Add support to expand glob patterns

cmd/root.go

@@ -36,6 +36,7 @@ import (
  "github.com/get-woke/woke/pkg/parser"
  "github.com/get-woke/woke/pkg/printer"
 
+ "github.com/bmatcuk/doublestar/v4"
  "github.com/mitchellh/go-homedir"
  "github.com/rs/zerolog"
  "github.com/rs/zerolog/log"
@@ -121,7 +122,11 @@ func rootRunE(cmd *cobra.Command, args []string) error {
  return err
  }
 
- findings := p.ParsePaths(print, parseArgs(args)...)
+ files, err := parseArgs(args)
+ if err != nil {
+ return err
+ }
+ findings := p.ParsePaths(print, files...)
 
  if exitOneOnFailure && findings > 0 {
  // We intentionally return an error if exitOneOnFailure is true, but don't want to show usage
@@ -162,16 +167,28 @@ func GetRootCmd() cobra.Command {
  return *rootCmd
 }
 
-func parseArgs(args []string) []string {
+// parseArgs parses the command-line positional arguments that contain file glob patterns.
+// If no argument is provided, return the default path (current directory).
+// Perform glob pattern expansion.
+func parseArgs(args []string) ([]string, error) {
  if len(args) == 0 {
  args = parser.DefaultPath
  }
-
+ var files []string
  if stdin {
- args = []string{os.Stdin.Name()}
+ files = []string{os.Stdin.Name()}
+ } else {
+ // Perform glob expansion.
+ for _, arg := range args {
+ f, err := doublestar.FilepathGlob(arg)
+ if err != nil {
+ return nil, err
+ }
+ files = append(files, f...)
+ }
  }
 
- return args
+ return files, nil
 }
 
 func setDebugLogLevel() {

cmd/root_test.go

@@ -2,14 +2,18 @@ package cmd
 
 import (
  "bytes"
+ "fmt"
  "io"
  "os"
+ "path/filepath"
  "regexp"
+ "strings"
  "testing"
 
  "github.com/get-woke/woke/pkg/output"
  "github.com/get-woke/woke/pkg/parser"
 
+ "github.com/bmatcuk/doublestar/v4"
  "github.com/mitchellh/go-homedir"
  "github.com/rs/zerolog"
  "github.com/spf13/cobra"
@@ -20,9 +24,12 @@ import (
 // run profiling with
 // go test -v -cpuprofile cpu.prof -memprofile mem.prof -bench=. ./cmd
 // memory:
-// go tool pprof mem.prof
+//
+// go tool pprof mem.prof
+//
 // cpu:
-// go tool pprof cpu.prof
+//
+// go tool pprof cpu.prof
 func BenchmarkRootRunE(b *testing.B) {
  zerolog.SetGlobalLevel(zerolog.NoLevel)
  output.Stdout = io.Discard
@@ -70,12 +77,147 @@ func TestParseArgs(t *testing.T) {
  t.Cleanup(func() {
  stdin = false
  })
- assert.Equal(t, parser.DefaultPath, parseArgs([]string{}))
- assert.Equal(t, []string{"../.."}, parseArgs([]string{"../.."}))
+ tests := []struct {
+ stdin bool
+ args []string
+ expectedArgs []string
+ expectedError error
+ }{
+ {
+ stdin: false,
+ args: []string{},
+ expectedArgs: parser.DefaultPath,
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../.."},
+ expectedArgs: []string{filepath.Join("..", "..")},
+ expectedError: nil,
+ },
+
+ // Test glob expansion
+ {
+ stdin: false,
+ args: []string{"../testdata/*.yml"},
+ expectedArgs: []string{
+ filepath.Join("..", "testdata/bad.yml"),
+ filepath.Join("..", "testdata/good.yml"),
+ },
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/g??d.yml"}, // matches any single non-separator character
+ expectedArgs: []string{filepath.Join("..", "testdata/good.yml")},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/[a-z]ood.yml"}, // character range
+ expectedArgs: []string{filepath.Join("..", "testdata", "good.yml")},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/[^abc]ood.yml"}, // character class with negation.
+ expectedArgs: []string{filepath.Join("..", "testdata", "good.yml")},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/[!abc]ood.yml"}, // character class with negation.
+ expectedArgs: []string{filepath.Join("..", "testdata", "good.yml")},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/[^g]ood.yml"}, // character class with negation.
+ expectedArgs: nil,
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/*/*.yml"},
+ expectedArgs: []string{
+ filepath.Join("..", "testdata", "subdir1", "bad.yml"),
+ filepath.Join("..", "testdata", "subdir1", "good.yml"),
+ },
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/**/*.yml"},
+ expectedArgs: []string{
+ filepath.Join("..", "testdata", "bad.yml"),
+ filepath.Join("..", "testdata", "good.yml"),
+ filepath.Join("..", "testdata", "subdir1", "bad.yml"),
+ filepath.Join("..", "testdata", "subdir1", "good.yml"),
+ filepath.Join("..", "testdata", "subdir1", "subdir2", "bad.yml"),
+ filepath.Join("..", "testdata", "subdir1", "subdir2", "good.yml"),
+ },
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/**/{good,bad}.yml"}, // Alternate pattern
+ expectedArgs: []string{
+ filepath.Join("..", "testdata", "bad.yml"),
+ filepath.Join("..", "testdata", "good.yml"),
+ filepath.Join("..", "testdata", "subdir1", "bad.yml"),
+ filepath.Join("..", "testdata", "subdir1", "good.yml"),
+ filepath.Join("..", "testdata", "subdir1", "subdir2", "bad.yml"),
+ filepath.Join("..", "testdata", "subdir1", "subdir2", "good.yml"),
+ },
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/**/?ood.yml"},
+ expectedArgs: []string{
+ filepath.Join("..", "testdata", "good.yml"),
+ filepath.Join("..", "testdata", "subdir1", "good.yml"),
+ filepath.Join("..", "testdata", "subdir1", "subdir2", "good.yml"),
+ },
+ expectedError: nil,
+ },
+
+ // Bad glob pattern
+ {
+ stdin: false,
+ args: []string{"r[.go"}, // Invalid character class
+ expectedArgs: nil,
+ expectedError: doublestar.ErrBadPattern,
+ },
+ {
+ stdin: false,
+ args: []string{"{.go"}, // Bad alternate pattern
+ expectedArgs: nil,
+ expectedError: doublestar.ErrBadPattern,
+ },
 
- stdin = true
- assert.Equal(t, []string{os.Stdin.Name()}, parseArgs([]string{}))
- assert.Equal(t, []string{os.Stdin.Name()}, parseArgs([]string{"../.."}))
+ {
+ stdin: true,
+ args: []string{},
+ expectedArgs: []string{os.Stdin.Name()},
+ expectedError: nil,
+ },
+ {
+ stdin: true,
+ args: []string{"../.."},
+ expectedArgs: []string{os.Stdin.Name()},
+ expectedError: nil,
+ },
+ }
+ for _, tt := range tests {
+ t.Run(strings.Join(tt.args, " "), func(t *testing.T) {
+ stdin = tt.stdin
+ files, err := parseArgs(tt.args)
+ assert.ErrorIs(t, err, tt.expectedError,
+ fmt.Sprintf("arguments: %v. Expected '%v', Got '%v'", tt.args, err, tt.expectedError))
+ assert.Equal(t, tt.expectedArgs, files)
+ })
+ }
 }
 
 func TestRunE(t *testing.T) {
@@ -109,7 +251,7 @@ func TestRunE(t *testing.T) {
  assert.Equal(t, expected, got)
  })
 
- t.Run("findings w error", func(t *testing.T) {
+ t.Run("findings with inclusive language issues", func(t *testing.T) {
  exitOneOnFailure = true
  // don't ignore testdata folder
  noIgnore = true
@@ -122,6 +264,19 @@ func TestRunE(t *testing.T) {
  assert.Regexp(t, regexp.MustCompile(`^files with findings: \d`), err.Error())
  })
 
+ t.Run("findings with invalid glob pattern", func(t *testing.T) {
+ exitOneOnFailure = true
+ // don't ignore testdata folder
+ noIgnore = true
+
+ t.Cleanup(func() {
+ exitOneOnFailure = false
+ })
+ err := rootRunE(new(cobra.Command), []string{"../testdata/**/[.yml"})
+ assert.Error(t, err)
+ assert.Regexp(t, regexp.MustCompile(`syntax error in pattern`), err.Error())
+ })
+
  t.Run("no rules enabled", func(t *testing.T) {
  disableDefaultRules = true
  t.Cleanup(func() {

docs/usage.md

@@ -28,10 +28,30 @@ No findings found.
 ### File globs
 
 By default, `woke` will run against all text files in your current directory.
-To change this, supply a space-separated list of globs as the first argument.
+To change this, supply a space-separated list of file glob patterns.
+`woke` supports the following glob pattern:
+
+```console
+pattern:
+ { term }
+term:
+ * matches any sequence of non-separator characters
+ ? matches any single non-separator character
+ /**/ matches zero or more directories
+ [class] matches any single non-path-separator character against a class of characters
+ {alt1,...} matches a sequence of characters if one of the comma-separated alternatives matches
+
+characters classes:
+ [abc] matches any single character within the set
+ [a-z] matches any single character in the range
+ [^class] matches any single character which does not match the class
+ [!class] same as ^: negates the class
+```
 
 This can be something like `**/*.go`, or a space-separated list of filenames.
 
+If `woke` is invoked from a shell, the invoking shell performs file glob pattern expansion according to the shell glob rules.
+
 ```bash
 $ woke test.txt
 test.txt:2:2-11: `Blacklist` may be insensitive, use `denylist`, `blocklist` instead (warning)

go.mod

@@ -3,6 +3,7 @@ module github.com/get-woke/woke
 go 1.18
 
 require (
+ github.com/bmatcuk/doublestar/v4 v4.3.2
  github.com/caitlinelfring/go-env-default v1.1.0
  github.com/fatih/color v1.13.0
  github.com/get-woke/fastwalk v1.0.0

go.sum

@@ -45,6 +45,8 @@ github.com/acomagu/bufpipe v1.0.3 h1:fxAGrHZTgQ9w5QqVItgzwj235/uYZYgbXitB+dLupOk
 github.com/acomagu/bufpipe v1.0.3/go.mod h1:mxdxdup/WdsKVreO5GpW4+M/1CE2sMG4jeGJ2sYmHc4=
 github.com/anmitsu/go-shlex v0.0.0-20161002113705-648efa622239/go.mod h1:2FmKhYUyUczH0OGQWaF5ceTx0UBShxjsH6f8oGKYe2c=
 github.com/armon/go-socks5 v0.0.0-20160902184237-e75332964ef5/go.mod h1:wHh0iHkYZB8zMSxRWpUBQtwG5a7fFgvEO+odwuTv2gs=
+github.com/bmatcuk/doublestar/v4 v4.3.2 h1:jhwioE79ok+L6P4xulmrNBfCD7yEkrR/BcGSCNDeVag=
+github.com/bmatcuk/doublestar/v4 v4.3.2/go.mod h1:xBQ8jztBU6kakFMg+8WGxn0c6z1fTSPVIjEY1Wr7jzc=
 github.com/caitlinelfring/go-env-default v1.1.0 h1:bhDfXmUolvcIGfQCX8qevQX8wxC54NGz0aimoUnhvDM=
 github.com/caitlinelfring/go-env-default v1.1.0/go.mod h1:tESXPr8zFPP/cRy3cwxrHBmjJIf2A1x/o4C9CET2rEk=
 github.com/census-instrumentation/opencensus-proto v0.2.1/go.mod h1:f6KPmirojxKA12rnyqOA5BBL4O983OfeGPqjHWSTneU=

pkg/parser/parser.go

@@ -54,9 +54,6 @@ func (p *Parser) ParsePaths(print printer.Printer, paths ...string) int {
  return r.Len()
  }
 
- if len(paths) == 0 {
- paths = DefaultPath
- }
  var wg sync.WaitGroup
 
  done := make(chan bool)

testdata/subdir1/bad.yml

@@ -0,0 +1,2 @@
+---
+this file also has a whitelist finding

testdata/subdir1/good.yml

@@ -0,0 +1,2 @@
+---
+this file has no findings.

testdata/subdir1/subdir2/bad.yml

@@ -0,0 +1,2 @@
+---
+this file also has a whitelist finding

testdata/subdir1/subdir2/good.yml

@@ -0,0 +1,2 @@
+---
+this file has no findings.