Prefix the stdlib · Pull Request #1010 · ocaml/ocaml

ghost · 2017-01-13T16:26:08Z

This patch changes the compilation scheme of the stdlib so that every module uses a long compilation unit name that:

is much less likely to clash with modules from other projects
doesn't systematically pollute the environment with many short names

Motivations

1. we can add new modules to the stdlib without guilt

With this patch, we can add new modules to the stdlib without worrying about clashing with a module from another project. #964 and #1002 will benefit immediately from this.

2. it's easier to use an alternative stdlib

Currently, when using an alternative standard library such as Base, you still get all the default modules and values from the stdlib in scope. We do terrible things in Base to at least deprecate them but it is not very satisfactory.

With this patch, all we need is an option to prevent the compiler from opening Stdlib (formely Pervasives) by default and we will get a clean initial environment. We'll still get the long names, but it is not as bad.

User visible changes

module X from the stdlib can be referred as Stdlib.X
element x from Pervasives can be referred as Stdlib.x
names of exceptions defined in the stdlib change in backtraces. For instance Pervasives.Exit becomes Stdlib__pervasives.Exit

Patch details

This patch uses the same technique we use for Jane Street libraries. The -o option is used to change the compilation unit name. For instance list.ml compiles to stdlib__list.cmo and we use module aliases to get back the short names.

The aliases from long names to short names live in the new Stdlib module added by this pull request. This module is now the one initially opened instead of Pervasives, so that we get all the short names in scope as before. This change requires a bootstrap.

stdlib/stdlib.ml looks like this:

module Arg = Arg
module Array = Array
...
module Pervasives = Pervasives
...

include Pervasives

The build system rewrites module FooBar = FooBar to module FooBar = Stdlib__fooBar and compiles the resulting file with -no-alias-dep -w -49.

Type printing

To avoid printing paths of the form Stdlib__map in error messages, this patch adds a simple heuristic to the type printer: given a path of the form <x>__<y>, if <x>.<y> is an alias for <x>__<y> then use <x>.<y> instead.

Contrary to the implementation -short-paths, this is simple and fast so it is enabled by default.

c-cube · 2017-01-13T16:35:25Z

Note: #640 would also benefit (a IO module would be awesome)

Drup · 2017-01-13T18:23:32Z

typing/printtyp.ml

  | Pident id ->
      Oide_ident (ident_name id)
-  | Pdot(Pident id, s, _pos) when Ident.same id ident_pervasive ->
+  | Pdot(Pident id, s, _pos) when is_pervasive_or_stdlib_ident id ->


This doesn't catch Stdlib.Pervasives.foo. Is that a problem ?

Indeed, I updated the code

Drup · 2017-01-13T18:32:09Z

I like this very much.

The fact that type errors can refer to Stdlib__* is, imho, quite a big problem. These modules shouldn't even really exist, we only need them because of limitations of module aliases (or pack). Also,-short-path is not enabled by default.

v-gb · 2017-01-14T16:03:11Z

A tiny benefit of this change is that it allows the code generated by ppx rewriters to refer to the stdlib (say List.map) with less risk of shadowing, by calling Stdlib.List.map.
See https://github.com/whitequark/ppx_deriving/blob/master/src/ppx_deriving_runtime.ml for an example of not being able to do that now.

This is a bit non-backwards compability, because module type of Queue is a signature with an abtract type before this feature, but a concrete type type 'a t = 'a Queue.t after this feature.
Using module type of this way is bad idea because of how fragile it is, but some people likely do that.

hcarty · 2017-01-14T16:27:25Z

Regarding the ppx_deriving improvements, this would also help with error messages when using something like ppx_deriving. The modules that are used to avoid shadowing can lead to some difficult to decipher error messages, particularly when not using short types.

ghost · 2017-01-16T12:04:51Z

I added a trivial heuristic to avoid printing the Stdlib__ names in error messages and in the toplevel. Now the only place they can appear is in backtraces. I also simplified the build to avoid the need for the Stdlib__ module.

Chris00 · 2017-01-16T12:23:25Z

Can it be a general rule? I.e. when X__Y and (an alias) X.Y refer to the same module, then favor X.Y in paths. This would be useful for user libraries and would encourage uniform naming of "inner" libraries.

ghost · 2017-01-16T12:55:15Z

Yes, that's how I implemented it. I updated the description of this PR

bobzhang · 2017-01-16T13:36:01Z

Can the heuristic be made general to single underscore?

ghost · 2017-01-16T14:36:49Z

I don't think it's desirable to do that for every underscore, what is your use-case?

Drup · 2017-01-16T21:38:46Z

@diml I like your solution to the alias problem and it's consistent with the behavior of short-path. Not sure if it's really documented though. The implementation looks correct, but it's unfortunate that it's completely distinct to the one for short-path.

Drup · 2017-01-16T21:39:29Z

typing/printtyp.ml

+    else
+      loop (i + 1)
+  in
+  loop 0


There is pretty much the same code in Printyp.penalty.

Indeed, I factorized the code

ghost · 2017-01-17T10:36:35Z

I updated the example in the manual using module aliases as a replacement for -pack to use double underscores and documented the heuristic.

Octachron · 2017-01-17T14:19:47Z

I wonder if it would be worthwhile to generalize the heuristic to multiple level of nested sublibraries, i.e. try to turn Lib__Sublib1__…M into Lib.Sublib1….M?

ghost · 2017-01-18T12:21:32Z

I'm not sure it's worthwhile as I've never seen such a hierarchy in an OCaml project. In any case since it's not really in the scope of this PR, I'd rather keep the simple heuristic here and we can discuss more complex ones latter

Octachron · 2017-01-18T21:01:45Z

I'm not sure it's worthwhile as I've never seen such a hierarchy in an OCaml project.

The fact that such a hierarchy is quite hard to set up currently may well be the cause of this inexistence.
But I agree that such change is utterly not necessary in the scope of this PR.

alainfrisch · 2017-01-23T13:01:26Z

Changing the name of existing files in the stdlib (such as list.cmi) might break existing projects that rely on such filenames (for instance to build an explicit set of allowed interfaces for compiling addins). A less invasive approach would be to rely on aliases only for new modules, with aliases such as module Float = Stdlib_float added to Pervasives (which is already opened). This could also be used as a first step to check that all build systems and external tools out there work ok with the use of module aliases in the standard library.

ghost · 2017-01-23T14:26:44Z

It seems cleaner to me to do the same for every module. Fixing such projects would be trivial

dra27 · 2017-01-23T14:44:52Z

Trivial, but experience in previous versions seems to suggest we should be avoiding barriers to upgrading as far as possible. A halfway-house where the new modules/values are always in Stdlib and old ones are only done this way if a configure-time option is given would at least allow a period of time (1 or 2 releases) to review the impact? We had a regression on missing files (which is a related thing to renamed files!) in 4.03.0.

I must confess that the need to alter so many test cases makes me lean towards wondering if the last two drawbacks - especially names in backtraces - should be addressed at once? An uncaught exception error message is already not great, and it starts to look even worse with Stdlib__ in it. Fragile or not, a brief look on GitHub suggests module type of SomeModuleFromStdLib is certainly used and removing that is not trivial if you've got to support multiple versions of OCaml. It feels like we should be ensuring that noone ever has to write Stdlib__ in anything.

I have a non-critical nitpick about the way stdlib__.ml is generated - it's a massive piece of code duplication. Why not make $(P) a sentinel value (i.e. illegal in both module + filename) and use STDLIB_MODULES to generate the entire file?

In particular, for Lacaml__M modules that are aliased to submodules Lacaml.M, the latter path should eventually be used by the type printer. See ocaml/ocaml#1010 Fixes https://github.com/mmottl/lacaml

In particular, for Lacaml__M modules that are aliased to submodules Lacaml.M, the latter path should eventually be used by the type printer. See ocaml/ocaml#1010 Fixes mmottl#25

ghost · 2017-01-24T10:20:22Z

I'll run some tests on opam so that we can get a clear idea of the breakage.

Regarding exception names, what we need is to do is change the module name used in extension constructor names. I can think of two options:

add a command line argument
add the same heuristic as for type printing: if in the initial environment (after interpreting the -open arguments) Foo.Bar is an alias for Foo__bar then use the former instead of the latter

The second option seems reasonable to me and in-line with the heuristic for type printing.

For the module type of problem, it seems to me that it would best if aliases were resolved before abstracting the signature. At least at Jane Street we have been bitten several times by the fact tbat adding an alias somewhere breaks a use of module type of somewhere else. @garrigue, what do you think?

About the last point, I agree that the duplication is unfortunate. However i think it should be the other way around: stdlib.ml is the source file that people should look at instead of the build system

dra27 · 2017-01-25T09:03:10Z

I agree that using the Foo__bar heuristic for exceptions seems a good (and natural) plan. Obviously doing alias resolution earlier deals with the problem completely, but if we don't end up doing that, is there the possibility of a warning along the lines of @lpw25's (caml-devel) suggestion of (module type of struct include Queue end)? Even if there's no possibility of a warning, that pattern feels better for the test cases to me than having Stdlib__ prefixes ended up in code. If in the future we come up with an even better way of doing this, it would be nice to be able to be able to say that no one should have ever used the prefixes anywhere at any time!

That StdlibModules file is used elsewhere - there'd also need to be a solution for the Camlinternal modules which aren't mentioned in stdlib.ml. I also don't particularly like either the unnecessary double-typing of each module name module List = List or the magic transformation of the second half (but I realise that it means that the file is always syntactically valid). It would also feel a little odd to be having the build system extract information automatically from stdlib.ml, at least to me.

ghost · 2017-01-25T09:57:57Z

Maybe it's possible to implement the warning, although it would be simpler to just resolve aliases. I need to try again the module type of struct include ... in the testsuite.

Regarding StdlibModules, AFAICT it only used to expunge the toplevel. The duplication between stdlib/Makefile and stdlib/StdlibModules can certainly be avoided and I was planning to look at it after as it's not specific to this PR. Technically in stdlib/Makefile we could just use $(wildcard *.ml) to list the standard library modules. The only problem is the ordering for linking.

To me the current stdlib.ml file states the interface of the standard library, so it's valuable to have it as it is now. The magic transformation to add the Stdlib__ prefixes is only there so that this stays a detail of the build system rather than something the user should know about.

To build JS libraries we do things slightly differently: the special alias file is named lib__.ml-gen and is indeed generated by the build system. lib.ml is just a normal module, expect that it is not prefixed. In the case of the stdlib, since Pervasives has no dependency on other modules of the standard library, we can use stdlib.ml direcrtly as the alias module. I though it was better this way as it avoids the need for a Stdlib__ module.

lpw25 · 2017-01-25T10:04:09Z

For the module type of problem, it seems to me that it would best if aliases were resolved before abstracting the signature.

I'm not keen on this idea. module type of already does some ad-hoc resolving of aliases, and it is a huge PITA (I plan to do a PR to remove it soon). Adding more ad-hoc treatment will likely make the problem worse.

The issue here is not with module aliases, or how you are using them in this PR, but with module type of and the fact that some people use it in cases where they really shouldn't. If module type of had had the strengthening semantics from the beginning then none of this would be a problem.

My point about module type of struct include M end is that you should only use module type of in situations where that extended form would still type-check. In those cases the change in this PR will not break anything.

ghost · 2018-02-14T14:45:13Z

If you get a lot of Reference to undefined ... when running the tests, try cleaning testsuite/tools. It should fix the issue

hhugo · 2018-02-17T15:45:16Z

ocamlfind needs to be patched. I've opened https://gitlab.camlcity.org/gerd/lib-findlib/merge_requests/10

ghost · 2018-02-19T09:53:03Z

Thanks!

gasche · 2018-02-20T09:32:07Z

On my machine, building a fresh trunk ( 2b25681 ) then running

cd testsuite
make all

fails with the following tests failing:

List of failed tests:
    tests/typing-poly/'poly.ml' with 1 (expect)
    tests/typing-modules/'aliases.ml' with 1 (expect)
    tests/typing-gadts/'pr5689.ml' with 1 (expect)
    tests/messages/'precise_locations.ml' with 1 (expect)
    tests/typing-objects/'Tests.ml' with 1 (expect)
    tests/typing-gadts/'omega07.ml' with 1 (expect)
    tests/typing-gadts/'dynamic_frisch.ml' with 1 (expect)
    tests/typing-objects/'Exemples.ml' with 1 (expect)
    tests/tool-ocamldoc-html/Module_whitespace.ml
    tests/typing-misc/'pr7668_bad.ml' with 1 (expect)

I believe that this is caused by this PR, given that the proposed expect-diffs look as follows:

 [%%expect{|
-val partition_map :
-  ('a -> [< `Left of 'b | `Right of 'c ]) -> 'a list -> 'b list * 'c list =
-  <fun>
-Line _, characters 35-96:
-  ...................................partition_map (fun x -> if x then `Left ()
-  else `Right ()) xs
-Error: This expression has type unit list * unit list
-       but an expression was expected of type int list * int list
-       Type unit is not compatible with type int
+Line _:
+Error: Reference to undefined global `List'
 |}]

shindere · 2018-02-20T09:38:43Z

Ow, right now I have absolutely no idea why I didn't see this error neither on my local machine, nor on CI. @gasche: could it be that, because of a lack of dependencies, say, your expect tool has not been rebuilt correctly?

gasche · 2018-02-20T09:49:09Z

I can reproduce the failures from a fresh git clone on my machine, so I don't think it is a dependency issue. Can someone else reproduce the failure from a fresh git clone?

shindere · 2018-02-20T10:10:58Z

@gasche just trying, will report here ASAP

lpw25 · 2018-02-20T10:14:18Z

@gasche I had those errors the other day, and they were hard to get rid of. Eventually, I git cleaned and make distcleaned enough to fix the problem, but I don't know exactly what it was that was causing the issues.

One theory I had was that it was the install directory. Have you tried deleting the directory that you have configured it to install in (or configuring it to install in a different directory)?

gasche · 2018-02-20T10:19:58Z

Tests must work without any installation action, and independently of the install directory. In particular, tests must work without running"make install". If this property is broken, then it is a bug that must be fixed. The bug may have been introduced somewhere outside this pull-request, however.

gasche · 2018-02-20T10:25:33Z

I just tried and make install fixes the issue. The problem must come from test scripts relying on the installed environment, so it is probably independent of the present PR.

shindere · 2018-02-20T10:32:36Z

Gabriel Scherer (2018/02/20 02:25 -0800):

I just tried and `make install` fixes the issue. The problem must come from test scripts relying on the installed environment, so it is probably independent of the present PR.

Hmm that gets interesting! Thanks for having found htis out! So normally, ocamltest tries very hard to not use anything installed. Could it be that expect_test itself when testing programs uses an installed version of the stdlib rather than the one that comes with the sources?

shindere · 2018-02-20T10:33:51Z

Gabriel Scherer (2018/02/20 10:20 +0000):

Tests must work without any installation action, and independently of the install directory. In particular, tests must work without running"make install". If this property is broken, then it is a bug that must be fixed. The bug may have been introduced somewhere outside this pull-request, however.

Well accordiing to GNU build system recommendaitons, what you write is semi-true or, say, not complete. It is true that one should be able to test without installing. But, is may also be desirable to run the tests not only from the just compiled sources but also against a freshly installed version. This is for instance provided by automake. And by the way: couldn't reproduce the problem with a fresh clone but didn't try to install anything. This is where it would be handy to be able to build / test iside a container.

ghost · 2018-02-20T10:42:15Z

Could it be that expect_test itself when testing programs uses an
installed version of the stdlib rather than the one that comes with the
sources?

I checked, and it's only calling Toploop.initialize_toplevel_env. In particular it is not calling Compmisc.init_path, so it the installed stdlib directory should in theory not be added to the search path.

gasche · 2018-02-20T10:57:05Z

I can reliably reproduce the issue by building and installing an old trunk, then building the current trunk and running the tests from there. I don't know which change is responsible for this breakage, but in any case it should be fixed. (Given that it may be irrelevant to the present PR, I may create a devel- thread later in the day.)

shindere · 2018-02-20T10:57:07Z

@gasche now trying with a fresh clone but also as a different user that has
no opam installed, on a system where OCaml is not installed.

ghost · 2018-02-20T13:38:27Z

I wrote an expect test that prints Config.load_path and it indeed contains the installed stdlib directory

ghost · 2018-02-20T13:48:13Z

This should be fixed after this: #1621

--------------- (Changes that can break existing programs are marked with a "*") ### Code generation and optimizations: - GPR#2278: Remove native code generation support for 32-bit Intel macOS, iOS and other Darwin targets. (Mark Shinwell, review by Nicolas Ojeda Bar and Xavier Leroy) ### Compiler user-interface and warnings: * GPR#2276: Remove support for compiler plugins and hooks (also adds [Dynlink.unsafe_get_global_value]) (Mark Shinwell, Xavier Clerc, review by Nicolás Ojeda Bär, Florian Angeletti, David Allsopp and Xavier Leroy) ### Internal/compiler-libs changes: - GPR#1579: Add a separate types for clambda primitives (Pierre Chambart, review by Vincent Laviron and Mark Shinwell) - GPR#1965: remove loop constructors in Cmm and Mach (Vincent Laviron) - GPR#1973: fix compilation of catches with multiple handlers (Vincent Laviron) - GPR#2228: refactoring the handling of .cmi files by moving it from Env to a new module Persistent_env (Gabriel Scherer, review by Jérémie Dimino and Thomas Refis) - GPR#2229: Env: remove prefix_idents cache (Thomas Refis, review by Frédéric Bour and Gabriel Scherer) - GPR#2237: Reorder linearisation of Trywith to avoid a call instruction (Vincent Laviron and Greta Yorsh, additional review by Mark Shinwell) - GPR#2265: Add bytecomp/opcodes.mli (Mark Shinwell, review by Nicolas Ojeda Bar) - GPR#2280: Don't make more Clambda constants after starting Cmmgen (Mark Shinwell, review by Vincent Laviron) - GPR#2283: Add [is_prefix] and [find_and_chop_longest_common_prefix] to [Misc.Stdlib.List] (Mark Shinwell, review by Alain Frisch and Stephen Dolan) - GPR#2284: Add various utility functions to [Misc] and remove functions from [Misc.Stdlib.Option] that are now in [Stdlib.Option] (Mark Shinwell, review by Thomas Refis) - GPR#2286: Functorise [Consistbl] (Mark Shinwell, review by Gabriel Radanne) - GPR#2308: More debugging information on [Cmm] terms (Mark Shinwell, review by Stephen Dolan) - GPR#7878: Replaced TypedtreeIter with tast_iterator (Isaac "Izzy" Avram) ### Runtime system: - GPR#1725: Deprecate Obj.set_tag (Stephen Dolan, review by Gabriel Scherer and Damien Doligez) * GPR#2240: Constify "identifier" in struct custom_operations (Cedric Cellier, review by Xavier Leroy) ### Standard library: - GPR#2262: take precision (.<n>) and flags ('+' and ' ') into account in printf %F (Pierre Roux, review by Gabriel Scherer) ### Other libraries: - MPR#7903, GPR#2306: Make Thread.delay interruptible by signals again (Xavier Leroy, review by Jacques-Henri Jourdan and Edwin Török) - GPR#2248: Unix alloc_sockaddr: Fix read of uninitialized memory for an unbound Unix socket. Add support for receiving abstract (Linux) socket paths. (Tim Cuthbertson, review by Sébastien Hinderer and Jérémie Dimino) - GPR#2289: Delete the vmthreads library. This library was deprecated in 4.08.0. (Jérémie Dimino) ### Compiler user-interface and warnings: - GPR#2301: Hint on type error on int literal (Jules Aguillon, review by Nicolás Ojeda Bär , Florian Angeletti, Gabriel Scherer and Armaël Guéneau) - GPR#2307: Hint on type error on int's operators (Jules Aguillon, with help from Armaël Guéneau, review by Gabriel Scherer and Florian Angeletti) ### Bug fixes: - MPR#7937, GPR#2287: fix uncaught Unify exception when looking for type declaration (Florian Angeletti, review by Jacques Garrigue) OCaml 4.08.0 ------------ ### Language features: - MPR#6422, MPR#7083, GPR#305, GPR#1568: Allow "exception" under or-patterns (Thomas Refis, with help and review from Alain Frisch, Gabriel Scherer, Jeremy Yallop, Leo White and Luc Maranget) - GPR#1506, GPR#2147, GPR#2166, GPR#2167: Extended open to arbitrary module expression in structures and to applicative paths in signatures (Runhang Li, review by Alain Frisch, Florian Angeletti, Jeremy Yallop, Leo White and Thomas Refis) - GPR#1705: Allow @@attributes on exceptions. (Hugo Heuzard, review by Gabriel Radanne and Thomas Refis) - GPR#1804: New notion of "alerts" that generalizes the deprecated warning (Alain Frisch, review by Leo White and Damien Doligez) - GPR#1892: Allow shadowing of items coming from an include (Thomas Refis, review by Gabriel Radanne) - GPR#1947: Add "monadic" let operators (Leo White, review by Thomas Refis) * GPR#2106: .~ is now a reserved keyword, and is no longer available for use in extended indexing operators (Jeremy Yallop, review by Gabriel Scherer, Florian Angeletti, and Damien Doligez) - GPR#2110: Allow GADT constructors to introduce equations and existential types under or-patterns. (Thomas Refis and Leo White, review by Jacques Garrigue) - GPR#2122: Introduce local substitutions in signatures: "type t := type_expr" and "module M := Extended(Module).Path" (Thomas Refis, with help and review from Leo White, and Alain Frisch) ### Type system: - GPR#1826: allow expanding a type to a private abbreviation instead of abstracting when removing references to an identifier. (Thomas Refis and Leo White, review by Jacques Garrigue) - GPR#1909: unsharing pattern types (Thomas Refis, with help from Leo White, review by Jacques Garrigue) - GPR#1942, GPR#2244: simplification of the static check for recursive definitions (Alban Reynaud and Gabriel Scherer, review by Jeremy Yallop, Armaël Guéneau and Damien Doligez) * MPR#7814, GPR#2041, GPR#2235: allow modules from include directories to shadow other ones, even in the toplevel (Jérémie Dimino, review by Alain Frisch and David Allsopp) ### Standard library: - GPR#2128: Add Fun module. (Many fine eyes) - GPR#2010: Add Bool module. (Many fine eyes) - GPR#2011: Add Int module. (Many fine eyes) - GPR#1940: Add Option module and Format.pp_print_option (Many fine eyes) - GPR#1956: Add Result module and Format.pp_print_result (Many fine eyes) - MPR#2533, GPR#1839, GPR#1949: added Unix.fsync (Francois Berenger, Nicolás Ojeda Bär, review by Daniel Bünzli, David Allsopp and ygrek) - MPR#5072, MPR#6655, GPR#1876: add aliases in Stdlib for built-in types and exceptions. (Jeremy Yallop, reports by Pierre Letouzey and David Sheets, review by Valentin Gatien-Baron, Gabriel Scherer and Alain Frisch) - MPR#6701, GPR#1185, GPR#1803: make float_of_string and string_of_float locale-independent. (ygrek, review by Xavier Leroy and Damien Doligez) - GPR#1590: ocamllex-generated lexers can be instructed not to update their lex_curr_p/lex_start_p fields, resulting in a significant performance gain when those fields are not required. (Alain Frisch, Jérémie Dimino) - MPR#7795, GPR#1782: Fix off-by-one error in Weak.create. (KC Sivaramakrishnan) - GPR#1731: Format, use raise_notrace to preserve backtraces. (Frédéric Bour, report by Jules Villard, review by Gabriel Scherer) - GPR#1792, MPR#7794: Add Unix.open_process_args{,_in,_out,_full} similar to Unix.open_process{,_in,_out,_full}, but passing an explicit argv array. (Nicolás Ojeda Bär, review by Jérémie Dimino, request by Volker Diels-Grabsch) - GPR#1999: Add Unix.process{,_in,_out,_full}_pid to retrieve opened process's pid. (Romain Beauxis, review by Nicolás Ojeda Bär) - GPR#2220: Remove duplicate process management code in otherlibs/threads/unix.ml (Romain Beauxis, review by Gabriel Scherer and Alain Frisch) - GPR#1182: Add new Printf formats %#d %#Ld %#ld %#nd (idem for %i and %u) for alternative integer formatting. (ygrek, review by Gabriel Scherer) - MPR#7235: Format, flush err_formatter at exit. (Pierre Weis, request by Jun Furuse) - GPR#1857, MPR#7812: Remove Sort module, deprecated since 2000 and emitting a deprecation warning since 4.02. (whitequark) - GPR#1923: Arg module sometimes misbehaved instead of rejecting invalid -keyword=arg inputs (Valentin Gatien-Baron, review by Gabriel Scherer) - GPR#1957: Add Stack.{top_opt,pop_opt} and Queue.{peek_opt,take_opt}. (Vladimir Keleshev, review by Nicolás Ojeda Bär and Gabriel Scherer) - GPR#1959: Add Format.dprintf, a printing function which outputs a closure usable with %t. (Gabriel Radanne, request by Armaël Guéneau, review by Florian Angeletti and Gabriel Scherer) - GPR#1959: Small simplification and optimization to Format.ifprintf (Gabriel Radanne, review by Gabriel Scherer) - GPR#1855, GPR#2118: Add `Fun.protect ~finally` for enforcing local invariants whether a function raises or not, similar to `unwind-protect` in Lisp and `FINALLY` in Modula-2. It is careful about preserving backtraces and treating exceptions in finally as errors. (Marcello Seri and Guillaume Munch-Maccagnoni, review by Daniel Bünzli, Gabriel Scherer, François Bobot, Nicolás Ojeda Bär, Xavier Clerc, Boris Yakobowski, Damien Doligez, and Xavier Leroy) - GPR#1986, MPR#6450: Add Set.disjoint (Nicolás Ojeda Bär, review by Gabriel Scherer) * GPR#1605: Deprecate Stdlib.Pervasives. Following #1010, Pervasives is no longer needed and Stdlib should be used instead. (Jérémie Dimino, review by Nicolás Ojeda Bär) - GPR#2004: Use common standard library path `lib/ocaml` for Windows, for consistency with OSX & Linux. Previously was located at `lib`. (Bryan Phelps, Jordan Walke, review by David Allsopp) - GPR#1966: Add Format semantic tags using extensible sum types. (Gabriel Radanne, review by Nicolás Ojeda Bär) - GPR#1794: Add constants zero, one, minus_one and functions succ, pred, is_finite, is_infinite, is_nan, is_integer, trunc, round, next_after, sign_bit, min, max, min_max, min_num, max_num, min_max_num to module Float. (Christophe Troestler, review by Alain Frish, Xavier Clerc and Daniel Bünzli) - GPR#1354, GPR#2177: Add fma support to Float module. (Laurent Thévenoux, review by Alain Frisch, Jacques-Henri Jourdan, Xavier Leroy) - GPR#2119: clarify the documentation of Set.diff (Gabriel Scherer, suggestion by John Skaller) - MPR#7812, GPR#2125: Add Filename.chop_suffix_opt (Alain Frisch, review by Nicolás Ojeda Bär, suggestion by whitequark) - GPR#1864: Extend Bytes and Buffer with functions to read/write binary representations of numbers (Alain Frisch and Daniel Bünzli) - GPR#1458: Add unsigned operations unsigned_div, unsigned_rem, unsigned_compare and unsigned_to_int to modules Int32, Int64, Nativeint. (Nicolás Ojeda Bär, review by Daniel Bünzli, Alain Frisch and Max Mouratov) - GPR#2145: Deprecate the mutability of Gc.control record fields (Damien Doligez, review by Alain Frisch) - GPR#2159, MPR#7874: annotate {String,Bytes}.equal as being [@@noalloc]. (Pierre-Marie Pédrot, review by Nicolás Ojeda Bär) - GPR#1936: Add module Float.Array (Damien Doligez, review by Xavier Clerc and Alain Frisch) - GPR#2002: Add Format.pp_print_custom_break, a new more general kind of break hint that can emit non-whitespace characters. (Vladimir Keleshev and Pierre Weis, review by Josh Berdine, Gabriel Radanne) - GPR#2183: Fix segfault in Array.create_float with -no-flat-float-array (Damien Doligez, review by Gabriel Scherer and Jeremy Yallop) - GPR#2185: Add `List.filter_map` (Thomas Refis, review by Alain Frisch and Gabriel Scherer) - GPR#1525: Make function set_max_indent respect documentation (Pierre Weis, Richard Bonichon, review by Florian Angeletti) ### Other libraries: - GPR#2222: Set default status in waitpid when pid is zero. Otherwise, status value is undefined. (Romain Beauxis and Xavier Leroy, review by Stephen Dolan) * GPR#2104, GPR#2211, PR#4127, PR#7709: Fix Thread.sigmask. When system threads are loaded, Unix.sigprocmask is now an alias for Thread.sigmask. This changes the behavior at least on MacOS, where Unix.sigprocmask used to change the masks of all threads. (Jacques-Henri Jourdan, review by Jérémie Dimino) - GPR#1061: Add ?follow parameter to Unix.link. This allows hardlinking symlinks. (Christopher Zimmermann, review by Xavier Leroy, Damien Doligez, David Allsopp, David Sheets) - GPR#2038: Deprecate vm threads (Jérémie Dimino) * PR#4208, PR#4229, PR#4839, PR#6462, PR#6957, PR#6950, GPR#1063, GPR#2176, GPR#2297: Make (nat)dynlink sound. (Mark Shinwell, Leo White, Nicolás Ojeda Bär, Pierre Chambart) - GPR#2263: Delete the deprecated Bigarray.*.map_file functions in favour of `*_of_genarray (Unix.map_file ...)` functions instead. The `Unix.map_file` function was introduced in OCaml 4.06.0 onwards. (Jérémie Dimino, reviewed by David Allsopp and Anil Madhavapeddy) ### Compiler user-interface and warnings: - GPR#2133: Warn on literal patterns found anywhere in a constructor's arguments. (Jeremy Yallop, review by Gabriel Scherer) - PR#2096: Add source highlighting for errors & warnings in batch mode (Armaël Guéneau, review by Gabriel Scherer and Jérémie Dimino) - PR#2091: Add a warning triggered by type declarations "type t = ()" (Armaël Guéneau, report by linse, review by Florian Angeletti and Gabriel Scherer) - PR#6416, GPR#1120: unique printed names for identifiers (Florian Angeletti, review by Jacques Garrigue) - MPR#7116, GPR#1430: new -config-var option to get the value of a single configuration variable in scripts. (Gabriel Scherer, review by Sébastien Hinderer and David Allsopp, request by Adrien Nader) - GPR#1691: add shared_libraries to ocamlc -config exporting SUPPORTS_SHARED_LIBRARIES from Makefile.config. (David Allsopp, review by Gabriel Scherer and Mark Shinwell) - GPR#1720: Improve error reporting for missing 'rec' in let-bindings. (Arthur Charguéraud and Armaël Guéneau, with help and advice from Gabriel Scherer, Frédéric Bour, Xavier Clerc and Leo White) - GPR#1733,1993,1998,2058,2094,2140: Typing error message improvements - GPR#1733, change the perspective of the unexpected existential error message. - GPR#1993, expanded error messages for universal quantification failure - GPR#1998, more context for unbound type parameter error - GPR#2058, full explanation for unsafe cycles in recursive module definitions (suggestion by Ivan Gotovchits) - GPR#2094, rewording for "constructor has no type" error - MPR#7565, GPR#2140, more context for universal variable escape in method type (Florian Angeletti, reviews by Jacques Garrique, Armaël Guéneau, Gabriel Radanne, Gabriel Scherer and Jeremy Yallop) - GPR#1748: do not error when instantiating polymorphic fields in patterns. (Thomas Refis, review by Gabriel Scherer) - MPR#6913, GPR#1786: new -match-context-rows option to control the degree of optimization in the pattern matching compiler. (Dwight Guth, review by Gabriel Scherer and Luc Maranget) - GPR#1822: keep attributes attached to pattern variables from being discarded. (Nicolás Ojeda Bär, review by Thomas Refis) - GPR#1845: new `-dcamlprimc` option to keep the generated C file containing the information about primitives; pass `-fdebug-prefix-map` to the C compiler when supported, for reproducible builds (Xavier Clerc, review by Jérémie Dimino) - GPR#1856, GPR#1869: use `BUILD_PATH_PREFIX_MAP` when compiling primitives in order to make builds reproducible if code contains uses of `__FILE__` or `__LOC__` (Xavier Clerc, review by Gabriel Scherer and Sébastien Hinderer) - GPR#1906: created warning 64 when using -unsafe and a -pp argument that returns a syntax tree, to replace the print that was there already (Valentin Gatien-Baron) - GPR#1913: new flag -dump-into-file to print debug output like -dlambda into a file named after the file being built, instead of on stderr. (Valentin Gatien-Baron, review by Thomas Refis) - GPR#1921: in the compilation context passed to ppx extensions, add more configuration options related to type-checking: -rectypes, -principal, -alias-deps, -unboxed-types, -unsafe-string (Gabriel Scherer, review by Gabriel Radanne, Xavier Clerc and Frédéric Bour) - GPR#1925: Print error locations more consistently between batch mode, toplevel and expect tests (Armaël Guéneau, review by Thomas Refis, Gabriel Scherer and François Bobot) - GPR#1930: pass the elements from `BUILD_PATH_PREFIX_MAP` to the assembler (Xavier Clerc, review by Gabriel Scherer, Sébastien Hinderer, and Xavier Leroy) - GPR#1945, GPR#2032: new "-stop-after [parsing|typing]" option to stop compilation after the parsing or typing pass (Gabriel Scherer, review by Jérémie Dimino) - GRP#1953: Add locations to attributes in the parsetree. (Hugo Heuzard, review by Gabriel Radanne) - GPR#1954: Add locations to toplevel directives. (Hugo Heuzard, review by Gabriel Radanne) - GPR#1976: Better error messages for extension constructor type mismatches (Thomas Refis, review by Gabriel Scherer) * GPR#1979: Remove support for TERM=norepeat when displaying errors (Armaël Guéneau, review by Gabriel Scherer and Florian Angeletti) - GPR#1960: The parser keeps previous location when relocating ast node. (Hugo Heuzard, review by Jérémie Dimino) - GPR#1841, MPR#7808: the environment variable OCAMLTOP_INCLUDE_PATH can now specify a list of additional include directories for the ocaml toplevel. (Nicolás Ojeda Bär, request by Daniel Bünzli, review by Daniel Bünzli and Damien Doligez) - MPR#7864, GPR#2109: remove duplicates from spelling suggestions. (Nicolás Ojeda Bär, review by Armaël Guéneau) - MPR#6638, GPR#1110: introduced a dedicated warning to report unused "open!" statements (Alain Frisch, report by dwang, review by and design from Leo White) - GPR#1974: Trigger warning 5 in "let _ = e" and "ignore e" if e is of function type and syntactically an application. (For the case of "ignore e" the warning already existed, but used to be triggered even when e was not an application.) (Nicolás Ojeda Bär, review by Alain Frisch and Jacques Garrigue) ### Code generation and optimizations: - MPR#7725, GPR#1754: improve AFL instrumentation for objects and lazy values. (Stephen Dolan) - GPR#1631: AMD64 code generator: emit shorter instruction sequences for the sign-extension operations. (LemonBoy, review by Alain Frisch and Xavier Leroy) - MPR#7246, GPR#2146: make a few int64 primitives use [@@unboxed] stubs on 32bits (Jérémie Dimino) - GPR#1917: comballoc: ensure object allocation order is preserved (Stephen Dolan) - MPR#6242, GPR#2143: optimize some local functions (Alain Frisch, review by Gabriel Scherer) - GPR#2082: New options [-insn-sched] and [-no-insn-sched] to control instruction scheduling. (Mark Shinwell, review by Damien Doligez) - GPR#2239: Fix match miscompilation with flambda (Leo White, review by Alain Frisch) ### Runtime system: - MPR#7198, MPR#7750, GPR#1738: add a function (caml_custom_alloc_mem) and three GC parameters to give the user better control of the out-of-heap memory retained by custom values; use the function to allocate bigarrays and I/O channels. (Damien Doligez, review by Alain Frisch) - MPR#7676, GPR#2144: Remove old GC heuristic (Damien Doligez, report and review by Alain Frisch) - GPR#1793: add the -m and -M command-line options to ocamlrun. (Sébastien Hinderer, review by Xavier Clerc and Damien Doligez) - GPR#1723: Remove internal Meta.static_{alloc,free} primitives. (Stephen Dolan, review by Gabriel Scherer) - GPR#1867: Remove the C plugins mechanism. (Xavier Leroy, review by David Allsopp, Damien Doligez, Sébastien Hinderer) - GPR#1895: Printexc.get_callstack would return only one frame in native code in threads other then the initial one (Valentin Gatien-Baron, review by Xavier Leroy) - GPR#1900, MPR#7814: avoid exporting non-prefixed identifiers in the debug and instrumented runtimes. (Damien Doligez, report by Gabriel Scherer) - GPR#2079: Avoid page table lookup in Pervasives.compare with no-naked-pointers (Sam Goldman, review by Gabriel Scherer, David Allsopp, Stephen Dolan) ### Tools - GPR#2182: Split Emacs caml-mode as an independent project. (Christophe Troestler, review by Gabriel Scherer) - MPR#7843, GPR#2013: ocamldoc, better handling of {{!label}text} in the latex backend. (Florian Angeletti, review by Nicolás Ojeda Bär and Gabriel Scherer) - MPR#7844, GPR#2040: Emacs, use built-in detection of comments, fixes an imenu crash. (Wilfred Hughes, review by Christophe Troestler) - MPR#7850: Emacs, use symbol boundaries in regular expressions, fixes an imenu crash. (Wilfred Hughes, review by Christophe Troestler) - GPR#1711: the new 'open' flag in OCAMLRUNPARAM takes a comma-separated list of modules to open as if they had been passed via the command line -open flag. (Nicolás Ojeda Bär, review by Mark Shinwell) - GPR#1865: support dark themes in Emacs, and clean up usage of deprecated Emacs APIs (Wilfred Hughes, review by Clément Pit-Claudel) - GPR#2000: ocamdoc, extended support for "include module type of ..." (Florian Angeletti, review by Jérémie Dimino) - GPR#2045: ocamlmklib now supports options -args and -args0 to provide extra command-line arguments in a file. (Nicolás Ojeda Bär, review by Gabriel Scherer and Daniel Bünzli) - GPR#2189: change ocamldep Makefile-output to print each dependency on a new line, for more readable diffs of versioned dependencies. (Gabriel Scherer, review by Nicolás Ojeda Bär) - GPR#2223: ocamltest: fix the "bsd" and "not-bsd" built-in actions to recognize all BSD variants (Damien Doligez, review by Sébastien Hinderer and David Allsopp) ### Manual and documentation: - MPR#7548: printf example in the tutorial part of the manual (Kostikova Oxana, rewiew by Gabriel Scherer, Florian Angeletti, Marcello Seri and Armaël Guéneau) - MPR#7546, GPR#2020: preambles and introduction for compiler-libs. (Florian Angeletti, review by Daniel Bünzli, Perry E. Metzger and Gabriel Scherer) - MPR#7547, GPR#2273: Tutorial on Lazy expressions and patterns in OCaml Manual (Ulugbek Abdullaev, review by Florian Angeletti and Gabriel Scherer) - MPR#7720, GPR#1596, precise the documentation of the maximum indentation limit in Format. (Florian Angeletti, review by Richard Bonichon and Pierre Weis) - MPR#7825: html manual split compilerlibs from stdlib in the html index of modules (Florian Angeletti, review by Perry E. Metzger and Gabriel Scherer) - GPR#1209, GPR#2008: in the Extension section, use the caml_example environment (uses the compiler to check the example code). This change was made possible by a lot of tooling work from Florian Angeletti: GPR#1702, GPR#1765, GPR#1863, and Gabriel Scherer's GPR#1903. (Gabriel Scherer, review by Florian Angeletti) - GPR#1788, 1831, 2007, 2198, 2232, move language extensions to the core chapters: - GPR#1788: quoted string description - GPR#1831: local exceptions and exception cases - GPR#2007: 32-bit, 64-bit and native integer literals - GPR#2198: lazy patterns - GPR#2232: short object copy notation (Florian Angeletti, review by Xavier Clerc, Perry E. Metzger, Gabriel Scherer and Jeremy Yallop) - GPR#1863: caml-tex2, move to compiler-libs (Florian Angeletti, review by Sébastien Hinderer and Gabriel Scherer) - GPR#2105: Change verbatim to caml_example in documentation (Maxime Flin, review by Florian Angeletti) - GPR#2114: ocamldoc, improved manpages for documentation inside modules (Florian Angeletti, review by Gabriel Scherer) - GPR#2117: stdlib documentation, duplicate the operator precedence table from the manual inside a separate "OCaml_operators" module. (Florian Angeletti, review by Daniel Bünzli, Perry E. Metzger and Gabriel Scherer) - GPR#2187: document "exception A | pat" patterns (Florian Angeletti, review by Perry E. Metzger and Jeremy Yallop) ### Compiler distribution build system: - GPR#1776: add -no-install-bytecode-programs and related configure options to control (non-)installation of ".byte" executables. (Mark Shinwell, review by Sébastien Hinderer and Gabriel Scherer) - GPR#1777: add -no-install-source-artifacts and related configure options to control installation of .cmt, .cmti, .mli and .ml files. (Mark Shinwell, review by Nicolás Ojeda Bär and Sébastien Hinderer) - GPR#1781: cleanup of the manual's build process. (steinuil, review by Marcello Seri, Gabriel Scherer and Florian Angeletti) - GPR#1797: remove the deprecated Makefile.nt files. (Sébastien Hinderer, review by Nicolas Ojeda Bar) - GPR#1805: fix the bootstrap procedure and its documentation. (Sébastien Hinderer, Xavier Leroy and Damien Doligez; review by Gabriel Scherer) - GPR#1840: build system enhancements. (Sébastien Hinderer, review by David Allsopp, Xavier Leroy and Damien Doligez) - GPR#1852: merge runtime directories (Sébastien Hinderer, review by Xavier Leroy and Damien Doligez) - GPR#1854: remove the no longer defined BYTECCCOMPOPTS build variable. (Sébastien Hinderer, review by Damien Doligez) - GPR#2024: stop supporting obsolete platforms: Rhapsody (old beta version of MacOS X, BeOS, alpha*-*-linux*, mips-*-irix6*, alpha*-*-unicos, powerpc-*-aix, *-*-solaris2*, mips*-*-irix[56]*, i[3456]86-*-darwin[89].*, i[3456]86-*-solaris*, *-*-sunos* *-*-unicos. (Sébastien Hinderer, review by Xavier Leroy, Damien Doligez, Gabriel Scherer and Armaël Guéneau) - GPR#2053: allow unix, vmthreads and str not to be built. (David Allsopp, review by Sébastien Hinderer) * GPR#2059: stop defining OCAML_STDLIB_DIR in s.h. (Sébastien Hinderer, review by David Allsopp and Damien Doligez) * GPR#2066: remove the standard_runtime configuration variable. (Sébastien Hinderer, review by Xavier Leroy, Stephen Dolan and Damien Doligez) * GPR#2139: use autoconf to generate the compiler's configuration script (Sébastien Hinderer, review by Damien Doligez and David Allsopp) - GPR#2148: fix a parallel build bug involving CamlinternalLazy. (Stephen Dolan, review by Gabriel Scherer and Nicolas Ojeda Bar) - GPR#2264, MPR#7904: the configure script now sets the Unicode handling mode under Windows according to the value of the variable WINDOWS_UNICODE_MODE. If WINDOWS_UNICODE_MODE is "ansi" then it is assumed to be the current code page encoding. If WINDOWS_UNICODE_MODE is "compatible" or empty or not set at all, then encoding is UTF-8 with code page fallback. (Nicolás Ojeda Bär, review by Sébastien Hinderer and David Allsopp) - GPR#2266: ensure Cygwin ports configure with `EXE=.exe`, or the compiler is unable to find the camlheader files (subtle regression of GPR#2139/2041) (David Allsopp, report and review by Sébastien Hinderer) - MPR#7919, GPR#2311: Fix assembler detection in configure (Sébastien Hinderer, review by David Allsopp) ### Internal/compiler-libs changes: - MPR#7918, GPR#1703, GPR#1944, GPR#2213, GPR#2257: Add the module Compile_common, which factorizes the common part in Compile and Optcompile. This also makes the pipeline more modular. (Gabriel Radanne, help from Gabriel Scherer and Valentin Gatien-Baron, review by Mark Shinwell and Gabriel Radanne, regression spotted by Clément Franchini) - GPR#292: use Menhir as the parser generator for the OCaml parser. Satellite GPRs: GPR#1844, GPR#1846, GPR#1853, GPR#1850, GPR#1934, GPR#2151, GPR#2174 (Gabriel Scherer, Nicolás Ojeda Bär, Frédéric Bour, Thomas Refis and François Pottier, review by Nicolás Ojeda Bär, Leo White and David Allsopp) - GPR#374: use Misc.try_finally for resource cleanup in the compiler codebase. This should fix the problem of catch-and-reraise `try .. with` blocks destroying backtrace information. (François Bobot, help from Gabriel Scherer and Nicolás Ojeda Bär, review by Gabriel Scherer) - GPR#1148, GPR#1287, GPR#1288, GPR#1874: significant improvements of the tools/check-typo script used over the files of the whole repository; contributors are now expected to check that check-typo passes on their pull requests; see CONTRIBUTING.md for more details. (David Allsopp, review by Damien Doligez and Sébastien Hinderer) - GPR#1610, GPR#2252: Remove positions from paths (Leo White, review by Frédéric Bour and Thomas Refis) - GPR#1745: do not generalize the type of every sub-pattern, only of variables. (Thomas Refis, review by Leo White) - GPR#1746: remove unreachable error variant: Make_seltype_nongen. (Florian Angeletti, review by Gabriel Radanne) - GPR#1747: type_cases: always propagate. (Thomas Refis, review by Jacques Garrigue) - GPR#1811: shadow the polymorphic comparison in the middle-end (Xavier Clerc, review by Pierre Chambart) - GPR#1833: allow non-val payloads in CMM Ccatch handlers (Simon Fowler, review by Xavier Clerc) - GPR#1866: document the release process (Damien Doligez and Gabriel Scherer, review by Sébastien Hinderer, Perry E. Metzger, Xavier Leroy and David Allsopp) - GPR#1886: move the Location.absname reference to Clflags.absname (Armaël Guéneau, review by Jérémie Dimino) - GPR#1894: generalize highlight_dumb in location.ml to handle highlighting several locations (Armaël Guéneau, review by Gabriel Scherer) - GPR#1903: parsetree, add locations to all nodes with attributes (Gabriel Scherer, review by Thomas Refis) - GPR#1905: add check-typo-since to check the files changed since a given git reference (Gabriel Scherer, review by David Allsopp) - GPR#1910: improve the check-typo use of .gitattributes (Gabriel Scherer, review by David Allsopp and Damien Doligez) - GPR#1938: always check ast invariants after preprocessing (Florian Angeletti, review by Alain Frisch and Gabriel Scherer) - GPR#1941: refactor the command line parsing of ocamlcp and ocamloptp (Valentin Gatien-Baron, review by Florian Angeletti) - GPR#1948: Refactor Stdlib.Format. Notably, use Stdlib.Stack and Stdlib.Queue, and avoid exceptions for control flow. (Vladimir Keleshev, review by Nicolás Ojeda Bär and Gabriel Scherer) * GPR#1952: refactor the code responsible for displaying errors and warnings `Location.report_error` is removed, use `Location.print_report` instead (Armaël Guéneau, review by Thomas Refis) - MPR#7835, GPR#1980: separate scope from stamp in idents (Thomas Refis, review by Leo White) - GPR#1996: expose Pprintast.longident to help compiler-libs users print Longident.t values. (Gabriel Scherer, review by Florian Angeletti and Thomas Refis) - GPR#2030: makefile targets to build AST files of sources for parser testing. See parsing/HACKING.adoc. (Gabriel Scherer, review by Nicolás Ojeda Bär) * GPR#2041: add a cache for looking up files in the load path (Jérémie Dimino, review by Alain Frisch and David Allsopp) - GPR#2047, GPR#2269: a new type for unification traces (Florian Angeletti, report by Leo White (GPR#2269), review by Thomas Refis and Gabriel Scherer) - GPR#2055: Add [Linearize.Lprologue]. (Mark Shinwell, review by Pierre Chambart) - GPR#2056: Use [Backend_var] rather than [Ident] from [Clambda] onwards; use [Backend_var.With_provenance] for variables in binding position. (Mark Shinwell, review by Pierre Chambart) - GPR#2060: "Phantom let" support for the Clambda language. (Mark Shinwell, review by Vincent Laviron) - GPR#2065: Add [Proc.destroyed_at_reloadretaddr]. (Mark Shinwell, review by Damien Doligez) - GPR#2060: "Phantom let" support for the Clambda language. (Mark Shinwell, review by Vincent Laviron) - GPR#2070: "Phantom let" support for the Cmm language. (Mark Shinwell, review by Vincent Laviron) - GPR#2072: Always associate a scope to a type (Thomas Refis, review by Jacques Garrigue and Leo White) - GPR#2074: Correct naming of record field inside [Ialloc] terms. (Mark Shinwell, review by Jérémie Dimino) - GPR#2076: Add [Targetint.print]. (Mark Shinwell) - GPR#2080: Add [Proc.dwarf_register_numbers] and [Proc.stack_ptr_dwarf_register_number]. (Mark Shinwell, review by Bernhard Schommer) - GPR#2088: Add [Clambda.usymbol_provenance]. (Mark Shinwell, review by Damien Doligez) - GPR#2152, GPR#2517: refactorize the fixpoint to compute type-system properties of mutually-recursive type declarations. (Gabriel Scherer and Rodolphe Lepigre, review by Armaël Guéneau) - GPR#2156: propagate more type information through Lambda and Clambda intermediate language, as a preparation step for more future optimizations (Pierre Chambart and Alain Frisch, cross-reviewed by themselves) - GPR#2160: restore --disable-shared support and ensure testsuite runs correctly when compiled without shared library support. (David Allsopp, review by Damien Doligez and Sébastien Hinderer) * GPR#2173: removed TypedtreeMap (Thomas Refis, review by Gabriel Scherer) - GPR#2112: Fix Thread.yield unfairness with busy threads yielding to each other. (Andrew Hunter, review by Jacques-Henri Jourdan, Spiros Eliopoulos, Stephen Weeks, & Mark Shinwell) - MPR#7867: Fix #mod_use raising an exception for filenames with no extension. (Geoff Gole) - GPR#2100: Fix Unix.getaddrinfo when called on strings containing null bytes; it would crash the GC later on. (Armaël Guéneau, report and fix by Joe, review by Sébastien Hinderer) - MPR#7847, GPR#2019: Fix an infinite loop that could occur when the (Menhir-generated) parser encountered a syntax error in a certain specific state. (François Pottier, report by Stefan Muenzel, review by Frédéric Bour, Thomas Refis, Gabriel Scherer) - GPR#1626: Do not allow recursive modules in `with module` (Leo White, review by Gabriel Radanne) - MPR#7726, GPR#1676: Recursive modules, equi-recursive types and stack overflow (Jacques Garrigue, report by Jeremy Yallop, review by Leo White) - MPR#7723, GPR#1698: Ensure `with module` and `with type` do not weaken module aliases. (Leo White, review by Gabriel Radanne and Jacques Garrigue) - GPR#1719: fix Pervasives.LargeFile functions under Windows. (Alain Frisch) - GPR#1739: ensure ocamltest waits for child processes to terminate on Windows. (David Allsopp, review by Sébastien Hinderer) - MPR#7554, GPR#1751: Lambda.subst: also update debug event environments (Thomas Refis, review by Gabriel Scherer) - MPR#7238, GPR#1825: in Unix.in_channel_of_descr and Unix.out_channel_of_descr, raise an error if the given file description is not suitable for character-oriented I/O, for example if it is a block device or a datagram socket. (Xavier Leroy, review by Jérémie Dimino and Perry E. Metzger) - MPR#7799, GPR#1820: fix bug where Scanf.format_from_string could fail when the argument string contained characters that require escaping. (Gabriel Scherer and Nicolás Ojeda Bär, report by Guillaume Melquiond, review by Gabriel Scherer) - GPR#1843: ocamloptp was doing the wrong thing with option -inline-max-unroll. (Github user @poechsel, review by Nicolás Ojeda Bär). - GPR#1890: remove last use of Ctype.unroll_abbrev (Thomas Refis, report by Leo White, review by Jacques Garrigue) - GPR#1893: dev-branch only, warning 40(name not in scope) triggered spurious warnings 49(missing cmi) with -no-alias-deps. (Florian Angeletti, report by Valentin Gatien-Baron, review by Gabriel Scherer) - GPR#1912: Allow quoted strings, octal/unicode escape sequences and identifiers containing apostrophes in ocamllex actions and comments. (Pieter Goetschalckx, review by Damien Doligez) - MPR#7828, GPR#1935: correct the conditions that generate warning 61, Unboxable_type_in_prim_decl (Stefan Muenzel) - GPR#1958: allow [module M(_:S) = struct end] syntax (Hugo Heuzard, review by Gabriel Scherer) - GPR#1970: fix order of floatting documentation comments in classes (Hugo Heuzard, review by Nicolás Ojeda Bär) - GPR#1977: [@@ocaml.warning "..."] attributes attached to type declarations are no longer ignored. (Nicolás Ojeda Bär, review by Gabriel Scherer) - MPR#7830, GPR#1987: fix ocamldebug crash when printing a value in the scope of an `open` statement for which the `.cmi` is not available. (Nicolás Ojeda Bär, report by Jocelyn Sérot, review by Gabriel Scherer) - MPR#7408, MPR#7846, GPR#2015: Check arity of primitives. (Hugo Heuzard, review by Nicolás Ojeda Bär) - MPR#7854, GPR#2062: fix an issue where the wrong locale may be used when using the legacy ANSI encoding under Windows. (Nicolás Ojeda Bär, report by Tiphaine Turpin) - GPR#2083: Fix excessively aggressive float unboxing and introduce similar fix as a preventative measure for boxed int unboxing. (Thomas Refis, Mark Shinwell, Leo White) - GPR#2130: fix printing of type variables with a quote in their name (Alain Frisch, review by Armaël Guéneau and Gabriel Scherer, report by Hugo Heuzard) - GPR#2131: fix wrong calls to Env.normalize_path on non-module paths (Alain Frisch, review by Jacques Garrigue) - GPR#2175: Apply substitution to all modules when packing (Leo White, review by Gabriel Scherer) - GPR#2231: Env: always freshen persistent signatures before using them (Thomas Refis and Leo White, review by Gabriel Radanne) - MPR#7923, GPR#2259: fix regression in FlexDLL bootstrapped build caused by refactoring the root Makefile for Dune in GPR#2093) (David Allsopp, report by Marc Lasson) - MPR#7929, GPR#2261: Subst.signature: call cleanup_types exactly once (Thomas Refis, review by Gabriel Scherer and Jacques Garrigue, report by Daniel Bünzli and Jon Ludlam) OCaml 4.07.1 (4 October 2018) ----------------------------- ### Bug fixes: - MPR#7815, GPR#1896: major GC crash with first-fit policy (Stephen Dolan and Damien Doligez, report by Joris Giovannangeli) * MPR#7818, GPR#2051: Remove local aliases in functor argument types, to prevent the aliasing of their target. (Jacques Garrigue, report by mandrykin, review by Leo White) - MPR#7820, GPR#1897: Fix Array.of_seq. This function used to apply a circular permutation of one cell to the right on the sequence. (Thierry Martinez, review by Nicolás Ojeda Bär) - MPR#7821, GPR#1908: make sure that the compilation of extension constructors doesn't cause the compiler to load more cmi files (Jérémie Dimino, review by Gabriel Scherer) - MPR#7824, GPR#1914: subtype_row: filter out absent fields when row is closed (Leo White and Thomas Refis, report by talex, review by Jacques Garrigue) - GPR#1915: rec_check.ml is too permissive for certain class declarations. (Alban Reynaud with Gabriel Scherer, review by Jeremy Yallop) - MPR#7833, GPR#1946: typecore: only 1k existential per match, not 100k (Thomas Refis, report by Jerome Simeon, review by Jacques Garrigue) - MPR#7838: -principal causes assertion failure in type checker (Jacques Garrigue, report by Markus Mottl, review by Thomas Refis) OCaml 4.07.0 (10 July 2018) --------------------------- (Changes that can break existing programs are marked with a "*") ### Language features: - MPR#6023, GPR#1648: Allow type-based selection of GADT constructors. (Thomas Refis and Leo White, review by Jacques Garrigue and Gabriel Scherer) - GPR#1546: Allow empty variants. (Runhang Li, review by Gabriel Radanne and Jacques Garrigue) ### Standard library: - MPR#4170, GPR#1674: add the constant `Float.pi`. (Christophe Troestler, review by Damien Doligez) - MPR#6139, GPR#1685: Move the Bigarray module to the standard library. Keep the bigarray library as on overlay adding the deprecated map_file functions. (Jérémie Dimino, review by Mark Shinwell) - MPR#7528, GPR#1500: add a Format.pp_set_geometry function to avoid memory effects in set_margin and set_max_indent. (Florian Angeletti, review by Richard Bonichon, Gabriel Radanne, Gabiel Scherer and Pierre Weis) - MPR#7690, GPR#1528: fix the float_of_string function for hexadecimal floats with very large values of the exponent. (Olivier Andrieu) - GPR#1002: add a new `Seq` module defining a list-of-thunks style iterator. Also add `{to,of}_seq` to several standard modules. (Simon Cruanes, review by Alain Frisch and François Bobot) * GPR#1010: pack all standard library modules into a single module Stdlib which is the default opened module (Stdlib itself includes Pervasives) to free up the global namespace for other standard libraries, while still allowing any OCaml standard library module to be referred to as Stdlib.Module). This is implemented efficiently using module aliases (prefixing all modules with Stdlib__, e.g. Stdlib__string). (Jérémie Dimino, David Allsopp and Florian Angeletti, review by David Allsopp and Gabriel Radanne) - GPR#1637: String.escaped is faster and does not allocate when called with a string that does not contain any characters needing to be escaped. (Alain Frisch, review by Xavier Leroy and Gabriel Scherer) - GPR#1638: add a Float module. (Nicolás Ojeda Bär, review by Alain Frisch and Jeremy Yallop) - GPR#1697: Tune [List.init] tailrec threshold so that it does not stack overflow when compiled with the Js_of_ocaml backend. (Hugo Heuzard, reviewed by Gabriel Scherer) ### Other libraries: - MPR#7745, GPR#1629: Graphics.open_graph displays the correct window title on Windows again (fault introduced by 4.06 Unicode changes). (David Allsopp) * GPR#1406: Unix.isatty now returns true in the native Windows ports when passed a file descriptor connected to a Cygwin PTY. In particular, compiler colors for the native Windows ports now work under Cygwin/MSYS2. (Nicolás Ojeda Bär, review by Gabriel Scherer, David Allsopp, Xavier Leroy) - GPR#1451: [getpwuid], [getgrgid], [getpwnam], [getgrnam] now raise Unix error instead of returning [Not_found] when interrupted by a signal. (Arseniy Alekseyev, review by Mark Shinwell and Xavier Leroy) - GPR#1477: raw_spacetime_lib can now be used in bytecode. (Nicolás Ojeda Bär, review by Mark Shinwell) - GPR#1533: (a) The implementation of Thread.yield for system thread now uses nanosleep(1) for enabling better preemption. (b) Thread.delay is now an alias for Unix.sleepf. (Jacques-Henri Jourdan, review by Xavier Leroy and David Allsopp) ### Compiler user-interface and warnings: - MPR#7663, GPR#1694: print the whole cycle and add a reference to the manual in the unsafe recursive module evaluation error message. (Florian Angeletti, report by Matej Košík, review by Gabriel Scherer) - GPR#1166: In OCAMLPARAM, an alternative separator can be specified as first character (instead of comma) in the set ":|; ," (Fabrice Le Fessant) - GPR#1358: Fix usage warnings with no mli file. (Leo White, review by Alain Frisch) - GPR#1428: give a non dummy location for warning 49 (no cmi found). (Valentin Gatien-Baron) - GPR#1491: Improve error reporting for ill-typed applicative functor types, F(M).t. (Valentin Gatien-Baron, review by Florian Angeletti and Gabriel Radanne) - GPR#1496: Refactor the code printing explanation for unification type errors, in order to avoid duplicating pattern matches. (Armaël Guéneau, review by Florian Angeletti and Gabriel Scherer) - GPR#1505: Add specific error messages for unification errors involving functions of type "unit -> _". (Arthur Charguéraud and Armaël Guéneau, with help from Leo White, review by Florian Angeletti and Gabriel Radanne) - GPR#1510: Add specific explanation for unification errors caused by type constraints propagated by keywords (such as if, while, for...). (Armaël Guéneau and Gabriel Scherer, original design by Arthur Charguéraud, review by Frédéric Bour, Gabriel Radanne and Alain Frisch) - GPR#1515: honor the BUILD_PATH_PREFIX_MAP environment variable to enable reproducible builds. (Gabriel Scherer, with help from Ximin Luo, review by Damien Doligez) - GPR#1534: Extend the warning printed when (*) is used, adding a hint to suggest using ( * ) instead. (Armaël Guéneau, with help and review from Florian Angeletti and Gabriel Scherer) - GPR#1552, GPR#1577: do not warn about ambiguous variables in guards (warning 57) when the ambiguous values have been filtered by a previous clause. (Gabriel Scherer and Thomas Refis, review by Luc Maranget) - GPR#1554: warnings 52 and 57: fix reference to manual detailed explanation. (Florian Angeletti, review by Thomas Refis and Gabriel Scherer) - GPR#1618: add the -dno-unique-ids and -dunique-ids compiler flags. (Sébastien Hinderer, review by Leo White and Damien Doligez) - GPR#1649: change compilation order of toplevel definitions, so that some warnings emitted by the bytecode compiler appear more in-order than before. (Luc Maranget, advice and review by Damien Doligez) - GPR#1806: add linscan to OCAMLPARAM options. (Raja Boujbel) ### Code generation and optimizations: - MPR#7630, GPR#1401: Faster compilation of large modules with Flambda. (Pierre Chambart, report by Emilio Jesús Gallego Arias, Pierre-Marie Pédrot and Paul Steckler, review by Gabriel Scherer and Leo White) - MPR#7630, GPR#1455: Disable CSE for the initialization function. (Pierre Chambart, report by Emilio Jesús Gallego Arias, review by Gabriel Scherer and Xavier Leroy) - GPR#1370: Fix code duplication in Cmmgen. (Vincent Laviron, with help from Pierre Chambart, reviews by Gabriel Scherer and Luc Maranget) - GPR#1486: ARM 32-bit port: add support for ARMv8 in 32-bit mode, a.k.a. AArch32. For this platform, avoid ITE conditional instruction blocks and use simpler IT blocks instead. (Xavier Leroy, review by Mark Shinwell) - GPR#1487: Treat negated float comparisons more directly. (Leo White, review by Xavier Leroy) - GPR#1573: emitcode: merge events after instructions reordering. (Thomas Refis and Leo White, with help from David Allsopp, review by Frédéric Bour) - GPR#1606: Simplify the semantics of Lambda.free_variables and Lambda.subst, including some API changes in bytecomp/lambda.mli. (Pierre Chambart, review by Gabriel Scherer) - GPR#1613: ensure that set-of-closures are processed first so that other entries in the let-rec symbol do not get dummy approximations. (Leo White and Xavier Clerc, review by Pierre Chambart) * GPR#1617: Make string/bytes distinguishable in the bytecode. (Hugo Heuzard, reviewed by Nicolás Ojeda Bär) - GPR#1627: Reduce cmx sizes by sharing variable names (Flambda only). (Fuyong Quah, Leo White, review by Xavier Clerc) - GPR#1665: reduce the size of cmx files in classic mode by dropping the bodies of functions that will not be inlined. (Fuyong Quah, review by Leo White and Pierre Chambart) - GPR#1666: reduce the size of cmx files in classic mode by dropping the bodies of functions that cannot be reached from the module block. (Fuyong Quah, review by Leo White and Pierre Chambart) - GPR#1686: Turn off by default flambda invariants checks. (Pierre Chambart) - GPR#1707: Add [Closure_origin.t] to trace inlined functions to prevent infinite loops from repeatedly inlining copies of the same function. (Fu Yong Quah) - GPR#1740: make sure startup.o is always linked in when using "-output-complete-obj". Previously, it was always linked in only on some platforms, making this option unusable on platforms where it wasn't. (Jérémie Dimino, review by Sébastien Hinderer and Xavier Leroy) ### Runtime system: - GPR#515 GPR#676 MPR#7173: Add a public C API for weak arrays and ephemerons. Update the documentation for a 4.03 change: finalisation functions are now run before the erasure of the corresponding values. (François Bobot and Jacques-Henri Jourdan, review by Mark Shinwell, Damien Doligez and Frédéric Bour) - MPR#6411, GPR#1535: don't compile everything with -static-libgcc on mingw32, only dllbigarray.dll and libbigarray.a. Allows the use of C++ libraries which raise exceptions. (David Allsopp) - MPR#7100, GPR#1476: trigger a minor GC when custom blocks accumulate in minor heap. (Alain Frisch, report by talex, review by Damien Doligez, Leo White, Gabriel Scherer) - GPR#1431: remove ocamlrun dependencies on curses/terminfo/termcap C library. (Xavier Leroy, review by Daniel Bünzli) - GPR#1478: The Spacetime profiler now works under Windows (but it is not yet able to collect profiling information from C stubs). (Nicolás Ojeda Bär, review by Xavier Leroy, Mark Shinwell) - GPR#1483: fix GC freelist accounting for chunks larger than the maximum block size. (David Allsopp and Damien Doligez) - GPR#1526: install the debug and instrumented runtimes (lib{caml,asm}run{d,i}.a). (Gabriel Scherer, reminded by Julia Lawall) - GPR#1563: simplify implementation of LSRINT and ASRINT. (Max Mouratov, review by Frédéric Bour) - GPR#1644: remove caml_alloc_float_array from the bytecode primitives list (it's a native code primitive). (David Allsopp) - GPR#1701: fix missing root bug in GPR#1476. (Mark Shinwell) - GPR#1752: do not alias function arguments to sigprocmask. (Anil Madhavapeddy) - GPR#1753: avoid potential off-by-one overflow in debugger socket path length. (Anil Madhavapeddy) * GPR#1683: Change Marshal format to make Custom_tag objects store their length. Old versions of OCaml will no longer be able to parse new marshalled files containing custom blocks, but old files will still parse. (Stephen Dolan) ### Tools: - MPR#7643, GPR#1377: ocamldep, fix an exponential blowup in presence of nested structures and signatures, e.g. "include struct … include(struct … end) … end" (Florian Angeletti, review by Gabriel Scherer, report by Christophe Raffalli) - MPR#7687, GPR#1653: deprecate -thread option, which is equivalent to -I +threads. (Nicolás Ojeda Bär, report by Daniel Bünzli) - MPR#7710: `ocamldep -sort` should exit with nonzero code in case of cyclic dependencies. (Xavier Leroy, report by Mantis user baileyparker) - GPR#1537: boot/ocamldep is no longer included in the source distribution; boot/ocamlc -depend can be used in its place. (Nicolás Ojeda Bär, review by Xavier Leroy and Damien Doligez) - GPR#1585: optimize output of "ocamllex -ml". (Alain Frisch, review by Frédéric Bour and Gabriel Scherer) - GPR#1667: add command-line options -no-prompt, -no-version, -no-time, -no-breakpoint-message and -topdirs-path to ocamldebug. (Sébastien Hinderer, review by Damien Doligez) - GPR#1695: add the -null-crc command-line option to ocamlobjinfo. (Sébastien Hinderer, review by David Allsopp and Gabriel Scherer) - GPR#1710: ocamldoc, improve the 'man' rendering of subscripts and superscripts. (Gabriel Scherer) - GPR#1771: ocamldebug, avoid out of bound access. (Thomas Refis) ### Manual and documentation: - MPR#7613: minor rewording of the "refutation cases" paragraph. (Florian Angeletti, review by Jacques Garrigue) - MPR#7647, GPR#1384: emphasize ocaml.org website and forum in README. (Yawar Amin, review by Gabriel Scherer) - MPR#7698, GPR#1545: improve wording in OCaml manual in several places, mostly in Chapter 1. This addresses the easier changes suggested in the PR. (Jim Fehrle, review by Florian Angeletti and David Allsopp) - GPR#1540: manual, decouple verbatim and toplevel style in code examples. (Florian Angeletti, review by Gabriel Scherer) - GPR#1556: manual, add a consistency test for manual references inside the compiler source code. (Florian Angeletti, review by Gabriel Scherer) - GPR#1647: manual, subsection on record and variant disambiguation. (Florian Angeletti, review by Alain Frisch and Gabriel Scherer) - GPR#1702: manual, add a signature mode for code examples. (Florian Angeletti, review by Gabriel Scherer) - GPR#1741: manual, improve typesetting and legibility in HTML output. (steinuil, review by Gabriel Scherer) - GPR#1757: style the html manual, changing type and layout. (Charles Chamberlain, review by Florian Angeletti, Xavier Leroy, Gabriel Radanne, Perry E. Metzger, and Gabriel Scherer) - GPR#1765: manual, ellipsis in code examples. (Florian Angeletti, review and suggestion by Gabriel Scherer) - GPR#1767: change html manual to use relative font sizes. (Charles Chamberlain, review by Daniel Bünzli, Perry E. Metzger, Josh Berdine, and Gabriel Scherer) - GPR#1779: integrate the Bigarray documentation into the main manual. (Perry E. Metzger, review by Florian Angeletti and Xavier Clerc) ### Type system: - MPR#7611, GPR#1491: reject the use of generative functors as applicative. (Valentin Gatien-Baron) - MPR#7706, GPR#1565: in recursive value declarations, track static size of locally-defined variables. (Gabriel Scherer, review by Jeremy Yallop and Leo White, report by Leo White) - MPR#7717, GPR#1593: in recursive value declarations, don't treat unboxed constructor size as statically known. (Jeremy Yallop, report by Pierre Chambart, review by Gabriel Scherer) - MPR#7767, GPR#1712: restore legacy treatment of partially-applied labeled functions in 'let rec' bindings. (Jeremy Yallop, report by Ivan Gotovchits, review by Gabriel Scherer) * MPR#7787, GPR#1652, GPR#1743: Don't remove module aliases in `module type of` and `with module`. The old behaviour can be obtained using the `[@remove_aliases]` attribute. (Leo White and Thomas Refis, review by Jacques Garrigue) - GPR#1468: Do not enrich type_decls with incoherent manifests. (Thomas Refis and Leo White, review by Jacques Garrigue) - GPR#1469: Use the information from [@@immediate] annotations when computing whether a type can be [@@unboxed]. (Damien Doligez, report by Stephan Muenzel, review by Alain Frisch) - GPR#1513: Allow compilation units to shadow sub-modules of Pervasives. For instance users can now use a largeFile.ml file in their project. (Jérémie Dimino, review by Nicolás Ojeda Bär, Alain Frisch and Gabriel Radanne) - GPR#1516: Allow float array construction in recursive bindings when configured with -no-flat-float-array. (Jeremy Yallop, report by Gabriel Scherer) - GPR#1583: propagate refined ty_arg to Parmatch checks. (Thomas Refis, review by Jacques Garrigue) - GPR#1609: Changes to ambivalence scope tracking. (Thomas Refis and Leo White, review by Jacques Garrigue) - GPR#1628: Treat reraise and raise_notrace as nonexpansive. (Leo White, review by Alain Frisch) * GPR#1778: Fix Soundness bug with non-generalized type variable and local modules. This is the same bug as MPR#7414, but using local modules instead of non-local ones. (Leo White, review by Jacques Garrigue) ### Compiler distribution build system: - MPR#5219, GPR#1680, GPR#1877: use 'install' instead of 'cp' in install scripts. (Gabriel Scherer, review by Sébastien Hinderer and Valentin Gatien-Baron) - MPR#7679: make sure .a files are erased before calling ar rc, otherwise leftover .a files from an earlier compilation may contain unwanted modules. (Xavier Leroy) - GPR#1571: do not perform architecture tests on 32-bit platforms, allowing 64-bit back-ends to use 64-bit specific constructs. (Xavier Clerc, review by Damien Doligez) ### Internal/compiler-libs changes: - MPR#7738, GPR#1624: Asmlink.reset also resets lib_ccobjs/ccopts. (Cedric Cellier, review by Gabriel Scherer) - GPR#1488, GPR#1560: Refreshing parmatch. (Gabriel Scherer and Thomas Refis, review by Luc Maranget) - GPR#1502: more command line options for expect tests. (Florian Angeletti, review by Gabriel Scherer) - GPR#1511: show code at error location in expect-style tests, using new Location.show_code_at_location function. (Gabriel Scherer and Armaël Guéneau, review by Valentin Gatien-Baron and Damien Doligez) - GPR#1519, GPR#1532, GRP#1570: migrate tests to ocamltest. (Sébastien Hinderer, review by Gabriel Scherer, Valentin Gatien-Baron and Nicolás Ojeda Bär) - GPR#1520: more robust implementation of Misc.no_overflow_mul. (Max Mouratov, review by Xavier Leroy) - GPR#1557: Organise and simplify translation of primitives. (Leo White, review by François Bobot and Nicolás Ojeda Bär) - GPR#1567: register all idents relevant for reraise. (Thomas Refis, review by Alain Frisch and Frédéric Bour) - GPR#1586: testsuite: 'make promote' for ocamltest tests. (The new "-promote" option for ocamltest is experimental and subject to change/removal). (Gabriel Scherer) - GPR#1619: expect_test: print all the exceptions, even the unexpected ones. (Thomas Refis, review by Jérémie Dimino) - GPR#1621: expect_test: make sure to not use the installed stdlib. (Jérémie Dimino, review by Thomas Refis) - GPR#1646: add ocamldoc test to ocamltest and migrate ocamldoc tests to ocamltest. (Florian Angeletti, review by Sébastien Hinderer) - GPR#1663: refactor flambda specialise/inlining handling. (Leo White and Xavier Clerc, review by Pierre Chambart) - GPR#1679: remove Pbittest from primitives in lambda. (Hugo Heuzard, review by Mark Shinwell) * GPR#1704: Make Ident.t abstract and immutable. (Gabriel Radanne, review by Mark Shinwell) - GPR#1699: Clean up Maps and Sets throughout the compiler. Remove the Tbl module in favor of dedicated Maps. (Gabriel Radanne, review by Mark Shinwell) ### Bug fixes: - MPR#4499, GPR#1479: Use native Windows API to implement Sys.getenv, Unix.getenv and Unix.environment under Windows. (Nicolás Ojeda Bär, report by Alain Frisch, review by David Allsopp, Xavier Leroy) - MPR#5250, GPR#1435: on Cygwin, when ocamlrun searches the path for a bytecode executable file, skip directories and other non-regular files, like other Unix variants do. (Xavier Leroy) - MPR#6394, GPR#1425: fix fatal_error from Parmatch.get_type_path. (Virgile Prevosto, review by David Allsopp, Thomas Refis and Jacques Garrigue) * MPR#6604, GPR#931: Only allow directives with filename and at the beginning of the line. (Tadeu Zagallo, report by Roberto Di Cosmo, review by Hongbo Zhang, David Allsopp, Gabriel Scherer, Xavier Leroy) - MPR#7138, MPR#7701, GPR#1693: Keep documentation comments even in empty structures and signatures. (Leo White, Florian Angeletti, report by Anton Bachin) - MPR#7178, MPR#7253, MPR#7796, GPR#1790: Make sure a function registered with "at_exit" is executed only once when the program exits. (Nicolás Ojeda Bär and Xavier Leroy, review by Max Mouratov) - MPR#7391, GPR#1620: Do not put a dummy method in object types. (Thomas Refis, review by Jacques Garrigue) - PR#7660, GPR#1445: Use native Windows API to implement Unix.utimes in order to avoid unintended shifts of the argument timestamp depending on DST setting. (Nicolás Ojeda Bär, review by David Allsopp, Xavier Leroy) - MPR#7668: -principal is broken with polymorphic variants. (Jacques Garrigue, report by Jun Furuse) - MPR#7680, GPR#1497: Incorrect interaction between Matching.for_let and Simplif.simplify_exits. (Alain Frisch, report and review by Vincent Laviron) - MPR#7682, GPR#1495: fix [@@unboxed] for records with 1 polymorphic field. (Alain Frisch, report by Stéphane Graham-Lengrand, review by Gabriel Scherer) - MPR#7695, GPR#1541: Fatal error: exception Ctype.Unify(_) with field override (Jacques Garrigue, report by Nicolás Ojeda Bär) - MPR#7704, GPR#1564: use proper variant tag in non-exhaustiveness warning. (Jacques Garrigue, report by Thomas Refis) - MPR#7711, GPR#1581: Internal typechecker error triggered by a constraint on self type in a class type. (Jacques Garrigue, report and review by Florian Angeletti) - MPR#7712, GPR#1576: assertion failure with type abbreviations. (Thomas Refis, report by Michael O'Connor, review by Jacques Garrigue) - MPR#7747: Type checker can loop infinitely and consume all computer memory. (Jacques Garrigue, report by kantian) - MPR#7751, GPR#1657: The toplevel prints some concrete types as abstract. (Jacques Garrigue, report by Matej Kosik) - MPR#7765, GPR#1718: When unmarshaling bigarrays, protect against integer overflows in size computations. (Xavier Leroy, report by Maximilian Tschirschnitz, review by Gabriel Scherer) - MPR#7760, GPR#1713: Exact selection of lexing engine, that is correct "Segfault in ocamllex-generated code using 'shortest'". (Luc Maranget, Frédéric Bour, report by Stephen Dolan, review by Gabriel Scherer) - MPR#7769, GPR#1714: calls to Stream.junk could, under some conditions, be ignored when used on streams based on input channels. (Nicolás Ojeda Bär, report by Michael Perin, review by Gabriel Scherer) - MPR#7793, GPR#1766: the toplevel #use directive now accepts sequences of ';;' tokens. This fixes a bug in which certain files accepted by the compiler were rejected by ocamldep. (Nicolás Ojeda Bär, report by Hugo Heuzard, review by Hugo Heuzard) - GPR#1517: More robust handling of type variables in mcomp. (Leo White and Thomas Refis, review by Jacques Garrigue) - GPR#1530, GPR#1574: testsuite, fix 'make parallel' and 'make one DIR=...' to work on ocamltest-based tests. (Runhang Li and Sébastien Hinderer, review by Gabriel Scherer) - GPR#1550, GPR#1555: Make pattern matching warnings more robust to ill-typed columns. (Thomas Refis, with help from Gabriel Scherer and Luc Maranget) - GPR#1614: consider all bound variables when inlining, fixing a compiler fatal error. (Xavier Clerc, review by Pierre Chambart, Leo White) - GPR#1622: fix bug in the expansion of command-line arguments under Windows which could result in some elements of Sys.argv being truncated in some cases. (Nicolás Ojeda Bär, review by Sébastien Hinderer) - GPR#1623: Segfault on Windows 64 bits when expanding wildcards in arguments. (Marc Lasson, review by David Allsopp, Alain Frisch, Sébastien Hinderer, Xavier Leroy, Nicolas Ojeda Bar) - GPR#1661: more precise principality warning regarding record fields disambiguation. (Thomas Refis, review by Leo White) - GPR#1687: fix bug in the printing of short functor types "(S1 -> S2) -> S3". (Pieter Goetschalckx, review by Gabriel Scherer) - GPR#1722: Scrape types in Typeopt.maybe_pointer. (Leo White, review by Thomas Refis) - GPR#1755: ensure that a bigarray is never collected while reading complex values. (Xavier Clerc, Mark Shinwell and Leo White, report by Chris Hardin, reviews by Stephen Dolan and Xavier Leroy) - GPR#1764: in byterun/memory.c, struct pool_block, use C99 flexible arrays if available. (Xavier Leroy, review by Max Mouratov) - GPR#1774: ocamlopt for ARM could generate VFP loads and stores with bad offsets, rejected by the assembler. (Xavier Leroy, review by Mark Shinwell) - GPR#1808: handle `[@inlined]` attributes under a module constraint. (Xavier Clerc, review by Leo White) - GPR#1810: use bit-pattern comparison when meeting float approximations. (Xavier Clerc, report by Christophe Troestler, review by Nicolás Ojeda Bär and Gabriel Scherer) - GPR#1835: Fix off-by-one errors in Weak.get_copy and Weak.blit. (KC Sivaramakrishnan) - GPR#1849: bug in runtime function generic_final_minor_update() that could lead to crashes when Gc.finalise_last is used. (report and fix by Yuriy Vostrikov, review by François Bobot) OCaml 4.06.1 (16 Feb 2018): --------------------------- ### Bug fixes: - MPR#7661, GPR#1459: fix faulty compilation of patterns using extensible variants constructors (Luc Maranget, review by Thomas Refis and Gabriel Scherer, report by Abdelraouf Ouadjaout and Thibault Suzanne) - MPR#7702, GPR#1553: refresh raise counts when inlining a function (Vincent Laviron, Xavier Clerc, report by Cheng Sun) - MPR#7704, GPR#1559: Soundness issue with private rows and pattern-matching (Jacques Garrigue, report by Jeremy Yallop, review by Thomas Refis) - MPR#7705, GPR#1558: add missing bounds check in Bigarray.Genarray.nth_dim. (Nicolás Ojeda Bär, report by Jeremy Yallop, review by Gabriel Scherer) - MPR#7713, GPR#1587: Make pattern matching warnings more robust to ill-typed columns; this is a backport of GPR#1550 from 4.07+dev (Thomas Refis, review by Gabriel Scherer, report by Andreas Hauptmann) - GPR#1470: Don't commute negation with float comparison (Leo White, review by Xavier Leroy) - GPR#1538: Make pattern matching compilation more robust to ill-typed columns (Gabriel Scherer and Thomas Refis, review by Luc Maranget) OCaml 4.06.0 (3 Nov 2017): -------------------------- (Changes that can break existing programs are marked with a "*") ### Language features: - MPR#6271, MPR#7529, GPR#1249: Support "let open M in ..." in class expressions and class type expressions. (Alain Frisch, reviews by Thomas Refis and Jacques Garrigue) - GPR#792: fix limitations of destructive substitutions, by allowing "S with type t := type-expr", "S with type M.t := type-expr", "S with module M.N := path" (Valentin Gatien-Baron, review by Jacques Garrigue and Leo White) * GPR#1064, GPR#1392: extended indexing operators, add a new class of user-defined indexing operators, obtained by adding at least one operator character after the dot symbol to the standard indexing operators: e,g ".%()", ".?[]", ".@{}<-": let ( .%() ) = List.nth in [0; 1; 2].%(1) After this change, functions or methods with an explicit polymorphic type annotation and of which the first argument is optional now requires a space between the dot and the question mark, e.g. "<f:'a.?opt:int->unit>" m…

… to 4.07 Related: ocaml/ocaml#1010

* Return 404 on not found - Add missing 404 response code when - Package documentation not found - Package file not found - Define and use helper function html_opt. Streamlined processing of optinal data into Dream responses - Use html_opt as custom binder. - Formatting, this indents to the left a lot of code * rename let^ to let</>? and use it in all places where we can, html_opt -> http_or_404 * plain 404 response without HTML will do on static paths --------- Co-authored-by: Cuihtlauac ALVARADO <cuihtlauac@tarides.com>

ghost added the stdlib label Jan 13, 2017

Drup reviewed Jan 13, 2017

View reviewed changes

Drup reviewed Jan 16, 2017

View reviewed changes

gasche added the caml-weekly-news label Jan 23, 2017

Chris00 mentioned this pull request Jan 23, 2017

Change the names of internal modules to Lacaml__M mmottl/lacaml#25

Closed

c-cube mentioned this pull request Jan 24, 2017

Stdlib functional iterators #1002

Merged

dra27 mentioned this pull request Jan 27, 2017

add resource-handling IO functions in pervasives #640

Closed

nojb mentioned this pull request Feb 27, 2018

Add Stdlib.Float module #1638

Merged

gasche mentioned this pull request Mar 16, 2018

ocamldoc: honour nostdlib flag #1662

Merged

ghost mentioned this pull request Apr 11, 2018

ocamldoc Makefile: change target names to avoid useless recompilation #1717

Merged

damiendoligez mentioned this pull request Jun 7, 2018

Fixes for 4.07 ocaml-ppx/ppx_import#23

Closed

ghost mentioned this pull request Jul 16, 2018

Fix 7821 #1908

Merged

jakoblichtenberg mentioned this pull request Sep 6, 2018

Build against ocaml-4.06 and ocaml-4.07 fails xavierleroy/camlidl#6

Closed

dra27 mentioned this pull request Oct 3, 2019

Fix MSVC build for 4.07-4.09 ocaml/stdlib-shims#12

Merged

trefis mentioned this pull request Nov 8, 2019

Update creation of initial environment ocaml/merlin#1052

Merged

wictory added a commit to wictory/facile that referenced this pull request Dec 11, 2019

Change Pervasives. -> Stdlib. Bump the minimal required ocaml version…

0e7240b

… to 4.07 Related: ocaml/ocaml#1010

dra27 mentioned this pull request Jan 21, 2021

Change the Standard Library prefixing to match Dune's #10169

Merged

dbuenzli mentioned this pull request Mar 26, 2021

No synopsis rendered for modules for Stdlib modules ocaml/odoc#632

Closed

This pull request was closed.

Conversation

ghost commented Jan 13, 2017 • edited by ghost Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Motivations

1. we can add new modules to the stdlib without guilt

2. it's easier to use an alternative stdlib

User visible changes

Patch details

Type printing

Uh oh!

c-cube commented Jan 13, 2017

Uh oh!

Drup Jan 13, 2017

Choose a reason for hiding this comment

Uh oh!

ghost Jan 16, 2017

Choose a reason for hiding this comment

Uh oh!

Drup commented Jan 13, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

v-gb commented Jan 14, 2017

Uh oh!

hcarty commented Jan 14, 2017

Uh oh!

ghost commented Jan 16, 2017

Uh oh!

Chris00 commented Jan 16, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ghost commented Jan 16, 2017

Uh oh!

bobzhang commented Jan 16, 2017

Uh oh!

ghost commented Jan 16, 2017

Uh oh!

Drup commented Jan 16, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Drup Jan 16, 2017

Choose a reason for hiding this comment

Uh oh!

ghost Jan 17, 2017

Choose a reason for hiding this comment

Uh oh!

ghost commented Jan 17, 2017

Uh oh!

Octachron commented Jan 17, 2017

Uh oh!

ghost commented Jan 18, 2017

Uh oh!

Octachron commented Jan 18, 2017

Uh oh!

alainfrisch commented Jan 23, 2017

Uh oh!

ghost commented Jan 23, 2017

Uh oh!

dra27 commented Jan 23, 2017

Uh oh!

ghost commented Jan 24, 2017 • edited by damiendoligez Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dra27 commented Jan 25, 2017

Uh oh!

ghost commented Jan 25, 2017

Uh oh!

lpw25 commented Jan 25, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ghost commented Feb 14, 2018

Uh oh!

hhugo commented Feb 17, 2018

Uh oh!

ghost commented Feb 19, 2018

Uh oh!

gasche commented Feb 20, 2018

Uh oh!

shindere commented Feb 20, 2018 via email

Uh oh!

gasche commented Feb 20, 2018

ghost commented Jan 13, 2017 •

edited by ghost

Loading

Drup commented Jan 13, 2017 •

edited

Loading

Chris00 commented Jan 16, 2017 •

edited

Loading

Drup commented Jan 16, 2017 •

edited

Loading

ghost commented Jan 24, 2017 •

edited by damiendoligez

Loading

lpw25 commented Jan 25, 2017 •

edited

Loading

lpw25 commented Feb 20, 2018 •

edited

Loading