Skip to content

Update python/poetrylock to support line number location information in extracted packages. #2124

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

Open
copybara-service wants to merge 1 commit into
base: main
Choose a base branch
from
+76 −24
Open

Update python/poetrylock to support line number location information in extracted packages. #2124

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
63 changes: 57 additions & 6 deletions extractor/filesystem/language/python/poetrylock/poetrylock.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,10 +16,15 @@
package poetrylock

import (
"bufio"
"bytes"
"context"
"fmt"
"io"
"path/filepath"
"regexp"
"slices"
"strings"

"github.com/BurntSushi/toml"
"github.com/google/osv-scalibr/extractor"
Expand Down Expand Up @@ -97,22 +102,30 @@ func resolveGroups(pkg poetryLockPackage) []string {

// Extract extracts packages from poetry.lock files passed through the scan input.
func (e Extractor) Extract(ctx context.Context, input *filesystem.ScanInput) (inventory.Inventory, error) {
var parsedLockfile *poetryLockFile

_, err := toml.NewDecoder(input.Reader).Decode(&parsedLockfile)

content, err := io.ReadAll(input.Reader)
if err != nil {
return inventory.Inventory{}, fmt.Errorf("could not read file: %w", err)
}

var parsedLockfile *poetryLockFile
if err := toml.Unmarshal(content, &parsedLockfile); err != nil {
return inventory.Inventory{}, fmt.Errorf("could not extract: %w", err)
}

packageNames := make([]string, 0, len(parsedLockfile.Packages))
for _, p := range parsedLockfile.Packages {
packageNames = append(packageNames, p.Name)
}
lineNums := findPackageLineNumbers(content, packageNames)

packages := make([]*extractor.Package, 0, len(parsedLockfile.Packages))

for _, lockPackage := range parsedLockfile.Packages {
for i, lockPackage := range parsedLockfile.Packages {
pkgDetails := &extractor.Package{
Name: lockPackage.Name,
Version: lockPackage.Version,
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath(input.Path),
Location: extractor.LocationFromPathAndLine(input.Path, lineNums[i]),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: resolveGroups(lockPackage),
},
Expand All @@ -128,4 +141,42 @@ func (e Extractor) Extract(ctx context.Context, input *filesystem.ScanInput) (in
return inventory.Inventory{Packages: packages}, nil
}

var nameRegex = regexp.MustCompile(`^name\s*=\s*["']([^"']+)["']`)

func findPackageLineNumbers(content []byte, packageNames []string) []int {
lineNums := make([]int, len(packageNames))
scanner := bufio.NewScanner(bytes.NewReader(content))
currentLine := 0
pkgIdx := 0
inPackageBlock := false

for scanner.Scan() {
currentLine++
line := strings.TrimSpace(scanner.Text())

if line == "[[package]]" {
inPackageBlock = true
continue
}

if line == "[metadata]" {
break
}

if inPackageBlock && strings.HasPrefix(line, "[") && !strings.HasPrefix(line, "[[package]]") {
inPackageBlock = false
}

if inPackageBlock && pkgIdx < len(packageNames) {
matches := nameRegex.FindStringSubmatch(line)
if len(matches) > 1 && matches[1] == packageNames[pkgIdx] {
lineNums[pkgIdx] = currentLine
pkgIdx++
inPackageBlock = false
}
}
}
return lineNums
}

var _ filesystem.Extractor = Extractor{}
37 changes: 19 additions & 18 deletions extractor/filesystem/language/python/poetrylock/poetrylock_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,6 +82,7 @@ func TestExtractor_FileRequired(t *testing.T) {
}

func TestExtractor_Extract(t *testing.T) {
loc := extractor.LocationFromPathAndLine
tests := []extracttest.TestTableEntry{
{
Name: "invalid toml",
Expand All @@ -108,7 +109,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "numpy",
Version: "1.23.3",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/one-package.lock"),
Location: loc("testdata/one-package.lock", 2),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -125,7 +126,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "proto-plus",
Version: "1.22.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/two-packages.lock"),
Location: loc("testdata/two-packages.lock", 2),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -134,7 +135,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "protobuf",
Version: "4.21.5",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/two-packages.lock"),
Location: loc("testdata/two-packages.lock", 16),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -151,7 +152,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "emoji",
Version: "2.0.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/one-package-with-metadata.lock"),
Location: loc("testdata/one-package-with-metadata.lock", 2),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -168,7 +169,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "ike",
Version: "0.2.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/source-git.lock"),
Location: loc("testdata/source-git.lock", 2),
SourceCode: &extractor.SourceCodeIdentifier{
Commit: "cd66602cd29f61a2d2e7fb995fef1e61708c034d",
},
Expand All @@ -188,7 +189,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "appdirs",
Version: "1.4.4",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/source-legacy.lock"),
Location: loc("testdata/source-legacy.lock", 2),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -205,7 +206,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "numpy",
Version: "1.23.3",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/optional-package.lock"),
Location: loc("testdata/optional-package.lock", 2),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"optional"},
},
Expand All @@ -222,7 +223,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "async-timeout",
Version: "5.0.1",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 4),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"optional"},
},
Expand All @@ -231,7 +232,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "factory-boy",
Version: "3.3.1",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 17),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"dev"},
},
Expand All @@ -240,7 +241,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "faker",
Version: "33.3.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 36),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"dev", "test"},
},
Expand All @@ -249,7 +250,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "proto-plus",
Version: "1.22.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 52),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -258,7 +259,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "proto-plus",
Version: "1.23.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 71),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -267,7 +268,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "protobuf",
Version: "4.25.5",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 90),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -276,7 +277,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "python-dateutil",
Version: "2.9.0.post0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 111),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"dev", "test"},
},
Expand All @@ -285,7 +286,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "six",
Version: "1.17.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 146),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{},
},
Expand All @@ -294,7 +295,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "typing-extensions",
Version: "4.12.2",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 158),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"dev", "test"},
},
Expand All @@ -303,7 +304,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "urllib3",
Version: "2.3.0",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 170),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"dev"},
},
Expand All @@ -312,7 +313,7 @@ func TestExtractor_Extract(t *testing.T) {
Name: "redis",
Version: "5.2.1",
PURLType: purl.TypePyPi,
Location: extractor.LocationFromPath("testdata/multiple-packages.v2.lock"),
Location: loc("testdata/multiple-packages.v2.lock", 126),
Metadata: &osv.DepGroupMetadata{
DepGroupVals: []string{"optional"},
},
Expand Down
Loading