My aim is to take an existing function, foo, and create an exact copy of it called bar, which is simple enough with alias_method. I would then like to dynamically redefine foo such that it has the exact same type signature, so that I can call bar from it, among other reasons.
This requirements mean that I cannot just do something like
define_method(:foo) do |*args, &block|
send(:bar, *args, &block)
end
because it changes the type signature of foo.
I also don't see how I can use something like method(:foo).parameters as that will tell me what the type signature is, but will not specify, for example, the values of default arguments.
Any help is greatly appreaciated!
Ruby has no concept of manifest types and manifest type signatures. Since they don't exist, you obviously can't get them.
When doing Ruby programming, there is of course a latent concept of types and type signatures in the programmer's head. But that's exactly where that concept exists: in the programmer's head and only in the programmer's head.
It might also exist in documentation, but that is not guarateed. Also, there is no standard format for putting it in documentation. There are various different formats for expressing types in Ruby documentation, sometimes the types are not expressed using any form of (semi-)formal notation at all, but only in prose, and sometimes, they are implicit in the names of parameters. In some cases, the types are just part of the Ruby culture, everybody knows them, but they are never actually written down anywhere (the most obvious example is the each protocol that the Enumerable mixin depends on, which everybody "just knows" without being explicitly specified).
You are also asking about default arguments for optional parameters: these are evaluated dynamically, getting static information about them is simply impossible because of the Halting Problem, Rice's Theorem and all the other fun undecidability results in programming.
TL;DR
Ruby doesn't care about types. It only cares about whether or not an object can #respond_to? a message. As a result, any hard-wired expectations about types should be encoded in method/variable names, and in the documentation.
Use Duck-Typing to Wrap Methods
The "Ruby way" is to use duck-typing rather than strict "type signatures." While there's nothing wrong with wrapping a method, your methods should:
Use meaningful argument names, if the type of an argument matters.
Perform implicit or explicit coercion or define singleton methods in the cases where you need an object to #respond_to? a method it doesn't currently support.
Implement singleton methods when necessary to permit duck-typing.
For example:
def foo array
array.is_a? Array
end
def flexible_foo string_or_array
if string_or_array.respond_to? :split
array = string_or_array.split /,?\s+/
else
array = string_or_array
end
foo array
end
flexible_foo 'a, b, c'
#=> true
flexible_foo %w[a b c]
#=> true
In this example, #foo expects an array. By wrapping #foo, we create a work-alike method that coerces the value into an array if it responds to the :split message, which String does and Array does not.
Documentation
Both RDoc and YARD do a reasonable job of documenting method signatures "out of the box," but YARD also has support for using tags to document things like "type signatures" and return types.
If your code is written with fixed expectations about what kinds of objects can be passed as arguments, then you can document those expectations in comments which RDoc or YARD will dutifully report. However, this is considered to be the programmer's responsibility rather than the Ruby interpreter's, and you'll know if you've broken the implicit contract when Ruby raises a NoMethodError exception at runtime.
This is one reason the Ruby community embraces test-driven development: since Ruby can redefine methods and classes on the fly, the interpreter won't know until runtime whether the calling method has sent an invalid message or not. This is generally considered a Good Thing®, but your mileage and opinions may certainly vary.
Context: i argue that saying pass_by_reference when it's really pass_by_sharing is misleading
Here is the excerpt from the book "Effective Ruby" I'm arguing against
"Most objects are passed around as references and not as actual values. When these types of objects are inserted into a container the collection class is actually storing a reference to the object and not the object itself. (The notable exception to the rule is the Fixnum class whose objects are always passed by value and not by reference.)
The same is true when objects are passed as method arguments. The method will receive a reference to the object and not a new copy. This is great for efficiency but has a startling implication.
"
The 'call by value' and 'call by object sharing' terminology matches Ruby's behavior, and
the terminology is consistent with other object orientated languages that have the same
semantics.
'Call by value' and 'call by object sharing' basically mean the same thing in object orientated languages, so which one is used doesn't really matter. Someone just thought it would clarify the confusion in the terminology to add more terminology.
If 'call by reference' was implemented in Ruby though, it would be something like:
def f(byref x)
x = "CHANGED"
end
x = ""
f(x)
# X is "CHANGED"
Here, the value of x is changed. The value being which object x refers to.
Using terms 'call by reference' just creates confusion though because they mean
different things to different people. It's unnecessary in
languages like Ruby because you don't have a choice. In languages with different
calling mechanisms like C++ and C# it makes more sense to teach these terms because
they have a real effect on programs and we can come up with non hypothetical examples
of them.
When explaining parameters in Ruby, you don't need to use any of these terms though.
They're meaningless to people that don't already know the language. Just
describe the behavior itself without that terminology and avoid the baggage.
I would say if you insist on using these terms, then use 'call by value' because it's usually considered more correct. The 'Programming Ruby' book calls it 'call by value', as well as plenty of Ruby programmers. Using the term with a different meaning than its technical one isn't helpful.
You are right. Ruby is pass-by-value only. The semantics of passing and assigning in Ruby are exactly identical to those in Java. And Java is universally described (on Stack Overflow and the rest of the Internet) as pass-by-value only. Terms about languages such as pass-by-value and pass-by-reference must be consistently used across languages to be meaningful.
The thing that is often misunderstood by people who say Java, Ruby, etc. "pass objects by reference" is that "objects" are not values in these languages, and thus cannot be "passed". The value of every variable and result of every expression is a "reference", which is a pointer to an object. The expression for creating an object returns an object pointer; when you access an attribute through the dot notation, the left side takes an object pointer; when you assign one variable to another, you copy the pointer resulting in two pointers to the same object. You always deal with pointers to objects, never objects themselves.
This is made explicit in Java as the only types in Java are primitive types and reference types -- there are no "object types". So every value in Java that is not a primitive is a reference (a pointer to an object). Ruby is dynamically-typed, so variables don't have explicit types. But you can imagine a dynamically-typed language as just a statically-typed language having exactly one type; and for languages like Python and Ruby, if this type were described, it be a pointer-to-object type.
The issue ultimately boils down to a problem of definitions. People argue over things because there is no precise definition, or they each have slightly different definitions. Rather then argue over vaguely-defined things like what is the "value" of a variable, or whether named values are "variables" or "names", etc., we need to use a definition for pass-by-value and pass-by-reference that is based purely on semantics of a language structure. #fgb's answer provides a clear semantic test for pass-by-reference. In "true pass-by-reference", e.g. with & in C++ and PHP, or with ref or out in C#, simple assignment (i.e. =) to a parameter variable has the same effect as simple assignment to the passed variable in the original scope. In pass-by-value, simple assignment (i.e. =) to a parameter variable has no effect in the original scope. This is what we see in Java, Python, Ruby, and many other languages.
I dislike people coming up with new names like "pass by object sharing", when they don't understand that the semantics are covered by an existing term, pass-by-value. Adding a new term only adds more to the confusion rather than reduce it, because it does not resolve the definitions of existing terms, only adding a new term that also needs to be defined.
It seems to me like Ruby has a lot of syntactic flexibility, and many things may be written in several ways.
Are there any language features/syntactic sugar/coding conventions that you avoid as a Ruby programmer for clarity? I am asking about things that you choose not to use on purpose, not stuff you still have to learn.
If your answer is, "I use everything!," do you ever comment code that would be otherwise obvious if the reader knows the relevant Ruby syntax?
[I'm particularly interested in Ruby in a RoR context, but everything is welcome.]
The whole range of "$" globals (see Pickaxe2 pp333-336) mostly inherited from Perl, are pretty ghastly, although I have on occasion found myself using $: instead of $LOAD_PATH.
I generally refrain from going overboard with monkey patching because it could lead to some maintainability and readability issues. It's a great feature if used correctly, but it's easy to get carried away.
The for ... in ... loop. It directly compiles to obj.each (and throws a strange error message accordingly) and is totally unnecessary. I don't even see where it improves readability - if you've been around ruby for more than a week, #each should be natural.
This might be obvious but I generally avoid using eval if there's any alternative.
First off: I'll break many of these rules if it's for a short one-off script, or a one-liner on the command line, or in irb. But most of my time is spent in medium sized or larger scripts or applications. So:
Avoid:
using class << self code block for class methods. It's a cute trick, but no better than def self.foo, and less readable (especially after the first page).
for i in collection: Use collection.each instead.
proc {...}: usually lambda {...} is better.
class variables (e.g. ##foo). They are problematic and can usually be replaced with class-level instance variables without much effort.
Anything that causes a warning, and preferably anything that causes a warning when run via the more strict ruby -w. This is especially important if you are writing a gem to be used by others.
'else' on a 'begin ... rescue ... end' block. Personal preference: It's too much of an edge case and so few people even know it exists or how it works to be worth it.
ObjectSpace and GC. You probably don't need to go there. You definitely don't want to go there.
=begin and =end multi-line comments. Personal preference for line-wise comments. These just annoy the hell out of me.
Use it, but sparingly or as a last resort (and comment it appropriately):
eval (or class_eval, etc) when passing in a string. There are some metaprogramming tricks you just can't do without passing in a string. And occasionally, the string version performs dramatically better (and sometimes that matters). Otherwise, I prefer to send in blocks of actual ruby code for my metaprogramming. eval can be avoided entirely for many metaprogramming tasks.
Adding or redefining methods on classes which were not created by me and may be used by code outside my control; a.k.a. monkey-patching. This rule is mostly for larger codebases and libraries; I'll gladly and quickly make an exception for small one-off scripts. I'll also make an exception for fixing buggy third-party libraries (although you may shoot yourself in the foot when you upgrade!). Selector namespaces (or something similar) would go a long way to make ruby nice in this regard. That said, sometimes it is worth the trouble. ;-)
Global variables (except for classes). I'll even pass in $stdout as a parameter to my objects or methods, rather than use them directly. It makes re-use of the code much easier and safer. Sometimes you can't avoid it (e.g. $0, $:, $$ and other environmental variables, but even then you can limit your usage).
speaking of which, I prefer to limit my usage of the perlish symbol globals entirely, but if they need to be used more than a little bit, then require "English".
break, redo, next, try: Often they make a block, loop, or method much more elegant than it otherwise could be. Usually they just make you scratch your head for a few minutes when you haven't seen that code for a while.
__END__ for a data-block. Excellent for a small one-file script. Unhelpful for a multi-file app.
Don't use it, but don't really avoid it either:
try/catch
continuations
Things I use often, that others might not care for, or I don't see often:
'and' and 'or' keywords: they have different precedence from && and ||, so you need to be careful with them. I find their different precedence to be very useful.
regex black magic (provided I've got some examples in unit tests for it)
HEREDOC strings
One thing I really loathe is "improper" use of {} and do ... end for blocks. I can't seem to find exactly where I learnt the practice myself but it is generally accepted to do {} for single line blocks and do ... end for multiline blocks.
Proper use:
[1, 2, 3, 4].map {|n| n * n }.inject(1) { |n,product| n * product }
or
[1, 2, 3, 4].inject do |n,product|
n = n * n
product = n * product
end
Improper use:
[1,2,3,4].map do |n| n * n end.inject(1) do |n,product| n * product end
or
[1, 2, 3, 4].inject { |n,product|
n = n * n
product = n * product
}
All of which, of course, will execute giving 576
Avoid chaining too many method calls together. It is very common in Ruby to chain methods together.
user.friends.each {|friend| friend.invite_to_party}
This may seem ok, but breaks Law of Demeter:
More formally, the Law of Demeter for functions requires that a method M of an object O may only invoke the methods of the following kinds of objects:
O itself
M's parameters
any objects created/instantiated
within M
O's direct component objects
The example above isn't perfect and a better solution would be something like this:
user.invite_friends_to_party
The example's problem isn't Ruby's fault but it is very easy to produce code that breaks Law of Demeter and makes code unreadable.
In short, avoid features that decreases code readability. It is a very important that the code you produce is easy to read.
begin nil+234 rescue '' end
above syntax is valid, but you should never use it.
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 11 years ago.
Locked. This question and its answers are locked because the question is off-topic but has historical significance. It is not currently accepting new answers or interactions.
I've read some of the recent language vs. language questions with interest... Perl vs. Python, Python vs. Java, Can one language be better than another?
One thing I've noticed is that a lot of us have very superficial reasons for disliking languages. We notice these things at first glance and they turn us off. We shun what are probably perfectly good languages as a result of features that we'd probably learn to love or ignore in 2 seconds if we bothered.
Well, I'm as guilty as the next guy, if not more. Here goes:
Ruby: All the Ruby example code I see uses the puts command, and that's a sort of childish Yiddish anatomical term. So as a result, I can't take Ruby code seriously even though I should.
Python: The first time I saw it, I smirked at the whole significant whitespace thing. I avoided it for the next several years. Now I hardly use anything else.
Java: I don't like identifiersThatLookLikeThis. I'm not sure why exactly.
Lisp: I have trouble with all the parentheses. Things of different importance and purpose (function declarations, variable assignments, etc.) are not syntactically differentiated and I'm too lazy to learn what's what.
Fortran: uppercase everything hurts my eyes. I know modern code doesn't have to be written like that, but most example code is...
Visual Basic: it bugs me that Dim is used to declare variables, since I remember the good ol' days of GW-BASIC when it was only used to dimension arrays.
What languages did look right to me at first glance? Perl, C, QBasic, JavaScript, assembly language, BASH shell, FORTH.
Okay, now that I've aired my dirty laundry... I want to hear yours. What are your language hangups? What superficial features bother you? How have you gotten over them?
I hate Hate HATE "End Function" and "End IF" and "If... Then" parts of VB. I would much rather see a curly bracket instead.
PHP's function name inconsistencies.
// common parameters back-to-front
in_array(needle, haystack);
strpos(haystack, needle);
// _ to separate words, or not?
filesize();
file_exists;
// super globals prefix?
$GLOBALS;
$_POST;
I never really liked the keywords spelled backwards in some scripting shells
if-then-fi is bad enough, but case-in-esac is just getting silly
I just thought of another... I hate the mostly-meaningless URLs used in XML to define namespaces, e.g. xmlns="http://purl.org/rss/1.0/"
Pascal's Begin and End. Too verbose, not subject to bracket matching, and worse, there isn't a Begin for every End, eg.
Type foo = Record
// ...
end;
Although I'm mainly a PHP developer, I dislike languages that don't let me do enough things inline. E.g.:
$x = returnsArray();
$x[1];
instead of
returnsArray()[1];
or
function sort($a, $b) {
return $a < $b;
}
usort($array, 'sort');
instead of
usort($array, function($a, $b) { return $a < $b; });
I like object-oriented style. So it bugs me in Python to see len(str) to get the length of a string, or splitting strings like split(str, "|") in another language. That is fine in C; it doesn't have objects. But Python, D, etc. do have objects and use obj.method() other places. (I still think Python is a great language.)
Inconsistency is another big one for me. I do not like inconsistent naming in the same library: length(), size(), getLength(), getlength(), toUTFindex() (why not toUtfIndex?), Constant, CONSTANT, etc.
The long names in .NET bother me sometimes. Can't they shorten DataGridViewCellContextMenuStripNeededEventArgs somehow? What about ListViewVirtualItemsSelectionRangeChangedEventArgs?
And I hate deep directory trees. If a library/project has a 5 level deep directory tree, I'm going to have trouble with it.
C and C++'s syntax is a bit quirky. They reuse operators for different things. You're probably so used to it that you don't think about it (nor do I), but consider how many meanings parentheses have:
int main() // function declaration / definition
printf("hello") // function call
(int)x // type cast
2*(7+8) // override precedence
int (*)(int) // function pointer
int x(3) // initializer
if (condition) // special part of syntax of if, while, for, switch
And if in C++ you saw
foo<bar>(baz(),baaz)
you couldn't know the meaning without the definition of foo and bar.
the < and > might be a template instantiation, or might be less-than and greater-than (unusual but legal)
the () might be a function call, or might be just surrounding the comma operator (ie. perform baz() for size-effects, then return baaz).
The silly thing is that other languages have copied some of these characteristics!
Java, and its checked exceptions. I left Java for a while, dwelling in the .NET world, then recently came back.
It feels like, sometimes, my throws clause is more voluminous than my method content.
There's nothing in the world I hate more than php.
Variables with $, that's one extra odd character for every variable.
Members are accessed with -> for no apparent reason, one extra character for every member access.
A freakshow of language really.
No namespaces.
Strings are concatenated with ..
A freakshow of language.
All the []s and #s in Objective C. Their use is so different from the underlying C's native syntax that the first time I saw them it gave the impression that all the object-orientation had been clumsily bolted on as an afterthought.
I abhor the boiler plate verbosity of Java.
writing getters and setters for properties
checked exception handling and all the verbiage that implies
long lists of imports
Those, in connection with the Java convention of using veryLongVariableNames, sometimes have me thinking I'm back in the 80's, writing IDENTIFICATION DIVISION. at the top of my programs.
Hint: If you can automate the generation of part of your code in your IDE, that's a good hint that you're producing boilerplate code. With automated tools, it's not a problem to write, but it's a hindrance every time someone has to read that code - which is more often.
While I think it goes a bit overboard on type bureaucracy, Scala has successfully addressed some of these concerns.
Coding Style inconsistencies in team projects.
I'm working on a large team project where some contributors have used 4 spaces instead of the tab character.
Working with their code can be very annoying - I like to keep my code clean and with a consistent style.
It's bad enough when you use different standards for different languages, but in a web project with HTML, CSS, Javascript, PHP and MySQL, that's 5 languages, 5 different styles, and multiplied by the number of people working on the project.
I'd love to re-format my co-workers code when I need to fix something, but then the repository would think I changed every line of their code.
It irritates me sometimes how people expect there to be one language for all jobs. Depending on the task you are doing, each language has its advantages and disadvantages. I like the C-based syntax languages because it's what I'm most used to and I like the flexibility they tend to bestow on the developer. Of course, with great power comes great responsibility, and having the power to write 150 line LINQ statements doesn't mean you should.
I love the inline XML in the latest version of VB.NET although I don't like working with VB mainly because I find the IDE less helpful than the IDE for C#.
If Microsoft had to invent yet another C++-like language in C# why didn't they correct Java's mistake and implement support for RAII?
Case sensitivity.
What kinda hangover do you need to think that differentiating two identifiers solely by caSE is a great idea?
I hate semi-colons. I find they add a lot of noise and you rarely need to put two statements on a line. I prefer the style of Python and other languages... end of line is end of a statement.
Any language that can't fully decide if Arrays/Loop/string character indexes are zero based or one based.
I personally prefer zero based, but any language that mixes the two, or lets you "configure" which is used can drive you bonkers. (Apache Velocity - I'm looking in your direction!)
snip from the VTL reference (default is 1, but you can set it to 0):
# Default starting value of the loop
# counter variable reference.
directive.foreach.counter.initial.value = 1
(try merging 2 projects that used different counter schemes - ugh!)
In no particular order...
OCaml
Tuples definitions use * to separate items rather than ,. So, ("Juliet", 23, true) has the type (string * int * bool).
For being such an awesome language, the documentation has this haunting comment on threads: "The threads library is implemented by time-sharing on a single processor. It will not take advantage of multi-processor machines. Using this library will therefore never make programs run faster." JoCaml doesn't fix this problem.
^^^ I've heard the Jane Street guys were working to add concurrent GC and multi-core threads to OCaml, but I don't know how successful they've been. I can't imagine a language without multi-core threads and GC surviving very long.
No easy way to explore modules in the toplevel. Sure, you can write module q = List;; and the toplevel will happily print out the module definition, but that just seems hacky.
C#
Lousy type inference. Beyond the most trivial expressions, I have to give types to generic functions.
All the LINQ code I ever read uses method syntax, x.Where(item => ...).OrderBy(item => ...). No one ever uses expression syntax, from item in x where ... orderby ... select. Between you and me, I think expression syntax is silly, if for no other reason than that it looks "foreign" against the backdrop of all other C# and VB.NET code.
LINQ
Every other language uses the industry standard names are Map, Fold/Reduce/Inject, and Filter. LINQ has to be different and uses Select, Aggregate, and Where.
Functional Programming
Monads are mystifying. Having seen the Parser monad, Maybe monad, State, and List monads, I can understand perfectly how the code works; however, as a general design pattern, I can't seem to look at problems and say "hey, I bet a monad would fit perfect here".
Ruby
GRRRRAAAAAAAH!!!!! I mean... seriously.
VB
Module Hangups
Dim _juliet as String = "Too Wordy!"
Public Property Juliet() as String
Get
Return _juliet
End Get
Set (ByVal value as String)
_juliet = value
End Set
End Property
End Module
And setter declarations are the bane of my existence. Alright, so I change the data type of my property -- now I need to change the data type in my setter too? Why doesn't VB borrow from C# and simply incorporate an implicit variable called value?
.NET Framework
I personally like Java casing convention: classes are PascalCase, methods and properties are camelCase.
In C/C++, it annoys me how there are different ways of writing the same code.
e.g.
if (condition)
{
callSomeConditionalMethod();
}
callSomeOtherMethod();
vs.
if (condition)
callSomeConditionalMethod();
callSomeOtherMethod();
equate to the same thing, but different people have different styles. I wish the original standard was more strict about making a decision about this, so we wouldn't have this ambiguity. It leads to arguments and disagreements in code reviews!
I found Perl's use of "defined" and "undefined" values to be so useful that I have trouble using scripting languages without it.
Perl:
($lastname, $firstname, $rest) = split(' ', $fullname);
This statement performs well no matter how many words are in $fullname. Try it in Python, and it explodes if $fullname doesn't contain exactly three words.
SQL, they say you should not use cursors and when you do, you really understand why...
its so heavy going!
DECLARE mycurse CURSOR LOCAL FAST_FORWARD READ_ONLY
FOR
SELECT field1, field2, fieldN FROM atable
OPEN mycurse
FETCH NEXT FROM mycurse INTO #Var1, #Var2, #VarN
WHILE ##fetch_status = 0
BEGIN
-- do something really clever...
FETCH NEXT FROM mycurse INTO #Var1, #Var2, #VarN
END
CLOSE mycurse
DEALLOCATE mycurse
Although I program primarily in python, It irks me endlessly that lambda body's must be expressions.
I'm still wrapping my brain around JavaScript, and as a whole, Its mostly acceptable. Why is it so hard to create a namespace. In TCL they're just ugly, but in JavaScript, it's actually a rigmarole AND completely unreadable.
In SQL how come everything is just one, huge freekin SELECT statement.
In Ruby, I very strongly dislike how methods do not require self. to be called on current instance, but properties do (otherwise they will clash with locals); i.e.:
def foo()
123
end
def foo=(x)
end
def bar()
x = foo() # okay, same as self.foo()
x = foo # not okay, reads unassigned local variable foo
foo = 123 # not okay, assigns local variable foo
end
To my mind, it's very inconsistent. I'd rather prefer to either always require self. in all cases, or to have a sigil for locals.
Java's packages. I find them complex, more so because I am not a corporation.
I vastly prefer namespaces. I'll get over it, of course - I'm playing with the Android SDK, and Eclipse removes a lot of the pain. I've never had a machine that could run it interactively before, and now I do I'm very impressed.
Prolog's if-then-else syntax.
x -> y ; z
The problem is that ";" is the "or" operator, so the above looks like "x implies y or z".
Java
Generics (Java version of templates) are limited. I can not call methods of the class and I can not create instances of the class. Generics are used by containers, but I can use containers of instances of Object.
No multiple inheritance. If a multiple inheritance use does not lead to diamond problem, it should be allowed. It should allow to write a default implementation of interface methods, a example of problem: the interface MouseListener has 5 methods, one for each event. If I want to handle just one of them, I have to implement the 4 other methods as an empty method.
It does not allow to choose to manually manage memory of some objects.
Java API uses complex combination of classes to do simple tasks. Example, if I want to read from a file, I have to use many classes (FileReader, FileInputStream).
Python
Indentation is part of syntax, I prefer to use the word "end" to indicate end of block and the word "pass" would not be needed.
In classes, the word "self" should not be needed as argument of functions.
C++
Headers are the worst problem. I have to list the functions in a header file and implement them in a cpp file. It can not hide dependencies of a class. If a class A uses the class B privately as a field, if I include the header of A, the header of B will be included too.
Strings and arrays came from C, they do not provide a length field. It is difficult to control if std::string and std::vector will use stack or heap. I have to use pointers with std::string and std::vector if I want to use assignment, pass as argument to a function or return it, because its "=" operator will copy entire structure.
I can not control the constructor and destructor. It is difficult to create an array of objects without a default constructor or choose what constructor to use with if and switch statements.
In most languages, file access. VB.NET is the only language so far where file access makes any sense to me. I do not understand why if I want to check if a file exists, I should use File.exists("") or something similar instead of creating a file object (actually FileInfo in VB.NET) and asking if it exists. And then if I want to open it, I ask it to open: (assuming a FileInfo object called fi) fi.OpenRead, for example. Returns a stream. Nice. Exactly what I wanted. If I want to move a file, fi.MoveTo. I can also do fi.CopyTo. What is this nonsense about not making files full-fledged objects in most languages? Also, if I want to iterate through the files in a directory, I can just create the directory object and call .GetFiles. Or I can do .GetDirectories, and I get a whole new set of DirectoryInfo objects to play with.
Admittedly, Java has some of this file stuff, but this nonsense of having to have a whole object to tell it how to list files is just silly.
Also, I hate ::, ->, => and all other multi-character operators except for <= and >= (and maybe -- and ++).
[Disclaimer: i only have a passing familiarity with VB, so take my comments with a grain of salt]
I Hate How Every Keyword In VB Is Capitalized Like This. I saw a blog post the other week (month?) about someone who tried writing VB code without any capital letters (they did something to a compiler that would let them compile VB code like that), and the language looked much nicer!
My big hangup is MATLAB's syntax. I use it, and there are things I like about it, but it has so many annoying quirks. Let's see.
Matrices are indexed with parentheses. So if you see something like Image(350,260), you have no clue from that whether we're getting an element from the Image matrix, or if we're calling some function called Image and passing arguments to it.
Scope is insane. I seem to recall that for loop index variables stay in scope after the loop ends.
If you forget to stick a semicolon after an assignment, the value will be dumped to standard output.
You may have one function per file. This proves to be very annoying for organizing one's work.
I'm sure I could come up with more if I thought about it.