Follow synonyms in #show_module _type by gasche · Pull Request #11534 · ocaml/ocaml

gasche · 2022-09-03T19:32:36Z

This PR against 4.14 is intended as a fix for #11533.

At the time of writing, it contains ad-hoc logic in the implementation of #show_module_type to follow a chain of synonyms, mirroring similar logic in the implementation of #show_module to follow a chain of aliases.

Fixes: #11533

gasche · 2022-09-03T19:33:08Z

(Reviews are already welcome.)

gasche · 2022-09-03T20:10:47Z

I added some tests.

While playing with #show in presence of module aliases or module type synonyms I encountered some strange behaviors

# module U = Stdlib.Unit;;
module U = Unit
# #show U;;
module U = Unit
module U = Unit
module U :
  sig
    type t = unit = ()
    val equal : t -> t -> bool
    val compare : t -> t -> int
    val to_string : t -> string
  end

Note the repeated line module U = Unit in the #show U output, which is not part of the regression reported in #11533 -- but probably also an interaction with implicit strengthening during environment lookup. After the present PR, the same issue can now show up for module types as well.

This stuttering is independent of the present regression, I am not sure how to fix it, so I propose to leave it outside the scope of the present bugfix PR. (It is also a lesser problem, as it adds some useless noise in the output but does not hide the previously-useful output.)

gasche · 2022-09-06T12:45:29Z

This small PR is ripe for review. I you look at the code, it is useful to compare it with the code of #show_module above:

ocaml/toplevel/topdirs.ml

Lines 539 to 561 in adcf2cb

    
           let () = 
        
             reg_show_prim "show_module" 
        
               (fun env loc id lid -> 
        
                  let path, md = Env.lookup_module ~loc lid env in 
        
                  let id = match path with 
        
                    | Pident id -> id 
        
                    | _ -> id 
        
                  in 
        
                  let rec accum_aliases md acc = 
        
                    let acc rs = 
        
                      Sig_module (id, Mp_present, 
        
                                  {md with md_type = trim_signature md.md_type}, 
        
                                  rs, Exported) :: acc in 
        
                    match md.md_type with 
        
                    | Mty_alias path -> 
        
                        let md = Env.find_module path env in 
        
                        accum_aliases md (acc Trec_not) 
        
                    | Mty_ident _ | Mty_signature _ | Mty_functor _ -> 
        
                        List.rev (acc (is_rec_module id md)) 
        
                  in 
        
                  accum_aliases md [] 
        
               ) 
        
               "Print the signature of the corresponding module."

trefis · 2022-09-07T13:59:44Z

toplevel/topdirs.ml

+         | Some (Mty_ident path) ->
+             let mtd = Env.find_modtype path env in
+             accum_defs mtd acc
+         | None | Some (Mty_alias _ | Mty_signature _ | Mty_functor _) ->


I'd be tempted to extract the Some (Mty_alias _) case, to explicitly state that it can't happen.

I'm not sure it would really help.
This is just a choice we made of not having module aliases in the syntax of module types, but this code would be correct anyway.

xavierleroy · 2022-09-25T09:48:14Z

Ping? Who would be good reviewers for this PR?

garrigue

This looks basically fine to me.
If fixing stuttering is easy, I would prefer it to be done, otherwise I do not care so much.

garrigue · 2022-09-26T03:32:57Z

toplevel/topdirs.ml

+         | Some (Mty_ident path) ->
+             let mtd = Env.find_modtype path env in
+             accum_defs mtd acc
+         | None | Some (Mty_alias _ | Mty_signature _ | Mty_functor _) ->


I'm not sure it would really help.
This is just a choice we made of not having module aliases in the syntax of module types, but this code would be correct anyway.

testsuite/tests/tool-toplevel/show.ml

garrigue · 2022-09-26T04:32:35Z

toplevel/topdirs.ml

+         | _ -> id
+       in
+       let rec accum_defs mtd acc =
+         let acc = Sig_modtype (id, mtd, Exported) :: acc in


Couldn't you avoid stuttering by checking that you are not adding the same definition again?

Octachron · 2022-12-01T16:49:15Z

I propose to fix the stuttering at a later date so we can have the fix in 4.14 and 5.0 (this is a toplevel fix, which I am willing to test in the release candidate).

see ocaml#11533

gasche · 2022-12-01T21:51:41Z

Thanks for the ping! The increase in feature scope had somehow suspended this PR from my memory.

I integrated @garrigue's review comments, and I worked out a fix for the stuttering issue. (Guess what, it was caused by Stdlib__Unit. It's the second time in two weeks that this hidden prefixing adds some pain to my development life.)

gasche · 2022-12-01T21:52:13Z

This still needs an approval before we can merge.

Octachron

The code looks good to me (even if I would have fixed the stuttering in the reverse way).

gasche · 2022-12-02T10:34:27Z

By "the reverse way", do you mean that if the lookup chains finds first D1 and then D2, and in fact D2 double-underscore-simplifies into D1, you would rather keep D2 and throw away D1, rather than keep D1 and throw away D2 as I am doing? (That sounds slightly more difficult to implement for no obvious benefit.)

Octachron · 2022-12-02T10:43:34Z

As far as I can see, you are throwing away D1 and keeping D2 in your code since you only add a new item once you have confirmed that it would not be shadowed by a subsequent expansion:

if secretly_the_same_path env path new_path
then acc
else def :: acc

?

But I agree that it would be more difficult with dubious benefits.

gasche · 2022-12-02T12:36:36Z

We are talking about two successive aliases that would become module M = <path> and module M = <new_path> in the expansion trace. If path and new_path (which is an expansion of path) are the same modulo double-underscore-rewrite, then the def corresponding to module M = <new_path> is omitted from the result, but module M = <path> is already in the acc variable. (Or maybe I'm the one reading things wrong, but whatever, both would be fine anyway.)

…type Follow synonyms in #show_module _type (cherry picked from commit 85a0817)

a09392d Set Menhir version back to 20210419 again (ocaml#89) cc63992 Merge pull request ocaml#88 from mshinwell/flambda-backend-changes-2022-12-27 3e49df3 HACKING.jst.adoc 1866676 Merge flambda-backend changes e012992 Merge pull request ocaml#87 from mshinwell/merge-4.14.1 ac5c7c8 Merge tag '4.14.1' into main 3da21bc add a useful debug printer 83b7c72 Document the debug_printers script 98896e0 Remove a tiny code stutter I came across 99cb5d9 release 4.14.1 b49060f last commit before tagging 4.14.1 fae9aef Add documentation 708e5a9 Add tests c609eee Bootstrap 7f922d0 Polymorphic parameters 51aeb04 Keep generalized structure from patterns when typing let 4b68bb3 Add test of princiaplity from polymorphic type constraints 82c7afe fix wong raise aca252f x86: Force result of Icomp to be in a register (ocaml#11808) 985725b Add dynlink_compilerlibs.mli to .gitignore (ocaml#79) 2b1fa24 Regenerate parser (ocaml#80) 1bb6c79 Merge pull request ocaml#78 from mshinwell/flambda-backend-patches-2022-12-13 9029581 Update otherlibs/dynlink/Makefile 3e4f1b9 Revert toplevel/native/dune to ocaml-jst version 6061e4c Regenerate configure using autoconf 2.71 888d4b1 Back out patch which disables alloc-check in ocaml-jst a6d5796 Fix dynlink build 3e46daf Update .depend files a5c547e Bootstrap a6a9031 Merge flambda-backend changes 0ac7fdd temp fix for linker error (ocaml#77) 1018602 Remove references to 32-bit Cygwin (ocaml#11797) e2d0d9e Enable individual testing with Makefile.jst (ocaml#76) f10cbf6 increment version number after tagging 4.14.1~rc1 11c5ab7 release 4.14.1~rc1 e4c3920 last commit before tagging 4.14.1~rc1 9e598ca Merge pull request ocaml#11793 from dra27/then-than 2a7e501 Use a more relaxed mode for unification in Ctype.subst (ocaml#11771) (ocaml#73) 7b35ef7 Statically initialize `caml_global_data` with a valid value (ocaml#11788) cbd791a Allow immediates to cross modes (ocaml#58) 85a0817 Merge pull request ocaml#11534 from gasche/follow-synonyms-in-show-module-type 699f43c Changes e54e9bc fix the 'stuttering' issue in #show d9799d3 test comments fec3b23 follow synonyms when #show-ing module types 06a1ad7 regression tests for ocaml#11533 (still failing) 549d757 Run "misplaced attributes" check when compiling mlis (ocaml#72) b2b74bf Fix bug in `Mtype.strengthen_lazy` causing spurious typing errors (ocaml#11776) a6c0e75 Ensure that Ctype.nongen always calls remove_mode_variables (ocaml#70) 6c50831 array elements are global (ocaml#67) bc510ed Ensure that types from packed modules are always generalised (ocaml#11732) 4d47036 Fix ocaml#10768 8788ff6 Add/move some documentation 9891a36 Propagate location information to `local_` in expressions 988306d Add support for `global_` and `nonlocal_` constructor arguments (ocaml#50) 6729eb8 Missing CAMLparam in win32's Unix.stat (ocaml#11737) e7dd740 Add debug_printers.ml (ocaml#63) 65f2896 more entries in gitignore (ocaml#62) a9a84d0 Move `global_flag` to `Asttypes` (ocaml#60) fac5896 Minor attribute fixes from flambda-backend 75f402e Note about make install and Makefile.jst (ocaml#56) fb5b1e4 Remove the -force-tmc flag (ocaml#11661) bd87a61 ocamlmklib: use `ar rcs` instead of `ar rc` (ocaml#11670) 83762af Merge pull request ocaml#11622 from Octachron/fix_recursive_types_in_constructor_mismatch ca48730 Merge pull request ocaml#11609 from Octachron/pr11194_unbound_and_printing_context git-subtree-dir: ocaml git-subtree-split: a09392d

gasche force-pushed the follow-synonyms-in-show-module-type branch from a60a952 to 9469268 Compare September 3, 2022 20:04

trefis reviewed Sep 7, 2022

View reviewed changes

garrigue self-requested a review September 26, 2022 03:23

garrigue reviewed Sep 26, 2022

View reviewed changes

gasche added 3 commits December 1, 2022 22:16

regression tests for ocaml#11533 (still failing)

06a1ad7

follow synonyms when #show-ing module types

fec3b23

see ocaml#11533

test comments

d9799d3

gasche force-pushed the follow-synonyms-in-show-module-type branch from f371330 to 6453ea3 Compare December 1, 2022 21:46

gasche added 2 commits December 1, 2022 22:49

fix the 'stuttering' issue in #show

e54e9bc

Changes

699f43c

gasche force-pushed the follow-synonyms-in-show-module-type branch from 6453ea3 to 699f43c Compare December 1, 2022 21:49

Octachron approved these changes Dec 2, 2022

View reviewed changes

Octachron merged commit 85a0817 into ocaml:4.14 Dec 2, 2022

Octachron added a commit that referenced this pull request Dec 2, 2022

Merge pull request #11534 from gasche/follow-synonyms-in-show-module-…

6b4bbfe

…type Follow synonyms in #show_module _type (cherry picked from commit 85a0817)

Octachron added a commit that referenced this pull request Dec 2, 2022

Merge pull request #11534 from gasche/follow-synonyms-in-show-module-…

df142a3

…type Follow synonyms in #show_module _type (cherry picked from commit 85a0817)

Octachron mentioned this pull request Dec 6, 2022

OCaml 5.0.0, first release candidate ocaml/opam-repository#22615

Merged

Conversation

gasche commented Sep 3, 2022 • edited by xavierleroy Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

gasche commented Sep 3, 2022

Uh oh!

gasche commented Sep 3, 2022

Uh oh!

gasche commented Sep 6, 2022

Uh oh!

trefis Sep 7, 2022

Choose a reason for hiding this comment

Uh oh!

garrigue Sep 26, 2022

Choose a reason for hiding this comment

Uh oh!

xavierleroy commented Sep 25, 2022

Uh oh!

garrigue left a comment

Choose a reason for hiding this comment

Uh oh!

garrigue Sep 26, 2022

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

garrigue Sep 26, 2022

Choose a reason for hiding this comment

Uh oh!

Octachron commented Dec 1, 2022

Uh oh!

gasche commented Dec 1, 2022

Uh oh!

gasche commented Dec 1, 2022

Uh oh!

Octachron left a comment

Choose a reason for hiding this comment

Uh oh!

gasche commented Dec 2, 2022

Uh oh!

Octachron commented Dec 2, 2022

Uh oh!

gasche commented Dec 2, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

gasche commented Sep 3, 2022 •

edited by xavierleroy

Loading