Merge pull request #41 from deoxykev/refactor_rulesets

refactor rulesets into separate files and add a ruleset compiler cli …

cmd/main.go

@@ -8,6 +8,7 @@ import (
 	"strings"
 
 	"ladder/handlers"
+	"ladder/handlers/cli"
 
 	"github.com/akamensky/argparse"
 	"github.com/gofiber/fiber/v2"
@@ -17,6 +18,7 @@ import (
 
 //go:embed favicon.ico
 var faviconData string
+
 //go:embed styles.css
 var cssData embed.FS
 
@@ -40,7 +42,20 @@ func main() {
 
 	ruleset := parser.String("r", "ruleset", &argparse.Options{
 		Required: false,
-		Help:     "File, Directory or URL to a ruleset.yml. Overrides RULESET environment variable.",
+		Help:     "File, Directory or URL to a ruleset.yaml. Overrides RULESET environment variable.",
+	})
+
+	mergeRulesets := parser.Flag("", "merge-rulesets", &argparse.Options{
+		Required: false,
+		Help:     "Compiles a directory of yaml files into a single ruleset.yaml. Requires --ruleset arg.",
+	})
+	mergeRulesetsGzip := parser.Flag("", "merge-rulesets-gzip", &argparse.Options{
+		Required: false,
+		Help:     "Compiles a directory of yaml files into a single ruleset.gz Requires --ruleset arg.",
+	})
+	mergeRulesetsOutput := parser.String("", "merge-rulesets-output", &argparse.Options{
+		Required: false,
+		Help:     "Specify output file for --merge-rulesets and --merge-rulesets-gzip. Requires --ruleset and --merge-rulesets args.",
 	})
 
 	err := parser.Parse(os.Args)
@@ -48,6 +63,16 @@ func main() {
 		fmt.Print(parser.Usage(err))
 	}
 
+	// utility cli flag to compile ruleset directory into single ruleset.yaml
+	if *mergeRulesets || *mergeRulesetsGzip {
+		err = cli.HandleRulesetMerge(ruleset, mergeRulesets, mergeRulesetsGzip, mergeRulesetsOutput)
+		if err != nil {
+			fmt.Println(err)
+			os.Exit(1)
+		}
+		os.Exit(0)
+	}
+
 	if os.Getenv("PREFORK") == "true" {
 		*prefork = true
 	}

go.mod

@@ -26,4 +26,5 @@ require (
 	github.com/valyala/tcplisten v1.0.0 // indirect
 	golang.org/x/net v0.18.0 // indirect
 	golang.org/x/sys v0.14.0 // indirect
+	golang.org/x/term v0.14.0
 )

go.sum

@@ -68,6 +68,8 @@ golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9sn
 golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
 golang.org/x/term v0.5.0/go.mod h1:jMB1sMXY+tzblOD4FWmEbocvup2/aLOaQEp7JmGp78k=
 golang.org/x/term v0.7.0/go.mod h1:P32HKFT3hSsZrRxla30E9HqToFYAQPCMs/zFMBUFqPY=
+golang.org/x/term v0.14.0 h1:LGK9IlZ8T9jvdy6cTdfKUCltatMFOehAQo9SRC46UQ8=
+golang.org/x/term v0.14.0/go.mod h1:TySc+nGkYR6qt8km8wUhuFRTVSMIX3XPR58y2lC8vww=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
 golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=

handlers/cli/cli.go

@@ -0,0 +1,105 @@
+package cli
+
+import (
+	"fmt"
+	"io"
+	"io/fs"
+	"ladder/pkg/ruleset"
+	"os"
+
+	"golang.org/x/term"
+)
+
+// HandleRulesetMerge merges a set of ruleset files, specified by the rulesetPath or RULESET env variable, into either YAML or Gzip format.
+// Exits the program with an error message if the ruleset path is not provided or if loading the ruleset fails.
+//
+// Parameters:
+// - rulesetPath: A pointer to a string specifying the path to the ruleset file.
+// - mergeRulesets: A pointer to a boolean indicating if a merge operation should be performed.
+// - mergeRulesetsGzip: A pointer to a boolean indicating if the merge should be in Gzip format.
+// - mergeRulesetsOutput: A pointer to a string specifying the output file path. If empty, the output is printed to stdout.
+//
+// Returns:
+// - An error if the ruleset loading or merging process fails, otherwise nil.
+func HandleRulesetMerge(rulesetPath *string, mergeRulesets *bool, mergeRulesetsGzip *bool, mergeRulesetsOutput *string) error {
+	if *rulesetPath == "" {
+		*rulesetPath = os.Getenv("RULESET")
+	}
+	if *rulesetPath == "" {
+		fmt.Println("ERROR: no ruleset provided. Try again with --ruleset <ruleset.yaml>")
+		os.Exit(1)
+	}
+
+	rs, err := ruleset.NewRuleset(*rulesetPath)
+	if err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+
+	if *mergeRulesetsGzip {
+		return gzipMerge(rs, mergeRulesetsOutput)
+	}
+	return yamlMerge(rs, mergeRulesetsOutput)
+}
+
+// gzipMerge takes a RuleSet and an optional output file path pointer. It compresses the RuleSet into Gzip format.
+// If the output file path is provided, the compressed data is written to this file. Otherwise, it prints a warning
+// and outputs the binary data to stdout
+//
+// Parameters:
+// - rs: The ruleset.RuleSet to be compressed.
+// - mergeRulesetsOutput: A pointer to a string specifying the output file path. If empty, the output is directed to stdout.
+//
+// Returns:
+// - An error if compression or file writing fails, otherwise nil.
+func gzipMerge(rs ruleset.RuleSet, mergeRulesetsOutput *string) error {
+	gzip, err := rs.GzipYaml()
+	if err != nil {
+		return err
+	}
+
+	if *mergeRulesetsOutput != "" {
+		out, err := os.Create(*mergeRulesetsOutput)
+		defer out.Close()
+		_, err = io.Copy(out, gzip)
+		if err != nil {
+			return err
+		}
+	}
+
+	if term.IsTerminal(int(os.Stdout.Fd())) {
+		println("WARNING: binary output can mess up your terminal. Use '--merge-rulesets-output <ruleset.gz>' or pipe it to a file.")
+		os.Exit(1)
+	}
+	_, err = io.Copy(os.Stdout, gzip)
+	if err != nil {
+		return err
+	}
+	return nil
+}
+
+// yamlMerge takes a RuleSet and an optional output file path pointer. It converts the RuleSet into YAML format.
+// If the output file path is provided, the YAML data is written to this file. If not, the YAML data is printed to stdout.
+//
+// Parameters:
+// - rs: The ruleset.RuleSet to be converted to YAML.
+// - mergeRulesetsOutput: A pointer to a string specifying the output file path. If empty, the output is printed to stdout.
+//
+// Returns:
+// - An error if YAML conversion or file writing fails, otherwise nil.
+func yamlMerge(rs ruleset.RuleSet, mergeRulesetsOutput *string) error {
+	yaml, err := rs.Yaml()
+	if err != nil {
+		return err
+	}
+	if *mergeRulesetsOutput == "" {
+		fmt.Printf(yaml)
+		os.Exit(0)
+	}
+
+	err = os.WriteFile(*mergeRulesetsOutput, []byte(yaml), fs.FileMode(os.O_RDWR))
+	if err != nil {
+		return fmt.Errorf("ERROR: failed to write merged YAML ruleset to '%s'\n", *mergeRulesetsOutput)
+	}
+	return nil
+}

pkg/ruleset/ruleset.go

@@ -39,20 +39,20 @@ type Rule struct {
 		CSP           string `yaml:"content-security-policy,omitempty"`
 	} `yaml:"headers,omitempty"`
 	GoogleCache bool    `yaml:"googleCache,omitempty"`
-	RegexRules  []Regex `yaml:"regexRules"`
+	RegexRules  []Regex `yaml:"regexRules,omitempty"`
 
 	UrlMods struct {
-		Domain []Regex `yaml:"domain"`
+		Domain []Regex `yaml:"domain,omitempty"`
-		Path   []Regex `yaml:"path"`
+		Path   []Regex `yaml:"path,omitempty"`
-		Query  []KV    `yaml:"query"`
+		Query  []KV    `yaml:"queryomitempty"`
-	} `yaml:"urlMods"`
+	} `yaml:"urlMods,omitempty"`
 
 	Injections []struct {
-		Position string `yaml:"position"`
+		Position string `yaml:"position,omitempty"`
-		Append   string `yaml:"append"`
+		Append   string `yaml:"append,omitempty"`
-		Prepend  string `yaml:"prepend"`
+		Prepend  string `yaml:"prepend,omitempty"`
-		Replace  string `yaml:"replace"`
+		Replace  string `yaml:"replace,omitempty"`
-	} `yaml:"injections"`
+	} `yaml:"injections,omitempty"`
 }
 
 // NewRulesetFromEnv creates a new RuleSet based on the RULESET environment variable.

ruleset.yaml

@@ -1,194 +0,0 @@
-- domain: example.com
-  domains: 
-  - www.beispiel.de
-  googleCache: true
-  headers:
-    x-forwarded-for: none
-    referer: none
-    user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36
-    cookie: privacy=1
-  regexRules:
-    - match: <script\s+([^>]*\s+)?src="(/)([^"]*)"
-      replace: <script $1 script="/https://www.example.com/$3"
-  injections:
-    - position: head # Position where to inject the code
-      append: |
-        <script>
-          window.localStorage.clear();
-          console.log("test");
-          alert("Hello!");
-        </script>
-    - position: h1
-      replace: |
-        <h1>An example with a ladder ;-)</h1>
-- domain: www.americanbanker.com
-  paths:
-    - /news
-  injections:
-    - position: head
-      append: |
-        <script>
-          document.addEventListener("DOMContentLoaded", () => {
-            const inlineGate = document.querySelector('.inline-gate');
-            if (inlineGate) {
-              inlineGate.classList.remove('inline-gate');
-              const inlineGated = document.querySelectorAll('.inline-gated');
-              for (const elem of inlineGated) { elem.classList.remove('inline-gated'); }
-            }
-          });
-        </script>
-- domain: www.nzz.ch
-  paths:
-    - /international
-    - /sport
-    - /wirtschaft
-    - /technologie
-    - /feuilleton
-    - /zuerich
-    - /wissenschaft
-    - /gesellschaft
-    - /panorama
-    - /mobilitaet
-    - /reisen
-    - /meinung
-    - /finanze
-  injections:
-    - position: head
-      append: |
-        <script>
-          document.addEventListener("DOMContentLoaded", () => {
-            const paywall = document.querySelector('.dynamic-regwall');
-            removeDOMElement(paywall)
-          });
-        </script>
-- domains: 
-  - www.architecturaldigest.com
-  - www.bonappetit.com
-  - www.cntraveler.com
-  - www.epicurious.com
-  - www.gq.com
-  - www.newyorker.com
-  - www.vanityfair.com
-  - www.vogue.com
-  - www.wired.com
-  injections:
-    - position: head
-      append: |
-        <script>
-          document.addEventListener("DOMContentLoaded", () => {
-            const banners = document.querySelectorAll('.paywall-bar, div[class^="MessageBannerWrapper-"');
-            banners.forEach(el => { el.remove(); });
-          });
-        </script>
-- domains: 
-  - www.nytimes.com
-  - www.time.com
-  headers:
-    ueser-agent: Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)
-    cookie: nyt-a=; nyt-gdpr=0; nyt-geo=DE; nyt-privacy=1
-    referer: https://www.google.com/ 
-  injections:
-    - position: head
-      append: |
-        <script>
-          window.localStorage.clear();
-          document.addEventListener("DOMContentLoaded", () => {
-            const banners = document.querySelectorAll('div[data-testid="inline-message"], div[id^="ad-"], div[id^="leaderboard-"], div.expanded-dock, div.pz-ad-box, div[id="top-wrapper"], div[id="bottom-wrapper"]');
-            banners.forEach(el => { el.remove(); });
-          });
-        </script>
-- domains: 
-  - www.thestar.com
-  - www.niagarafallsreview.ca
-  - www.stcatharinesstandard.ca
-  - www.thepeterboroughexaminer.com
-  - www.therecord.com
-  - www.thespec.com
-  - www.wellandtribune.ca
-  injections:
-    - position: head
-      append: |
-        <script>
-          window.localStorage.clear();
-          document.addEventListener("DOMContentLoaded", () => {
-            const paywall = document.querySelectorAll('div.subscriber-offers');
-            paywall.forEach(el => { el.remove(); });
-            const subscriber_only = document.querySelectorAll('div.subscriber-only');
-            for (const elem of subscriber_only) {
-              if (elem.classList.contains('encrypted-content') && dompurify_loaded) {
-                const parser = new DOMParser();
-                const doc = parser.parseFromString('<div>' + DOMPurify.sanitize(unscramble(elem.innerText)) + '</div>', 'text/html');
-                const content_new = doc.querySelector('div');
-                elem.parentNode.replaceChild(content_new, elem);
-              }
-              elem.removeAttribute('style');
-              elem.removeAttribute('class');
-            }
-            const banners = document.querySelectorAll('div.subscription-required, div.redacted-overlay, div.subscriber-hide, div.tnt-ads-container');
-            banners.forEach(el => { el.remove(); });
-            const ads = document.querySelectorAll('div.tnt-ads-container, div[class*="adLabelWrapper"]');
-            ads.forEach(el => { el.remove(); });
-            const recommendations = document.querySelectorAll('div[id^="tncms-region-article"]');
-            recommendations.forEach(el => { el.remove(); });
-          });
-        </script>
-- domain: www.usatoday.com
-  injections:
-    - position: head
-      append: |
-        <script>
-          document.addEventListener("DOMContentLoaded", () => {
-            const banners = document.querySelectorAll('div.roadblock-container, .gnt_nb, [aria-label="advertisement"], div[id="main-frame-error"]');
-            banners.forEach(el => { el.remove(); });
-          });
-        </script>
-- domain: www.washingtonpost.com
-  injections:
-    - position: head
-      append: |
-        <script>
-          document.addEventListener("DOMContentLoaded", () => {
-            let paywall = document.querySelectorAll('div[data-qa$="-ad"], div[id="leaderboard-wrapper"], div[data-qa="subscribe-promo"]');
-            paywall.forEach(el => { el.remove(); });
-            const images = document.querySelectorAll('img');
-            images.forEach(image => { image.parentElement.style.filter = ''; });
-            const headimage = document.querySelectorAll('div .aspect-custom');
-            headimage.forEach(image => { image.style.filter = ''; });
-          });
-        </script>
-- domain: medium.com
-  headers:
-    referer: https://t.co/x?amp=1
-    x-forwarded-for: none
-    user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36
-    content-security-policy: script-src 'self';
-    cookie: 
-- domain: tagesspiegel.de
-  headers:
-    content-security-policy: script-src 'self';
-    user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36
-  urlMods:
-    query:
-      - key: amp
-        value: 1
-- domain: www.ft.com
-  headers:
-    referer: https://t.co/x?amp=1
-  injections:
-    - position: head
-      append: |
-        <script>
-          document.addEventListener("DOMContentLoaded", () => {
-            const styleTags = document.querySelectorAll('link[rel="stylesheet"]');
-            styleTags.forEach(el => { 
-              const href = el.getAttribute('href').substring(1);
-              const updatedHref = href.replace(/(https?:\/\/.+?)\/{2,}/, '$1/');
-              el.setAttribute('href', updatedHref);
-            });
-            setTimeout(() => {
-              const cookie = document.querySelectorAll('.o-cookie-message, .js-article-ribbon, .o-ads, .o-banner, .o-message, .article__content-sign-up');
-              cookie.forEach(el => { el.remove(); });
-            }, 1000);
-          })
-        </script>
-

rulesets/ca/_multi-metroland-media-group.yaml

@@ -0,0 +1,35 @@
+- domains: 
+  - www.thestar.com
+  - www.niagarafallsreview.ca
+  - www.stcatharinesstandard.ca
+  - www.thepeterboroughexaminer.com
+  - www.therecord.com
+  - www.thespec.com
+  - www.wellandtribune.ca
+  injections:
+    - position: head
+      append: |
+        <script>
+          window.localStorage.clear();
+          document.addEventListener("DOMContentLoaded", () => {
+            const paywall = document.querySelectorAll('div.subscriber-offers');
+            paywall.forEach(el => { el.remove(); });
+            const subscriber_only = document.querySelectorAll('div.subscriber-only');
+            for (const elem of subscriber_only) {
+              if (elem.classList.contains('encrypted-content') && dompurify_loaded) {
+                const parser = new DOMParser();
+                const doc = parser.parseFromString('<div>' + DOMPurify.sanitize(unscramble(elem.innerText)) + '</div>', 'text/html');
+                const content_new = doc.querySelector('div');
+                elem.parentNode.replaceChild(content_new, elem);
+              }
+              elem.removeAttribute('style');
+              elem.removeAttribute('class');
+            }
+            const banners = document.querySelectorAll('div.subscription-required, div.redacted-overlay, div.subscriber-hide, div.tnt-ads-container');
+            banners.forEach(el => { el.remove(); });
+            const ads = document.querySelectorAll('div.tnt-ads-container, div[class*="adLabelWrapper"]');
+            ads.forEach(el => { el.remove(); });
+            const recommendations = document.querySelectorAll('div[id^="tncms-region-article"]');
+            recommendations.forEach(el => { el.remove(); });
+          });
+        </script>

rulesets/ch/nzz-ch.yaml

@@ -0,0 +1,24 @@
+- domain: www.nzz.ch
+  paths:
+    - /international
+    - /sport
+    - /wirtschaft
+    - /technologie
+    - /feuilleton
+    - /zuerich
+    - /wissenschaft
+    - /gesellschaft
+    - /panorama
+    - /mobilitaet
+    - /reisen
+    - /meinung
+    - /finanze
+  injections:
+    - position: head
+      append: |
+        <script>
+          document.addEventListener("DOMContentLoaded", () => {
+            const paywall = document.querySelector('.dynamic-regwall');
+            removeDOMElement(paywall)
+          });
+        </script>

rulesets/de/tagesspiegel-de.yaml

@@ -0,0 +1,9 @@
+# loads amp version of page
+- domain: tagesspiegel.de
+  headers:
+    content-security-policy: script-src 'self';
+    user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36
+  urlMods:
+    query:
+      - key: amp
+        value: 1

rulesets/gb/ft-com.yaml

@@ -0,0 +1,20 @@
+- domain: www.ft.com
+  headers:
+    referer: https://t.co/x?amp=1
+  injections:
+    - position: head
+      append: |
+        <script>
+          document.addEventListener("DOMContentLoaded", () => {
+            const styleTags = document.querySelectorAll('link[rel="stylesheet"]');
+            styleTags.forEach(el => { 
+              const href = el.getAttribute('href').substring(1);
+              const updatedHref = href.replace(/(https?:\/\/.+?)\/{2,}/, '$1/');
+              el.setAttribute('href', updatedHref);
+            });
+            setTimeout(() => {
+              const cookie = document.querySelectorAll('.o-cookie-message, .js-article-ribbon, .o-ads, .o-banner, .o-message, .article__content-sign-up');
+              cookie.forEach(el => { el.remove(); });
+            }, 1000);
+          })
+        </script>

rulesets/us/_multi-conde-nast.yaml

@@ -0,0 +1,19 @@
+- domains: 
+  - www.architecturaldigest.com
+  - www.bonappetit.com
+  - www.cntraveler.com
+  - www.epicurious.com
+  - www.gq.com
+  - www.newyorker.com
+  - www.vanityfair.com
+  - www.vogue.com
+  - www.wired.com
+  injections:
+    - position: head
+      append: |
+        <script>
+          document.addEventListener("DOMContentLoaded", () => {
+            const banners = document.querySelectorAll('.paywall-bar, div[class^="MessageBannerWrapper-"');
+            banners.forEach(el => { el.remove(); });
+          });
+        </script>

rulesets/us/americanbanker-com.yaml

@@ -0,0 +1,16 @@
+- domain: americanbanker.com
+  paths:
+    - /news
+  injections:
+    - position: head
+      append: |
+        <script>
+          document.addEventListener("DOMContentLoaded", () => {
+            const inlineGate = document.querySelector('.inline-gate');
+            if (inlineGate) {
+              inlineGate.classList.remove('inline-gate');
+              const inlineGated = document.querySelectorAll('.inline-gated');
+              for (const elem of inlineGated) { elem.classList.remove('inline-gated'); }
+            }
+          });
+        </script>

rulesets/us/medium-com.yaml

@@ -0,0 +1,7 @@
+- domain: medium.com
+  headers:
+    referer: https://t.co/x?amp=1
+    x-forwarded-for: none
+    user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36
+    content-security-policy: script-src 'self';
+    cookie: 

rulesets/us/nytimes-com.yaml

@@ -0,0 +1,17 @@
+- domains: 
+  - www.nytimes.com
+  - www.time.com
+  headers:
+    ueser-agent: Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)
+    cookie: nyt-a=; nyt-gdpr=0; nyt-geo=DE; nyt-privacy=1
+    referer: https://www.google.com/ 
+  injections:
+    - position: head
+      append: |
+        <script>
+          window.localStorage.clear();
+          document.addEventListener("DOMContentLoaded", () => {
+            const banners = document.querySelectorAll('div[data-testid="inline-message"], div[id^="ad-"], div[id^="leaderboard-"], div.expanded-dock, div.pz-ad-box, div[id="top-wrapper"], div[id="bottom-wrapper"]');
+            banners.forEach(el => { el.remove(); });
+          });
+        </script>

rulesets/us/usatoday-com.yaml

@@ -0,0 +1,10 @@
+- domain: www.usatoday.com
+  injections:
+    - position: head
+      append: |
+        <script>
+          document.addEventListener("DOMContentLoaded", () => {
+            const banners = document.querySelectorAll('div.roadblock-container, .gnt_nb, [aria-label="advertisement"], div[id="main-frame-error"]');
+            banners.forEach(el => { el.remove(); });
+          });
+        </script>

rulesets/us/washingtonpost-com.yaml

@@ -0,0 +1,14 @@
+- domain: www.washingtonpost.com
+  injections:
+    - position: head
+      append: |
+        <script>
+          document.addEventListener("DOMContentLoaded", () => {
+            let paywall = document.querySelectorAll('div[data-qa$="-ad"], div[id="leaderboard-wrapper"], div[data-qa="subscribe-promo"]');
+            paywall.forEach(el => { el.remove(); });
+            const images = document.querySelectorAll('img');
+            images.forEach(image => { image.parentElement.style.filter = ''; });
+            const headimage = document.querySelectorAll('div .aspect-custom');
+            headimage.forEach(image => { image.style.filter = ''; });
+          });
+        </script>