Skip to content

Improving normalization patterns #968

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
Gwynbleidd0241 wants to merge 6 commits into
base: master
Choose a base branch
from
+44 −18
Open

Improving normalization patterns #968

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
9c258bf
upgrade filepath and datetime, change func normalizeByScanner, add te…
Gwynbleidd0241 Apr 23, 2026
6978047
add examples in genBenchInput
Gwynbleidd0241 Apr 23, 2026
7b35ab1
add commit to new logic skip unconsumed element
Gwynbleidd0241 Apr 24, 2026
b1cfc0f
fix
Gwynbleidd0241 Apr 29, 2026
24ce765
Merge branch 'master' into 967-upgrade-patterns
Gwynbleidd0241 May 6, 2026
a335f73
Merge branch 'master' into 967-upgrade-patterns
Gwynbleidd0241 May 20, 2026
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
18 changes: 14 additions & 4 deletions plugin/action/hash/normalize/token_normalizer.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -268,7 +268,17 @@ func (n *tokenNormalizer) normalizeByScanner(out []byte, scanner *lexmachine.Sca
prevEnd := 0
for tokRaw, err, eos := scanner.Next(); !eos; tokRaw, err, eos = scanner.Next() {
if ui, is := err.(*machines.UnconsumedInput); is {
scanner.TC = ui.FailTC // skip
// Jumping to FailTC may skip start of the next token.
// Example: part/offset = 54/5990:
// After matching 54, lexer reports unconsumed /5,
// FailTC moves past /5, so 5 is skipped and next scan starts from 990,
// Result normalization before: part/offset = <int>/5990.
//
// Using max(scanner.TC+1, ui.FailTC-1):
// FailTC-1 keeps last byte, which may be a token start,
// scanner.TC+1 ensures forward progress, since FailTC-1 can be equal to scanner.TC,
// Result normalization after: part/offset = <int>/<int>.
scanner.TC = max(scanner.TC+1, ui.FailTC-1)
continue
} else if err != nil {
out = out[:0]
Expand Down Expand Up @@ -484,7 +494,7 @@ var builtinTokenPatterns = []TokenPattern{
},
{
Placeholder: placeholderByPattern[pFilepath],
RE: `(/[a-zA-Z0-9-_.]+)+`,
RE: `(/[a-zA-Z-_.][a-zA-Z0-9-_.]*)+`,
mask: pFilepath,
},
{
Expand All @@ -511,10 +521,10 @@ var builtinTokenPatterns = []TokenPattern{
mask: pHash,
},
{
// RFC3339, RFC3339Nano, DateTime, DateOnly, TimeOnly, Go time with monotonic clock
// RFC3339, RFC3339Nano, DateTime, DateOnly, TimeOnly, Go time with optional monotonic clock
Placeholder: placeholderByPattern[pDatetime],
RE: fmt.Sprintf(`(%s)|(%s)|(%s)|(%s)`,
`\d\d\d\d-\d\d-\d\d \d\d:\d\d:\d\d\.\d+ [+\-]\d\d\d\d [A-Z]+ m=[+\-]\d+\.\d+`,
`\d\d\d\d-\d\d-\d\d \d\d:\d\d:\d\d\.\d+ [+\-]\d\d\d\d [A-Z]+( m=[+\-]\d+\.\d+)?`,
`\d\d\d\d-\d\d-\d\dT\d\d:\d\d:\d\d(\.\d+)?(Z|[\+\-]\d\d:\d\d)`,
`\d\d:\d\d:\d\d`,
`\d\d\d\d-\d\d-\d\d( \d\d:\d\d:\d\d)?`,
Expand Down
34 changes: 19 additions & 15 deletions plugin/action/hash/normalize/token_normalizer_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -265,6 +265,8 @@ func TestTokenNormalizerBuiltin(t *testing.T) {
"some 2025-01-13 20:58:04.019973588 +0000 UTC m=+1417512.275697914 here",
"some 2025-01-13 20:58:04.019973588 -0700 MST m=-123.456789012 here",
"some 2025-01-13 20:58:04.019973588 +0300 MSK m=+0.123456789 here",
"some 2025-01-13 20:58:04.019973588 -0700 MST here",
"some 2025-01-13 20:58:04.019973588 +0300 MSK here",
"some 2025-01-13T10:20:40Z here",
"some 2025-01-13T10:20:40.999999999Z here",
"some 2025-01-13T10:20:40-06:00 here",
Expand Down Expand Up @@ -470,9 +472,9 @@ func TestTokenNormalizerCustom(t *testing.T) {
},
},
inputs: []string{
`2006/01/02 15:04:05 error occurred, client: 10.125.172.251, upstream: "http://10.117.246.15:84/download", host: "mpm-youtube-downloader-38.name.com:84"`,
`2006/01/02 15:04:05 error occurred, client: 10.125.172.251, upstream: "http://10.117.246.15:84/download", host: "mpm-youtube-downloader-38.name.com:84", part/offset: 10117/2461584`,
},
want: "<nginx_datetime> error occurred, client: <ip>, upstream: <double_quoted>, host: <double_quoted>",
want: "<nginx_datetime> error occurred, client: <ip>, upstream: <double_quoted>, host: <double_quoted>, part/offset: <int>/<int>",
},
{
name: "empty_patterns",
Expand Down Expand Up @@ -515,19 +517,21 @@ func TestTokenNormalizerCustom(t *testing.T) {

func genBenchInput(count int) []byte {
var examples = []string{
"s1mple falsehood", // no match
"test@host1.host2.com", // email
"http://some.host.com/page1?a=1", // url
"hello-world-123.COM", // host
"7c1811ed-e98f-4c9c-a9f9-58c757ff494f", // uuid
"a94a8fe5ccb19ba61c4c0873d391e987982fbbd3", // sha1
"098f6bcd4621d373cade4e832627b4f6", // md5
"2025-01-13T10:20:40Z", // datetime
"1.2.3.4", // ip
"-1.2m5s", // duration
"0x13eb85e69dfbc0758b12acdaae36287d", // hex
"-4.56", // float
"123", // int
"48757ec9f04efe7faacec8722f3476339b125a6b6172b8a69ff3aa329e0bd0ff", // hash(sha256)
"a94a8fe5ccb19ba61c4c0873d391e987982fbbd3", // hash(sha1)
"098f6bcd4621d373cade4e832627b4f6", // hash(md5)
"s1mple falsehood", // no match
"test@host1.host2.com", // email
"http://some.host.com/page1?a=1", // url
"hello-world-123.COM", // host
"7c1811ed-e98f-4c9c-a9f9-58c757ff494f", // uuid
"/home/user/photos", // filepath
"2025-01-13T10:20:40Z", // datetime
"1.2.3.4", // ip
"-1.2m5s", // duration
"0x13eb85e69dfbc0758b12acdaae36287d", // hex
"-4.56", // float
"123", // int
"truE faLse",
}

Expand Down
Loading