Modernisation of elisp code and conformance with coding guidelines by purcell · Pull Request #1247 · ocaml/merlin

purcell · 2021-01-27T04:34:30Z

I set about tidying up the merlin elisp to match GNU and package conventions and best practices. By targeting a post-2015 Emacs, we also improve the out-of-the-box experience by automatically enabling the xref backend. Everything is split up into self-explanatory commits, and I'm happy to make adjustments if desired.

I've tested this locally, including its operation with the merlin-eldoc package. (I actually think the core eldoc behaviour should be included in merlin itself and I might make a follow-up PR to add that.)

purcell · 2021-01-27T04:41:22Z

Also: ultimately it should be possible to have a CI action for the elisp code, at least to make sure it byte-compiles cleanly and conforms to package conventions (ie. package-lint).

Company will always be loaded when this code is evaluated

This code will execute immediately anyway

auto-complete is likely used in other buffers too, so hard-coding this binding across all of them is inappropriate. Instead, we set it up locally in each buffer. This is still not entirely appropriate, so we only do it as part of the "easy" setup.

purcell · 2021-01-29T19:48:16Z

Rebased against conflicting changes in master.

Khady · 2021-01-30T05:18:12Z

Note that if there are changes in merlin-eldoc that would help with the elisp that lives in this repository, especially to reduce the maintenance cost, I am happy to accept them

Fourchaux · 2021-02-19T18:25:50Z

Eventually, I will add these missing sharp quotes:

merlin-ac.el:
142 (define-key ac-complete-mode-map (kbd "C-c C-l") 'ac-merlin-locate)
142 (define-key ac-complete-mode-map (kbd "C-c C-l") #'ac-merlin-locate)

151 (add-hook 'merlin-mode-hook 'merlin-ac--setup)
151 (add-hook 'merlin-mode-hook #'merlin-ac--setup)

merlin-cap.el:
80 (add-hook 'merlin-mode-hook 'merlin-cap--setup)
80 (add-hook 'merlin-mode-hook #'merlin-cap--setup)

merlin-company.el:
20 ;;   (add-hook 'after-init-hook 'global-company-mode)
20 ;;   (add-hook 'after-init-hook #'global-company-mode)

merlin.el:
1218 (add-hook 'pre-command-hook 'merlin--type-enclosing-reset-hooked))
1218 (add-hook 'pre-command-hook #'merlin--type-enclosing-reset-hooked))

1116 (define-key keymap (kbd "C-<up>") 'merlin-type-enclosing-go-up)
1116 (define-key keymap (kbd "C-<up>") #'merlin-type-enclosing-go-up)

1117 (define-key keymap (kbd "C-<down>") 'merlin-type-enclosing-go-down)
1117 (define-key keymap (kbd "C-<down>") #'merlin-type-enclosing-go-down)

1118 (define-key keymap (kbd "C-d") 'merlin-destruct-enclosing)
1118 (define-key keymap (kbd "C-d") #'merlin-destruct-enclosing)

1119 (define-key keymap (kbd "C-w") 'merlin-copy-enclosing)
1119 (define-key keymap (kbd "C-w") #'merlin-copy-enclosing)

1733 (define-key merlin-map (kbd "C-c C-x") 'merlin-error-next)
1733 (define-key merlin-map (kbd "C-c C-x") #'merlin-error-next)

1734 (define-key merlin-map (kbd "C-c C-l") 'merlin-locate)
1734 (define-key merlin-map (kbd "C-c C-l") #'merlin-locate)

1735 (define-key merlin-map (kbd "C-c &"  ) 'merlin-pop-stack)
1735 (define-key merlin-map (kbd "C-c &"  ) #'merlin-pop-stack)

1736 (define-key merlin-map (kbd "C-c C-r") 'merlin-error-check)
1736 (define-key merlin-map (kbd "C-c C-r") #'merlin-error-check)

1737 (define-key merlin-map (kbd "C-c C-t") 'merlin-type-enclosing)
1737 (define-key merlin-map (kbd "C-c C-t") #'merlin-type-enclosing)

1738 (define-key merlin-map (kbd "C-c C-d") 'merlin-destruct)
1738 (define-key merlin-map (kbd "C-c C-d") #'merlin-destruct)

1739 (define-key merlin-map (kbd "C-c C-n") 'merlin-phrase-next)
1739 (define-key merlin-map (kbd "C-c C-n") #'merlin-phrase-next)

1740 (define-key merlin-map (kbd "C-c C-p") 'merlin-phrase-prev)
1740 (define-key merlin-map (kbd "C-c C-p") #'merlin-phrase-prev)

purcell · 2021-02-21T03:27:40Z

Eventually, I will add these missing sharp quotes:

Sure, I can do that as part of this PR.

@Fourchaux

Thanks @Fourchaux

purcell · 2021-03-02T00:14:07Z

The build failures seem spurious here, by the way.

pmetzger · 2021-03-02T01:04:58Z

This looks cool. Is it going to be merged?

emacs/merlin.el

purcell · 2021-03-02T04:18:36Z

@purcell you should consider using the lsp frontend. There's not much motivation to maintain editor specific backends anymore.

Yes, I probably will start doing so before long. :-)

let-def · 2021-03-02T08:28:32Z

@trefis this PR looks good to me. I'll merge soon unless there's objections.

Can you wait a bit? Some users built custom extensions using merlin elisp functions, we need to make sure this is not going to break anything.

rgrinberg · 2021-03-02T16:07:25Z

Can you wait a bit? Some users built custom extensions using merlin elisp functions, we need to make sure this is not going to break anything.

Sure, I thought @Khady already expressed that he was happy with the changes. Let's hear from him again though.

trefis · 2021-03-02T16:12:38Z

We were thinking about people from janestreet actually. We've notified them offline.

bschommer · 2021-03-02T17:00:31Z

emacs/merlin-iedit.el

+                          "-identifier-at" (merlin-unmake-point (point)))))
      (when r
        (if (listp r)
            (flet ((iedit-printable (a)


How about replacing the flet here with a cl-letf that would remove a usage of a deprecated function?

Suggested change

(flet ((iedit-printable (a)

(cl-letf ((iedit-printable (a)

Thanks, that's not quite the right formulation because it's not a drop-in equivalent, but I've made a subsequent change in this spirit. 👍

The var will always be bound at this point, because the iedit feature has been required.

let-def · 2021-04-07T17:06:51Z

Thanks for your work :)

purcell · 2021-04-08T02:25:18Z

Thanks @let-def -- if anyone reports issues subsequently, please CC me and I will help. The packages on MELPA will be split according to the optional third-party package dependencies now: merlin, merlin-ac, merlin-ioccur and merlin-company. I'll submit a follow-up PR to update the README here accordingly.

See ocaml/merlin#1247

@Fourchaux

…idelines (ocaml#1247) * [emacs] Remove unnecessary eval-after-load Company will always be loaded when this code is evaluated * Require cl-lib instead of deprecated cl * Assume Emacs 24.3 and enable lexical-binding * Eliminate dynamically-scoped var * Fix warnings about unused lexical variables * Remove redundant eval-after-load This code will execute immediately anyway * Inline merlin-setup * Assume Emacs 25.1 (2016 release) and enable xref backend unconditionally * Don't modify the global ac-complete-mode-map auto-complete is likely used in other buffers too, so hard-coding this binding across all of them is inappropriate. Instead, we set it up locally in each buffer. This is still not entirely appropriate, so we only do it as part of the "easy" setup. * Make merlin-ac a separate package with correct dependencies * Make merlin-company a separate package with correct dependencies * Make merlin-iedit a separate package with correct dependencies * Avoid "merlin/" prefix, which breaks elisp naming conventions * Prefer cl-case / pcase to repetitive conds * Fix comment types * merlin-lookup is just "assoc-default" * Define obsolete function aliases to keep merlin-eldoc compatible * Fix prefix of bounds-of-ocaml-atom-at-point * Prefer to sharp-quote functions when adding hooks etc. Thanks @Fourchaux * Ignore .elc files * merlin-iedit: drop references to now-removed var and function in iedit * Prefer cl-letf to deprecated flet * Remove redundant bound-and-true-p test The var will always be bound at this point, because the iedit feature has been required.

@Fourchaux

…idelines (ocaml#1247) * [emacs] Remove unnecessary eval-after-load Company will always be loaded when this code is evaluated * Require cl-lib instead of deprecated cl * Assume Emacs 24.3 and enable lexical-binding * Eliminate dynamically-scoped var * Fix warnings about unused lexical variables * Remove redundant eval-after-load This code will execute immediately anyway * Inline merlin-setup * Assume Emacs 25.1 (2016 release) and enable xref backend unconditionally * Don't modify the global ac-complete-mode-map auto-complete is likely used in other buffers too, so hard-coding this binding across all of them is inappropriate. Instead, we set it up locally in each buffer. This is still not entirely appropriate, so we only do it as part of the "easy" setup. * Make merlin-ac a separate package with correct dependencies * Make merlin-company a separate package with correct dependencies * Make merlin-iedit a separate package with correct dependencies * Avoid "merlin/" prefix, which breaks elisp naming conventions * Prefer cl-case / pcase to repetitive conds * Fix comment types * merlin-lookup is just "assoc-default" * Define obsolete function aliases to keep merlin-eldoc compatible * Fix prefix of bounds-of-ocaml-atom-at-point * Prefer to sharp-quote functions when adding hooks etc. Thanks @Fourchaux * Ignore .elc files * merlin-iedit: drop references to now-removed var and function in iedit * Prefer cl-letf to deprecated flet * Remove redundant bound-and-true-p test The var will always be bound at this point, because the iedit feature has been required.

@Fourchaux

…idelines (#1247) * [emacs] Remove unnecessary eval-after-load Company will always be loaded when this code is evaluated * Require cl-lib instead of deprecated cl * Assume Emacs 24.3 and enable lexical-binding * Eliminate dynamically-scoped var * Fix warnings about unused lexical variables * Remove redundant eval-after-load This code will execute immediately anyway * Inline merlin-setup * Assume Emacs 25.1 (2016 release) and enable xref backend unconditionally * Don't modify the global ac-complete-mode-map auto-complete is likely used in other buffers too, so hard-coding this binding across all of them is inappropriate. Instead, we set it up locally in each buffer. This is still not entirely appropriate, so we only do it as part of the "easy" setup. * Make merlin-ac a separate package with correct dependencies * Make merlin-company a separate package with correct dependencies * Make merlin-iedit a separate package with correct dependencies * Avoid "merlin/" prefix, which breaks elisp naming conventions * Prefer cl-case / pcase to repetitive conds * Fix comment types * merlin-lookup is just "assoc-default" * Define obsolete function aliases to keep merlin-eldoc compatible * Fix prefix of bounds-of-ocaml-atom-at-point * Prefer to sharp-quote functions when adding hooks etc. Thanks @Fourchaux * Ignore .elc files * merlin-iedit: drop references to now-removed var and function in iedit * Prefer cl-letf to deprecated flet * Remove redundant bound-and-true-p test The var will always be bound at this point, because the iedit feature has been required.

@ulugbekna

CHANGES: Tue Apr 12 11:44:22 AM CET 2021 + merlin binary - external configuration reading: + use relative paths to communicate with Dune when possible. This solves issues related to symlinks on Unix and improve Windows support (ocaml/merlin#1271, fixes ocaml/merlin#1288) + make the `workdir` configuration value when using the `dune ocaml-merlin` configuration provider the same as when using `dot-merlin-reader` so that ppxes behaves in the same way as before (ocaml/merlin#1284, fixes ocaml/dune#4479, discussion in ocaml/merlin#1292) - destruct: + improve prefixing of generated constructors in Destruct by filtering opened modules (ocaml/merlin#1277) + make the destruct command more resilient to ill-typed expressions and when called without nodes (ocaml/merlin#1304, fixes ocaml/merlin#1300) - reintroduce some record recovery and improve completion (ocaml/merlin#1276) - introduce a new AST node for holes (`_`), allow correct typing of these holes and add a new `holes` command that returns the locations of all holes in the current file along with their types (ocaml/merlin#1242, ocaml/merlin#1289) - Mppx: don't restore cookies after invocation. Ppx are invoked only once so there is no need to manage cookies. This small change should increase performance and should not change any other behavior (ocaml/merlin#1309) - Windows: system command variant: do not open a window console when launching a ppx (ocaml/merlin#1270, fixes ocaml/merlin#714) - fix same file documentation bug (ocaml/merlin#1265 by @ulugbekna, fixes ocaml/merlin#1261) + editor modes - vim: Add `MerlinNextHole` and `MerlinPreviousHole` commands to navigate between holes. Jump to the first hole after destruct (ocaml/merlin#1287, ocaml/merlin#1303) - emacs: Add `merlin-next-hole` and `merlin-previous-hole` commands to navigate holes. Jump to the first hole after calling destruct. (ocaml/merlin#1291) - emacs: modernization of the elisp code and conformance with coding guidelines (ocaml/merlin#1247, ocaml/merlin#1310 by Steve Purcell ) - vim & emacs : new client-side "merlin use package" commands, restoring previous behavior (ocaml/merlin#1272, fixes ocaml/merlin#1191) + test suite - cover constructor disambiguation and record fields (ocaml/merlin#1276) - cover the new `holes` command and AST node (ocaml/merlin#1242, ocaml/merlin#1289) - cover the document fix (ocaml/merlin#1265, ocaml/merlin#1315)

@ulugbekna

CHANGES: Tue Apr 12 11:44:22 AM CET 2021 + merlin binary - external configuration reading: + use relative paths to communicate with Dune when possible. This solves issues related to symlinks on Unix and improve Windows support (ocaml/merlin#1271, fixes ocaml/merlin#1288) + make the `workdir` configuration value when using the `dune ocaml-merlin` configuration provider the same as when using `dot-merlin-reader` so that ppxes behaves in the same way as before (ocaml/merlin#1284, fixes ocaml/dune#4479, discussion in ocaml/merlin#1292) - destruct: + improve prefixing of generated constructors in Destruct by filtering opened modules (ocaml/merlin#1277) + make the destruct command more resilient to ill-typed expressions and when called without nodes (ocaml/merlin#1304, fixes ocaml/merlin#1300) - reintroduce some record recovery and improve completion (ocaml/merlin#1276) - introduce a new AST node for holes (`_`), allow correct typing of these holes and add a new `holes` command that returns the locations of all holes in the current file along with their types (ocaml/merlin#1242, ocaml/merlin#1289) - Mppx: don't restore cookies after invocation. Ppx are invoked only once so there is no need to manage cookies. This small change should increase performance and should not change any other behavior (ocaml/merlin#1309) - Windows: system command variant: do not open a window console when launching a ppx (ocaml/merlin#1270, fixes ocaml/merlin#714) - fix same file documentation bug (ocaml/merlin#1265 by @ulugbekna, fixes ocaml/merlin#1261) + editor modes - vim: Add `MerlinNextHole` and `MerlinPreviousHole` commands to navigate between holes. Jump to the first hole after destruct (ocaml/merlin#1287, ocaml/merlin#1303) - emacs: Add `merlin-next-hole` and `merlin-previous-hole` commands to navigate holes. Jump to the first hole after calling destruct. (ocaml/merlin#1291) - emacs: modernization of the elisp code and conformance with coding guidelines (ocaml/merlin#1247, ocaml/merlin#1310 by Steve Purcell ) - vim & emacs : new client-side "merlin use package" commands, restoring previous behavior (ocaml/merlin#1272, fixes ocaml/merlin#1191) + test suite - cover constructor disambiguation and record fields (ocaml/merlin#1276) - cover the new `holes` command and AST node (ocaml/merlin#1242, ocaml/merlin#1289) - cover the document fix (ocaml/merlin#1265, ocaml/merlin#1315)

@ulugbekna

CHANGES: Tue Apr 12 11:44:22 AM CET 2021 + merlin binary - external configuration reading: + use relative paths to communicate with Dune when possible. This solves issues related to symlinks on Unix and improve Windows support (ocaml/merlin#1271, fixes ocaml/merlin#1288) + make the `workdir` configuration value when using the `dune ocaml-merlin` configuration provider the same as when using `dot-merlin-reader` so that ppxes behaves in the same way as before (ocaml/merlin#1284, fixes ocaml/dune#4479, discussion in ocaml/merlin#1292) - destruct: + improve prefixing of generated constructors in Destruct by filtering opened modules (ocaml/merlin#1277) + make the destruct command more resilient to ill-typed expressions and when called without nodes (ocaml/merlin#1304, fixes ocaml/merlin#1300) - reintroduce some record recovery and improve completion (ocaml/merlin#1276) - introduce a new AST node for holes (`_`), allow correct typing of these holes and add a new `holes` command that returns the locations of all holes in the current file along with their types (ocaml/merlin#1242, ocaml/merlin#1289) - Mppx: don't restore cookies after invocation. Ppx are invoked only once so there is no need to manage cookies. This small change should increase performance and should not change any other behavior (ocaml/merlin#1309) - Windows: system command variant: do not open a window console when launching a ppx (ocaml/merlin#1270, fixes ocaml/merlin#714) - fix same file documentation bug (ocaml/merlin#1265 by @ulugbekna, fixes ocaml/merlin#1261) + editor modes - vim: Add `MerlinNextHole` and `MerlinPreviousHole` commands to navigate between holes. Jump to the first hole after destruct (ocaml/merlin#1287, ocaml/merlin#1303) - emacs: Add `merlin-next-hole` and `merlin-previous-hole` commands to navigate holes. Jump to the first hole after calling destruct. (ocaml/merlin#1291) - emacs: modernization of the elisp code and conformance with coding guidelines (ocaml/merlin#1247, ocaml/merlin#1310 by Steve Purcell ) - vim & emacs : new client-side "merlin use package" commands, restoring previous behavior (ocaml/merlin#1272, fixes ocaml/merlin#1191) + test suite - cover constructor disambiguation and record fields (ocaml/merlin#1276) - cover the new `holes` command and AST node (ocaml/merlin#1242, ocaml/merlin#1289) - cover the document fix (ocaml/merlin#1265, ocaml/merlin#1315)

@ulugbekna

CHANGES: Tue Apr 12 11:44:22 AM CET 2021 + merlin binary - external configuration reading: + use relative paths to communicate with Dune when possible. This solves issues related to symlinks on Unix and improve Windows support (ocaml/merlin#1271, fixes ocaml/merlin#1288) + make the `workdir` configuration value when using the `dune ocaml-merlin` configuration provider the same as when using `dot-merlin-reader` so that ppxes behaves in the same way as before (ocaml/merlin#1284, fixes ocaml/dune#4479, discussion in ocaml/merlin#1292) - destruct: + improve prefixing of generated constructors in Destruct by filtering opened modules (ocaml/merlin#1277) + make the destruct command more resilient to ill-typed expressions and when called without nodes (ocaml/merlin#1304, fixes ocaml/merlin#1300) - reintroduce some record recovery and improve completion (ocaml/merlin#1276) - introduce a new AST node for holes (`_`), allow correct typing of these holes and add a new `holes` command that returns the locations of all holes in the current file along with their types (ocaml/merlin#1242, ocaml/merlin#1289) - Mppx: don't restore cookies after invocation. Ppx are invoked only once so there is no need to manage cookies. This small change should increase performance and should not change any other behavior (ocaml/merlin#1309) - Windows: system command variant: do not open a window console when launching a ppx (ocaml/merlin#1270, fixes ocaml/merlin#714) - fix same file documentation bug (ocaml/merlin#1265 by @ulugbekna, fixes ocaml/merlin#1261) + editor modes - vim: Add `MerlinNextHole` and `MerlinPreviousHole` commands to navigate between holes. Jump to the first hole after destruct (ocaml/merlin#1287, ocaml/merlin#1303) - emacs: Add `merlin-next-hole` and `merlin-previous-hole` commands to navigate holes. Jump to the first hole after calling destruct. (ocaml/merlin#1291) - emacs: modernization of the elisp code and conformance with coding guidelines (ocaml/merlin#1247, ocaml/merlin#1310 by Steve Purcell ) - vim & emacs : new client-side "merlin use package" commands, restoring previous behavior (ocaml/merlin#1272, fixes ocaml/merlin#1191) + test suite - cover constructor disambiguation and record fields (ocaml/merlin#1276) - cover the new `holes` command and AST node (ocaml/merlin#1242, ocaml/merlin#1289) - cover the document fix (ocaml/merlin#1265, ocaml/merlin#1315)

@ulugbekna

CHANGES: Tue Apr 12 11:44:22 AM CET 2021 + merlin binary - external configuration reading: + use relative paths to communicate with Dune when possible. This solves issues related to symlinks on Unix and improve Windows support (ocaml/merlin#1271, fixes ocaml/merlin#1288) + make the `workdir` configuration value when using the `dune ocaml-merlin` configuration provider the same as when using `dot-merlin-reader` so that ppxes behaves in the same way as before (ocaml/merlin#1284, fixes ocaml/dune#4479, discussion in ocaml/merlin#1292) - destruct: make the destruct command more resilient to ill-typed expressions and when called without nodes (ocaml/merlin#1304, fixes ocaml/merlin#1300) - Mppx: don't restore cookies after invocation. Ppx are invoked only once so there is no need to manage cookies. This small change should increase performance and should not change any other behavior (ocaml/merlin#1309) - windows: + system command variant: do not open a window console when launching a ppx (ocaml/merlin#1270, fixes ocaml/merlin#714) + fix Emacs hanging when starting Merlin (ocaml/merlin#1263) + fix path canonicalization (ocaml/merlin#1254) - fix same file documentation bug (ocaml/merlin#1265 by @ulugbekna, fixes ocaml/merlin#1261) + editor modes - emacs: + modernization of the elisp code and conformance with coding guidelines (ocaml/merlin#1247, ocaml/merlin#1310 by Steve Purcell ) + use opam var where applicable (ocaml/merlin#1310) + fix "wrong number of argument" (ocaml/merlin#1250 by @atharvashukla, fixes ocaml/merlin#1234) + fix for Neovim's CursorMoved semantics (ocaml/merlin#1213 by @ddickstein) - vim & emacs : new client-side "merlin use package" commands, restoring previous behavior (ocaml/merlin#1272, fixes ocaml/merlin#1191) + test suite - cover the document fix (ocaml/merlin#1265, ocaml/merlin#1315)

This reverts commit 3c0a72f.

…oding guidelines (ocaml#1247)" This reverts commit 4e382a7.

…lines (ocaml#1247)" This reverts commit 30c52bf.

purcell added 12 commits January 30, 2021 08:46

[emacs] Remove unnecessary eval-after-load

cbd020d

Company will always be loaded when this code is evaluated

Require cl-lib instead of deprecated cl

ac4b20a

Assume Emacs 24.3 and enable lexical-binding

c2c3407

Eliminate dynamically-scoped var

75ec7e9

Fix warnings about unused lexical variables

7937fe0

Remove redundant eval-after-load

8730ca8

This code will execute immediately anyway

Inline merlin-setup

ac9fcfc

Assume Emacs 25.1 (2016 release) and enable xref backend unconditionally

252f4a8

Make merlin-ac a separate package with correct dependencies

f216f66

Make merlin-company a separate package with correct dependencies

dcacb23

Make merlin-iedit a separate package with correct dependencies

f7de2fb

purcell force-pushed the modernisation branch from 2bdf3fb to deed331 Compare January 29, 2021 19:48

purcell added 7 commits February 21, 2021 16:46

Avoid "merlin/" prefix, which breaks elisp naming conventions

883d593

Prefer cl-case / pcase to repetitive conds

f23e875

Fix comment types

00615e5

merlin-lookup is just "assoc-default"

8e30e76

Define obsolete function aliases to keep merlin-eldoc compatible

d7b4731

Fix prefix of bounds-of-ocaml-atom-at-point

f722b4d

Prefer to sharp-quote functions when adding hooks etc.

1469d12

Thanks @Fourchaux

purcell force-pushed the modernisation branch from 748c578 to 1469d12 Compare February 21, 2021 03:47

Ignore .elc files

cfefaad

rgrinberg reviewed Mar 2, 2021

View reviewed changes

emacs/merlin.el Show resolved Hide resolved

bschommer reviewed Mar 2, 2021

View reviewed changes

purcell added 3 commits March 3, 2021 11:12

merlin-iedit: drop references to now-removed var and function in iedit

e6fa7c7

Prefer cl-letf to deprecated flet

c1f88df

Remove redundant bound-and-true-p test

a5846be

The var will always be bound at this point, because the iedit feature has been required.

bschommer mentioned this pull request Mar 3, 2021

Use cl-letf instead of flet. (#1268) #1274

Closed

Drup mentioned this pull request Apr 5, 2021

Tentative fix for iedit. #1301

Closed

voodoos mentioned this pull request Apr 7, 2021

Prepare for 4.2 release #1299

Merged

32 tasks

Merge branch 'master' into modernisation

f24366f

let-def merged commit 30c52bf into ocaml:master Apr 7, 2021

purcell added a commit to melpa/melpa that referenced this pull request Apr 8, 2021

Split merlin into multiple packages

17598d9

See ocaml/merlin#1247

purcell mentioned this pull request Apr 8, 2021

Update Emacs usage docs where applicable #1310

Merged

voodoos added a commit that referenced this pull request Apr 8, 2021

[backport] Modernisation of elisp code (#1247)

3c0a72f

antalsz added a commit to antalsz/merlin that referenced this pull request Jul 14, 2021

Revert "[backport] Modernisation of elisp code (ocaml#1247)"

911a790

This reverts commit 3c0a72f.

antalsz added a commit to antalsz/merlin that referenced this pull request Jul 14, 2021

Revert "[backport] Modernisation of elisp code and conformance with c…

4e10287

…oding guidelines (ocaml#1247)" This reverts commit 4e382a7.

antalsz added a commit to antalsz/merlin that referenced this pull request Jul 15, 2021

Revert "Modernisation of elisp code and conformance with coding guide…

f3bad3f

…lines (ocaml#1247)" This reverts commit 30c52bf.

purcell deleted the modernisation branch February 2, 2022 10:35

Conversation

purcell commented Jan 27, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

purcell commented Jan 27, 2021

Uh oh!

purcell commented Jan 29, 2021

Uh oh!

Khady commented Jan 30, 2021

Uh oh!

Fourchaux commented Feb 19, 2021

Uh oh!

purcell commented Feb 21, 2021

Uh oh!

purcell commented Mar 2, 2021

Uh oh!

pmetzger commented Mar 2, 2021

Uh oh!

Uh oh!

purcell commented Mar 2, 2021

Uh oh!

let-def commented Mar 2, 2021

Uh oh!

rgrinberg commented Mar 2, 2021

Uh oh!

trefis commented Mar 2, 2021

Uh oh!

bschommer Mar 2, 2021

Choose a reason for hiding this comment

Uh oh!

purcell Mar 2, 2021

Choose a reason for hiding this comment

Uh oh!

let-def commented Apr 7, 2021

Uh oh!

purcell commented Apr 8, 2021

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

purcell commented Jan 27, 2021 •

edited

Loading