Named Booleans prevent C++ bugs and save you time

https://raymii.org/s/blog/Named_Booleans_prevent_bugs.html

by signa11 • 3 years ago

112 164 3 years ago

dietr1ch• 3 years ago

I think that these bundled conditions are bad, and that having negative booleans is terrible. Using and almost abusing early returns on bad input almost always leads to easier to follow code.

    if (someLongNamedVarIsNotUnknown && someLongNamedMapCountIsZero)
        return false;
    else
        return true;

This is not the same, but I'd suspect there might be a bug in the code above

    if (someLongNamedVarIsUnknown) {
      return false;  // Can't do much
    }
    
    if (someLongNamedMapCountIsZero) {
      return false;  // Nothing to do
    }
    
    // Do stuff
    return true

The named booleans suggested here help adding clarity, but more often than not they'll end up being single-use (const) variables that just paraphrase their definitions. In these cases I'd rather avoid introducing a variable, and rely in comments if I really need to explain something.

andix3• 3 years ago

True, functions with return values are so much better in such a case. I really miss if expressions and switch expressions in most languages. Most functional languages (and also c#) have them. Something like (let’s do JavaScript in this example):

  const value = if (a) {
    return true
  } else {
    return switch (b) {
      case 1: return false
      default: return undefined
    }
  }

baq• 3 years ago

Interestingly Rust does this and I agree JavaScript with const feels only 80% there without this feature.

tialaramex7• 3 years ago

Apparently C++ 26 might get statement expressions in some form, it sounds like Barry Revzin sketched something at Issaquah (the recent WG21 meeting) which can do this.

deathvyp3r• 3 years ago

you can do this in c++ as well with a lambda

  const value = [&]() {   
    if (a) {
      return true;
    }
    return switch (b) {
      case 1: return false;
      default: return undefinied;
    }
  }();

pyjarrett• 3 years ago

Ada calls these "if expression" and the switch-variant "case expressions". It's the same syntax as the control flow ones, just wrapped with parens.

    value := (if a then True
              else (
                 case b is
                    when 1 => False,
                    when others => Undefined
              ));

monocasa3• 3 years ago

The named booleans suggested here help adding clarity, but more often than not they'll end up being single-use (const) variables that just paraphrase their definitions. In these cases I'd rather avoid introducing a variable, and rely in comments if I really need to explain something.

Interesting. With the same axioms I come to the opposite conclusion. That comments are to be avoided except when strictly necessary, and giving names to concepts via intermediate const variables is one of the best weapons in that fight. Ie. make your code read as close to what the comment would be (but simply via naming) makes it very clear when the names no longer match the behavior, in contrast to comments which have a tendency to bit rot.

Sai_• 3 years ago

This is what I used to love about Obj-C - idiomatically written Obj-C code was always so much easier to read and understand without needing comments.

simplotek6• 3 years ago

The named booleans suggested here help adding clarity, but more often than not they'll end up being single-use (const) variables that just paraphrase their definitions.

That's ok. The worst tradeoff in the "named booleans" pattern proposed in the blog post is that it misses the whole point of chaining predicate evaluations, which is leveraging short-circuiting to prevent further predicates from being evaluated.

Chaining logical comparisons and the way they short-circuit is specially important when a predicate has preconditions and/or is expensive and/or has side-effects.

If your predicates are hard to read you don't fix that mess with "named booleans". You fix them by extracting them to a pure function like

   isNotUnknown(_someLongNamedVar) && isNotEmpty(_someLongNamedVar)

toxik4• 3 years ago

Nest the if statements then, perhaps in an else branch. Relying on short circuiting to avoid side effects or improving performance meaningfully is pretty shaky.

juiiiced• 3 years ago

I usually prefer to write more readable code, and then optimize later if necessary. If the short circuit becomes critical for performance then I would leave a comment, or as the parent suggests just order it with an early return.

finnh• 3 years ago

early return solves so many readability problems. i structure my code in favor of it.

Animats• 3 years ago

The only problem with this approach is that you evaluate all the conditions, even if you don't need to. Sometimes that's inefficient, and sometimes it's invalid, as in

    char *p = NULL;
    ...
    if (p && *p != ' ') { ... }

If you write

    bool p_non_null = (p != NULL);
    bool char_non_space = (*p != ' ');
    if p_non_null && char_ non_space { ... }

Oops.

bandika2• 3 years ago

The correct rewriting would be:

    bool char_non_space = p_non_null && (*p != ' ');

tpoacher• 3 years ago

but then 'char_not_space' is a misleading name, as it could denote the presence of a space due to p being null

wruza5• 3 years ago

Back in mid-00s I was thinking about languages and features. One of the keywords I still miss in mainstream is “alias”. Alias is a symbol that only evaluates lazily.

  alias not_null = (bool)p;
  alias not_space = (*p != ' ');

  if (not_null && not_space) {

This may bite you in another way, but living in a world where everything gets evaluated asap-y isn’t syntactically easy either.

Would be also nice to have this:

  alias __cache foo = <heavy(expr())>;

  if (foo && foo.length && foo[0].x) {

`foo` only evaluates once.

yunruse• 3 years ago

I was curious about the concept of inverting this – requiring that all variables lazily evaluate, and in the case that evaluation is needed now, some keyword obtains it. But when I think about it, that's sort of just `await`.

I'm curious as to if it's feasible to syncretise these two aspects of variable evaluation into a shared feature that makes async and sync code feel alike.

anlsh• 3 years ago

In any sane language you can just do this using a lambda. Unfortunately C is just broken so :(

mjevans• 3 years ago

Isn't alias handled by the C pre-processor's #DEFINE operation?

  #DEFINE NOT_NULL(x) (bool)x
  #DEFINE NOT_SPACE(x) ( ' ' != *x )
  // ...
  if ( NOT_NULL(v) && NOT_SPACE(v) ) { ... }

greggyb2• 3 years ago

It is not a direct implementation, and may end up being clunky, but I expect a fairly small library of helper functions could improve ergonomics quite a bit:

Define your booleans as functions of no args. A sample in Python:

    var = 'abc'
    var_contains_a = lambda: 'a' in var
    var_shorter_than_5 = lambda: len(var) < 5

    if var_contains_a() and var_shorter_than_5():
        print(var)

injidup• 3 years ago

I do this in c++ quite often with short lambdas

  auto not_null = [&]{ return (bool)p; };
  auto not_space = [&]{ return (*p != ' '); };

  if (not_null() && not_space()) {

It's not quite as ergonomic but it works.

As an aside. I was curious if the compiler will optimize in the short circuit even if you extract the eagerly evaluated named booleans. However I was pretty sure the rules of c++ are against you and tests show this to be the case

But if you simulate the alias keyword with lambdas then the result after optimization are the same

https://godbolt.org/z/3cczcnK4s

gdprrrr• 3 years ago

Scala has lazy val

cbarrick• 3 years ago

I understand your "invalid" point. Your example dereferences a potentially null pointer.

But I don't understand your "inefficient" point. At the end of the day, everything gets turned into SSA. So expression-per-line and nested-expressions generates the same code, right?

maccard• 3 years ago

It's the same idea, that the && short circuits.

    bool dirty = ...;
    bool success = doExpensiveCalculation(...);

    if (dirty && doExpensiveCalculation)
    {

saghm• 3 years ago

Not necessarily; in first case, the stuff after the `&&` will definitively not be run. In the second case, it depends on how the code is optimized, and there's no guarantee that things moved to intermediate variables won't still be evaluated even if they're never used. Since the compiler presumably can't know the value of the boolean at runtime, it would need to reorder things to avoid doing the unnecessary work, and I wouldn't think it's something that you could just assume.

ComplexSystems• 3 years ago

This is a good point. I think that if you really want something like these named bools but which keeps the same lazy evaluation behavior, you'd have to use preprocessor macros then.

MithrilTuxedo• 3 years ago

Don't they just want a function with a clear name that returns a boolean value, something to mask the implementation details so it's not part of your cognitive load at that level of abstraction.

rajnathani• 3 years ago

Future programming languages could have (IIRC some current PLs allow for such syntax, some of them I believe using 'let'):

    if (p_non_null = (p != NULL)) && (char_non_space = (*p != ' ')) { ... }

But where the IDE only shows the LHS of the expression as:

    if p_non_null && char_ non_space { ... }

But also where the IDE makes it easier to write the earlier full expression above.

Also, the initial code:

    bool p_non_null = (p != NULL);
    bool char_non_space = (*p != ' ');
    if p_non_null && char_ non_space { ... }

This could simply be analyzed by the compiler to not store the named boolean variables in the first place, and to evaluate them inline in the if-condition so as to allow for the short-circuiting to occur fine.

bakugo• 3 years ago

Personally, I prefer to just split long if statements into multiple lines.

    if (
        (_someLongNamedVar != FooLongNameEnum::Unknown) &&
        (_someLongNamedMap.count(_someLongNamedVar) == 0)
    ) {
        // do something
    }

seydor• 3 years ago

right, but putting the && at the beginning of the next line makes it more obvious this is a continuation statement

heleninboodler• 3 years ago

I actually like putting the && on its own line in between.

rzzzt• 3 years ago

For bonus alignment points and malleability, start with 'true' and place the closing parenthesis on a new line:

    if (true
        && condition1
        && condition2
        && condition3
    ) {
        ...
    }

If you ever need to change the set of conditions, the diff will cleanly add/remove whole lines.

zabzonk• 3 years ago

the problem is the length of the variable/enum names. if they were shorter, the problem would be much more obvious - as it is the code is unreadable.

the lengths of names of entities, in any programming language, should be related to their scope - small scope, small names. and as limiting scope is what you should be doing all the time, almost all names should be short.

also, unless you really know what you are doing, don't use leading underscores, or underscores in general, in c or c++.

here is something i wrote a while back, which admitedly does not too much address the single underscore issue: https://punchlet.wordpress.com/2009/12/01/letter-the-second/

throwbadubadu4• 3 years ago

Or just put parentheses around your operands defensively even if by precedence rules not necessary as some style guides demand... went from hating and finding it too verbose to accepting and seeing advantage in that.

if ((_someLongNamedVar != FooLongNameEnum::Unknown) && (_someLongNamedMap.count (_someLongNamedVar) == 0))

really not too bad? And easily enforcable and automated checkable by your favourite tool..

Named vars for subexpressions often helpful too.. but especially here you loose short-circuiting?

(Not very C++ specific topic btw..?)

secondcoming• 3 years ago

I'm a massive fan of using parentheses even when not strictly necessary. I find it helps me visually parse the code better then a long chain of boolean expressions. I don't bother with it when a term in the expression is simple, e.g

  if (b1 && (b2 != b3))

rather than

  if ((b1) && (b1 != b3))

mike_hock• 3 years ago

I think the discussion is silly.

99.9% when the condition is "too hard to read" it's because the condition is too hard to read, period, and the correct fix is to restructure the code to have readable branch conditions, not to throw parentheses on there to make a monster expression slightly less confusing.

It's like discussing the chemistry of air fresheners instead of taking the trash out more often.

usefulcat• 3 years ago

It’s not clear to me how “add more parens” is necessarily an effective solution to avoiding a misplaced paren, as was the case here.

throwbadubadu4• 3 years ago

See above, sure you should also structure/format appropriately, but it then similarly sticks out by the same argument that makes it easier by shortening another exoression to a name.

teo_zero• 3 years ago

The mistake in the article was a ")" slipped before "== 0". Exactly the same mistake can happen in your code. Your extra parentheses won't make it less likely.

throwbadubadu4• 3 years ago

Disagreed, I spotted that in a split second immediately by following that now ingrained rule.

fifticon• 3 years ago

I do this and sometimes also something else: A local enum as a return value, where you have to compare the enum to use its value (ie you can't just 0/1 it). I of course don't always do this. I do it, when the semantics of the function value is a bit tricky. It is to protect against bugs "oh i'm pretty sure I know what returning true/false here means. It both helps the caller, as well as the implementation. when you are writing the return statements for edge cases, it can be a help that you return not just 'true', but IREALLYINSISTTHISMEANTWECOULDPARSEITANDCONTINUE. Again, I don't do this for simple obvious code.

jasonzemos• 3 years ago

If one of the variables preceding the `if` has side-effects then it's inferior to the specified lazy-evaluation of the `if`. Any external function not annotated pure will be called to assign those variables which otherwise wouldn't have been if it was packed into the `if`.

This is actually something I think the C++ standard should address soon. There are cases when extern functions which aren't appropriately pure are still "optional." The example par excellence is debug logging, should one choose to not use macros.

zone411• 3 years ago

This article seems to be targeted towards beginners. From the title, I was expecting it to cover something like https://github.com/joboccara/NamedType, which is a technique I would recommend.

lpapez• 3 years ago

Alternative solution: just write a damn test. How did that inverted condition get past testing?

MaxBarraclough• 3 years ago

That would be a good addition, but I wouldn't call it an alternative. Good test coverage shouldn't be an excuse for unreadable style.

bandika2• 3 years ago

All those bools in the article should be const.

einpoklum• 3 years ago

My take:

1. Avoid overly long and complex condition expressions - break them up for readability.

2. If a part of a long expression has obvious meaning, assign it to a temporary (probably saving a comment); if it doesn't, break a line after it. I don't like assigning variables with arbitrary names that simply duplicate the expression calculating them.

3. `true` and `false` are magic numbers. Calling `my_function(my_data, false)` is quite the hazard in my view: It makes the reader rather likely to fail to notice the exact semantics. I would expect, say,

    enum : bool { dont_save_copy = false, save_copy_to_file = true };
    my_function(myData, dont_save_copy)

4. This is relevant to almost any language! Even functional ones, where you would be encouraged to use "let" or "with" instead of super-complicated unbroken expressions. If you're writing APL though... you need super-memory and a discipline of steel :-)

ahartmetz• 3 years ago

While at it, also don't do if (condition) return true; else return false;

Boolean is a perfectly cromulent data type that you can return from a function directly.

Twisol• 3 years ago

I agree most of the time, but it can make sense in certain contexts. One common example is comparison methods:

   if (x1 < y1) return true;
   if (x2 < y2) return true;
   if (x3 < y3) return true;
   return false;

I wouldn't want to replace the last two lines with `return (x3 < y3)`, even though that's functionally the same thing.

In this contrived example, we could just return a single compound condition, but hopefully it's clear that there are cases where doing so would make more of a mess, too.

wruza5• 3 years ago

For my mind it only works if a meaning of a return value is “compatible” with an expression. E.g. if a function returns a status, but an expression is an integer comparison, the flow of logic skips an important conversion step.

  bool is_db_open() {
    return (db != NULL);
  }

  bool is_time_to_report() {
    if (n_errors >= limit) {
      return true;
    }
    return false;

    // a little confusing way
    return n_errors >= limit;
  }

Also, if there’s a chance of an additional condition, they can be easily be added and documented.

  if (n_errors >= limit) {
    return true; // too many errors
  }
  if (now() >= deadline) {
    return true; // periodic
  }
  return false;

For the same reason I often do this:

  if (x == 0) {
    return 0; // not `x`

gpderetta2• 3 years ago

Many of these seen in the wild are due to workarounding an extremely annoying and useless MSVC performance warning.

ChrisMarshallNY• 3 years ago

This is a good technique. Not sure I’d consider it particularly novel, though. It’s pretty much boilerplate for how coding is taught.

I tend not to work that way, but it’s really a matter of personal style. I do it to reduce the verbosity of my code (which tends to be fairly verbose, in other ways –especially comments). I will, occasionally, encounter the bug he mentions, but not particularly frequently.

personjerry• 3 years ago

It's not really a new technique. It's pretty common in large scale software development to break down complex logic like this. The reader will note that the possible loss in performance is usually more than worth the gain in readability, and hence makes the code much more desirable at scale.

zooweemama• 3 years ago

Pretty sure that the compiler optimizes this and there is no performance loss.

meindnoch• 3 years ago

You're wrong.

  if (cheapCheck() && expensiveCheck()) { ... }

is not equivalent to

  bool cheap = cheapCheck();
  bool expensive = expensiveCheck();
  if (cheap && expensive) { ... }

unless we're in a lazy pure functional language.

pwdisswordfishc• 3 years ago

The original bug would have been prevented by not overusing parentheses. This:

    if (_someLongNamedVar != FooLongNameEnum::Unknown && _someLongNamedMap.count(_someLongNamedVar) == 0)

does exactly what it’s supposed to.

LudwigNagasena2• 3 years ago

That is what happens when people zealously try to not be "too clever". They make everything as explicit as possible for "clarity" even when it obviously hinders it for anyone who cared to learn the most basic things about the language they use.

shrimp_emoji• 3 years ago

the most basic things about the language they use

This is C++.

The most basic thing has been deprecated but is still supported for backwards compatibility.

rwalle• 3 years ago

So the topic here really has nothing to do with .count, STL, C++ 20 or even C++ itself. It could happen in almost just every language.

A very confusing article.

TheHideout• 3 years ago

This could have also been prevented using Yoda Notation with 0 == _someLongNamedMap.count(...).

https://en.wikipedia.org/wiki/Yoda_conditions

sidlls• 3 years ago

First thing I thought of when I saw the original wrong code. I prefer the author's approach, though, as I do believe it is more clear, unless there is something performance critical about it that requires short-circuits for maximum efficiency.

dcuthbertson• 3 years ago

Thanks for the reference to Yoda conditions. I hadn't heard of Yoda code until a few weeks ago when a Go linter complained about some code I wrote a few years ago. I realized I've been writing Yoda code in C/C++ for quite a while and that habit had continued (somewhat inconsistently) into Go programming. It's nice to skim through pages of code and see the value that will cause the computer to enter an if-block. The linter made me feel like Yoda code was somehow wrong. It's good to know that others think it can help in some situations.

cesarb2• 3 years ago

Since nobody seems to have done the traditional "blame the programming language" yet, let me get it out of the way: this particular bug wouldn't have happened in a language which doesn't have implicit conversion between boolean and number. The compiler would reject the code both because the numeric result of ".count(...)" is being treated as a boolean by the "&&", and because the boolean result of the "&&" is being treated as a number by the "== 0".

One example of a language without that implicit conversion is Java. It's fairly annoying to have to put a "!= 0" whenever you want to convert a number to a boolean, but it does avoid a lot of programming mistakes within if conditions.

tester756• 3 years ago

That's the way to go.

But it is sad that in $current_year, we're still fighting boring, solved problems.

toxik4• 3 years ago

Meanwhile in Python land, this is not a common class of bugs even though this implicit boolification happens.

LudwigNagasena2• 3 years ago

If you want, you can disallow implicit conversion between boolean and number using a linter.

quietbritishjim• 3 years ago

I imagine that you'd get a load of false positives due to textual inclusion of header files.

orwin2• 3 years ago

Or with -Wimplicit.

mentos• 3 years ago

Is it possible to enable a warning in C++ for these cases?

TonyTrapp• 3 years ago

Given that there is a multitude of C++ compilers: Depends on your compiler. But in general, most compilers have a warning for this sort of thing.

inetknght4• 3 years ago

1. Not in GCC.

2. This would be caught by static analyzer

See Compiler Explorer example here [0]

[0] https://gcc.godbolt.org/z/nK7MdEcMj

s9w• 3 years ago

real-world cpp lives together with static analysis, which obviously catches this

orwin2• 3 years ago

Or you compile with -Wall -Werror, as you should do for serious projects.

I'm not saying C++ is a great language, it's not, but if you remove the secu then shoot you in the foot, it's your fault.

inetknght4• 3 years ago

Or you compile with -Wall -Werror, as you should do for serious projects.

-Wall -Werror does not catch the error described in the blog.

It does get caught by a static analyzer though [0].

[0] https://gcc.godbolt.org/z/nK7MdEcMj

knorker• 3 years ago

You may also need -Wconversion

NoOneSpecialII• 3 years ago

If only there was a language that defines explicit conversions between booleans and numbers.

remram2• 3 years ago

Don't they all have == and != though?

Smar• 3 years ago

Linters like cland-tidy help a lot with finding implicit bool conversions, among other common mistakes. Could even hook to patchset-received to ensure linters are passing ;)

andrepd• 3 years ago

Isn't this (1) the most trivial programming advice ever (give names to things), and (2) not equivalent because of short circuiting?

kgeist• 3 years ago

As an alternative, you can also create private helper functions to evaluate conditions. You get the best of two worlds: function names are self-explanatory and it supports lazy evaluation.

osigurdson2• 3 years ago

There may not be anything wrong with it in isolation, but cognitive load is increased when every identifier is effectively a sentence or paragraph.

rerdavies• 3 years ago

But cognitive load also increases when you have no idea what the significance of a variable is.

osigurdson2• 3 years ago

Agree, but it is an optimization problem: maximize meaning while minimizing identifier length. Recognize that it is possible to have identifiers that are too long.

jeffrallen2• 3 years ago

Rust and Go also avoid C++ bugs.

Why are we even still talking about C++?

anthomtb• 3 years ago

Neither Rust or Go hit Github's Top 10 languages (though they are among the fastest growing): https://octoverse.github.com/2022/top-programming-languages.

There is a lot of code out there already written in C++ and a lot of people working on that code. Not to mention an active standards body.

Expect to be hearing about C++ for a long time to come

jeffrallen2• 3 years ago

Dommage.

kqr2• 3 years ago

You can also sometimes combine the techniques in the article above to skip booleans entirely. Instead of passing in a `bool primeMember` that gets used in a calculation like

    deliveryTime = 3;
    if (primeMember) {
        deliveryTime = 1;
    }

you can pass in a `Membership memberType` value that contains the required information, so the above conditional turns into just the assignment

    deliveryTime = memberType.deliveryTime;

This is more amenable to extension and contraction: you can add or remove different classes of delivery times without changing the logic in which it's used.

You can even use private constructors and public static fields to create a list of "constants" containing the known member types, like the `Membership.REGULAR` and `Membership.PRIME`, retaining the property of the code not being able to use an invalid membership type.

This can be used also for more complicated conditionals like the one in the article, by having a factory method that copies the right "constant" composite value based on whatever logic is desired.

AgentOrange1234• 3 years ago

This assignment-then-conditional-reassignment is admirably compact and, in languages like C/Java might well be the most elegant way to handle this.

On the downside, this breaks single-assignment, binding style variables. That is, now deliveryTime can’t be final/val/const.

I wish I had a better solution. Lisp did this so well with if/cond/etc as expressions.

overgard• 3 years ago

Related, one technique I like to use even though it's more verbose is to replace booleans with enumerations. For instance, maybe you have something that could be selected or not. I generally think instead of having "true" be selected and "false" be not selected, it's better just to have ListItem::Selected and ListItem::Unselected. That way when you pass things to a function, instead of reading a lot of true/false things in random places, you can see "ListItem::Selected" instead. Small upfront cost for a lot of readability down the line.

Typescript is nice for this in that you can use strings but enforce that they're in some set.

msm_• 3 years ago

I agree, but without pattern matching converting "boolean-like" enums get tedious. For example:

    ConnectionMode connectionMode;
    if (useHttpsCheckbox.selected == ListItem::Selected) {
       connectionMode = ConnectionMode::Https;
    } else if (useHttpsCheckbox.selected == ListItem::Unselected) {
       connectionMode = ConnectionMode::Http;
    } else {
       shouldNeverHappen();
    }

In the real world there are often even longer chains like this - I found that bool-like variables ofter mean slightly different things at different layers of abstraction.

And also, you could simplify that by ignoring the `else` entirely, but there are no guarantees that it won't break later. For example another contributor may add ConnectionMode::TryHttpsElseHttp, and compiler won't help us. Languages with pattern matching don't have this particular problem.

gpderetta2• 3 years ago

You can of course use a switch statement and the compiler will warn if the cases are not exhaustive for the enum.

arximboldi• 3 years ago

The post peaks with this example is this one:

  bool usersHairColorMatchesThisMonthsAdsColour = _user.hair == HairColour::Red;
  bool userFeetSizeFitsInShoe = _user.feetSize <= _requestedShoe.size;
  bool shoePriceFitsInUserBudget = _requestedShoe.price <= BudgetHelpers::Calculator(_user);

  bool shoeIsProbablyOkayForUser = usersHairColorMatchesThisMonthsAdsColour && userFeetSizeFitsInShoe && shoePriceFitsInUserBudget;

  if(shoeIsProbablyOkayForUser)

That they compare, with the "uglier":

  // this months ad campaign color is red 
  if(_user.hair == HairColor::Red && _user.feetSize <= _requestedShoe.size && _requestedShoe.price <= BudgetHelpers::Calculator(_user))

I very much prefer the latter, I'm sorry (except it needs some new lines, of course). It performs better (that `BudgetHelper::Calculator` doesn't look very cheap ;) but crucially it is, to me personally, more "readable". First, I find a comment easier to read than a variable name, particularly when you use the not so readable dromedaryCaseToCrambleWordsInLongSentences. Thus, a variable adds no value when it's only used once (arguably, it adds more noise, since one has to go twice through the same text). And it is often debatable how a programmers poor translation of code into English adds any value: in fact, I find `_user.feetSize <= _requestedShoe.size` to be way more succint, precise and easier to parse than `userFeetSizeFitsInShoe`.

We really need to get over this school of thought that "the more wordy your variables the more readable" because it is simply not true. You don't have to go all the other way to the other end and write code using greek letters like in math. Meet somewhere in the middle, be aware of context and use the tools of the language (e.g. lexical scoping) to your power.

Unlike many of these bloggers want us to believe, you can't just boil "readability" down to some simple universal rules, and only experience makes the master. Taste, preference, purpose and context take a big part. As such, I find it useless to talk about "readable" code: readable for who? A law text, a maths paper, a school maths book, a user manual, a tutorial, a novel, a HN comment... they may all be written in English and with great skill in consideration for their "readability" by their target audiences, yet manifest completely different styles even when discussing the same topics.

For me, I like my code like I often enjoy most good writting: succint and precise. Wordy variables that add no value or abstraction are of little use to me.

DangitBobby• 3 years ago

I find that people want to find a name for some condition or computed result so they can box away a concept and use it without thinking about it anymore. That's great when you're writing software and you have lots of time to consider and remember your names, but not always so great when you revisit software and now you need to actually _know what those things are and whether they are logically correct_. Sometimes even the best name isn't enough to describe what it is you have in your hands right now, and instead of seeing it where it's used, you have to go find where it was originally defined, and you potentially have to repeat this process for several named objects at once, keeping them all in your head. IMO you need a better reason to name something than the example from the article, or you actually sacrifice clarity and readability.

alpaca1282• 3 years ago

I tend to prefer the method of defining clearly named variables, but I agree those names are ridiculous.

    hair_col_matches && size_fits && in_budget

is less than half the length and imho more readable because it doesn't turn each variable intoNeedlesslyWordyCamelCaseSentences.

you can't just boil "readability" down to some simple universal rules

Yes, in the end blindly following rules doesn't turn out as well as understanding why those best practices exist in the first place.

cassepipe• 3 years ago

I use this technique a lot to make the code more readable and less error prone but I have never bothered checking if this was trivially optimized or not. Do you know ?

fooker• 3 years ago

Isn't this partly why `enum class` was introduced, and is now the idiomatic way to express enums?

GnarfGnarf• 3 years ago

Excellent suggestion. I use this style myself in many places. Makes the code more legible.

As has been pointed out, there are cases where it's not efficient or correct, but in general it's very useful. (Been coding since 1965, not always C++ obviously).

pflanze• 3 years ago

Effectively the if statement was inverted.

If this wording means to convey that the complete test in the wrong code version is always giving the inverse of the correct version, then that statement is wrong. The truth tables, using 0 and 1 for false and true ("false" and "true" look too messy here):

The buggy code:

      0 1+ <- second condition
    0 1 1
    1 1 0
    ^- first condition

The fixed code:

      0 1+ <- second condition
    0 0 0
    1 1 0
    ^- first condition

i.e. the test as a whole is not inverted if the first subcondition is true.

_0w8t• 3 years ago

The bug in question would not happen if a reviewer demanded to split the complicated if. Yes, it would add a few lines depending on the overall control flow, but code in question is hard to read and must be written in a less dense way.

sebazzz• 3 years ago

Code Complete 2 is still relevant and advocates useful practices like these.

nuancebydefault3• 3 years ago

I read comments about need for lazy evaluation. In fact all constructs that ensure lazy evaluation try to solve a not really existing problem,in a hard to review manner. What the heck is wrong with just explicit nested if statements? Not cool enough? ``` if (p!=nullptr) { if (*p == ' ') { } else { } } ``` keep it simple

Named Booleans prevent C++ bugs and save you time

Related Stories