Library: shorten display of unsigned types

I changed the rendering of unsigned types in diagnostic output
to use the short notation, e.g. `uint` instead of `unsigned int`.
This dramatically improves the legibility of verification strings.

Moreover, I took the opportunigy to look through the existing page
with codeing style guides to explicitly write down some conventions
formed over years of usage.

I did not just »make up« those light heartedly, rather these conventions
are the result of a craftsman's ''attentive observation and self-reflection.''
This commit is contained in:
Fischlurch 2024-11-21 18:07:30 +01:00
parent 26bf32525b
commit dcbde6d163
13 changed files with 109 additions and 60 deletions

View file

@ -1,16 +1,20 @@
Coding Guidelines
=================
:Date: Spring 2023
:Date: Autumn 2024
:toc:
_this page summarises some style and coding guidelines for the Lumiera code base_
.Basic Tenet
Code is written for humans. Use your judgement and common sense above all.
Style Guide
-----------
The Lumiera project uses GNU indentation style with slight adaptations.
- *no tabs* please. The typical ``semi indent'' of GNU style thus becomes 2 spaces.
- *no tabs* please. +
The typical ``semi indent'' of GNU style thus becomes 2 spaces.
- maximum line length is rather around *110 characters*.footnote:[This is not a hard
limit, rather a guideline -- however, you should never try to stuff several distinct
topics into a single line...]
@ -20,26 +24,60 @@ The Lumiera project uses GNU indentation style with slight adaptations.
* the braces for a class scope are indented by 2 spaces
* the access modifiers start at this brace level, while all declarations and definitions
within the class are again indented by 2 spaces
* the braces for free-standing functions however are not indented and thus placed directly
below the name of the function in the definition. This holds also for member functions
defined out-line and not within the class body.
* the line breaking rules are relaxed. Definitions and statements may be written as single line,
provided that the length remains below 110 chars and the result _remains legible_. Otherwise,
we'll fall back to the default and wrap the lines. More specifically
provided that the length remains below 110 chars and the result _remains legible_.
More specifically...
** function declarations may be written in one line
** same for definitions with just a single statement in the body
** same for simple if-statements without else-branch.
** If in doubt however, we prefer to wrap the lines.
* it is OK to write if-then-else without braces, but only as long as the structure
is uttermost clear and poses no danger for maintenance. Use your common sense on this one.
Often, a ternary operator is a viable alternative. When a `?` ... `:` becomes long and
hard to read, the alternative clause starting with `:` is placed directly below the `?`
* the space between function name and opening parenthesis of the argument list is not
enforced when this doesn't make sense, esp. for argument-less functions, chained calls
enforced when this doesn't make sense, notably for argument-less functions, chained calls
or constructor syntax. But in all other cases, we really value this additional space,
it improves legibility.
* template argument declarations are _always_ written on a separate line, above the
return type declaration. This rule holds even if the rest of a definition can be
written within a single line.
* the body of λ-functions is indented beyond the capture clause, i.e. it is rather
starkly offset with respect to the regular code. When capture clauses become complicated
beyond just a short list of names -- especially when including initialisers -- they are
preferably written on a separate line, directly above the argument list in parenthesis
* the opening brace of namespaces is placed on the same line. Optionally, the namespace
body may be indented, as long as this helps underpinning the nesting structure. But
there is no need to use 3 indents on a 3 level nested namespace. One level is enough
to highlight the presence of a nesting.
- the shorthand type names `uint`, `ulong`, `uchar`, `short`, `ushort`, `llong`, `ullong`
are preferred, as it is clearer to write the fundamental type in a single word. Index
variables _should always be unsigned._ Use `size_t` whenever applicable.footnote:[yet it is OK
to use `uint` in a for-loop where the limits are clear; likewise it is acceptable to use
a direct conversion signed ⟷ unsigned when the situation is simple and obvious. No need
to be pedantic on everything]
When the bit-size matters, please use explicit designations like `int32_t`, `uint64_t`.
- we prefer to highlight the _typing as distinct_ from the names and entities; notably
the `*` pointer and `&` reference modifier are attached post-fix to the type, as opposed
to sticking it to the name: write ```int* thing`'', not ```int *thing`''. Whenever types
become difficult to parse, we introduce a type alias (`using Alias =`... with an uppercase
type alias name). This is especially relevant with function pointers, where we prefer
to define a type alias for the _function signature_ first.
- in a similar vein, we distinguish between an ```const object`'' which _is really in-itself_
unmodifiable, as opposed to _taking a const-reference,_ which is always spelled out
post-fix as `object const&`, or a constant-pointer spelled as `const*`.
- we use the textual form of the boolean operators `and`, `or` and `not`, because this
allows to make the code resemble complete meaningful sentences in human language.
We _do use_ the »line noise« variant however for all low-level bit manipulation,
`&`,`|`, `^`, sometimes also for `!` -- using those is taken as an indication of
entering the ``danger zone''...
Naming conventions
~~~~~~~~~~~~~~~~~~
Naming conventions are used to characterise the kind of element at hand and give a visual
@ -159,7 +197,7 @@ Recommendations at code level
* always consider the _ownership_ and _lifecycle_ -- value objects can not have any, should
ideally be immutable, and stored inline or avoid heap allocation. If in doubt, decompose
and turn unclear and changing parts into a service / dependency, attached by (value) handle
* objects with reference semantics should be made _noncopyable_, while value objects should use
* objects with reference semantics should be made _non-copyable_, while value objects should use
default copy semantics. Use 'lib/nocopy.hpp' to express the flavours of move only / no copy.
* equality comparisons of ref objects are to be based on their identity solely, while for
value objects, all properties must be included which are tangible and independently variable.
@ -175,7 +213,7 @@ Recommendations at code level
+
* if is something is ``just a number'' yet has some specific meaning,
better use a lightweight wrapper object to tag it with semantics
* if a common pattern works involving distinct, unrelated entities,
* if a common pattern works by involving distinct, unrelated entities,
then better use generic programming or even higher-kinded types,
instead of forcing unrelated types to inherit from some supertype.
* avoid downcasts, `void*` and switch-on-type programming; this

View file

@ -207,6 +207,11 @@ apologies for that."
"|lumiera::"
, regex::ECMAScript | regex::optimize};
static regex lolong {"long long"
, regex::ECMAScript | regex::optimize};
static regex unSigned {"unsigned (\\w+)"
, regex::ECMAScript | regex::optimize};
static regex stdString {"(__cxx11::)?basic_string<char, char_traits<char>, allocator<char>\\s*>(\\s+\\B)?"
, regex::ECMAScript | regex::optimize};
@ -219,7 +224,7 @@ apologies for that."
static regex uniquePtr {"unique_ptr<(\\w+), default_delete<\\1>\\s*"
, regex::ECMAScript | regex::optimize};
static regex lumieraP {"P<(\\w+), shared_ptr<\\1>\\s*"
static regex lumieraP {"P<(\\w+), shared_ptr<\\1>\\s*"
, regex::ECMAScript | regex::optimize};
@ -227,6 +232,8 @@ apologies for that."
auto end = typeName.end();
end = regex_replace(pos, pos, end, commonPrefixes, "");
end = regex_replace(pos, pos, end, lolong, "llong");
end = regex_replace(pos, pos, end, unSigned, "u$1");
end = regex_replace(pos, pos, end, stdString, "string");
end = regex_replace(pos, pos, end, stdAllocator, "$1");
end = regex_replace(pos, pos, end, mapAllocator, "$1");

View file

@ -26,8 +26,12 @@
#include <cstdint>
/* === minimal common place === */
using uchar = unsigned char;
using uint = unsigned int;
using uchar = unsigned char;
using ulong = unsigned long int;
using llong = long long int;
using ullong = unsigned long long int;
using ushort = unsigned short int;
using f128 = long double;
static_assert(10 <= sizeof(f128));

View file

@ -495,19 +495,19 @@ namespace util {
/* === Literals for common size designations === */
inline uint
operator""_KiB (unsigned long long const siz)
operator""_KiB (ullong const siz)
{
return uint(siz) * 1024u;
}
inline uint
operator""_MiB (unsigned long long const siz)
operator""_MiB (ullong const siz)
{
return uint(siz) * 1024u*1024u;
}
inline unsigned long long
operator""_GiB (unsigned long long const siz)
inline ullong
operator""_GiB (ullong const siz)
{
return siz * 1024uLL*1024uLL*1024uLL;
}

View file

@ -162,7 +162,7 @@ namespace util {
* \endcode
*/
inline util::Rat
operator""_r (unsigned long long num)
operator""_r (ullong num)
{
return util::Rat{num};
}

View file

@ -257,7 +257,7 @@ namespace time {
return *this;
}
/// Support mixing with plain long int arithmetics
/// Support mixing with plain 64bit int arithmetics
operator gavl_time_t() const { return t_; }
/// Support for micro-tick precise time arithmetics
operator FSecs() const { return FSecs{t_, TimeValue::SCALE}; }

View file

@ -69,10 +69,10 @@ namespace util {
};
template<>
struct IDiv<long long>
struct IDiv<llong>
: lldiv_t
{
IDiv<long long> (long long num, long long den)
IDiv<llong> (llong num, llong den)
: lldiv_t(lldiv (num,den))
{ }
};

View file

@ -160,8 +160,8 @@ out-lit: Digxel<int, digxel::Formatter<int> >--empty-- 0--(val=123)--123|
out-lit: Digxel<double, digxel::Formatter<double> >--empty--00.000--(val=123.457)--123.457|
out-lit: Digxel<int, digxel::SexaFormatter>--empty--00--(val=42)--42|
out-lit: Digxel<int, digxel::SexaFormatter>--empty--00--(val=-5)---5|
out-lit: Digxel<unsigned int, digxel::HexaFormatter>--empty--00--(val=12)--0C|
out-lit: Digxel<unsigned int, digxel::HexaFormatter>--empty--00--(val=111)--6F|
out-lit: Digxel<uint, digxel::HexaFormatter>--empty--00--(val=12)--0C|
out-lit: Digxel<uint, digxel::HexaFormatter>--empty--00--(val=111)--6F|
out-lit: Digxel<long, digxel::CountFormatter>--empty--0000--(val=-1234567890)---1234567890|
return: 0
END

View file

@ -694,7 +694,7 @@ namespace test{
CHECK(not ii.getRestElms());
CHECK (materialise(ii.getGroupedElms()) == "23-22-21-20-19"_expect);
CHECK ( test::showType<decltype(*ii)>()== "array<unsigned int, 5ul>&"_expect);
CHECK ( test::showType<decltype(*ii)>()== "array<uint, 5ul>&"_expect);
uint s = *(ii.getGroupedElms());
for ( ; ii; ++ii)

View file

@ -53,14 +53,14 @@ namespace test{
double funny (char, char, string);
void funky() const;
short fuzzy (float, float);
long long fuzzy();
llong fuzzy();
double fully;
};
class Fishy
{
/** @note private function can never be detected */
long long fuzzy();
llong fuzzy();
/** @note type Fishy exposes an extension point `fun` */
friend void fun (Fishy&);
@ -94,7 +94,7 @@ namespace test{
META_DETECT_EXTENSION_POINT (fun);
META_DETECT_FUNCTION (double, funny, (char, char, string));
META_DETECT_FUNCTION (long long, fuzzy, (void));
META_DETECT_FUNCTION (llong, fuzzy, (void));
META_DETECT_FUNCTION_NAME (funny);
META_DETECT_FUNCTION_NAME (funky);
META_DETECT_FUNCTION_NAME (fuzzy);

View file

@ -133,7 +133,7 @@ namespace test {
// this is how the key trick of the _Fun traits template works:
// for anything "function like" we retrieve a member-pointer to the function call operator
// which we then pass on to the dedicated overload for member pointers
CHECK ("int (Functor::*)(unsigned int)" == typeStr<decltype(&Functor::operator())>());
CHECK ("int (Functor::*)(uint)" == typeStr<decltype(&Functor::operator())>());
Func f1{freeFun};
@ -150,60 +150,60 @@ namespace test {
Func f5{lambda};
CHECK ("int (unsigned int)" == showSig (freeFun));
CHECK ("int (unsigned int)" == showSig (&freeFun));
CHECK ("int (unsigned int)" == showSig (Functor::staticFun));
CHECK ("int (unsigned int)" == showSig (lambda));
CHECK ("int (unsigned int)" == showSig (f5));
CHECK ("int (uint)" == showSig (freeFun));
CHECK ("int (uint)" == showSig (&freeFun));
CHECK ("int (uint)" == showSig (Functor::staticFun));
CHECK ("int (uint)" == showSig (lambda));
CHECK ("int (uint)" == showSig (f5));
CHECK ("int (unsigned int)" == showSigRef (freeFun));
CHECK ("int (unsigned int)" == showSigRef (lambda));
CHECK ("int (unsigned int)" == showSigRef (f5));
CHECK ("int (uint)" == showSigRef (freeFun));
CHECK ("int (uint)" == showSigRef (lambda));
CHECK ("int (uint)" == showSigRef (f5));
CHECK ("int (unsigned int)" == showSigCRef (freeFun));
CHECK ("int (unsigned int)" == showSigCRef (lambda));
CHECK ("int (unsigned int)" == showSigCRef (f5));
CHECK ("int (uint)" == showSigCRef (freeFun));
CHECK ("int (uint)" == showSigCRef (lambda));
CHECK ("int (uint)" == showSigCRef (f5));
CHECK ("int (unsigned int)" == showSigRRef (move(lambda)));
CHECK ("int (unsigned int)" == showSigRRef (move(f5)));
CHECK ("int (uint)" == showSigRRef (move(lambda)));
CHECK ("int (uint)" == showSigRRef (move(f5)));
CHECK ("int (unsigned int)" == showSig (move(&freeFun)));
CHECK ("int (unsigned int)" == showSig (move(lambda)));
CHECK ("int (unsigned int)" == showSig (move(f5)));
CHECK ("int (uint)" == showSig (move(&freeFun)));
CHECK ("int (uint)" == showSig (move(lambda)));
CHECK ("int (uint)" == showSig (move(f5)));
Func& funRef = f1;
Functor& funkyRef = funk;
Func const& funCRef = f1;
Functor const& funkyCRef = funk;
CHECK ("int (unsigned int)" == showSig (funRef));
CHECK ("int (unsigned int)" == showSig (funkyRef));
CHECK ("int (unsigned int)" == showSig (funCRef));
CHECK ("int (unsigned int)" == showSig (funkyCRef));
CHECK ("int (uint)" == showSig (funRef));
CHECK ("int (uint)" == showSig (funkyRef));
CHECK ("int (uint)" == showSig (funCRef));
CHECK ("int (uint)" == showSig (funkyCRef));
CHECK ("int (unsigned int)" == typeStr<_Fun<int(uint)>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Func&>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Func&&>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Func const&>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Functor&>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Functor&&>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Functor const&>::Sig>());
CHECK ("int (uint)" == typeStr<_Fun<int(uint)>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Func&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Func&&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Func const&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Functor&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Functor&&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Functor const&>::Sig>());
using Siggy = _Fun<Func>::Sig;
CHECK ("int (unsigned int)" == typeStr<_Fun<Siggy&>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Siggy&&>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<Siggy const&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Siggy&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Siggy&&>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<Siggy const&>::Sig >());
auto memfunP = &Functor::fun;
FuncF fM{memfunP};
Func fMF{bind (fM, funk, _1)};
CHECK ("int (unsigned int)" == typeStr<_Fun<decltype(memfunP)>::Sig>());
CHECK ("int (Functor&, unsigned int)" == typeStr<_Fun<decltype(fM)>::Sig >());
CHECK ("int (unsigned int)" == typeStr<_Fun<decltype(fMF)>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<decltype(memfunP)>::Sig>());
CHECK ("int (Functor&, uint)" == typeStr<_Fun<decltype(fM)>::Sig >());
CHECK ("int (uint)" == typeStr<_Fun<decltype(fMF)>::Sig >());
// _Fun<F> can be used for metaprogramming with enable_if

View file

@ -113,7 +113,7 @@ namespace test{
auto breed = Result{evil, 55ll}; // an odd number...
VERIFY_ERROR (STATE, breed.maybeThrow() );
CHECK (Type(breed) == "Result<long long>"_expect);
CHECK (Type(breed) == "Result<llong>"_expect);
auto dead = Result{[]{ throw 55; }};
auto deed = Result{[]{ /* :-) */ }};

View file

@ -114,7 +114,7 @@ namespace test {
verifyIntegerTypes<long>();
verifyIntegerTypes<short>();
verifyIntegerTypes<int64_t>();
verifyIntegerTypes<long long int>();
verifyIntegerTypes<llong>();
if (!isnil (arg))
runPerformanceTest();