1
0
Fork 0
mirror of https://github.com/gocsaf/csaf.git synced 2025-12-22 18:15:42 +01:00
gocsaf/csaf/advisories.go
Paul Schwabauer 5709b14650
Extend structured logging usage in aggregator (#622)
* Extend structured logging usage in aggregator

* Use structured logging in advisories processor

* Remove unnecessary inner function

* Format

* Feat: Add verbose flag to example aggregator toml (in comment)

---------

Co-authored-by: JanHoefelmeyer <jan.hoefelmeyer@intevation.de>
2025-03-19 09:04:19 +01:00

391 lines
9.5 KiB
Go

// This file is Free Software under the Apache-2.0 License
// without warranty, see README.md and LICENSES/Apache-2.0.txt for details.
//
// SPDX-License-Identifier: Apache-2.0
//
// SPDX-FileCopyrightText: 2022 German Federal Office for Information Security (BSI) <https://www.bsi.bund.de>
// Software-Engineering: 2022 Intevation GmbH <https://intevation.de>
package csaf
import (
"context"
"encoding/csv"
"fmt"
"io"
"log/slog"
"net/http"
"net/url"
"strings"
"time"
"github.com/gocsaf/csaf/v3/util"
)
// AdvisoryFile constructs the urls of a remote file.
type AdvisoryFile interface {
slog.LogValuer
URL() string
SHA256URL() string
SHA512URL() string
SignURL() string
IsDirectory() bool
}
// PlainAdvisoryFile contains all relevant urls of a remote file.
type PlainAdvisoryFile struct {
Path string
SHA256 string
SHA512 string
Sign string
}
// URL returns the URL of this advisory.
func (paf PlainAdvisoryFile) URL() string { return paf.Path }
// SHA256URL returns the URL of SHA256 hash file of this advisory.
func (paf PlainAdvisoryFile) SHA256URL() string { return paf.SHA256 }
// SHA512URL returns the URL of SHA512 hash file of this advisory.
func (paf PlainAdvisoryFile) SHA512URL() string { return paf.SHA512 }
// SignURL returns the URL of signature file of this advisory.
func (paf PlainAdvisoryFile) SignURL() string { return paf.Sign }
// IsDirectory returns true, if was fetched via directory feeds.
func (paf PlainAdvisoryFile) IsDirectory() bool { return false }
// LogValue implements [slog.LogValuer]
func (paf PlainAdvisoryFile) LogValue() slog.Value {
return slog.GroupValue(slog.String("url", paf.URL()))
}
// DirectoryAdvisoryFile only contains the base file path.
// The hash and signature files are directly constructed by extending
// the file name.
type DirectoryAdvisoryFile struct {
Path string
}
// URL returns the URL of this advisory.
func (daf DirectoryAdvisoryFile) URL() string { return daf.Path }
// SHA256URL returns the URL of SHA256 hash file of this advisory.
func (daf DirectoryAdvisoryFile) SHA256URL() string { return daf.Path + ".sha256" }
// SHA512URL returns the URL of SHA512 hash file of this advisory.
func (daf DirectoryAdvisoryFile) SHA512URL() string { return daf.Path + ".sha512" }
// SignURL returns the URL of signature file of this advisory.
func (daf DirectoryAdvisoryFile) SignURL() string { return daf.Path + ".asc" }
// IsDirectory returns true, if was fetched via directory feeds.
func (daf DirectoryAdvisoryFile) IsDirectory() bool { return true }
// LogValue implements [slog.LogValuer]
func (daf DirectoryAdvisoryFile) LogValue() slog.Value {
return slog.GroupValue(slog.String("url", daf.URL()))
}
// AdvisoryFileProcessor implements the extraction of
// advisory file names from a given provider metadata.
type AdvisoryFileProcessor struct {
AgeAccept func(time.Time) bool
Log func(loglevel slog.Level, format string, args ...any)
client util.Client
expr *util.PathEval
doc any
base *url.URL
}
// NewAdvisoryFileProcessor constructs a filename extractor
// for a given metadata document.
func NewAdvisoryFileProcessor(
client util.Client,
expr *util.PathEval,
doc any,
base *url.URL,
) *AdvisoryFileProcessor {
return &AdvisoryFileProcessor{
client: client,
expr: expr,
doc: doc,
base: base,
}
}
// empty checks if list of strings contains at least one none empty string.
func empty(arr []string) bool {
for _, s := range arr {
if s != "" {
return false
}
}
return true
}
// Process extracts the advisory filenames and passes them with
// the corresponding label to fn.
func (afp *AdvisoryFileProcessor) Process(
fn func(TLPLabel, []AdvisoryFile) error,
) error {
lg := afp.Log
if lg == nil {
lg = func(loglevel slog.Level, format string, args ...any) {
slog.Log(context.Background(), loglevel, "AdvisoryFileProcessor.Process: "+format, args...)
}
}
// Check if we have ROLIE feeds.
rolie, err := afp.expr.Eval(
"$.distributions[*].rolie.feeds", afp.doc)
if err != nil {
lg(slog.LevelError, "rolie check failed", "err", err)
return err
}
fs, hasRolie := rolie.([]any)
hasRolie = hasRolie && len(fs) > 0
if hasRolie {
var feeds [][]Feed
if err := util.ReMarshalJSON(&feeds, rolie); err != nil {
return err
}
lg(slog.LevelInfo, "Found ROLIE feed(s)", "length", len(feeds))
for _, feed := range feeds {
if err := afp.processROLIE(feed, fn); err != nil {
return err
}
}
} else {
// No rolie feeds -> try to load files from index.txt
directoryURLs, err := afp.expr.Eval(
"$.distributions[*].directory_url", afp.doc)
var dirURLs []string
if err != nil {
lg(slog.LevelError, "extracting directory URLs failed", "err", err)
} else {
var ok bool
dirURLs, ok = util.AsStrings(directoryURLs)
if !ok {
lg(slog.LevelError, "directory_urls are not strings")
}
}
// Not found -> fall back to PMD url
if empty(dirURLs) {
baseURL, err := util.BaseURL(afp.base)
if err != nil {
return err
}
dirURLs = []string{baseURL}
}
for _, base := range dirURLs {
if base == "" {
continue
}
// Use changes.csv to be able to filter by age.
files, err := afp.loadChanges(base, lg)
if err != nil {
return err
}
// XXX: Is treating as white okay? better look into the advisories?
if err := fn(TLPLabelWhite, files); err != nil {
return err
}
}
} // TODO: else scan directories?
return nil
}
// loadChanges loads baseURL/changes.csv and returns a list of files
// prefixed by baseURL/.
func (afp *AdvisoryFileProcessor) loadChanges(
baseURL string,
lg func(slog.Level, string, ...any),
) ([]AdvisoryFile, error) {
base, err := url.Parse(baseURL)
if err != nil {
return nil, err
}
changesURL := base.JoinPath("changes.csv").String()
resp, err := afp.client.Get(changesURL)
if err != nil {
return nil, err
}
if resp.StatusCode != http.StatusOK {
return nil, fmt.Errorf("fetching %s failed. Status code %d (%s)",
changesURL, resp.StatusCode, resp.Status)
}
defer resp.Body.Close()
var files []AdvisoryFile
c := csv.NewReader(resp.Body)
const (
pathColumn = 0
timeColumn = 1
)
for line := 1; ; line++ {
r, err := c.Read()
if err == io.EOF {
break
}
if err != nil {
return nil, err
}
if len(r) < 2 {
lg(slog.LevelError, "Not enough columns", "line", line)
continue
}
t, err := time.Parse(time.RFC3339, r[timeColumn])
if err != nil {
lg(slog.LevelError, "Invalid time stamp in line", "url", changesURL, "line", line, "err", err)
continue
}
// Apply date range filtering.
if afp.AgeAccept != nil && !afp.AgeAccept(t) {
continue
}
path := r[pathColumn]
if _, err := url.Parse(path); err != nil {
lg(slog.LevelError, "Contains an invalid URL", "url", changesURL, "path", path, "line", line)
continue
}
files = append(files,
DirectoryAdvisoryFile{Path: base.JoinPath(path).String()})
}
return files, nil
}
func (afp *AdvisoryFileProcessor) processROLIE(
labeledFeeds []Feed,
fn func(TLPLabel, []AdvisoryFile) error,
) error {
for i := range labeledFeeds {
feed := &labeledFeeds[i]
if feed.URL == nil {
continue
}
up, err := url.Parse(string(*feed.URL))
if err != nil {
slog.Error("Invalid URL in feed", "feed", *feed.URL, "err", err)
continue
}
feedURL := afp.base.ResolveReference(up)
slog.Info("Got feed URL", "feed", feedURL)
fb, err := util.BaseURL(feedURL)
if err != nil {
slog.Error("Invalid feed base URL", "url", fb, "err", err)
continue
}
feedBaseURL, err := url.Parse(fb)
if err != nil {
slog.Error("Cannot parse feed base URL", "url", fb, "err", err)
continue
}
res, err := afp.client.Get(feedURL.String())
if err != nil {
slog.Error("Cannot get feed", "err", err)
continue
}
if res.StatusCode != http.StatusOK {
slog.Error("Fetching failed",
"url", feedURL, "status_code", res.StatusCode, "status", res.Status)
continue
}
rfeed, err := func() (*ROLIEFeed, error) {
defer res.Body.Close()
return LoadROLIEFeed(res.Body)
}()
if err != nil {
slog.Error("Loading ROLIE feed failed", "err", err)
continue
}
var files []AdvisoryFile
resolve := func(u string) string {
if u == "" {
return ""
}
p, err := url.Parse(u)
if err != nil {
slog.Error("Invalid URL", "url", u, "err", err)
return ""
}
return feedBaseURL.ResolveReference(p).String()
}
rfeed.Entries(func(entry *Entry) {
// Filter if we have date checking.
if afp.AgeAccept != nil {
if t := time.Time(entry.Updated); !t.IsZero() && !afp.AgeAccept(t) {
return
}
}
var self, sha256, sha512, sign string
for i := range entry.Link {
link := &entry.Link[i]
lower := strings.ToLower(link.HRef)
switch link.Rel {
case "self":
self = resolve(link.HRef)
case "signature":
sign = resolve(link.HRef)
case "hash":
switch {
case strings.HasSuffix(lower, ".sha256"):
sha256 = resolve(link.HRef)
case strings.HasSuffix(lower, ".sha512"):
sha512 = resolve(link.HRef)
}
}
}
if self == "" {
return
}
var file AdvisoryFile
switch {
case sha256 == "" && sha512 == "":
slog.Error("No hash listed on ROLIE feed", "file", self)
return
case sign == "":
slog.Error("No signature listed on ROLIE feed", "file", self)
return
default:
file = PlainAdvisoryFile{self, sha256, sha512, sign}
}
files = append(files, file)
})
var label TLPLabel
if feed.TLPLabel != nil {
label = *feed.TLPLabel
} else {
label = "unknown"
}
if err := fn(label, files); err != nil {
return err
}
}
return nil
}