Loading ...
Sorry, an error occurred while loading the content.

Re: Funcref and script local functions

Expand Messages
  • Eric Arnold
    ... The problem with this is that you can no longer have private object function refs. I d be interested to hear from Bram about what the intent was here. I
    Message 1 of 12 , Jul 3, 2006
    • 0 Attachment
      On 7/3/06, Yakov Lerner <iler.ml@...> wrote:
      > On 7/3/06, Yakov Lerner <iler.ml@...> wrote:
      > > On 6/30/06, Hari Krishna Dara <hari_vim@...> wrote:
      > > > ... The Funcref obtained via function('s:T') can't be
      > > > called from outside the script ... [unexpectedly]
      > >
      > > I agree, Hari. I'd expect funcref function('s:T') to be callable
      > > outside of the script, too.
      >
      > To make myself more clear. I expect g:Xxx() to be callable
      > from global scope of from another script in this example:
      >
      > " --- scritp x.vim
      > function! s:XXX()
      > echo "func XXX"
      > endfunction
      >
      > let g:Xxx=function('s:XXX')
      >
      > Yakov
      >


      The problem with this is that you can no longer have private object
      function refs.

      I'd be interested to hear from Bram about what the intent was here.

      I think I can see some method in the maddness.

      1) let ref = function('s:XXX')

      This mains the standard scope rules for the func ref, and so it
      stays local even if the variable holding it is global. I can image
      situations where this could be useful.

      2) let ref = function('<SNR>66_XXX')

      This forces the function to be available globally because it is
      explicitely defined, and there is little chance of mistakes about
      that.

      3) function obj.funcref() dict

      Again, I think the intent for having the object function in the global
      scope is unknown (to me at least). I think it left global because you
      don't really need specific scoping for it, as when it is used, it
      *should* be used via the object name, which is scoped by the user.

      function s:obj.funcref() dict

      Hari, can you give an example of why function('s:T') should be
      globally scoped? I can't see a need for it, given all the
      possibilites for obtaining a ref.
    • Yakov Lerner
      ... Of course you can .: let s:Xxx=function( s:XXX ) script-private funcref let l:Xxx=function( s:XXX ) function-private funcref let
      Message 2 of 12 , Jul 3, 2006
      • 0 Attachment
        On 7/3/06, Eric Arnold <eric.p.arnold@...> wrote:
        > On 7/3/06, Yakov Lerner <iler.ml@...> wrote:
        > > On 7/3/06, Yakov Lerner <iler.ml@...> wrote:
        > > > On 6/30/06, Hari Krishna Dara <hari_vim@...> wrote:
        > > > > ... The Funcref obtained via function('s:T') can't be
        > > > > called from outside the script ... [unexpectedly]
        > > >
        > > > I agree, Hari. I'd expect funcref function('s:T') to be callable
        > > > outside of the script, too.
        > >
        > > To make myself more clear. I expect g:Xxx() to be callable
        > > from global scope of from another script in this example:
        > >
        > > " --- scritp x.vim
        > > function! s:XXX()
        > > echo "func XXX"
        > > endfunction
        > >
        > > let g:Xxx=function('s:XXX')
        > >
        > > Yakov
        > >
        >
        >
        > The problem with this is that you can no longer have private object
        > function refs.

        Of course you can .:

        let s:Xxx=function('s:XXX') " script-private funcref

        let l:Xxx=function('s:XXX') " function-private funcref

        let g:Xxx=function('s:XXX') " globally-accessible funcref

        Yakov
      • Eric Arnold
        ... No. These are variables where you have managed their scopes, but the underlying func ref *contained* would lose it s ability to maintain it s scope, i.e.
        Message 3 of 12 , Jul 3, 2006
        • 0 Attachment
          On 7/3/06, Yakov Lerner <iler.ml@...> wrote:
          > On 7/3/06, Eric Arnold <eric.p.arnold@...> wrote:
          > > On 7/3/06, Yakov Lerner <iler.ml@...> wrote:
          > > > On 7/3/06, Yakov Lerner <iler.ml@...> wrote:
          > > > > On 6/30/06, Hari Krishna Dara <hari_vim@...> wrote:
          > > > > > ... The Funcref obtained via function('s:T') can't be
          > > > > > called from outside the script ... [unexpectedly]
          > > > >
          > > > > I agree, Hari. I'd expect funcref function('s:T') to be callable
          > > > > outside of the script, too.
          > > >
          > > > To make myself more clear. I expect g:Xxx() to be callable
          > > > from global scope of from another script in this example:
          > > >
          > > > " --- scritp x.vim
          > > > function! s:XXX()
          > > > echo "func XXX"
          > > > endfunction
          > > >
          > > > let g:Xxx=function('s:XXX')
          > > >
          > > > Yakov
          > > >
          > >
          > >
          > > The problem with this is that you can no longer have private object
          > > function refs.
          >
          > Of course you can .:


          No. These are variables where you have managed their scopes, but the
          underlying func ref *contained* would lose it's ability to maintain
          it's scope, i.e. .....


          > let s:Xxx=function('s:XXX') " script-private funcref


          You could do

          let g:Xxx = s:Xxx

          and now the private/script-local function s:XXX() can be called
          globally through g:Xxx . I'm not sure whether this would be a
          good or bad thing. It's sort of an ease-of-use vs strict scoping
          rules.

          Also, I think there might be a semantics problem here about the term
          "function reference". It seems to be concurrently the variable name,
          which can be called as a function directly, vs the underlying
          reference itself. I *think* that the func ref var actually does
          contain a reference, rather than *being* a reference, especially since
          you can pass it around, and maintain the original function name

          Anyway, I think there are, or should be two terms to talk about this
          kind of issue:

          - function reference

          - function reference variable



          > let l:Xxx=function('s:XXX') " function-private funcref
          >
          > let g:Xxx=function('s:XXX') " globally-accessible funcref
          >
          > Yakov
          >


          I'm still not sure that I understand the utility of this. If you want
          to create a global func ref variable which can call its content, which
          is a script-local function ref, then why not just make the function
          global in the first place? Are you worried about a global namespace
          conflict of the global function [ref] name? I guess the proposed
          change/request would allow creating an alias in the form of a global
          func ref var, bypass a global namespace conflict.

          Detouring away from the namespace question, here are two examples to consider:


          let s:obj = {}
          function s:obj.tstfunc() dict
          return 'here'
          endfunction
          echo s:obj.tstfunc()
          let g:Ref = s:obj
          " Yank and put this into the command line to test global scope:
          echo g:Ref.tstfunc()


          or


          function! s:tstfunc2()
          return 'here2'
          endfunction

          let s:ref = function( 's:tstfunc2' )
          echo s:ref()
          let g:Global_ref = s:ref
          " Yank and put this into the command line to test global scope, it should fail:
          echo g:Global_ref()


          So, my question is: Why should this be allowed? In the above test
          script, if you made function('s:XXX') global, then it is true in
          your examples, that you can create local function ref variables which
          hold that ref. However, in your example,

          > let g:Xxx=function('s:XXX') " globally-accessible funcref

          here you have violated the explicit scope set on the function s:XXX
          . Actually, I personally don't care which happens, since I use
          scoping in Vim only for namespace separation, not data hiding (where I
          think the violation might cause some problem that I can't think of
          :-).

          Lastly, we've got an inconsistency: dict functions created as
          numbered function refs which are always global, vs the above which
          maintains scope.
        • Hari Krishna Dara
          On Mon, 3 Jul 2006 at 6:35am, Eric Arnold wrote: [snip] numbered functions. [/snip] ... I think it is best to leave the discussion about numbered functions. It
          Message 4 of 12 , Jul 3, 2006
          • 0 Attachment
            On Mon, 3 Jul 2006 at 6:35am, Eric Arnold wrote:

            [snip]
            numbered functions.
            [/snip]
            >
            > I'm still not getting it, I think. Do you have a case where the
            > numbered function scheme will break down, or is it about the
            > callbacks, described below?

            I think it is best to leave the discussion about numbered functions. It
            hasn't got much value to the current discussion of whether the functions
            references to the script local functions should be callable outside the
            script in which the functions are defined.

            > > >
            > > >
            > > > On 6/29/06, Hari Krishna Dara <hari_vim@...> wrote:
            > > > >
            > > > > When Funcref's were introduced in Vim7, I expected them to work for
            > > > > script-local functions, across scripts. The documentation didn't say
            > > > > that, but it didn't say that it wouldn't either, and I thought that
            that
            > > > > is one of its biggest uses (other than the actual intended
            > > > > functionality, which is for implementing numbered functions). However,
            I
            >
            > I'm not sure that there is a problem. As with C code, if you have
            > the option of declaring a function global/local, public/private, etc.
            > I think Vim script is allowing these options.
            >
            > Are you saying that you want to override the private script
            > declarations by declaring a function reference to a low enough level
            > pointer that it goes under the scope checker?

            As Yakov Lerner already clarified, this should be allowed by Vim. As we
            already know, Vim's script-local scoping for functions is not realy
            intended to prevent external callers. In fact, Vim allows you to use
            these functions in all your mappings etc., and the calls then originate
            from outside outside the script context. This is probably why Vim allows
            a script local function be called in general, using the <SNR> syntax.
            All that I am saying is that this should be more formalized for
            Funcrefs (see below).

            >
            >
            > > > > found that the Funcref references for such functions can't actually be
            > > > > passed out to other scripts. This reduces the usefulness of this
            feature
            > > > > as we can't register private functions to receive callbacks from other
            > > > > scripts.
            >
            >
            > I think this is probably a request that it be more object oriented
            > than it is, ie. you really want object-scoped functions, not
            > script-scoped. You seem to want the script localized, so it can't be
            > access generally, but then be public for registering callbacks. This
            > seems like an object-scope problem.

            The script-local functions in Vim is more of a means to distinguish the
            public API from private. They are not really intended to prevent outside
            callers. We are not going to build a security system using Vim scripting
            after all.

            >
            > I think that the numbered functions are allowed globally, probably
            > because they are intended to be used as you describe, for callbacks
            > from other scripts, since they are only created for the 'dict'
            > "object" functions, as far as I can tell.
            >
            > > > >
            > > > > What is weird is that the the Funcref() actually behaves exactly like
            > > > > the function name itself. Say you have a function called s:T() and say
            > > > > the script id is 60. The Funcref obtained via function('s:T') can't be
            > > > > called from outside the script, but if the Funcref is obtained using
            > > > > function('<SNR>60_T'), then it will be fine. Also, a Funcref obtained
            >
            >
            > Both of these examples seem reasonable to me. If you declare a
            > function reference to a script-local object, s:T then you don't
            > want it being accessed outside the script. If you declare a
            > '<SNR>60_T' reference, then you probably wanted to use it outside
            > the script, otherwise you wouldn't have gone through the trouble of
            > finding the script id.

            You can't declare a function using the <SNR> prefix. When you use the s:
            or <SID> prefix, vim automatically changes it to the <SNR> form using
            the id of the script. When you want to register a callback, through an
            autocommand or a map, you don't need to go through the trouble (which
            actually is not a big trouble) of finding the script id, as Vim takes
            care of expand <SID> to the right value. This doesn't work the same,
            when you want to register the same function as a callback with another
            plugin, so you have to explicitly pass the function name with the <SNR>
            form (ie., <SNR>60_T not <SID>T). If Funcref's on s:T is same as
            <SNR>60_T, then this makes it simpler (and more formal) (again, see
            below).

            > > > > using these two methods will not be to the same object, though you
            would
            > > > > expect them to be. The below echoes 0:
            >
            >
            > How did you test what the object was? Actually, I wouldn't expect it
            > to be the same object in any case, since each reference to it should
            > crease a new instance. They both might refer to the same function
            > definition stored internally, but I don't know.

            As my code sample below shows, the "is" operator on two separate
            references of "s:T" gives you "true", but if one reference is to the
            <SNR> form, you get "false". This is counter-intuitive, as you
            referencing the same exact function.

            >
            > Also, we aren't talking about true "objects", just to be clear, but an
            > enhancment that allows object-oriented-like functional access. This
            > limits the expectations we can have.

            This limitation is not applicable to my argument.

            [snip]

            > >
            > > No, that is not what I was saying. If there is a way to declare
            > > variables without initializing them, I would have said something like:
            > >
            > > Funcref var
            > >
            > > However, the equivalent of the above is to say:
            > >
            > > let var = function('somefunc')
            > >
            > > The alternative is of course to just initialize the variable as and when
            > > it is required, but I generally don't like this approach, as it is not
            > > clear which variables are being used.
            > >
            >
            > I want to have it auto-initialize whether in a "let" statement, or an
            > implicit setting via a function call argument. And I'd like to be
            > able to test a value that hasn't been explicitly initialized, i.e.
            >
            > if var1[1].elem1 > 0
            > ...
            >
            > where nothing about "var1" has been defined, but is used such that
            > automatic NULL elements would have to be created manually in any case.
            > If I really want it to fail if [1] or .elem1 are not
            > defined, I can use 'exists()' or 'has()'. As I said elsewhere, it
            > should be an option, for backward compatibility.

            I am not talking about "initialization" here, just declaration.

            --
            Thanks,
            Hari

            __________________________________________________
            Do You Yahoo!?
            Tired of spam? Yahoo! Mail has the best spam protection around
            http://mail.yahoo.com
          • Bram Moolenaar
            ... The whole purpose of script-local functions is that they will only be used locally in the script. Thus they can t be called from outside the script
            Message 5 of 12 , Jul 8, 2006
            • 0 Attachment
              Hari Krishna Dara wrote:

              > When Funcref's were introduced in Vim7, I expected them to work for
              > script-local functions, across scripts. The documentation didn't say
              > that, but it didn't say that it wouldn't either, and I thought that that
              > is one of its biggest uses (other than the actual intended
              > functionality, which is for implementing numbered functions). However, I
              > found that the Funcref references for such functions can't actually be
              > passed out to other scripts. This reduces the usefulness of this feature
              > as we can't register private functions to receive callbacks from other
              > scripts.

              The whole purpose of script-local functions is that they will only be
              used locally in the script. Thus they can't be called from outside the
              script (without the trick to obtain the script nr). Also not when using
              a Funcref.

              > What is weird is that the the Funcref() actually behaves exactly like
              > the function name itself.

              Yes, that basically sums it up.

              > There are other aspects of the new features that are very
              > counter-intuitive to me, whether I think in terms of Python or generic
              > "objects" in any language. The one which gets me the most is the
              > implicit typing of variables based on the initializer. For basic types
              > prior to Vim7 (integer and string), you could easily switch the value of
              > the variable from integer to string or vice versa, and the type() of the
              > variable would change, suggesting that it behaves like "duck typing" (as
              > per (wikipedia). But this observation can't be extended to the newer
              > object types, as the below will fail:
              >
              > let a = {}
              > let a = []
              >
              > If the type of value determines the type of the variable, and if we are
              > merely dealing with references (assigning references instead of copying
              > objects), then why should the second statement above generate the below
              > error?
              >
              > E706: Variable type mismatch for: a
              >
              > Is there a standard for this type of language behavior? I didn't find
              > anything at this page: http://en.wikipedia.org/wiki/Dynamically_typed

              I don't know if there is a standard and generally I don't care. There
              is nothing against inventing something new. The main point is that it's
              very easy to make mistakes with using the wrong type for a variable.
              Using declarations is the normal way to have the compiler or interpreter
              check the type. But declarations are annoying and certainly don't fit
              with Vim 6.x script. Therefore I used the "sticky types" mechanism: the
              first time a variable is assigned a value that's the type it has. Very
              simple and efficient. You just need to remember that variables are not
              dynamically typed. Except for int/string, for backwards compatibility.

              --
              Back off man, I'm a scientist.
              -- Peter, Ghostbusters

              /// Bram Moolenaar -- Bram@... -- http://www.Moolenaar.net \\\
              /// sponsor Vim, vote for features -- http://www.Vim.org/sponsor/ \\\
              \\\ download, build and distribute -- http://www.A-A-P.org ///
              \\\ help me help AIDS victims -- http://ICCF-Holland.org ///
            • Hari Krishna Dara
              ... I often provide public functions that are useful for other plugins to interface, so I don t like mixing them up with the functions that are not really
              Message 6 of 12 , Jul 8, 2006
              • 0 Attachment
                On Sat, 8 Jul 2006 at 3:10pm, Bram Moolenaar wrote:

                >
                > Hari Krishna Dara wrote:
                >
                > > When Funcref's were introduced in Vim7, I expected them to work for
                > > script-local functions, across scripts. The documentation didn't say
                > > that, but it didn't say that it wouldn't either, and I thought that that
                > > is one of its biggest uses (other than the actual intended
                > > functionality, which is for implementing numbered functions). However, I
                > > found that the Funcref references for such functions can't actually be
                > > passed out to other scripts. This reduces the usefulness of this feature
                > > as we can't register private functions to receive callbacks from other
                > > scripts.
                >
                > The whole purpose of script-local functions is that they will only be
                > used locally in the script. Thus they can't be called from outside the
                > script (without the trick to obtain the script nr). Also not when using
                > a Funcref.

                I often provide public functions that are useful for other plugins to
                interface, so I don't like mixing them up with the functions that are
                not really meant for getting called directly. Can the support for
                calling the script local functions from outside using SNR form be ever
                dropped? If not, I don't mind using this form to keep my plugin
                interface clean, as a workaround (I actually use this trick all the
                time, but always wondered if this is a feature or a bug, and in case it
                is a bug, if it ever will be fixed).

                --
                Thanks,
                Hari

                >
                > > What is weird is that the the Funcref() actually behaves exactly like
                > > the function name itself.
                >
                > Yes, that basically sums it up.
                >
                > > There are other aspects of the new features that are very
                > > counter-intuitive to me, whether I think in terms of Python or generic
                > > "objects" in any language. The one which gets me the most is the
                > > implicit typing of variables based on the initializer. For basic types
                > > prior to Vim7 (integer and string), you could easily switch the value of
                > > the variable from integer to string or vice versa, and the type() of the
                > > variable would change, suggesting that it behaves like "duck typing" (as
                > > per (wikipedia). But this observation can't be extended to the newer
                > > object types, as the below will fail:
                > >
                > > let a = {}
                > > let a = []
                > >
                > > If the type of value determines the type of the variable, and if we are
                > > merely dealing with references (assigning references instead of copying
                > > objects), then why should the second statement above generate the below
                > > error?
                > >
                > > E706: Variable type mismatch for: a
                > >
                > > Is there a standard for this type of language behavior? I didn't find
                > > anything at this page: http://en.wikipedia.org/wiki/Dynamically_typed
                >
                > I don't know if there is a standard and generally I don't care. There
                > is nothing against inventing something new. The main point is that it's
                > very easy to make mistakes with using the wrong type for a variable.
                > Using declarations is the normal way to have the compiler or interpreter
                > check the type. But declarations are annoying and certainly don't fit
                > with Vim 6.x script. Therefore I used the "sticky types" mechanism: the
                > first time a variable is assigned a value that's the type it has. Very
                > simple and efficient. You just need to remember that variables are not
                > dynamically typed. Except for int/string, for backwards compatibility.
                >
                >

                __________________________________________________
                Do You Yahoo!?
                Tired of spam? Yahoo! Mail has the best spam protection around
                http://mail.yahoo.com
              Your message has been successfully submitted and would be delivered to recipients shortly.