golangci: try to set up golangci

The ebpf libraries are not helping at all.

Fix alegrey91#28

.github/workflows/tests.yaml

@@ -47,6 +47,34 @@ jobs:
           name: unit-test
           path: /tmp/unit/
 
+  linting:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@f43a0e5ff2bd294095638e18286ca9a3d1956744 # ratchet:actions/checkout@v3
+
+      - name: Set up Go
+        uses: actions/setup-go@93397bea11091df50f3d7e59dc26a7711a8bcfbe # ratchet:actions/setup-go@v4
+        with:
+          go-version: '1.21'
+
+      - name: Install Dependencies
+        run: |
+          sudo apt update
+          sudo apt install -y clang
+          sudo apt install -y libbpf-dev
+          sudo apt install -y libseccomp-dev
+
+      - name: Build coverage-instrumented binary
+        run: |
+          make build-static-libbpfgo
+          make build-bpf
+
+      - name: Golangci-lint
+        uses: golangci/[email protected]
+        with:
+          # TODO: find a way to pass this the same way we pass it to go test
+          args: internal/analyzer internal/archiver internal/elfreader internal/embeddable internal/executor internal/metadata internal/privileged internal/seccomputils internal/writer
+
   integration-test:
 
     runs-on: ubuntu-latest
@@ -76,7 +104,7 @@ jobs:
           mkdir -p /tmp/integration
           # we have to run integration tests one-by-one
           # otherwhise they will run in parallel.
-          # since harpoon apply network forwards, these could 
+          # since harpoon apply network forwards, these could
           # interact with each other and make the test fail.
           go test \
             -exec sudo \

.golangci.yml

@@ -0,0 +1,164 @@
+---
+# golangci-lint configuration file made by @ccoVeille
+# Source: https://github.com/ccoVeille/golangci-lint-config-examples/
+# Author: @ccoVeille
+# License: MIT
+#
+issues:
+  # Maximum issues count per one linter.
+  # Set to 0 to disable the limit (Default: 50)
+  max-issues-per-linter: 0
+  # Maximum count of issues with the same text.
+  # Set to 0 to disable the limit (Default: 3)
+  max-same-issues: 0
+
+run:
+  # Exit code when at least one issue was found (Default: 1)
+  # Using 0 allow to report all issues, without blocking the CI
+  issues-exit-code: 0
+
+linters:
+  # some linters are enabled by default
+  # https://golangci-lint.run/usage/linters/
+  #
+  # enable some extra linters
+  enable:
+    # Errcheck is a program for checking for unchecked errors in Go code.
+    - errcheck
+
+    # Linter for Go source code that specializes in simplifying code.
+    - gosimple
+
+    # Vet examines Go source code and reports suspicious constructs.
+    - govet
+
+    # Detects when assignments to existing variables are not used.
+    - ineffassign
+
+    # It's a set of rules from staticcheck. See https://staticcheck.io/
+    - staticcheck
+
+    # Fast, configurable, extensible, flexible, and beautiful linter for Go.
+    # Drop-in replacement of golint.
+    - revive
+
+    # check imports order and makes it always deterministic.
+    - gci
+
+    # make sure to use t.Helper() when needed
+    - thelper
+
+    # mirror suggests rewrites to avoid unnecessary []byte/string conversion
+    - mirror
+
+    # detect the possibility to use variables/constants from the Go standard library.
+    - usestdlibvars
+
+    # Detects common programming mistakes.
+    - errorlint
+
+    # Finds commonly misspelled English words.
+    - misspell
+
+    # Checks for duplicate words in the source code.
+    - dupword
+
+    # Errorlint is a linter for that can be used to find code that will cause problems with the error wrapping scheme introduced
+    - errorlint
+
+linters-settings:
+  gci:  # define the section orders for imports
+    sections:
+      # Standard section: captures all standard packages.
+      - standard
+      # Default section: catchall that is not standard or custom
+      - default
+      # linters that related to local tool, so they should be separated
+      - localmodule
+
+  revive:
+    rules:
+      # these are the default revive rules
+      # you can remove the whole "rules" node if you want
+      # BUT
+      # ! /!\ they all need to be present when you want to add more rules than the default ones
+      # otherwise, you won't have the default rules, but only the ones you define in the "rules" node
+
+      # Blank import should be only in a main or test package, or have a comment justifying it.
+      - name: blank-imports
+
+      # context.Context() should be the first parameter of a function when provided as argument.
+      - name: context-as-argument
+        arguments:
+          - allowTypesBefore: "*testing.T"
+
+      # Basic types should not be used as a key in `context.WithValue`
+      - name: context-keys-type
+
+      # Importing with `.` makes the programs much harder to understand
+      - name: dot-imports
+
+      # Empty blocks make code less readable and could be a symptom of a bug or unfinished refactoring.
+      - name: empty-block
+
+      # for better readability, variables of type `error` must be named with the prefix `err`.
+      - name: error-naming
+
+      # for better readability, the errors should be last in the list of returned values by a function.
+      - name: error-return
+
+      # for better readability, error messages should not be capitalized or end with punctuation or a newline.
+      - name: error-strings
+
+      # report when replacing `errors.New(fmt.Sprintf())` with `fmt.Errorf()` is possible
+      - name: errorf
+
+      # incrementing an integer variable by 1 is recommended to be done using the `++` operator
+      - name: increment-decrement
+
+      # highlights redundant else-blocks that can be eliminated from the code
+      - name: indent-error-flow
+
+      # This rule suggests a shorter way of writing ranges that do not use the second value.
+      - name: range
+
+      # receiver names in a method should reflect the struct name (p for Person, for example)
+      - name: receiver-naming
+
+      # redefining built in names (true, false, append, make) can lead to bugs very difficult to detect.
+      - name: redefines-builtin-id
+
+      # redundant else-blocks that can be eliminated from the code.
+      - name: superfluous-else
+
+      # prevent confusing name for variables when using `time` package
+      - name: time-naming
+
+      # warns when an exported function or method returns a value of an un-exported type.
+      - name: unexported-return
+
+      # spots and proposes to remove unreachable code. also helps to spot errors
+      - name: unreachable-code
+
+      # Functions or methods with unused parameters can be a symptom of an unfinished refactoring or a bug.
+      - name: unused-parameter
+
+      # report when a variable declaration can be simplified
+      - name: var-declaration
+
+      # warns when initialism, variable or package naming conventions are not followed.
+      - name: var-naming
+
+      - name: comment-spacings
+
+      - name: unhandled-error
+
+  dupword:
+    # Keywords used to ignore detection.
+    # Default: []
+    ignore:
+    #  - "blah" # this will accept "blah blah …" as a valid duplicate word
+
+  misspell:
+    # Setting locale to US will correct the British spelling of 'colour' to 'color'.
+    locale: US