Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix(python): support escaped characters in strings #1599

Merged
merged 1 commit into from
May 16, 2024
Merged

fix(python): support escaped characters in strings #1599

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions internal/languages/javascript/detectors/string/string.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ package string

import (
"fmt"
"strconv"

"github.com/bearer/bearer/internal/scanner/ast/query"
"github.com/bearer/bearer/internal/scanner/ast/tree"
"github.com/bearer/bearer/internal/scanner/ruleset"
"github.com/bearer/bearer/internal/util/stringutil"

"github.com/bearer/bearer/internal/scanner/detectors/common"
"github.com/bearer/bearer/internal/scanner/detectors/types"
Expand Down Expand Up @@ -37,7 +37,7 @@ func (detector *stringDetector) DetectAt(
IsLiteral: true,
}}, nil
case "escape_sequence":
value, err := strconv.Unquote(fmt.Sprintf(`"%s"`, node.Content()))
value, err := stringutil.Unescape(node.Content())
if err != nil {
return nil, fmt.Errorf("failed to decode escape sequence: %w", err)
}
Expand Down
122 changes: 77 additions & 45 deletions internal/languages/python/detectors/.snapshots/TestPythonString-string_literal
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
type: module
id: 0
range: 1:1 - 11:1
range: 1:1 - 12:1
dataflow_sources:
- 1
- 5
Expand All @@ -10,7 +10,8 @@ dataflow_sources:
- 25
- 33
- 37
- 44
- 42
- 49
children:
- type: expression_statement
id: 1
Expand Down Expand Up @@ -186,79 +187,103 @@ children:
range: 7:6 - 7:7
- type: expression_statement
id: 37
range: 9:1 - 9:15
range: 8:1 - 8:7
dataflow_sources:
- 38
children:
- type: boolean_operator
- type: string
id: 38
range: 9:1 - 9:15
alias_of:
range: 8:1 - 8:7
dataflow_sources:
- 39
- 40
- 41
children:
- type: "false"
- type: '"""'
id: 39
range: 9:1 - 9:6
content: "False"
- type: '"or"'
range: 8:1 - 8:2
- type: escape_sequence
id: 40
range: 9:7 - 9:9
- type: string
range: 8:3 - 8:5
content: \n
- type: '"""'
id: 41
range: 9:10 - 9:15
dataflow_sources:
- 42
- 43
children:
- type: '"""'
id: 42
range: 9:10 - 9:11
- type: '"""'
id: 43
range: 9:14 - 9:15
range: 8:6 - 8:7
- type: expression_statement
id: 44
id: 42
range: 10:1 - 10:15
dataflow_sources:
- 45
- 43
children:
- type: boolean_operator
id: 45
id: 43
range: 10:1 - 10:15
alias_of:
- 44
- 46
- 50
children:
- type: "false"
id: 44
range: 10:1 - 10:6
content: "False"
- type: '"or"'
id: 45
range: 10:7 - 10:9
- type: string
id: 46
range: 10:1 - 10:6
range: 10:10 - 10:15
dataflow_sources:
- 47
- 48
children:
- type: '"""'
id: 47
range: 10:1 - 10:2
range: 10:10 - 10:11
- type: '"""'
id: 48
range: 10:5 - 10:6
- type: '"or"'
id: 49
range: 10:7 - 10:9
range: 10:14 - 10:15
- type: expression_statement
id: 49
range: 11:1 - 11:15
dataflow_sources:
- 50
children:
- type: boolean_operator
id: 50
range: 11:1 - 11:15
alias_of:
- 51
- 55
children:
- type: string
id: 50
range: 10:10 - 10:15
id: 51
range: 11:1 - 11:6
dataflow_sources:
- 51
- 52
- 53
children:
- type: '"""'
id: 51
range: 10:10 - 10:11
- type: '"""'
id: 52
range: 10:14 - 10:15
range: 11:1 - 11:2
- type: '"""'
id: 53
range: 11:5 - 11:6
- type: '"or"'
id: 54
range: 11:7 - 11:9
- type: string
id: 55
range: 11:10 - 11:15
dataflow_sources:
- 56
- 57
children:
- type: '"""'
id: 56
range: 11:10 - 11:11
- type: '"""'
id: 57
range: 11:14 - 11:15

- node: 2
content: '''a'''
Expand Down Expand Up @@ -296,16 +321,23 @@ children:
value: a\n
isliteral: true
- node: 38
content: '"a\nb"'
data:
value: |-
a
b
isliteral: true
- node: 43
content: False or "foo"
data:
value: foo
isliteral: true
- node: 45
- node: 50
content: '"hey" or "foo"'
data:
value: hey
isliteral: true
- node: 45
- node: 50
content: '"hey" or "foo"'
data:
value: foo
Expand All @@ -320,17 +352,17 @@ children:
data:
value: b
isliteral: true
- node: 41
- node: 46
content: '"foo"'
data:
value: foo
isliteral: true
- node: 46
- node: 51
content: '"hey"'
data:
value: hey
isliteral: true
- node: 50
- node: 55
content: '"foo"'
data:
value: foo
Expand Down
15 changes: 13 additions & 2 deletions internal/languages/python/detectors/string/string.go
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,11 +1,13 @@
package string

import (
"fmt"
"regexp"

"github.com/bearer/bearer/internal/scanner/ast/query"
"github.com/bearer/bearer/internal/scanner/ast/tree"
"github.com/bearer/bearer/internal/scanner/ruleset"
"github.com/bearer/bearer/internal/util/stringutil"

"github.com/bearer/bearer/internal/scanner/detectors/common"
"github.com/bearer/bearer/internal/scanner/detectors/types"
Expand Down Expand Up @@ -73,10 +75,19 @@ func handleTemplateString(node *tree.Node, detectorContext types.Context) ([]int
var childIsLiteral bool
namedChildren := child.NamedChildren()

if len(namedChildren) == 0 {
switch {
case child.Type() == "escape_sequence":
value, err := stringutil.Unescape(child.Content())
if err != nil {
return fmt.Errorf("failed to decode escape sequence: %w", err)
}

childValue = value
childIsLiteral = true
case len(namedChildren) == 0:
childValue = ""
childIsLiteral = true
} else {
default:
var err error
childValue, childIsLiteral, err = common.GetStringValue(namedChildren[0], detectorContext)
if err != nil {
Expand Down
1 change: 1 addition & 0 deletions internal/languages/python/detectors/testdata/string_literal.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@
'''a'''
f'{foo} a'
r'a\n'
"a\nb"

False or "foo"
"hey" or "foo"
10 changes: 9 additions & 1 deletion internal/util/stringutil/stringutil.go
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,10 @@
package stringutil

import "strings"
import (
"fmt"
"strconv"
"strings"
)

func SliceContains(slice []string, value string) bool {
for _, sliceValue := range slice {
Expand All @@ -16,3 +20,7 @@ func StripQuotes(input string) string {
output := strings.Trim(input, `"`)
return strings.Trim(output, `'`)
}

func Unescape(value string) (string, error) {
return strconv.Unquote(fmt.Sprintf(`"%s"`, value))
}
Loading