From 9d88509780b308cd857e273872649799b3c7cff7 Mon Sep 17 00:00:00 2001
From: Shane St Savage <shane@axds.co>
Date: Fri, 4 Oct 2024 16:50:15 -0700
Subject: [PATCH] Initial commit

---
 LICENSE                           |  23 +++++
 README.md                         |  68 +++++++++++++
 agent-allowlist.txt               |  12 +++
 cleaner.go                        | 164 ++++++++++++++++++++++++++++++
 cleaner_test.go                   |  53 ++++++++++
 go.mod                            |   5 +
 go.sum                            |   2 +
 testdata/expected-crawler.log     |   4 +
 testdata/expected-error.log       |   1 +
 testdata/expected-non-crawler.log |   3 +
 testdata/extra-crawler-agents.txt |   2 +
 testdata/web.log                  |   8 ++
 12 files changed, 345 insertions(+)
 create mode 100644 LICENSE
 create mode 100644 README.md
 create mode 100644 agent-allowlist.txt
 create mode 100644 cleaner.go
 create mode 100644 cleaner_test.go
 create mode 100644 go.mod
 create mode 100644 go.sum
 create mode 100644 testdata/expected-crawler.log
 create mode 100644 testdata/expected-error.log
 create mode 100644 testdata/expected-non-crawler.log
 create mode 100644 testdata/extra-crawler-agents.txt
 create mode 100644 testdata/web.log

diff --git a/LICENSE b/LICENSE
new file mode 100644
index 0000000..6c40a95
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,23 @@
+The MIT License (MIT)
+=====================
+
+Permission is hereby granted, free of charge, to any person
+obtaining a copy of this software and associated documentation
+files (the “Software”), to deal in the Software without
+restriction, including without limitation the rights to use,
+copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the
+Software is furnished to do so, subject to the following
+conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+OTHER DEALINGS IN THE SOFTWARE.
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..621d972
--- /dev/null
+++ b/README.md
@@ -0,0 +1,68 @@
+# crawler-cleaner 🕷️🧹✨
+
+Processes json web log input from stdin (one json log object per line),
+removing any user agents determined to be crawlers/bot/scrapers.
+
+The project uses the
+[crawler-user-agents](https://github.com/monperrus/crawler-user-agents)
+project as the user agent database.
+
+By default crawler-cleaner looks for the user agent in the top level
+`http_user_agent` field in each json log. This may be configured using the
+`-user-agent-key` flag (but must be top level).
+
+Detected crawler logs can be discarded (default) or written to a separate
+file/stream. JSON parse errors (error messages between json logs etc)
+can also be written to a separate file. The following strings have
+special meaning for the output files;
+
+* `0`, `/dev/null`, `null` - discard output
+* `-`, `/dev/stdout`, `stdout` - write output to stdout
+- `+`, `/dev/stderr`, `stderr` - write output to stderr
+
+## Example usage
+
+```
+$ ./crawler-cleaner -help
+Usage of ./crawler-cleaner:
+  -crawler-output string
+        File to write crawler output to (default "/dev/null")
+  -error-output string
+        File to write unparsable json iput to (default "/dev/null")
+  -extra-crawler-agents-file string
+        File containing additional crawler user agent patterns, one per line
+  -non-crawler-output string
+        File to write non-crawler output to (default "/dev/stdout")
+  -user-agent-key string
+        Json key for user agent (default "http_user_agent")
+```
+
+```
+$ cat web.log | ./crawler-cleaner -crawler-output ./crawlers.log \
+  -non-crawler-output ./legit.log -error-output errors.log
+```
+
+or
+
+```
+$ <web.log ./crawler-cleaner -non-crawler-output stderr > ./legit.log 2> ./crawlers.log
+```
+
+## Reviewing results
+
+After running, it's useful to examine the user agents in both non-crawler
+and crawler outputs to identify any adjustments needed. Example command
+to view counts of user agents using [`jq`](https://jqlang.github.io/jq/):
+
+```
+<non-crawler.log jq '.http_user_agent' | sort | uniq -c | sort -n | less
+```
+
+## Adding extra crawler agents
+
+To add crawler agents to the set obtained from
+[crawler-user-agents](https://github.com/monperrus/crawler-user-agents/blob/master/crawler-user-agents.json),
+you may create a text file (default `./extra-user-agents.txt` and add user agent patterns,
+one per line. Note that these patterns can be regular expressions, but forward slashes `/` do
+not need to be escaped as they are in the `crawler-user-agents.json` file
+(i.e. use `meta-externalagent/` and not `meta-externalagent\\/`.
diff --git a/agent-allowlist.txt b/agent-allowlist.txt
new file mode 100644
index 0000000..ffe766d
--- /dev/null
+++ b/agent-allowlist.txt
@@ -0,0 +1,12 @@
+aiohttp
+Apache-HttpClient
+^curl
+Go-http-client
+http_get
+httpx
+libwww-perl
+node-fetch
+okhttp
+python-requests
+Python-urllib
+[wW]get
diff --git a/cleaner.go b/cleaner.go
new file mode 100644
index 0000000..27c5e2e
--- /dev/null
+++ b/cleaner.go
@@ -0,0 +1,164 @@
+package main
+
+import (
+	"bufio"
+	"bytes"
+	_ "embed"
+	"encoding/json"
+	"flag"
+	"fmt"
+	"io"
+	"os"
+	"regexp"
+	"slices"
+	"strings"
+
+	"github.com/monperrus/crawler-user-agents"
+)
+
+var (
+	userAgentKeyConfig     string
+	extraCrawlerAgentsFile string
+	nonCrawlerOutput       string
+	crawlerOutput          string
+	errorOutput            string
+)
+
+//go:embed agent-allowlist.txt
+var agentAllowListBytes []byte
+
+var allowedAgentOverrides = func() []string {
+	var allowedAgentOverrides []string
+	s := bufio.NewScanner(bytes.NewReader(agentAllowListBytes))
+	s.Split(bufio.ScanLines)
+	for s.Scan() {
+		allowedAgent := strings.TrimSpace(s.Text())
+		if len(allowedAgent) == 0 {
+			continue
+		}
+		allowedAgentOverrides = append(allowedAgentOverrides, allowedAgent)
+	}
+	return allowedAgentOverrides
+}()
+
+// adapted from crawler-user-agents/validate.go
+var crawlerRegexps = func() []*regexp.Regexp {
+	regexps := make([]*regexp.Regexp, 0, len(agents.Crawlers))
+	for _, crawler := range agents.Crawlers {
+		if !slices.Contains(allowedAgentOverrides, crawler.Pattern) {
+			regexps = append(regexps, regexp.MustCompile(crawler.Pattern))
+		}
+	}
+	return regexps
+}()
+
+// adapted from crawler-user-agents/validate.go
+// Returns if User Agent string matches any of crawler patterns.
+var isCrawler = func(userAgent string) bool {
+	for _, re := range crawlerRegexps {
+		if re.MatchString(userAgent) {
+			return true
+		}
+	}
+	return false
+}
+
+const defaultExtraCrawlerAgentsFile = "extra-crawler-agents.txt"
+const defaultUserAgentKey = "http_user_agent"
+
+func getWriter(w string) *os.File {
+	switch w {
+	case "-", "/dev/stdout", "stdout":
+		//stdout
+		return os.Stdout
+	case "+", "/dev/stderr", "stderr":
+		//stderr
+		return os.Stderr
+	case "0", "/dev/null", "null":
+		//devnull
+		return os.NewFile(0, os.DevNull)
+	default:
+		//file path
+		file, err := os.OpenFile(w, os.O_CREATE|os.O_TRUNC|os.O_WRONLY, 0644)
+		if err != nil {
+			panic(err)
+		}
+		return file
+	}
+}
+
+func addExtraCrawlerAgents(extraAgentsReader io.Reader) {
+	extraAgentsScanner := bufio.NewScanner(extraAgentsReader)
+	extraAgentsScanner.Split(bufio.ScanLines)
+
+	for extraAgentsScanner.Scan() {
+		extraAgent := strings.TrimSpace(extraAgentsScanner.Text())
+		if len(extraAgent) == 0 {
+			continue
+		}
+		crawlerRegexps = append(crawlerRegexps, regexp.MustCompile(extraAgent))
+	}
+}
+
+func cleanCrawlers(userAgentKey string, logReader io.Reader, nonCrawlerWriter io.Writer,
+	crawlerWriter io.Writer, errorWriter io.Writer) {
+	s := bufio.NewScanner(logReader)
+	for s.Scan() {
+		var v map[string]interface{}
+		if err := json.Unmarshal(s.Bytes(), &v); err != nil {
+			// json parse error
+			fmt.Fprintln(errorWriter, s.Text())
+			continue
+		}
+
+		agent, ok := v[userAgentKey]
+		// assume its ok if we don't have agent info
+		if ok && isCrawler(agent.(string)) {
+			// crawler detected
+			fmt.Fprintln(crawlerWriter, s.Text())
+		} else {
+			// crawler not detected
+			fmt.Fprintln(nonCrawlerWriter, s.Text())
+		}
+	}
+}
+
+func main() {
+	flag.StringVar(&extraCrawlerAgentsFile, "extra-crawler-agents-file", "", "File containing additional crawler user agent patterns, one per line")
+	flag.StringVar(&userAgentKeyConfig, "user-agent-key", defaultUserAgentKey, "Json key for user agent")
+	flag.StringVar(&nonCrawlerOutput, "non-crawler-output", "/dev/stdout", "File to write non-crawler output to")
+	flag.StringVar(&crawlerOutput, "crawler-output", "/dev/null", "File to write crawler output to")
+	flag.StringVar(&errorOutput, "error-output", "/dev/null", "File to write unparsable json iput to")
+	flag.Parse()
+
+	//use default extra agents if not specified and default file exists
+	if len(extraCrawlerAgentsFile) == 0 {
+		if _, err := os.Stat(defaultExtraCrawlerAgentsFile); err == nil {
+			extraCrawlerAgentsFile = defaultExtraCrawlerAgentsFile
+		}
+	}
+
+	//load extra agents file if set
+	if len(extraCrawlerAgentsFile) > 0 {
+		if _, err := os.Stat(extraCrawlerAgentsFile); err == nil {
+			extraAgents, err := os.Open(extraCrawlerAgentsFile)
+			if err != nil {
+				fmt.Println(err)
+			}
+			defer extraAgents.Close()
+
+			addExtraCrawlerAgents(extraAgents)
+		} else {
+			fmt.Println("Error loading extra agents file", extraCrawlerAgentsFile, err)
+		}
+	}
+
+	nonCrawlerWriter := getWriter(nonCrawlerOutput)
+	defer nonCrawlerWriter.Close()
+	crawlerWriter := getWriter(crawlerOutput)
+	defer crawlerWriter.Close()
+	errorWriter := getWriter(errorOutput)
+	defer errorWriter.Close()
+
+	cleanCrawlers(userAgentKeyConfig, os.Stdin, nonCrawlerWriter, crawlerWriter, errorWriter)
+}
diff --git a/cleaner_test.go b/cleaner_test.go
new file mode 100644
index 0000000..e1c2d26
--- /dev/null
+++ b/cleaner_test.go
@@ -0,0 +1,53 @@
+package main
+
+/*
+To generate test data:
+
+<testdata/web.log go run cleaner.go \
+  -crawler-output ./testdata/expected-crawler.log \
+  -non-crawler-output ./testdata/expected-non-crawler.log \
+  -error-output ./testdata/expected-error.log \
+  -extra-crawler-agents-file testdata/extra-crawler-agents.txt
+*/
+
+import (
+	"bytes"
+	"io"
+	"os"
+	"testing"
+)
+
+func readFile(path string) []byte {
+	data, err := os.ReadFile(path)
+	if err != nil {
+		panic(err)
+	}
+	return data
+}
+
+var extraCrawlerAgentsBytes []byte = readFile("testdata/extra-crawler-agents.txt")
+var rawLogBytes []byte = readFile("testdata/web.log")
+var expectedNonCrawlerBytes []byte = readFile("testdata/expected-non-crawler.log")
+var expectedCrawlerBytes []byte = readFile("testdata/expected-crawler.log")
+var expectedErrorBytes []byte = readFile("testdata/expected-error.log")
+
+func TestCleaner(t *testing.T) {
+	var nonCrawlerBytes, crawlerBytes, errorBytes bytes.Buffer
+
+	addExtraCrawlerAgents(bytes.NewReader(extraCrawlerAgentsBytes))
+
+	cleanCrawlers(defaultUserAgentKey, bytes.NewReader(rawLogBytes),
+		io.Writer(&nonCrawlerBytes), io.Writer(&crawlerBytes), io.Writer(&errorBytes))
+
+	if !bytes.Equal(expectedNonCrawlerBytes, nonCrawlerBytes.Bytes()) {
+		t.Fatal("Non crawler logs did not match expected")
+	}
+
+	if !bytes.Equal(expectedCrawlerBytes, crawlerBytes.Bytes()) {
+		t.Fatal("Crawler logs did not match expected")
+	}
+
+	if !bytes.Equal(expectedErrorBytes, errorBytes.Bytes()) {
+		t.Fatal("Error logs did not match expected")
+	}
+}
diff --git a/go.mod b/go.mod
new file mode 100644
index 0000000..6283d45
--- /dev/null
+++ b/go.mod
@@ -0,0 +1,5 @@
+module github.com/axiom-data-science/crawler-cleaner
+
+go 1.21
+
+require github.com/monperrus/crawler-user-agents v0.0.0-20240925083149-6c0133b66cc2
diff --git a/go.sum b/go.sum
new file mode 100644
index 0000000..ff10a74
--- /dev/null
+++ b/go.sum
@@ -0,0 +1,2 @@
+github.com/monperrus/crawler-user-agents v0.0.0-20240925083149-6c0133b66cc2 h1:+A6DL2F8K/8xq7YOXZVAbzXjP6GXQAAJfjGdpi0Zq0I=
+github.com/monperrus/crawler-user-agents v0.0.0-20240925083149-6c0133b66cc2/go.mod h1:GfRyKbsbxSrRxTPYnVi4U/0stQd6BcFCxDy6i6IxQ0M=
diff --git a/testdata/expected-crawler.log b/testdata/expected-crawler.log
new file mode 100644
index 0000000..6699a2c
--- /dev/null
+++ b/testdata/expected-crawler.log
@@ -0,0 +1,4 @@
+{"body_bytes_sent":1169,"content_type":"-","http_host":"some.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/600.2.5 (KHTML, like Gecko) Version/8.0.2 Safari/600.2.5 (Amazonbot/0.1; +https://developer.amazon.com/support/amazonbot)","scheme":"https","request_uri":"/path/to.html","remote_addr":"1.2.3.4","request_method":"GET","request_time":0.002,"sent_http_content_type":"application/json","status":"200","time_iso8601":"2024-09-30T23:59:59+00:00"}
+{"body_bytes_sent":1059,"content_type":"-","http_host":"other.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; bingbot/2.0; +http://www.bing.com/bingbot.htm) Chrome/116.0.1938.76 Safari/537.36","scheme":"https","request_uri":"/yet/another/path.html","remote_addr":"2.4.6.8","request_method":"GET","request_time":0.091,"sent_http_content_type":"text/html;charset=UTF-8","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
+{"body_bytes_sent":1059,"content_type":"-","http_host":"other.host.com","http_referer":"-","http_user_agent":"add-this-crawler to be detected","scheme":"https","request_uri":"/yet/another/path.html","remote_addr":"2.4.6.8","request_method":"GET","request_time":0.091,"sent_http_content_type":"text/html;charset=UTF-8","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
+{"body_bytes_sent":1059,"content_type":"-","http_host":"other.host.com","http_referer":"-","http_user_agent":"Yes, andThisBot As Well","scheme":"https","request_uri":"/yet/another/path.html","remote_addr":"2.4.6.8","request_method":"GET","request_time":0.091,"sent_http_content_type":"text/html;charset=UTF-8","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
diff --git a/testdata/expected-error.log b/testdata/expected-error.log
new file mode 100644
index 0000000..ac392ab
--- /dev/null
+++ b/testdata/expected-error.log
@@ -0,0 +1 @@
+ERROR THIS IS A GARBAGE LOG LINE (NON-JSON)
diff --git a/testdata/expected-non-crawler.log b/testdata/expected-non-crawler.log
new file mode 100644
index 0000000..a1d81ea
--- /dev/null
+++ b/testdata/expected-non-crawler.log
@@ -0,0 +1,3 @@
+{"body_bytes_sent":1176,"content_type":"-","http_host":"some.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/600.2.5 (KHTML, like Gecko) Version/8.0.2 Safari/600.2.5","scheme":"https","request_uri":"/other/path.html","remote_addr":"5.6.7.8","request_method":"GET","request_time":0.002,"sent_http_content_type":"text/html","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
+{"body_bytes_sent":1176,"content_type":"-","http_host":"hosting.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36","scheme":"https","request_uri":"/pathy/path.html","remote_addr":"1.2.5.6","request_method":"GET","request_time":0.002,"sent_http_content_type":"text/html","status":"200","time_iso8601":"2024-09-30T23:59:57+00:00"}
+{"body_bytes_sent":1176,"content_type":"-","http_host":"your.host.net","http_referer":"-","http_user_agent":"python-requests/2.28.1","scheme":"https","request_uri":"/final.html","remote_addr":"1.3.5.7","request_method":"GET","request_time":0.002,"sent_http_content_type":"application/json","status":"200","time_iso8601":"2024-09-30T23:59:56+00:00"}
diff --git a/testdata/extra-crawler-agents.txt b/testdata/extra-crawler-agents.txt
new file mode 100644
index 0000000..1a04924
--- /dev/null
+++ b/testdata/extra-crawler-agents.txt
@@ -0,0 +1,2 @@
+add-this-crawler
+andThisBot As Well
diff --git a/testdata/web.log b/testdata/web.log
new file mode 100644
index 0000000..e385952
--- /dev/null
+++ b/testdata/web.log
@@ -0,0 +1,8 @@
+{"body_bytes_sent":1169,"content_type":"-","http_host":"some.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/600.2.5 (KHTML, like Gecko) Version/8.0.2 Safari/600.2.5 (Amazonbot/0.1; +https://developer.amazon.com/support/amazonbot)","scheme":"https","request_uri":"/path/to.html","remote_addr":"1.2.3.4","request_method":"GET","request_time":0.002,"sent_http_content_type":"application/json","status":"200","time_iso8601":"2024-09-30T23:59:59+00:00"}
+{"body_bytes_sent":1176,"content_type":"-","http_host":"some.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/600.2.5 (KHTML, like Gecko) Version/8.0.2 Safari/600.2.5","scheme":"https","request_uri":"/other/path.html","remote_addr":"5.6.7.8","request_method":"GET","request_time":0.002,"sent_http_content_type":"text/html","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
+{"body_bytes_sent":1059,"content_type":"-","http_host":"other.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; bingbot/2.0; +http://www.bing.com/bingbot.htm) Chrome/116.0.1938.76 Safari/537.36","scheme":"https","request_uri":"/yet/another/path.html","remote_addr":"2.4.6.8","request_method":"GET","request_time":0.091,"sent_http_content_type":"text/html;charset=UTF-8","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
+{"body_bytes_sent":1059,"content_type":"-","http_host":"other.host.com","http_referer":"-","http_user_agent":"add-this-crawler to be detected","scheme":"https","request_uri":"/yet/another/path.html","remote_addr":"2.4.6.8","request_method":"GET","request_time":0.091,"sent_http_content_type":"text/html;charset=UTF-8","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
+ERROR THIS IS A GARBAGE LOG LINE (NON-JSON)
+{"body_bytes_sent":1059,"content_type":"-","http_host":"other.host.com","http_referer":"-","http_user_agent":"Yes, andThisBot As Well","scheme":"https","request_uri":"/yet/another/path.html","remote_addr":"2.4.6.8","request_method":"GET","request_time":0.091,"sent_http_content_type":"text/html;charset=UTF-8","status":"200","time_iso8601":"2024-09-30T23:59:58+00:00"}
+{"body_bytes_sent":1176,"content_type":"-","http_host":"hosting.host.com","http_referer":"-","http_user_agent":"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36","scheme":"https","request_uri":"/pathy/path.html","remote_addr":"1.2.5.6","request_method":"GET","request_time":0.002,"sent_http_content_type":"text/html","status":"200","time_iso8601":"2024-09-30T23:59:57+00:00"}
+{"body_bytes_sent":1176,"content_type":"-","http_host":"your.host.net","http_referer":"-","http_user_agent":"python-requests/2.28.1","scheme":"https","request_uri":"/final.html","remote_addr":"1.3.5.7","request_method":"GET","request_time":0.002,"sent_http_content_type":"application/json","status":"200","time_iso8601":"2024-09-30T23:59:56+00:00"}