuples/Variadics}
2790
2791Many programming languages have some form of tuple construct and/or variadic functions, \eg SETL, C, KW-C, \CC, D, Go, Java, ML, and Scala.
2792SETL~\cite{SETL} is a high-level mathematical programming language, with tuples being one of the primary data types.
2793Tuples in SETL allow subscripting, dynamic expansion, and multiple assignment.
2794C provides variadic functions through @va_list@ objects, but the programmer is responsible for managing the number of arguments and their types, so the mechanism is type unsafe.
2795KW-C~\cite{Buhr94a}, a predecessor of \CFA, introduced tuples to C as an extension of the C syntax, taking much of its inspiration from SETL.
2796The main contributions of that work were adding MRVF, tuple mass and multiple assignment, and record-field access.
2797\CCeleven introduced @std::tuple@ as a library variadic template structure.
2798Tuples are a generalization of @std::pair@, in that they allow for arbitrary length, fixed-size aggregation of heterogeneous values.
2799Operations include @std::get<N>@ to extract values, @std::tie@ to create a tuple of references used for assignment, and lexicographic comparisons.
2800\CCseventeen proposes \emph{structured bindings}~\cite{Sutter15} to eliminate pre-declaring variables and use of @std::tie@ for binding the results.
2801This extension requires the use of @auto@ to infer the types of the new variables, so complicated expressions with a non-obvious type must be documented with some other mechanism.
2802Furthermore, structured bindings are not a full replacement for @std::tie@, as it always declares new variables.
2803Like \CC, D provides tuples through a library variadic-template structure.
2804Go does not have tuples but supports MRVF.
2805Java's variadic functions appear similar to C's but are type-safe using homogeneous arrays, which are less useful than \CFA's heterogeneously-typed variadic functions.
2806Tuples are a fundamental abstraction in most functional programming languages, such as Standard ML~\cite{sml} and~\cite{Scala}, which decompose tuples using pattern matching.
2807
2808
2809\subsection{C Extensions}
2810
2811\CC is the best known C-based language, and is similar to \CFA in that both are extensions to C with source and runtime backwards compatibility.
2812Specific difference between \CFA and \CC have been identified in prior sections, with a final observation that \CFA has equal or fewer tokens to express the same notion in many cases.
2813The key difference in design philosophies is that \CFA is easier for C programmers to understand by maintaining a procedural paradigm and avoiding complex interactions among extensions.
2814\CC, on the other hand, has multiple overlapping features (such as the three forms of polymorphism), many of which have complex interactions with its object-oriented design.
2815As a result, \CC has a steep learning curve for even experienced C programmers, especially when attempting to maintain performance equivalent to C legacy-code.
2816
2817There are several other C extension-languages with less usage and even more dramatic changes than \CC.
2818Objective-C and Cyclone are two other extensions to C with different design goals than \CFA, as discussed above.
2819Other languages extend C with more focused features.
2820$\mu$\CC~\cite{uC++book}, CUDA~\cite{Nickolls08}, ispc~\cite{Pharr12}, and Sierra~\cite{Leissa14} add concurrent or data-parallel primitives to C or \CC;
2821data-parallel features have not yet been added to \CFA, but are easily incorporated within its design, while concurrency primitives similar to those in $\mu$\CC have already been added~\cite{Delisle18}.
2822Finally, CCured~\cite{Necula02} and Ironclad \CC~\cite{DeLozier13} attempt to provide a more memory-safe C by annotating pointer types with garbage collection information; type-checked polymorphism in \CFA covers several of C's memory-safety issues, but more aggressive approaches such as annotating all pointer types with their nullability or requiring runtime garbage collection are contradictory to \CFA's backwards compatibility goals.
2823
2824
2825\section{Conclusion and Future Work}
2826
2827The goal of \CFA is to provide an evolutionary pathway for large C development-environments to be more productive and safer, while respecting the talent and skill of C programmers.
2828While other programming languages purport to be a better C, they are in fact new and interesting languages in their own right, but not C extensions.
2829The purpose of this paper is to introduce \CFA, and showcase language features that illustrate the \CFA type-system and approaches taken to achieve the goal of evolutionary C extension.
2830The contributions are a powerful type-system using parametric polymorphism and overloading, generic types, tuples, advanced control structures, and extended declarations, which all have complex interactions.
2831The work is a challenging design, engineering, and implementation exercise.
2832On the surface, the project may appear as a rehash of similar mechanisms in \CC.
2833However, every \CFA feature is different than its \CC counterpart, often with extended functionality, better integration with C and its programmers, and always supporting separate compilation.
2834All of these new features are being used by the \CFA development-team to build the \CFA runtime-system.
2835Finally, we demonstrate that \CFA performance for some idiomatic cases is better than C and close to \CC, showing the design is practically applicable.
2836
2837There is ongoing work on a wide range of \CFA features, including arrays with size, runtime type-information, virtual functions, user-defined conversions, concurrent primitives, and modules.
2838While all examples in the paper compile and run, a public beta-release of \CFA will take another 8--12 months to finalize these extensions.
2839There are also interesting future directions for the polymorphism design.
2840Notably, \CC template functions trade compile time and code bloat for optimal runtime of individual instantiations of polymorphic functions.
2841\CFA polymorphic functions use dynamic virtual-dispatch;
2842the runtime overhead of this approach is low, but not as low as inlining, and it may be beneficial to provide a mechanism for performance-sensitive code.
2843Two promising approaches are an @inline@ annotation at polymorphic function call sites to create a template-specialization of the function (provided the code is visible) or placing an @inline@ annotation on polymorphic function-definitions to instantiate a specialized version for some set of types (\CC template specialization).
2844These approaches are not mutually exclusive and allow performance optimizations to be applied only when necessary, without suffering global code-bloat.
2845In general, we believe separate compilation, producing smaller code, works well with loaded hardware-caches, which may offset the benefit of larger inlined-code.
2846
2847
2848\section{Acknowledgments}
2849
2850The authors would like to recognize the design assistance of Glen Ditchfield, Richard Bilson, Thierry Delisle, Andrew Beach and Brice Dobry on the features described in this paper, and thank Magnus Madsen for feedback on the writing.
2851This work is supported by a corporate partnership with Huawei Ltd.\ (\url{http://www.huawei.com}), and Aaron Moss and Peter Buhr are partially funded by the Natural Sciences and Engineering Research Council of Canada.
2852
2853
2854\bibliography{pl}
2855
2856
2857\appendix
2858
2859\section{Benchmark Stack Implementations}
2860\label{sec:BenchmarkStackImplementations}
2861
2862Throughout, @/***/@ designates a counted redundant type annotation; code reformatted slightly for brevity.
2863
2864
2865\subsection{C}
2866
2867\begin{flushleft}
2868\lstDeleteShortInline@%
2869\begin{tabular}{@{}l@{\hspace{1.8\parindentlnth}}|@{\hspace{\parindentlnth}}l@{}}
2870\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
2871typedef struct node {
2872        void * value;
2873        struct node * next;
2874} node;
2875typedef struct stack {
2876        struct node * head;
2877} stack;
2878void copy_stack( stack * s, const stack * t,
2879                                void * (*copy)( const void * ) ) {
2880        node ** cr = &s->head;
2881        for (node * nx = t->head; nx; nx = nx->next) {
2882                *cr = malloc( sizeof(node) ); /***/
2883                (*cr)->value = copy( nx->value );
2884                cr = &(*cr)->next;
2885        }
2886        *cr = NULL;
2887}
2888void clear_stack( stack * s, void (* free_el)( void * ) ) {
2889        for ( node * nx = s->head; nx; ) {
2890                node * cr = nx;
2891                nx = cr->next;
2892                free_el( cr->value );
2893                free( cr );
2894        }
2895        s->head = NULL;
2896}
2897\end{cfa}
2898&
2899\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
2900stack new_stack() {
2901        return (stack){ NULL }; /***/
2902}
2903stack * assign_stack( stack * s, const stack * t,
2904                                void * (*copy_el)( const void * ),
2905                                void (*free_el)( void * ) ) {
2906        if ( s->head == t->head ) return s;
2907        clear_stack( s, free_el ); /***/
2908        copy_stack( s, t, copy_el ); /***/
2909        return s;
2910}
2911_Bool stack_empty( const stack * s ) {
2912        return s->head == NULL;
2913}
2914void push_stack( stack * s, void * v ) {
2915        node * n = malloc( sizeof(node) ); /***/
2916        *n = (node){ v, s->head }; /***/
2917        s->head = n;
2918}
2919void * pop_stack( stack * s ) {
2920        node * n = s->head;
2921        s->head = n->next;
2922        void * v = n->value;
2923        free( n );
2924        return v;
2925}
2926\end{cfa}
2927\end{tabular}
2928\lstMakeShortInline@%
2929\end{flushleft}
2930
2931
2932\subsection{\CFA}
2933\label{s:CforallStack}
2934
2935\begin{flushleft}
2936\lstDeleteShortInline@%
2937\begin{tabular}{@{}l|@{\hspace{\parindentlnth}}l@{}}
2938\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
2939forall( otype T ) {
2940        struct node {
2941                T value;
2942                node(T) * next;
2943        };
2944        struct stack { node(T) * head; };
2945        void ?{}( stack(T) & s, stack(T) t ) { // copy
2946                node(T) ** cr = &s.head;
2947                for ( node(T) * nx = t.head; nx; nx = nx->next ) {
2948                        *cr = alloc();
2949                        ((*cr)->value){ nx->value };
2950                        cr = &(*cr)->next;
2951                }
2952                *cr = 0;
2953        }
2954        void clear( stack(T) & s ) with( s ) {
2955                for ( node(T) * nx = head; nx; ) {
2956                        node(T) * cr = nx;
2957                        nx = cr->next;
2958                        ^(*cr){};
2959                        free( cr );
2960                }
2961                head = 0;
2962        }
2963
2964\end{cfa}
2965&
2966\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
2967        void ?{}( stack(T) & s ) { (s.head){ 0 }; }
2968        void ^?{}( stack(T) & s) { clear( s ); }
2969        stack(T) ?=?( stack(T) & s, stack(T) t ) {
2970                if ( s.head == t.head ) return s;
2971                clear( s );
2972                s{ t };
2973                return s;
2974        }
2975        _Bool empty( const stack(T) & s ) {
2976                return s.head == 0;
2977        }
2978        void push( stack(T) & s, T value ) with( s ) {
2979                node(T) * n = alloc();
2980                (*n){ value, head };
2981                head = n;
2982        }
2983        T pop( stack(T) & s ) with( s ) {
2984                node(T) * n = head;
2985                head = n->next;
2986                T v = n->value;
2987                ^(*n){};
2988                free( n );
2989                return v;
2990        }
2991}
2992\end{cfa}
2993\end{tabular}
2994\lstMakeShortInline@%
2995\end{flushleft}
2996
2997
2998\subsection{\CC}
2999
3000\begin{flushleft}
3001\lstDeleteShortInline@%
3002\begin{tabular}{@{}l|@{\hspace{\parindentlnth}}l@{}}
3003\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
3004template<typename T> struct stack {
3005        struct node {
3006                T value;
3007                node * next;
3008                node( const T & v, node * n = nullptr ) :
3009                        value( v ), next( n ) {}
3010        };
3011        node * head;
3012        void copy( const stack<T> & o ) {
3013                node ** cr = &head;
3014                for ( node * nx = o.head; nx; nx = nx->next ) {
3015                        *cr = new node{ nx->value }; /***/
3016                        cr = &(*cr)->next;
3017                }
3018                *cr = nullptr;
3019        }
3020        void clear() {
3021                for ( node * nx = head; nx; ) {
3022                        node * cr = nx;
3023                        nx = cr->next;
3024                        delete cr;
3025                }
3026                head = nullptr;
3027        }
3028\end{cfa}
3029&
3030\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
3031        stack() : head( nullptr ) {}
3032        stack( const stack<T> & o ) { copy( o ); }
3033        ~stack() { clear(); }
3034        stack & operator=( const stack<T> & o ) {
3035                if ( this == &o ) return *this;
3036                clear();
3037                copy( o );
3038                return *this;
3039        }
3040        bool empty() const {
3041                return head == nullptr;
3042        }
3043        void push( const T & value ) {
3044                head = new node{ value, head };  /***/
3045        }
3046        T pop() {
3047                node * n = head;
3048                head = n->next;
3049                T v = std::move( n->value );
3050                delete n;
3051                return v;
3052        }
3053};
3054
3055\end{cfa}
3056\end{tabular}
3057\lstMakeShortInline@%
3058\end{flushleft}
3059
3060
3061\subsection{\CCV}
3062
3063\begin{flushleft}
3064\lstDeleteShortInline@%
3065\begin{tabular}{@{}l|@{\hspace{\parindentlnth}}l@{}}
3066\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
3067struct stack {
3068        struct node {
3069                ptr<object> value;
3070                node * next;
3071                node( const object & v, node * n = nullptr ) :
3072                                value( v.new_copy() ), next( n ) {}
3073        };
3074        node * head;
3075        void copy( const stack & o ) {
3076                node ** cr = &head;
3077                for ( node * nx = o.head; nx; nx = nx->next ) {
3078                        *cr = new node{ *nx->value }; /***/
3079                        cr = &(*cr)->next;
3080                }
3081                *cr = nullptr;
3082        }
3083        void clear() {
3084                for ( node * nx = head; nx; ) {
3085                        node * cr = nx;
3086                        nx = cr->next;
3087                        delete cr;
3088                }
3089                head = nullptr;
3090        }
3091\end{cfa}
3092&
3093\begin{cfa}[xleftmargin=0pt,aboveskip=0pt,belowskip=0pt]
3094        stack() : head( nullptr ) {}
3095        stack( const stack & o ) { copy( o ); }
3096        ~stack() { clear(); }
3097        stack & operator=( const stack & o ) {
3098                if ( this == &o ) return *this;
3099                clear();
3100                copy( o );
3101                return *this;
3102        }
3103        bool empty() const {
3104                return head == nullptr;
3105        }
3106        void push( const object & value ) {
3107                head = new node{ value, head }; /***/
3108        }
3109        ptr<object> pop() {
3110                node * n = head;
3111                head = n->next;
3112                ptr<object> v = std::move( n->value );
3113                delete n;
3114                return v;
3115        }
3116};
3117
3118\end{cfa}
3119\end{tabular}
3120\lstMakeShortInline@%
3121\end{flushleft}
3122
3123
3124\end{document}
3125
3126% Local Variables: %
3127% tab-width: 4 %
3128% compile-command: "make" %
3129% End: %
Note: See TracBrowser for help on using the repository browser.