working-conventions/git/submodules.md

@@ -12,7 +12,7 @@ General information about Git's submodule support:
 
 Initial cloning of GHC HEAD (into the folder `./ghc`) is a simple as:
 
-```plaintext
+```
 git clone --recursive https://gitlab.haskell.org/ghc/ghc
 ```
 
@@ -24,7 +24,7 @@ Sometimes when you pull in new commits, the authors updated a submodule. After p
 
 At the top-level of `ghc.git` working copy:
 
-```plaintext
+```
 git pull --rebase
 git submodule update --init
 ```
@@ -61,7 +61,7 @@ git -C libraries/Cabal checkout -b wip/$BRANCH_NAME
 
 We can now proceed with making the desired changes in the submodule and commit them as usual:
 
-```plaintext
+```
 # perform modifications and as many `git {add,rm}`s as you deem necessary
 $EDITOR libraries/Cabal/src/somefile.hs
 # commit
@@ -138,13 +138,13 @@ To land a GHC MR that requires changes to Haddock, you will need to coordinate y
 * Get your MR in a mergeable state: approved and passing CI. The `lint-submods` test should give a warning at this point, but it will not cause the CI to fail.
 * Open an MR against the `ghc-head` branch of Haddock **on GitLab**, and get it merged.
 * If necessary, update the GHC MR to point to the correct ghc-head haddock commit. If the commit hash didn't change, then you can skip this step.
-* At this point, the GHC MR should be passing CI, **including** the `lint-submods` test. It can then be assigned to `marge-bot` and merged.HH
+* At this point, the GHC MR should be passing CI, **including** the `lint-submods` test. It can then be assigned to `marge-bot` and merged.
 
 # Troubleshooting
 
 Git sometimes produces a rather cryptic error messages when pushing the GHC repo with unpushed submodule changes:
 
-```plaintext
+```
 fatal: remote 'myremote' not configured
 fatal: process for submodule 'utils/haddock' failed
 fatal: the remote end hung up unexpectedly