Add support to expand glob patterns

cmd/root.go

@@ -26,6 +26,7 @@ import (
  "errors"
  "fmt"
  "os"
+ "path/filepath"
  "runtime"
  "strings"
  "time"
@@ -36,6 +37,7 @@ 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 +123,11 @@ 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 +168,40 @@ 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
  }
 
  if stdin {
  args = []string{os.Stdin.Name()}
  }
+ // Perform glob expansion.
+ var files []string
+ for _, arg := range args {
+ var f []string
+ var err error
+ if strings.Contains(arg, "**") {
+ // Double star glob expansion.
+ base, pattern := doublestar.SplitPattern(arg)
+ fsys := os.DirFS(base)
+ f, err = doublestar.Glob(fsys, pattern)
+ for i := range f {
+ f[i] = filepath.Join(base, f[i])
+ }
+ } else {
+ f, err = filepath.Glob(arg)
+ }
+ if err != nil {
+ return nil, err
+ }
+ files = append(files, f...)
+ }
 
- return args
+ return files, nil
 }
 
 func setDebugLogLevel() {

cmd/root_test.go

@@ -4,7 +4,9 @@ import (
  "bytes"
  "io"
  "os"
+ "path/filepath"
  "regexp"
+ "strings"
  "testing"
 
  "github.com/get-woke/woke/pkg/output"
@@ -70,12 +72,93 @@ 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{"../.."},
+ expectedError: nil,
+ },
+
+ // Test glob expansion
+ {
+ stdin: false,
+ args: []string{"../testdata/*.yml"},
+ expectedArgs: []string{"../testdata/good.yml", "../testdata/whitelist.yml"},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/g??d.yml"},
+ expectedArgs: []string{"../testdata/good.yml"},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/[a-z]ood.yml"},
+ expectedArgs: []string{"../testdata/good.yml"},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/*/*.yml"},
+ expectedArgs: []string{"../testdata/subdir1/good.yml", "../testdata/subdir1/whitelist.yml"},
+ expectedError: nil,
+ },
+ {
+ stdin: false,
+ args: []string{"../testdata/**/*.yml"},
+ expectedArgs: []string{
+ "../testdata/good.yml",
+ "../testdata/whitelist.yml",
+ "../testdata/subdir1/good.yml",
+ "../testdata/subdir1/whitelist.yml",
+ "../testdata/subdir1/subdir2/good.yml",
+ "../testdata/subdir1/subdir2/whitelist.yml",
+ },
+ expectedError: nil,
+ },
 
- stdin = true
- assert.Equal(t, []string{os.Stdin.Name()}, parseArgs([]string{}))
- assert.Equal(t, []string{os.Stdin.Name()}, parseArgs([]string{"../.."}))
+ // Bad glob pattern
+ {
+ stdin: false,
+ args: []string{"r[.go"},
+ expectedArgs: nil,
+ expectedError: filepath.ErrBadPattern,
+ },
+
+ {
+ 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)
+ assert.Equal(t, tt.expectedArgs, files)
+ })
+ }
 }
 
 func TestRunE(t *testing.T) {

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:
+
+```
+pattern:
+ { term }
+term:
+ '*' matches any sequence of non-Separator characters
+ '?' matches any single non-Separator character
+ '[' [ '^' ] { character-range } ']'
+ character class (must be non-empty)
+ c matches character c (c != '*', '?', '\\', '[')
+ '\\' c matches character c
+
+character-range:
+ c matches character c (c != '\\', '-', ']')
+ '\\' c matches character c
+ lo '-' hi matches character c for lo <= c <= hi
+```
 
 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

@@ -19,6 +19,7 @@ require (
 
 require (
  github.com/acomagu/bufpipe v1.0.3 // indirect
+ github.com/bmatcuk/doublestar/v4 v4.0.2 // indirect
  github.com/cpuguy83/go-md2man/v2 v2.0.0 // indirect
  github.com/davecgh/go-spew v1.1.1 // indirect
  github.com/fsnotify/fsnotify v1.5.1 // indirect

go.sum

@@ -60,6 +60,8 @@ github.com/armon/go-radix v1.0.0/go.mod h1:ufUuZ+zHj4x4TnLV4JWEpy2hxWSpsRywHrMgI
 github.com/armon/go-socks5 v0.0.0-20160902184237-e75332964ef5/go.mod h1:wHh0iHkYZB8zMSxRWpUBQtwG5a7fFgvEO+odwuTv2gs=
 github.com/bgentry/speakeasy v0.1.0/go.mod h1:+zsyZBPWlz7T6j88CTgSN5bM796AkVf0kBD4zp0CCIs=
 github.com/bketelsen/crypt v0.0.4/go.mod h1:aI6NrJ0pMGgvZKL1iVgXLnfIFJtfV+bKCoqOes/6LfM=
+github.com/bmatcuk/doublestar/v4 v4.0.2 h1:X0krlUVAVmtr2cRoTqR8aDMrDqnB36ht8wpWTiQ3jsA=
+github.com/bmatcuk/doublestar/v4 v4.0.2/go.mod h1:xBQ8jztBU6kakFMg+8WGxn0c6z1fTSPVIjEY1Wr7jzc=
 github.com/caitlinelfring/go-env-default v1.0.0 h1:DY8qY3OKb9PrGe+sjop7dw7tOKh6kV8cdZONlESSbZc=
 github.com/caitlinelfring/go-env-default v1.0.0/go.mod h1:vY8iS64s+wIBKayqiNGJsWMwc19NrxaNNTyWzuPvE44=
 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/good.yml

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

testdata/subdir1/subdir2/good.yml

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

testdata/subdir1/subdir2/whitelist.yml

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

testdata/subdir1/whitelist.yml

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