Failed to parse language config: duplicate field comment-tokens

[ofzo]

there is an error for parse language.toml when start helix, I try to use the default language config , and it works good, I don't know where has an error.

$ hx .
Failed to parse language config: duplicate field `comment-tokens`
in `language`

Press <ENTER> to continue with default language config

the content of my language config is too long, I put it on https://gist.github.com/ofzo/2c631daecad4b21ba9c0b59c4af38cd1

[the-mikedavis]

See #11666, this is most likely from adding a comment-tokens configuration for a language which has a comment-token configuration in the languages.toml in the root of this repo. (Or vice versa, languages.toml could set comment-tokens while custom language config sets comment-token.)

[mapsgeek]

i have the same issue without knowing where's the doublication coming from!! i just copy the languages.toml in the repo!
can you please elaborate how to debug this issue?! there's no other custom language configs

[pascalkuthe]

You should never copy the config file from the repo. Its inckuded in the binary. Only add what you want to overwrite in your config. The configs are merged automatically.

[Drakx]

See #11666, this is most likely from adding a comment-tokens configuration for a language which has a comment-token configuration in the languages.toml in the root of this repo. (Or vice versa, languages.toml could set comment-tokens while custom language config sets comment-token.)

Using this info I was able to resolve by going through each 'comment-token' and removing one by one until it was solved. Thanks.

[lenninlc]

I was able to solve it by narrowing down the 5 languages that I needed. It seems that the issue comes from comment-tokens entry somewhere in the languages.toml file(https://github.com/helix-editor/helix/blob/master/languages.toml). Hope it helps.

[rcortezk9]

I'm getting the same error and from what I'm understanding comment-tokens should be changed to comment-tokens is this correct or is this still an ongoing issue?

[Fotkurz]

Having this issue as well with the languages.toml from master.
It would be cool to know which block is the issue in but Idk if this is possible.

[rcortezk9]

Thank you for replying. I took the advice to add the languages I was working with and that resolved it.

[Hina-Kagiyama]

I somehow figured out the problematic configuration: on line 2059, the configuration of R language.

# [[language]]
# name = "r"
# scope = "source.r"
# injection-regex = "(r|R)"
# file-types = ["r", "R", { glob = ".Rprofile" }, { glob = "Rprofile.site" }, { glob = ".RHistory" }]
# shebangs = ["r", "R"]
# comment-tokens = ["#", "#'"]
# indent = { tab-width = 2, unit = "  " }
# language-servers = [ "r" ]

# [[grammar]]
# name = "r"
# source = { git = "https://github.com/r-lib/tree-sitter-r", rev = "cc04302e1bff76fa02e129f332f44636813b0c3c" }

Got this by binary search manually.

[Hina-Kagiyama]

More specifically, the "duplicated field" is related to the [[language]] part, and the [[grammar]] part works right.

I don't know why either.

[danilasar]

Deleting or commenting out these lines didn't helped me Apparently, it not enough, there're other bugs.

[scapoor]

In addition to R language, there is issue with gleam language.
After commenting both in languages.toml, it was working. However, why it is not working before commenting is still a mystery.

This is valid for the latest languages.toml in main repository.

diff --git a/languages.toml b/languages.toml
index 8f8b9ac..8a34141 100644
--- a/languages.toml
+++ b/languages.toml
@@ -2087,16 +2087,16 @@ language-servers = [ "solc" ]
 name = "solidity"
 source = { git = "https://github.com/JoranHonig/tree-sitter-solidity", rev = "f7f5251a3f5b1d04f0799b3571b12918af177fc8" }

-[[language]]
-name = "gleam"
-scope = "source.gleam"
-injection-regex = "gleam"
-file-types = ["gleam"]
-roots = ["gleam.toml"]
-comment-tokens = ["//", "///", "////"]
-indent = { tab-width = 2, unit = "  " }
-language-servers = [ "gleam" ]
-auto-format = true
+# [[language]]
+# name = "gleam"
+# scope = "source.gleam"
+# injection-regex = "gleam"
+# file-types = ["gleam"]
+# roots = ["gleam.toml"]
+# comment-tokens = ["//", "///", "////"]
+# indent = { tab-width = 2, unit = "  " }
+# language-servers = [ "gleam" ]
+# auto-format = true

 [[grammar]]
 name = "gleam"
@@ -2138,15 +2138,15 @@ language-servers = [ "robotframework_ls" ]
 name = "robot"
 source = { git = "https://github.com/Hubro/tree-sitter-robot", rev = "322e4cc65754d2b3fdef4f2f8a71e0762e3d13af" }

-[[language]]
-name = "r"
-scope = "source.r"
-injection-regex = "(r|R)"
-file-types = ["r", "R", { glob = ".Rprofile" }, { glob = "Rprofile.site" }, { glob = ".RHistory" }]
-shebangs = ["r", "R"]
-comment-tokens = ["#", "#'"]
-indent = { tab-width = 2, unit = "  " }
-language-servers = [ "r" ]
+# [[language]]
+# name = "r"
+# scope = "source.r"
+# injection-regex = "(r|R)"
+# file-types = ["r", "R", { glob = ".Rprofile" }, { glob = "Rprofile.site" }, { glob = ".RHistory" }]
+# shebangs = ["r", "R"]
+# comment-tokens = ["#", "#'"]
+# indent = { tab-width = 2, unit = "  " }
+# language-servers = [ "r" ]

 [[grammar]]
 name = "r"

Additionally, simply replacing the comment-tokens with comment-token for both gleam and r fixes this as well.