Grouping operator ( ) in JavaScript
The grouping operator ( ) controls the precedence of evaluation in expressions.
Does the functionality ( ) in JavaScript itself differ from Haskell or any other programming languages?
In other words,
Is the functionality ( ) in programming languages itself affected by evaluation strategies ?
Perhaps we can share the code below:
a() * (b() + c())
to discuss the topic here, but not limited to the example.
Please feel free to use your own examples to illustrate. Thanks.
Grouping parentheses mean the same thing in Haskell as they do in high school mathematics. They group a sub-expression into a single term. This is also what they mean in Javascript and most other programming language1, so you don't have to relearn this for Haskell coming from other common languages, if you have learnt it the right way.
Unfortunately, this grouping is often explained as meaning "the expression inside the parentheses must be evaluated before the outside". This comes from the order of steps you would follow to evaluate the expression in a strict language (like high school mathematics). However the grouping really isn't really about the order in which you evaluate things, even in that setting. Instead it is used to determine what the expression actually is at all, which you need to know before you can do anythign at all with the expression, let alone evaluate it. Grouping is generally resolved as part of parsing the language, totally separate from the order in which any runtime evaluation takes place.
Let's consider the OP's example, but I'm going to declare that function call syntax is f{} rather than f() just to avoid using the same symbol for two purposes. So in my newly-made-up syntax, the OP's example is:
a{} * (b{} + c{})
This means:
a is called on zero arguments
b is called on zero arguments
c is called on zero arguments
+ is called on two arguments; the left argument is the result of b{}, and the right argument is the result of c{}
* is called on two arguments: the left argument is the result of a{}, and the right argument is the result of b{} + c{}
Note I have not numbered these. This is just an unordered list of sub-expressions that are present, not an order in which we must evaluate them.
If our example had not used grouping parentheses, it would be a{} * b{} + c{}, and our list of sub-expressions would instead be:
a is called on zero arguments
b is called on zero arguments
c is called on zero arguments
+ is called on two arguments; the left argument is the result of a{} * b{}, and the right argument is the result of c{}
* is called on two arguments: the left argument is the result of a{}, and the right argument is the result of b{}
This is simply a different set of sub-expressions from the first (because the overall expression doesn't mean the same thing). That is all that grouping parentheses do; they allow you to specify which sub-expressions are being passed as arguments to other sub-expressions2.
Now, in a strict language "what is being passed to what" does matter quite a bit to evaluation order. It is impossible in a strict language to call anything on "the result of a{} + b{} without first having evaluated a{} + b{} (and we can't call + without evaluating a{} and b{}). But even though the grouping determines what is being passed to what, and that partially determines evaluation order3, grouping isn't really "about" evaluation order. Evaluation order can change as a result of changing the grouping in our expression, but changing the grouping makes it a different expression, so almost anything can change as a result of changing grouping!
Non-strict languages like Haskell make it especially clear that grouping is not about order of evaluation, because in non-strict languages you can pass something like "the result of a{} + b{}" as an argument before you actually evaluate that result. So in my lists of subexpressions above, any order at all could potentially be possible. The grouping doesn't determine it at all.
A language needs other rules beyond just the grouping of sub-expressions to pin down evaluation order (if it wants to specify the order), whether it's strict or lazy. So since you need other rules to determine it anyway, it is best (in my opinion) to think of evaluation order as a totally separate concept than grouping. Mixing them up seems like a shortcut when you're learning high school mathematics, but it's just a handicap in more general settings.
1 In languages with roughly C-like syntax, parentheses are also used for calling functions, as in func(arg1, arg2, arg3). The OP themselves has assumed this syntax in their a() * (b() + c()) example, where this is presumably calling a, b, and c as functions (passing each of them zero arguments).
This usage is totally unrelated to grouping parentheses, and Haskell does not use parentheses for calling functions. But there can be some confusion because the necessity of using parentheses to call functions in C-like syntax sometimes avoids the need for grouping parentheses e.g. in func(2 + 3) * 6 it is unambiguous that 2 + 3 is being passed to func and the result is being multiplied by 6; in Haskell syntax you would need some grouping parentheses because func 2 + 3 * 6 without parentheses is interpreted as the same thing as (func 2) + (3 * 6), which is not func (2 + 3) * 6.
C-like syntax is not alone in using parentheses for two totally unrelated purposes; Haskell overloads parentheses too, just for different things in addition to grouping. Haskell also uses them as part of the syntax for writing tuples (e.g. (1, True, 'c')), and the unit type/value () which you may or may not want to regard as just an "empty tuple".
2 Which is also what associativity and precedence rules for operators do. Without knowing that * is higher precedence than +, a * b + c is ambiguous; there would be no way to know what it means. With the precedence rules, we know that a * b + c means "add c to the result of multiplying a and b", but we now have no way to write down what we mean when we want "multiply a by the result of adding b and c" unless we also allow grouping parentheses.
3 Even in a strict language the grouping only partially determines evaluation order. If you look at my "lists of sub-expressions" above it's clear that in a strict language we need to have evaluated a{}, b{}, and c{} early on, but nothing determines whether we evaluate a{} first and then b{} and then c{}, or c{} first, and then a{} and then b{}, or any other permutation. We could even evaluate only the two of them in the innermost +/* application (in either order), and then the operator application before evaluating the third named function call, etc etc.
Even in a strict language, the need to evaluate arguments before the call they are passed to does not fully determine evaluation order from the grouping. Grouping just provides some constraints.
4 In general in a lazy language evaluation of a given call happens a bit at a time, as it is needed, so in fact in general all of the sub-evaluations in a given expression could be interleaved in a complicated fashion (not happening precisely one after the other) anyway.
To clarify the dependency graph:
Answer by myself (the Questioner), however, I am willing to be examined, and still waiting for your answer (not opinion based):
Grouping operator () in every language share the common functionality to compose Dependency graph.
In mathematics, computer science and digital electronics, a dependency graph is a directed graph representing dependencies of several objects towards each other. It is possible to derive an evaluation order or the absence of an evaluation order that respects the given dependencies from the dependency graph.
dependency graph 1
dependency graph 2
the functionality of Grouping operator () itself is not affected by evaluation strategies of any languages.
In Spacy pattern matching, I know that we can use Kleene operator for ranges. For example,
pattern = [{"LOWER": "hello"},{ "OP": "*"}]. Here the star, known as kleene operator, means match against zero or any number of tokens. How can I modify the rule such that only 4 or 5 tokens are matched after the token "hello"?
In other NLP applications, for example,in GATE application, we can use some pattern like {Token.string == "hello"}({Token})[4,5] for the above task. Does Spacy have any such mechanism?
Thanks
This isn't currently supported, see the feature request: https://github.com/explosion/spaCy/issues/5603.
In v3.0.6+, you can use the new match_alignments to filter matches in post-processing: https://spacy.io/api/matcher. The matcher will still be slow if your patterns with just * end up with a lot of long/overlapping matches.
I'm trying to write a lexer rule that would match following strings
a
aa
aaa
bbbb
the requirement here is all characters must be the same
I tried to use this rule:
REPEAT_CHARS: ([a-z])(\1)*
But \1 is not valid in antlr4. is it possible to come up with a pattern for this?
You can’t do that in an ANTLR lexer. At least, not without target specific code inside your grammar. And placing code in your grammar is something you should not do (it makes it hard to read, and the grammar is tied to that language). It is better to do those kind of checks/validations inside a listener or visitor.
Things like back-references and look-arounds are features that krept in regex-engines of programming languages. The regular expression syntax available in ANTLR (and all parser generators I know of) do not support those features, but are true regular languages.
Many features found in virtually all modern regular expression libraries provide an expressive power that far exceeds the regular languages. For example, many implementations allow grouping subexpressions with parentheses and recalling the value they match in the same expression (backreferences). This means that, among other things, a pattern can match strings of repeated words like "papa" or "WikiWiki", called squares in formal language theory.
-- https://en.wikipedia.org/wiki/Regular_expression#Patterns_for_non-regular_languages
Is there a way to index while performing REGEX_TEST() on a string to field to retrieve documents in ArangoDB?
Also if there is any way to optimize this please let me know
There is no index acceleration available for the REGEX_TEST() AQL function, and it is unlikely to come in the future. Not because there is no interest from users and developers, but because it's not really possible to build any sort of index data structure that would allow to speed up regular expression evaluation.
Regular expressions as supported by ArangoDB allow for many different types of expressions, but because they can differ so much, there is almost no chance to have a suitable index. For equality comparisons there are hash indexes, which are probably the fastest kind of index. For range queries there are skiplist indexes, and there are of course quite a few more index types known in computer science, but I'm not aware of a single one that could speed up arbitrary regex.
If your expression allows, maybe there is a chance add a filter criterion before REGEX_TEST() which might utilize an index? This will mostly be limited to case-sensitive prefix matching, e.g. FILTER REGEX_TEST(doc.str, "a[a-z]*") could be extended to FILTER doc.str >= "a" AND doc.str < "b" AND REGEX_TEST(doc.str, "a[a-z]*") and allow for a skiplist index being used to only evaluate the regex on documents where str starts with a. Or some simple regex like [fm]oo|bar could be rewritten to a set of equality comparisons: FILTER doc.str IN ["foo","moo","bar"]. Also have a look at ArangoSearch.
The GRAPH_TRAVERSAL has an option called filterVertices, which the documentation states will be used to only allow those vertices matching the examples to go through. Is there any negative version of this, e.g. to allow everything except those matching the filter?
In many cases this would be useful, e.g. traverse everything except those marked disabled (or old-version) or something like that. Of course this can be done with a JS function, but why not built-in?
You're right, its currently not possible, and if you want to use GRAPH_TRAVERSAL you have to write your own visitor function.
However, the recommended way is to use the new pattern matching where you can use FILTER statements like this:
db._query("FOR vertex IN 1..3 OUTBOUND 'circles/A' GRAPH
'traversalGraph' FILTER vertex._key != 'G' return v._key")
.toArray();
so you can use arbitrary filter expressions on vertices, edges and paths and their sub-parts.
In general our development focus will be on the pattern matching traversals and doing as much as is possible in AQL. If you like to implement such a feature for the general graph module, Contributions are always welcome.