Comment by lerno
4 days ago
Two reasons, the second being the important: (1) If I read "io.print", is this "the print function in the module io" or "the print method for the variable io". There tends to be an overlap in naming here so that's a downside (2) parsing and semantic checking is much easier if the namespace is clear from the grammar.
In particular, C3's "path shortening", where you're allowed to write `file::open("foo.txt")` rather than having to use the full `std::io::file::open("foo.txt")` is only made possible because the namespace is distinct at the grammar level.
If we play with changing the syntax because it isn't as elegant as `file.open("foo.txt")`, we'd have to pay by actually writing `std.io.file.open("foo.txt")` or change to a flat module system. That is a fairly steep semantic cost to pay for a nicer namespace separator.
I might have overlooked some options, if so - let me know.
I have never found either (1) or (2) to be a problem in hundreds of thousands of lines of Python.
> In particular, C3's "path shortening" ... we'd have to pay by actually writing `std.io.file.open("foo.txt")` or change to a flat module system.
You can easily and explicitly shorten paths in other languages. For example, in Python "from mypackage.mysubpackage import mymodule; mymodule.myfunc()"
Python even gracefully handles name collisions by allowing you to change the name of the local alias, e.g. "from my_other_package.mysubpackage import mymodule as other_module"
I find the "from .. import" to be really handy to understand touchpoints for other modules, and it is not very verbose, because you can have a comma-separated list of things that you are aliasing into the current namespace.
(You can also use "from some_module import *" to bring everything in, which is highly useful for exploratory programming but is an anti-pattern for production software.)
Of course you can explicitly shorten paths. I was talking about C3's path shortening which is doing this for you. This means you do not need to alias imports, which is otherwise how languages do it.
I don't want to get too far into details, but it's understandable that people misunderstand it if they haven't used it, as it's a novel approach not used by any other language.
Oh, I understand it. I just think that (a) explicit is better than implicit; and (b) the amount of characters that Python requires to keep imports explicit is truly minimal, and is a huge aid to figuring out where things came from.
> (1) If I read "io.print", is this "the print function in the module io" or "the print method for the variable io"
I don't see the issue. Just look up the id ? Moreover, if modules are seen as objects, the meaning is quite the same.
> checking is much easier if the namespace is clear from the grammar.
Again (this time by the checker) just look up the symbol table ?
Let's say you have find foo::bar(), then we know that the path is <some path>::foo, the function is `bar` consequently we search for all modules matching the substring ::foo, and depending on whether (1) we get multiple matches (2) we only get a match that is not properly visible (3) we get a match that isn't imported, (4) we get no match or (5) we get a visible match, we print different things. In the case 1-4, we give good errors to allow the user to take the proper action.
If instead we had foo.bar(), we cannot know if this is the method "bar" on local or global foo, or a function "bar()" in a path matching the substring "foo". Consequently we cannot properly issue 4, since we don't know what the intent was.
So far, not so bad. Let's say it's instead foo::baz::bar(). In the :: case, we don't have any change in complexity, we simply match ::foo::baz instead.
However, for foo.baz.bar(), we get more cases, and let us also bring in the possibility of a function pointer being invoked: 1. It is invoking the method bar() on the global baz is a module that ends with "foo" 2. It is calling a function pointer stored in member bar on the global variable baz is a module that ends with "foo" 3. It is calling the function bar() in a module that ends with "foo.baz" 4. It is calling the function pointer stored in the global bar in a module that ends with "foo.baz" 5. It is invoking the method bar on the member baz of the local foo 6. It is calling a function pointer stored in the member bar in the member baz of the local foo
This might seem doable, but note that for every module we have that has a struct, we need to speculatively dive into it to see if it might give a match. And then give a good error message to the user if everything fails.
Note here that if we had yet another level, `foo.bar.baz.abc()` then the number of combinations to search increases yet again.
I think you are overcomplicating this.
This is exactly the syntax Python uses, and there is no "search" per se.
Either an identifier is in the current namespace or not.
And if it is in the current namespace, there can only be one.
The only time multiple namespaces are searched is when you are scoped within a function or class which might have a local variable or member of the same name.
> find foo::bar(), then we know that the path is <some path>::foo, the function is `bar` consequently we search for all modules matching the substring ::foo,
The only reason you need to have a search and think about all the possibilities is that you are deliberately allowing implicit lookups. Again, in Python:
1) Everything is explicit; but 2) you can easily create shorthand aliases when you want.
> note that for every module we have that has a struct, we need to speculatively dive into it to see if it might give a match. And then give a good error message to the user if everything fails.
Only if you rely on search, as opposed to, you know, if you 'import foo' then 'foo' refers to what you imported.