-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
refactor: implement lexing and multi-mode parsing engine #21
base: master
Are you sure you want to change the base?
refactor: implement lexing and multi-mode parsing engine #21
Conversation
Impressive !
Le mar. 8 oct. 2019 à 06:45, Paulo Valente <notifications@github.com> a
écrit :
… This is still work in progress, but the main idea is already implemented.
TODO.md cites what is still missing (and should be done by the end of the
week).
The core idea here is that we'll be able to parse list comprehensions
differently than normal code, and comments are now dealt with as they
should be. In both cases, I relied on the concept of parsing modes, which
ultimately change the way certain tokens are parsed. For example, in
default mode, an ìftoken would expect anendtoken to close its block,
while incomprehensionmode, it wouldn't, and while incomment` mode, it
would be ignored altogether.
------------------------------
You can view, comment on, or merge this pull request online at:
#21
Commit Summary
- refactor: clean-up string matching engine
- fix: single line ignore
- fix: match elixir properly
- wip
File Changes
- *A* TODO.md
<https://github.com/noplay/vscode-rainbow-end/pull/21/files#diff-0>
(6)
- *M* package.json
<https://github.com/noplay/vscode-rainbow-end/pull/21/files#diff-1>
(4)
- *M* src/extension.ts
<https://github.com/noplay/vscode-rainbow-end/pull/21/files#diff-2>
(238)
- *M* src/languages.ts
<https://github.com/noplay/vscode-rainbow-end/pull/21/files#diff-3>
(53)
- *A* src/tokenizer.ts
<https://github.com/noplay/vscode-rainbow-end/pull/21/files#diff-4>
(116)
Patch Links:
- https://github.com/noplay/vscode-rainbow-end/pull/21.patch
- https://github.com/noplay/vscode-rainbow-end/pull/21.diff
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#21>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AACUKXKIWO2W2I7Z7MRLPQDQNQF45ANCNFSM4I6M3TXA>
.
|
@noplay I'm gonna need some help with the language definitions. I think that maybe the Other than that, I'm opening the PR to review! |
One improvement that needs addressing before merging is that the nesting for comprehensions is not implemented properly (only counting instead of using stacks). |
Any update? :) |
I've been kinda busy (and sick) these last weeks, but I intend to get back to this as soon as possible! |
Sorry it took so long, guys. I got sick, and then was busy with other things. Code is now ready for review (and possible merging). |
Finally got time to test sorry for the long time . I found two issue. One for ruby. if and unless are not highlited correctly . It's normal because they where implemented in a special list ignoreInDelimiters to support: puts "Hello" if a == 3 def test
if a == 9000 || b < 3
if c > 5
if b == 7
elsif x == 3
end
end
todo.each do |dd|
end
puts "Hello" if a == 3
puts "World" if a == 3
end
end The second issue is after changing multiple time to different tab the extension stop working. I'm not sure why. My demo env use a lot of different language (verilog, vhdl...). I'm still trying to figure what is wrong. |
Awesome work ! |
In other news, I was thinking of looking into LSP. This should solve most of our problems, if we are able to fetch code block starters and such -- however, I'd still need to look if LSP provides all needed info. |
Any news? |
The core idea here is that we'll be able to parse list comprehensions differently than normal code, and comments are now dealt with as they should be. In both cases, I relied on the concept of parsing modes, which ultimately change the way certain tokens are parsed. For example, in
default
mode, an ìftoken would expect an
endtoken to close its block, while in
comprehensionmode, it wouldn't, and while in
comment` mode, it would be ignored altogether.