Replace Python core with pure-Vimscript core ("vimcore") #119

cxw42 · 2019-04-19T14:12:21Z

Here it is - let me know what you think! Passes all tests except Windows UTF-8, which @ppalaga says might not be possible (editorconfig/editorconfig-core-c#31 (comment)).

Closes editorconfig/editorconfig#383 .
Closes #48 (replaces it).

Also closes the open Python issues:
Closes #57.
Closes #101.
Closes #116.

May help with #108.
May affect #111.

Part of the discussion in vim/vim#2286 .

cxw42 · 2019-04-20T11:23:21Z

The CI failure appears to be related to rake/bundler/vimrunner versions. Would someone with some Ruby expertise be able to give me some suggestions?

xuhdev · 2019-04-20T18:35:20Z

I guess you have generated the lock file in a new version of bundler. The easiest way to resolve this is probably use bundler less than 2.0 to generate Gemfile.lock on your system.

This reverts commit e1c2557. Thanks to editorconfig#119 (comment) for the suggestion. All tests now pass with `bundle _1.15.4_ exec rake`.

cxw42 · 2019-04-22T12:15:41Z

@xuhdev Thanks - that worked! I was able to run the tests on Cygwin from editorconfig-vim/ with

export BUNDLE_GEMFILE=tests/Gemfile
bundle _1.15.4_ exec rspec tests/spec/editorconfig_spec.rb

Please let me know if you would like any changes, or if you have any concerns.

cxw42 · 2019-04-26T15:49:44Z

@xuhdev Any thoughts? I am wondering if we can merge this one. After that, I would like to work with @rakus to see if we can merge some of his plugin changes into this official plugin. I think removing the Python should come before any Vimscript enhancements. What say you?

xuhdev

Do you have CI set up?

xuhdev · 2019-04-26T22:27:53Z

I remember you said something about number range ({-3...100}) and this core depends on that. I wanted to start a voting for that I couldn't find that thread. Could you remind me where it is?

cxw42 · 2019-04-28T01:16:08Z

@xuhdev

CI

The CI (Travis) is unchanged in this PR. The core is copied from https://github.com/cxw42/editorconfig-core-vimscript , which has its own CI (Appveyor) set up.

number range

The number range is editorconfig/editorconfig#371 . Thank you for following up on it!

I changed this PR so it does not depend on that issue. It has the current behaviour (e.g., {-10..10} rejects 0). Therefore, I think this can go ahead even before the vote on number ranges.

xuhdev · 2019-04-28T02:18:59Z

I understand you haven't changed anything about CI. What I meant is that, if you add a core library to the repo, we need CI to test that core library for guaranteeing its continuously correct functioning...

cxw42 · 2019-04-28T02:25:56Z

@xuhdev Sorry for misunderstanding! I have not but am happy to. However, the vimscript core requires a number of support files in order to be testable by ctest (e.g., bash scripts and bat files). May I add a separate repo submodule for those so plugin users don't get files they don't need?

xuhdev · 2019-04-28T02:32:00Z

I personally feel it's OK to add editorconfig/editorconfig-core-test as a submodule, but all other scripts IMO are integrated part of this subproject. Users in general get testing scripts, same for most other cases if they are supposed to get things from source. In addition, test scripts are unlikely to use a lot of spaces anyway.

cxw42 · 2019-04-28T11:35:32Z

@xuhdev OK - I'll update this repo. I may ask to revisit this point later if Bram decides he's willing to bundle this plugin with Vim.

This reverts commit e1c2557. Thanks to editorconfig#119 (comment) for the suggestion. All tests now pass with `bundle _1.15.4_ exec rake`.

xuhdev

Regarding to your new changes, I've made another comment. You also seem to have changed EOL of a few files. Just wanna make sure you intended it. After resolving this comment, it should be fine to merge now. We can make adjustment later once the specification version format is determined.

autoload/editorconfig_core.vim

cxw42 · 2019-05-19T13:39:39Z

@xuhdev #119 (review)

Regarding to your new changes, I've made another comment. You also seem to have changed EOL of a few files. Just wanna make sure you intended it.

Would you please let me know which files you are looking at? I just looked through git diff upstream/master and didn't see any EOL changes. All files currently have Unix (LF) line endings, which I think is reasonable. (Except for the crlf.in file from editorconfig-core-test, which needs to have CRLF to function! :) )

There are a number of renames. For example, I moved tests/ to tests/plugin so that I could create tests/core. Is that perhaps what you are looking at?

xuhdev · 2019-05-20T06:41:48Z

Perhaps I'm getting confused. Could you tell me, for example, what has changed in tests/core/ecvimlib.ps1 since then? I've seen a whole file change here.

cxw42 · 2019-05-20T10:34:56Z

Ah! Sure. Compared to master, that is a new file. Everything intests/core is copied in from editorconfig-core-vimscript per your earlier request to combine the repos. But I see that GitHub, at least on mobile, does not make that clear at all!

I'll take a look at that specific commit. I think I was using crlf in the other repo and regularized it to lf here. Edit That is indeed the case - I was using CRLF for files on Windows systems while developing the core. My thought at the time was that Windows users who opened Windows-specific files in Notepad would be able to read those files if they were CRLF. I will change those back and clean up the history a bit.

- Added "vim_core" mode. - Use the Vimscript core as the default if it is installed and the user has not specified a different core. - Also, for robustness, accept core modes case-insensitively. - Added VimScript core to documentation

@xuhdev

Per @xuhdev in editorconfig/editorconfig#383 (comment) From cxw42/editorconfig-core-vimscript@4495efd Removed caching code, which is not yet used, per @xuhdev in editorconfig#119 (comment)

- Remove the Python core from this plugin - Remove all plugin code related to modes other than vim_core - Update the documentation and README accordingly Note on tests: All tests now pass with `bundle _1.15.4_ exec rake`. Thanks to editorconfig#119 (comment) for the suggestion on how to make this work.

:EditorConfigEnable, :EditorConfigDisable

- This makes room for adding core tests. - Also, renamed and updated plugin_tests submodule

Scripts are modified from cxw42/editorconfig-core-vimscript@5d85d10 Windows-specific files have CRLF line endings; other files have LF line endings.

- Added the core tests to the existing Travis configuration - Added Appveyor configuration to permit testing on Windows. The configuration is modified from cxw42/editorconfig-core-vimscript@5d85d10 - Updated the README to include both badges. For now, they point to cxw42/editorconfig-vim; that will need to be updated when this branch is merged.

cxw42 · 2019-05-20T11:30:49Z

@xuhdev I cleaned up and squashed some. The CRLF files are now consistent throughout the commits in which they occur, and the .editorconfig file specifies those line ends appropriately. The previous commits are still available in cxw42/vimcore-old if you want to compare.

Also, fixed typo in Appveyor configuration.

xuhdev · 2019-05-21T03:45:29Z

Thank you for your hard work!

cxw42 · 2019-05-21T14:06:34Z

@xuhdev Happy to help!

When you have a chance, would you please set up Appveyor for this repo under editorconfig and update README.md? It currently points to my Appveyor account. No hurry, but I think it would be a good idea to continue running those tests.

xuhdev · 2019-05-21T20:44:55Z

Done!

xuhdev · 2019-06-04T20:48:08Z

Sorry for the late discussion on this issue, but I realized that we should have retained the ability to use an external core library as an option (the one that calls the external editorconfig executable, not the Python builtin). Could you restore the availability of that option, if this is not too difficult? We have done something similar to Emacs before: editorconfig/editorconfig-emacs#46 (comment) I'm OK with defaulting to the internal one, but an option to use an external one might still be important.

Lots of updates, including moving to pure vimscript (vimcore) See also: editorconfig/editorconfig-vim#119

xuhdev reviewed Apr 26, 2019

View reviewed changes

cxw42 force-pushed the vimcore branch from bee10bc to b7f348b Compare April 28, 2019 14:32

cxw42 force-pushed the vimcore branch 16 times, most recently from c4c183e to 5079f24 Compare April 30, 2019 14:59

cxw42 force-pushed the vimcore branch 2 times, most recently from b3126ad to 6549e37 Compare May 8, 2019 13:57

cxw42 force-pushed the vimcore branch from 6549e37 to 53d117a Compare May 16, 2019 11:10

xuhdev reviewed May 17, 2019

View reviewed changes

autoload/editorconfig_core.vim Outdated Show resolved Hide resolved

Chris White and others added 7 commits May 20, 2019 06:56

Added support for Vimscript core

c151747

- Added "vim_core" mode. - Use the Vimscript core as the default if it is installed and the user has not specified a different core. - Also, for robustness, accept core modes case-insensitively. - Added VimScript core to documentation

Copied in editorconfig-core-vimscript

7051a8d

Per @xuhdev in editorconfig/editorconfig#383 (comment) From cxw42/editorconfig-core-vimscript@4495efd Removed caching code, which is not yet used, per @xuhdev in editorconfig#119 (comment)

Add commands to enable/disable plugin

9059c63

:EditorConfigEnable, :EditorConfigDisable

Moved plugin tests from tests/ to tests/plugin/

62e60a0

- This makes room for adding core tests. - Also, renamed and updated plugin_tests submodule

Added core_tests submodule and scripts

1b3c91f

Scripts are modified from cxw42/editorconfig-core-vimscript@5d85d10 Windows-specific files have CRLF line endings; other files have LF line endings.

cxw42 force-pushed the vimcore branch from 6a229d0 to e5b2a4b Compare May 20, 2019 11:15

cxw42 force-pushed the vimcore branch from e5b2a4b to 823d61d Compare May 20, 2019 12:23

Updated license and documentation

7d48003

Also, fixed typo in Appveyor configuration.

cxw42 force-pushed the vimcore branch from 823d61d to 7d48003 Compare May 20, 2019 16:56

xuhdev approved these changes May 21, 2019

View reviewed changes

xuhdev merged commit 37bedf8 into editorconfig:master May 21, 2019

This was referenced Jun 5, 2019

optional graceful failure to load core #62

Closed

Add back in the option of using an external core #121

Closed

krishnakumarg1984 mentioned this pull request Jun 24, 2019

The claim on README.md is no longer true (editorconfig-vim requires if_python interface or external python interpreter.) sgur/vim-editorconfig#45

Closed

tony added a commit to tony/vim-config-framework that referenced this pull request Feb 23, 2020

Use editorconfig/editorconfig-vim again

4d3f47a

Lots of updates, including moving to pure vimscript (vimcore) See also: editorconfig/editorconfig-vim#119

cxw42 deleted the vimcore branch October 30, 2022 20:03

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Replace Python core with pure-Vimscript core ("vimcore") #119

Replace Python core with pure-Vimscript core ("vimcore") #119

cxw42 commented Apr 19, 2019

cxw42 commented Apr 20, 2019

xuhdev commented Apr 20, 2019

cxw42 commented Apr 22, 2019

cxw42 commented Apr 26, 2019

xuhdev left a comment

xuhdev commented Apr 26, 2019

cxw42 commented Apr 28, 2019

xuhdev commented Apr 28, 2019

cxw42 commented Apr 28, 2019

xuhdev commented Apr 28, 2019

cxw42 commented Apr 28, 2019

xuhdev left a comment

cxw42 commented May 19, 2019

xuhdev commented May 20, 2019

cxw42 commented May 20, 2019 •

edited

Loading

cxw42 commented May 20, 2019

xuhdev commented May 21, 2019

cxw42 commented May 21, 2019

xuhdev commented May 21, 2019

xuhdev commented Jun 4, 2019

Replace Python core with pure-Vimscript core ("vimcore") #119

Replace Python core with pure-Vimscript core ("vimcore") #119

Conversation

cxw42 commented Apr 19, 2019

cxw42 commented Apr 20, 2019

xuhdev commented Apr 20, 2019

cxw42 commented Apr 22, 2019

cxw42 commented Apr 26, 2019

xuhdev left a comment

Choose a reason for hiding this comment

xuhdev commented Apr 26, 2019

cxw42 commented Apr 28, 2019

xuhdev commented Apr 28, 2019

cxw42 commented Apr 28, 2019

xuhdev commented Apr 28, 2019

cxw42 commented Apr 28, 2019

xuhdev left a comment

Choose a reason for hiding this comment

cxw42 commented May 19, 2019

xuhdev commented May 20, 2019

cxw42 commented May 20, 2019 • edited Loading

cxw42 commented May 20, 2019

xuhdev commented May 21, 2019

cxw42 commented May 21, 2019

xuhdev commented May 21, 2019

xuhdev commented Jun 4, 2019

cxw42 commented May 20, 2019 •

edited

Loading