Add more icons from eza #37
Conversation
WalkthroughThis pull request introduces a new submodule for the "eza" project and adds infrastructure for generating and managing JSON files containing icon mappings for file names and extensions. The changes include adding a new Git submodule, creating Perl and shell scripts to generate JSON files, introducing a Nix shell configuration, and updating Vim autoload scripts to incorporate additional icon mappings from the newly generated JSON files. Changes
Poem
Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media? 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
CodeRabbit Configuration File (
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🧹 Nitpick comments (4)
scripts/generate-extension.json.pl (1)
15-17: Add input validation for line transformations.The current regex transformations assume specific input format. Consider adding validation:
+next if /^\s*$/; # Skip empty lines +next if /^\s*\/\//; # Skip comment-only lines +die "Malformed input line: $_" unless /=>/; # Validate line format s|\s*=>.*//\s*|: "|; s|$|",|; print $_;assets/json/basename.json (1)
2-62: Consider grouping related entries and adding missing icons.The dotfile entries could benefit from better organization:
- Group shell-related files (
.bash_*,.zsh_*)- Group version control files (
.git*)- Add missing icons for config files
assets/json/extension.json (2)
88-88: Consider using Windows-specific icons for Windows-related extensionsThe icons for Windows-specific extensions like
exeanddllcurrently use generic icons. Consider using Windows-specific icons to better represent these file types.- "dll": "", + "dll": "", - "exe": "", + "exe": ""Also applies to: 116-116
386-386: Standardize icons for subtitle-related extensionsThe subtitle-related extensions (srt, ssa, sub) currently use "" which seems more related to Scala/SBT. Consider using a more appropriate media-related icon for consistency.
- "srt": "", + "srt": "", - "ssa": "", + "ssa": "", - "sub": "", + "sub": ""Also applies to: 388-388, 393-393
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (8)
.gitmodules(1 hunks)assets/json/basename.json(1 hunks)assets/json/extension.json(3 hunks)eza(1 hunks)scripts/generate-basename.json.pl(1 hunks)scripts/generate-extension.json.pl(1 hunks)scripts/generate-json.sh(1 hunks)shell.nix(1 hunks)
✅ Files skipped from review due to trivial changes (4)
- eza
- .gitmodules
- shell.nix
- scripts/generate-json.sh
🔇 Additional comments (3)
assets/json/extension.json (3)
66-66: LGTM! Good use of download icons
The use of download icons ("") for temporary download files (*.part, *.crdownload) is appropriate and consistent.
Also applies to: 294-294
370-374: LGTM! Consistent use of hash icons
Good consistency in using the hash icon ("") across all SHA-related file extensions.
Line range hint 1-482: Verify JSON validity and key uniqueness
Let's ensure the JSON is valid and there are no duplicate keys.
✅ Verification successful
JSON structure verified successfully
The file is well-structured with valid JSON syntax and contains no duplicate keys.
🏁 Scripts executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash
# Description: Verify JSON validity and check for duplicate keys
# Check JSON validity
jq '.' assets/json/extension.json > /dev/null
if [ $? -eq 0 ]; then
echo "JSON is valid"
else
echo "Invalid JSON"
exit 1
fi
# Check for duplicate keys
DUPLICATES=$(jq -r 'to_entries | group_by(.key) | map(select(length > 1)) | map(.[0].key)[]' assets/json/extension.json)
if [ -n "$DUPLICATES" ]; then
echo "Found duplicate keys:"
echo "$DUPLICATES"
exit 1
else
echo "No duplicate keys found"
fi
Length of output: 285
| #!/usr/bin/env -S perl -n | ||
| BEGIN { | ||
| print '{'; | ||
| } |
There was a problem hiding this comment.
🛠️ Refactor suggestion
Consider adding error handling and input validation.
The script assumes well-formed input but should handle potential errors:
- Invalid input format
- Premature EOF
- Missing constant definition
#!/usr/bin/env -S perl -n
+use strict;
+use warnings;
+
BEGIN {
print '{';
+ $found_constant = 0; # Track if we found the constant
}Committable suggestion skipped: line range outside the PR's diff.
scripts/generate-basename.json.pl
Outdated
| #!/usr/bin/env -S perl -n | ||
| BEGIN { | ||
| print '{'; | ||
| } | ||
|
|
||
| END { | ||
| print ' ".": ""' . "\n}"; | ||
| } | ||
| if (/^const (?:DIRECTORY_ICONS|FILENAME_ICONS):/) { | ||
| $is_enabled = 1; | ||
| next; | ||
| } | ||
| $is_enabled = 0 if (/^};$/); | ||
| next unless $is_enabled; | ||
| s|\s*=>.*//\s*|: "|; | ||
| s|$|",|; | ||
| print $_; |
There was a problem hiding this comment.
🛠️ Refactor suggestion
Consider refactoring to reduce code duplication.
This script shares significant code with generate-extension.json.pl. Consider creating a shared module:
#!/usr/bin/env -S perl -n
+use strict;
+use warnings;
+use lib 'scripts'; # Add scripts directory to module search path
+use IconProcessor; # New shared module
-BEGIN {
- print '{';
-}
-
-END {
- print ' ".": ""' . "\n}";
-}
-if (/^const (?:DIRECTORY_ICONS|FILENAME_ICONS):/) {
- $is_enabled = 1;
- next;
-}
-$is_enabled = 0 if (/^};$/);
-next unless $is_enabled;
-s|\s*=>.*//\s*|: "|;
-s|$|",|;
-print $_;
+IconProcessor::process_input(
+ constants => ['DIRECTORY_ICONS', 'FILENAME_ICONS'],
+ output_file => 'assets/json/basename.json'
+);Committable suggestion skipped: line range outside the PR's diff.
assets/json/basename.json
Outdated
| "cron.d": "", | ||
| "cron.daily": "", | ||
| "cron.hourly": "", | ||
| "cron.minutely": "", | ||
| "cron.monthly": "", | ||
| "cron.weekly": "", | ||
| "crontab": "", | ||
| "crypttab": "", | ||
| "csh.cshrc": "", | ||
| "csh.login": "", | ||
| "csh.logout": "", | ||
| "docker-compose.yaml": "", | ||
| "docker-compose.yml": "", | ||
| "dune": "", | ||
| "dune-project": "", | ||
| "environment": "", | ||
| "etc": "", | ||
| "go.mod": "", | ||
| "go.sum": "", | ||
| "go.work": "", | ||
| "gradle": "", | ||
| "gradle.properties": "", | ||
| "gradlew": "", | ||
| "gradlew.bat": "", | ||
| "group": "", | ||
| "gruntfile.coffee": "", | ||
| "gruntfile.js": "", | ||
| "gruntfile.ls": "", | ||
| "gshadow": "", | ||
| "gulpfile.coffee": "", | ||
| "gulpfile.js": "", | ||
| "gulpfile.ls": "", | ||
| "hidden": "", | ||
| "license": "", | ||
| "heroku.yml": "", | ||
| "hi": "", | ||
| "hidden": "", | ||
| "home": "", | ||
| "hostname": "", | ||
| "id_dsa": "", | ||
| "id_ecdsa": "", | ||
| "id_ecdsa_sk": "", | ||
| "id_ed25519": "", | ||
| "id_ed25519_sk": "", | ||
| "id_rsa": "", | ||
| "include": "", | ||
| "inputrc": "", | ||
| "jsconfig.json": "", | ||
| "justfile": "", | ||
| "known_hosts": "", | ||
| "localized": "", | ||
| "makefile": "", | ||
| "mix.lock": "", | ||
| "npmignore": "", | ||
| "package.json": "", | ||
| "pkgbuild": "", | ||
| "procfile": "", | ||
| "rakefile": "", | ||
| "react.jsx": "", | ||
| "renovate.json": "", | ||
| "requirements.txt": "", | ||
| "localtime": "", | ||
| "makefile": "", | ||
| "node_modules": "", | ||
| "npm-shrinkwrap.json": "", | ||
| "npm_cache": "", | ||
| "npmrc": "", | ||
| "package-lock.json": "", | ||
| "package.json": "", | ||
| "pam.d": "", | ||
| "passwd": "", | ||
| "php.ini": "", | ||
| "platformio.ini": "", | ||
| "pom.xml": "", | ||
| "profile": "", | ||
| "pyproject.toml": "", | ||
| "release.toml": "", | ||
| "requirements.txt": "", | ||
| "robots.txt": "", | ||
| "rubydoc": "", | ||
| "tsconfig.json": "", | ||
| "yarn.lock": "" | ||
| "rvmrc": "", | ||
| "settings.gradle.kts": "", | ||
| "shadow": "", | ||
| "shells": "", | ||
| "ssh": "", | ||
| "sudoers": "", | ||
| "sudoers.d": "", | ||
| "timezone": "", | ||
| "tsconfig.json": "", | ||
| "webpack.config.js": "", | ||
| "xbps.d": "", | ||
| "xorg.conf.d": "", | ||
| "yarn.lock": "", | ||
| "zlogin": "", | ||
| "zlogout": "", | ||
| "zprofile": "", | ||
| "zshenv": "", | ||
| "zshrc": "" |
There was a problem hiding this comment.
💡 Codebase verification
Icon inconsistencies found and missing development files need to be added
.prettierrcis already handled in pattern.json with icon , but missing in basename.json.editorconfigis missing in both pattern.json and basename.json- License files (COPYING, LICENSE) and Makefile variants use empty icons "", which is consistent with the project's style
- Add entries for:
.editorconfig: "".prettierrc: "" (matching pattern.json)
🔗 Analysis chain
Verify icon consistency across similar file types.
Some observations:
- License files (COPYING, LICENSE) use empty icons
- Build files (Makefile variants) could use consistent icons
- Some common development files are missing (e.g.,
.editorconfig,.prettierrc)
Let me check the pattern.json file to see how these development files are handled there.
🏁 Scripts executed
The following scripts were executed for the analysis:
Script:
#!/bin/bash
# Check for common development files that might be missing
rg -l '\.editorconfig|\.prettierrc' || echo "Files not found in codebase"
Length of output: 62
Script:
#!/bin/bash
# Check how development files are handled in pattern.json
cat assets/json/pattern.json | grep -A 5 -B 5 '\.editorconfig\|\.prettierrc'
Length of output: 466
|
I think the idea of using eza’s icons is excellent. However, incorporating this approach as it is in the current version poses challenges in terms of compatibility with previous versions. Since this plugin is licensed under MIT, you are free to modify it as long as the license requirements are met. How about forking it into a completely new plugin, such as vim-nerdfont-eza, and hosting it in the eza-community repository? |
|
fork shouldn't be a good choice. It will bring confusion and split the power of contributors.
Currently, eza have more icons than this repo. we can utilize it to provide better user experience.
|
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
scripts/generate-basename.json.pl (2)
1-1: Limit usage of the -n option in the shebang.
Currently the script is invoked with “-n” which processes the input line-by-line, printing automatically. If you want fine-grained control, confirm whether you need it. Otherwise, you can remove or adapt it for clarity.
2-4: Encapsulate JSON opening brace.
Keeping the JSON printing logic within BEGIN/END blocks is fine but consider wrapping them in a function or module for testability.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (10)
.gitmodules(1 hunks)assets/json/extra/basename.json(1 hunks)assets/json/extra/extension.json(1 hunks)autoload/nerdfont/path/basename.vim(1 hunks)autoload/nerdfont/path/extension.vim(1 hunks)eza(1 hunks)scripts/generate-basename.json.pl(1 hunks)scripts/generate-extension.json.pl(1 hunks)scripts/generate-json.sh(1 hunks)shell.nix(1 hunks)
✅ Files skipped from review due to trivial changes (1)
- assets/json/extra/basename.json
🚧 Files skipped from review as they are similar to previous changes (5)
- eza
- .gitmodules
- scripts/generate-extension.json.pl
- scripts/generate-json.sh
- shell.nix
🔇 Additional comments (3)
autoload/nerdfont/path/extension.vim (1)
2-2: Confirm consistency between 'basename' and 'extension' merges.
You're merging 'extension' with 'extra/basename'. Make sure the combination is intentional and does not cause key collisions or inconsistent icon associations.
✅ Verification successful
Let me gather information about the JSON files being merged and their potential key collisions.
This will help us:
- See the structure and content of the JSON files being merged
- Find any related files that might provide context about the intended merge behavior
Let me try a different approach to find and examine the JSON files being used.
This will help us:
- Understand where and how the JSON files are loaded
- Locate the actual JSON files being merged
- See the full context of the extension.vim file
Let's examine the contents of these JSON files to check for potential collisions.
This will help us:
- See the actual content of both JSON files being merged
- Identify any potential key collisions
- Understand the structure and purpose of each file
Merge is intentionally designed and safe to use
After examining both JSON files, the merge is safe and intentional because:
extension.jsoncontains file extension mappings (e.g., ".js", ".py")extra/basename.jsoncontains specific filename mappings (e.g., ".gitignore", "package.json")- The few overlapping keys (like ".bashrc", ".zshrc") in both files use the same icons, ensuring consistent behavior
- The merge enhances functionality by combining extension-based and filename-based icon mappings
🏁 Scripts executed
The following scripts were executed for the analysis:
Error: No shell scripts found
Script:
#!/bin/bash
# Find and show contents of JSON files related to extension and basename
fd -e json -x cat {} \; | grep -A 20 -B 20 '"extension"\|"basename"'
# Additionally, let's look for any related configuration or test files
rg -l "extension.*basename|basename.*extension"
Length of output: 160
Script:
#!/bin/bash
# Look for the nerdfont#get_json function implementation to understand where these files are located
ast-grep --pattern 'function! nerdfont#get_json'
# Search for the actual JSON files in the repository
fd 'extension.json|basename.json' --type f
# Look at the implementation of the extension.vim file
cat autoload/nerdfont/path/extension.vim
Length of output: 894
Script:
#!/bin/bash
# Check contents of the relevant JSON files
echo "=== extension.json ==="
cat assets/json/extension.json
echo -e "\n=== extra/basename.json ==="
cat assets/json/extra/basename.json
Length of output: 8628
scripts/generate-basename.json.pl (2)
9-17: Refactor repeated logic with a shared module.
This script likely shares logic with the extension generator. You could put $is_enabled checking, line rewriting, and the final JSON formatting in a shared module.
6-8: Preserve consistent trailing commas.
Including a trailing comma for the last key-value pair in JSON might break strict JSON parsers. Confirm that downstream usage is flexible enough to handle this.
| @@ -1,5 +1,5 @@ | |||
| let g:nerdfont#path#basename#customs = get(g:, 'nerdfont#path#basename#customs', {}) | |||
| let g:nerdfont#path#basename#defaults = nerdfont#get_json('basename') | |||
| let g:nerdfont#path#basename#defaults = extend(nerdfont#get_json('basename'), nerdfont#get_json('extra/basename')) | |||
There was a problem hiding this comment.
💡 Codebase verification
The concern is valid - no error handling for missing JSON files
The implementation of nerdfont#get_json doesn't include any error handling for missing JSON files. Both basename.json and extra/basename.json exist in the repository at assets/json/, but the function uses readfile() and json_decode() without any try-catch block, which would throw errors if the files are missing or corrupted.
autoload/nerdfont.vim: Add error handling aroundreadfile()andjson_decode()calls to return an empty dictionary when JSON files are missing.
🔗 Analysis chain
Ensure fallback behavior for extra JSON data.
While extending defaults with extra JSON data is beneficial, consider verifying if 'extra/basename' exists or is accessible. Otherwise, the plugin might fail if the JSON file is missing.
Let me verify the concern by checking the implementation of nerdfont#get_json function and how it handles missing JSON files.
I found the implementation of nerdfont#get_json, but let me check if there's any error handling for missing JSON files in the codebase.
Let me check the actual location of JSON files in the repository.
🏁 Scripts executed
The following scripts were executed for the analysis:
Error: No shell scripts found
Script:
#!/bin/bash
# Search for the implementation of nerdfont#get_json function
ast-grep --pattern 'function! nerdfont#get_json($_) $_'
# Also search for any error handling related to JSON files
rg -g '*.vim' 'nerdfont#get_json' -A 5
Length of output: 2922
Script:
#!/bin/bash
# Check if there's any try-catch or error handling around nerdfont#get_json
rg -g '*.vim' 'try.*nerdfont#get_json|catch' -A 3
# Also check if the JSON files exist in the repository
fd -e json basename extra/basename
Length of output: 486
Script:
#!/bin/bash
# Find all JSON files in the repository
fd -e json
# Check the content of nerdfont#get_json function with more context
rg -g '*.vim' 'function! nerdfont#get_json' -B 5 -A 10
Length of output: 1154
| "sql": "", | ||
| "sqlite3": "", | ||
| "sr": "", | ||
| "srt": "", | ||
| "ss": "", | ||
| "ssa": "", | ||
| "stl": "", | ||
| "sty": "", | ||
| "styl": "", | ||
| "stylus": "", | ||
| "sub": "", | ||
| "sublime-build": "", | ||
| "sublime-keymap": "", | ||
| "sublime-menu": "", | ||
| "sublime-options": "", | ||
| "sublime-package": "", | ||
| "sublime-project": "", | ||
| "sublime-session": "", | ||
| "sublime-settings": "", | ||
| "sublime-snippet": "", | ||
| "sublime-theme": "", | ||
| "svelte": "", | ||
| "svg": "", | ||
| "swf": "", | ||
| "swift": "", | ||
| "t": "", | ||
| "tape": "", | ||
| "tar": "", | ||
| "taz": "", | ||
| "tbz": "", | ||
| "tbz2": "", | ||
| "tc": "", | ||
| "tex": "", | ||
| "tf": "", | ||
| "tfstate": "", | ||
| "tfvars": "", | ||
| "tgz": "", | ||
| "tif": "", | ||
| "tiff": "", | ||
| "tlz": "", | ||
| "tml": "", | ||
| "toml": "", | ||
| "torrent": "", | ||
| "tres": "", | ||
| "ts": "", | ||
| "tscn": "", | ||
| "tsv": "", | ||
| "tsx": "", | ||
| "ttc": "", | ||
| "ttf": "", | ||
| "twig": "", | ||
| "txt": "", | ||
| "txz": "", | ||
| "typ": "𝐭", | ||
| "tz": "", | ||
| "tzo": "", | ||
| "unity": "", | ||
| "unity3d": "", | ||
| "v": "", | ||
| "vdi": "", | ||
| "vhd": "", | ||
| "vhs": "", | ||
| "video": "", | ||
| "vim": "", | ||
| "vmdk": "", | ||
| "vob": "", | ||
| "vue": "", | ||
| "war": "", | ||
| "wav": "", | ||
| "webm": "", | ||
| "webmanifest": "", | ||
| "webp": "", | ||
| "whl": "", | ||
| "windows": "", | ||
| "wma": "", | ||
| "wmv": "", | ||
| "woff": "", | ||
| "woff2": "", | ||
| "wv": "", | ||
| "xcf": "", | ||
| "xci": "", | ||
| "xhtml": "", | ||
| "xlr": "", | ||
| "xls": "", | ||
| "xlsm": "", | ||
| "xlsx": "", | ||
| "xml": "", | ||
| "xpm": "", | ||
| "xul": "", | ||
| "xz": "", | ||
| "yaml": "", | ||
| "yml": "", | ||
| "z": "", | ||
| "z64": "", | ||
| "zig": "", | ||
| "zip": "", | ||
| "zsh": "", | ||
| "zsh-theme": "", | ||
| "zst": "" | ||
| } |
There was a problem hiding this comment.
💡 Codebase verification
After analyzing the output, I can now generate the final response as I've found the main extension mappings in assets/json/extension.json to compare against the extra entries.
Several icon collisions detected between built-in and extra extension mappings
The following extensions have conflicting icon definitions:
.(root): Both files define it with different icons ("" vs "")android,apk,avi,avif,bash,bat, and many others have duplicate entries- Some extensions in the extra file have different icons than the built-in ones (e.g.,
cs,vue)
This could lead to inconsistent icon display depending on which file takes precedence during the merge. Consider:
- Removing duplicate entries from the extra extension file
- Ensuring intentional overwrites are documented
- Maintaining a single source of truth for extension mappings
🔗 Analysis chain
Check for collisions between built-in and extra entries.
Merging extension icons from multiple sources can lead to collisions or overwrites. Confirm that these entries don’t conflict with the main JSON.
Let me gather information about potential collisions between built-in and extra entries.
🏁 Scripts executed
The following scripts were executed for the analysis:
Error: No shell scripts found
Script:
#!/bin/bash
# Find all JSON files that might contain extension definitions
fd -e json -t f | while read -r file; do
if [ "$file" != "assets/json/extra/extension.json" ]; then
echo "=== $file ==="
# Use jq to check if it's an object with string values (likely extension mappings)
if jq -e 'select(type == "object" and all(.[]; type == "string"))' "$file" > /dev/null 2>&1; then
cat "$file"
fi
fi
done
Length of output: 13870
Use scripts/generate-json.sh to generate: - assets/json/extension.json - assets/json/basename.json
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
assets/json/extra/extension.json (1)
8-8: Well-organized icon assignments with good consistencyThe icon assignments show good consistency across related file types. For example:
- Encryption-related files (.age, .gpg) use security-themed icons
- Terraform-related files (.tf, .tfstate, .tfvars) use consistent Terraform icons
This makes the icon set intuitive and user-friendly.
Consider adding icons for these modern file types if they're supported by eza:
- Container-related: .dockerfile, .containerfile
- Package manager files: package.json, Cargo.toml
- GitHub workflow files: .github/workflows/*.yml
Also applies to: 151-151, 416-418
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (10)
.gitmodules(1 hunks)assets/json/extra/basename.json(1 hunks)assets/json/extra/extension.json(1 hunks)autoload/nerdfont/path/basename.vim(1 hunks)autoload/nerdfont/path/extension.vim(1 hunks)eza(1 hunks)scripts/generate-basename.json.pl(1 hunks)scripts/generate-extension.json.pl(1 hunks)scripts/generate-json.sh(1 hunks)shell.nix(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (9)
- eza
- shell.nix
- assets/json/extra/basename.json
- scripts/generate-basename.json.pl
- scripts/generate-extension.json.pl
- .gitmodules
- autoload/nerdfont/path/extension.vim
- scripts/generate-json.sh
- autoload/nerdfont/path/basename.vim
🔇 Additional comments (2)
assets/json/extra/extension.json (2)
1-482: LGTM: Well-structured JSON format
The JSON structure is well-formed with consistent formatting and proper key-value pairs.
1-482: Verify consistency with eza's icon mappings
Since these icons are sourced from eza, let's verify they match eza's current icon set.
| @@ -0,0 +1,482 @@ | |||
| { | |||
| ".": "", | |||
There was a problem hiding this comment.
Critical: Icon collisions with built-in mappings detected
As noted in previous reviews, there are icon collisions with the built-in mappings. For example:
- Root directory "." has a different icon
- "android" and "apk" have duplicate entries
This could lead to inconsistent icon display depending on which mapping takes precedence.
Consider one of these approaches:
- Remove duplicate entries and only keep icons that don't exist in the built-in mappings
- Document which mappings intentionally override built-in ones and why
- Implement a clear precedence rule in the code that loads these mappings
Also applies to: 14-14, 16-16
|
It depends on the purpose of this PR. If the purpose of this PR is simply to “add icons,” it would be welcomed. In that case, including a mechanism for synchronization wouldn’t be necessary. Based on the changes in this PR, which include a synchronization script, I assumed it aligns with the latter purpose and hence suggested forking. |
|
OK, I'll remove the mechanism for synchronization. |
Add more icons from eza
Use scripts/generate-json.sh to generate:
https://github.com/orgs/eza-community/discussions/1270
Perhaps upstrem can provide json files directly.
Summary by CodeRabbit
New Features
Bug Fixes
Chores