fixed a first round of bugs and rewrote the URL rewriter using the old algorithm

htmlcheck/checker.go

@@ -158,9 +158,12 @@ func AmNewHTMLChecker(configName string) (HTMLChecker, error) {
 	if !ok {
 		return nil, fmt.Errorf("configuration %s not found", configName)
 	}
-	tset, ok := tagSetNameToSet[config.TagSet]
-	if !ok {
-		return nil, fmt.Errorf("tag set %s not found", config.TagSet)
+	var tset *bitset.BitSet = nil
+	if config.TagSet != "" {
+		tset, ok = tagSetNameToSet[config.TagSet]
+		if !ok {
+			return nil, fmt.Errorf("tag set %s not found", config.TagSet)
+		}
 	}
 	rc := htmlCheckerImpl{
 		config:             config,
@@ -617,10 +620,12 @@ func (ht *htmlCheckerImpl) handleAsHTML() bool {
 		// it's a closing tag and this tag doesn't permit the "close" form
 		return false
 	}
-	tagSetID := tagIndexToSetId[tagIndex]
-	if !ht.tagSet.Test(uint(tagSetID)) {
-		// the tag is not allowed - discard it, if one of the flags is set in the config
-		return ht.config.DiscardHTML || ht.config.DiscardRejected
+	if ht.tagSet != nil {
+		tagSetID := tagIndexToSetId[tagIndex]
+		if !ht.tagSet.Test(uint(tagSetID)) {
+			// the tag is not allowed - discard it, if one of the flags is set in the config
+			return ht.config.DiscardHTML || ht.config.DiscardRejected
+		}
 	}
 	if !ht.config.DiscardHTML && tag.balanceTags {
 		// this tag needs to be balanced - here's where we manipulate the stack

htmlcheck/emoticons.yaml

@@ -23,7 +23,7 @@ emoticons:
     replace: "😳"
   - name: biggrin
     patterns:
-      ":D"
+      - ":D"
     replace: "😁"
   - name: wink
     patterns:
@@ -48,9 +48,9 @@ emoticons:
     replace: "😡"
   - name: eek
     patterns:
-      ":eek:"
+      - ":eek:"
     replace: "😱"
   - name: confused
     patterns:
-      ":confused:"
+      - ":confused:"
     replace: "😕"

htmlcheck/rewriter.go

@@ -51,12 +51,10 @@ var rewriterRegistry = make(map[string]rewriter)
 func init() {
 	r1 := emailRewriter{}
 	rewriterRegistry[r1.Name()] = &r1
-	r2 := urlRewriter{}
+	r2 := postLinkRewriter{}
 	rewriterRegistry[r2.Name()] = &r2
-	r3 := postLinkRewriter{}
+	r3 := userLinkRewriter{}
 	rewriterRegistry[r3.Name()] = &r3
-	r4 := userLinkRewriter{}
-	rewriterRegistry[r4.Name()] = &r4
 }
 
 // emailRewriter is an implementation of Rewriter that recognizes E-mail addresses.
@@ -97,63 +95,6 @@ func (rw *emailRewriter) Rewrite(data string, svc rewriterServices) *markupData
 		rescan:      false}
 }
 
-// urlRewriter is an implementation of Rewriter that recognizes URLs.
-type urlRewriter struct{}
-
-// Name returns the rewriter's name.
-func (rw *urlRewriter) Name() string {
-	return "url"
-}
-
-/* Rewrite rewrites the given string data and adds markup before and after if needed.
- * Parameters:
- *     data - The data to be rewritten.
- *     svc - Services interface we can use.
- * Returns:
- *     Pointer to markup data, or nil.
- */
-func (rw *urlRewriter) Rewrite(data string, svc rewriterServices) *markupData {
-	url, err := url.Parse(data)
-	if err != nil {
-		secondChance := ""
-		if strings.HasPrefix(data, "www.") {
-			secondChance = "http://" + data
-		} else if strings.HasPrefix(data, "ftp.") {
-			secondChance = "ftp://" + data
-		} else if strings.HasPrefix(data, "gopher.") {
-			secondChance = "gopher://" + data
-		}
-		if secondChance == "" {
-			return nil
-		}
-		url, err = url.Parse(secondChance)
-		if err != nil {
-			return nil
-		}
-	}
-
-	if url.Scheme == "http" || url.Scheme == "https" {
-		svc.addExternalRef(url)
-	}
-
-	var openA strings.Builder
-	openA.WriteString("<a href=\"")
-	openA.WriteString(url.String())
-	openA.WriteString("\"")
-	catenate := svc.rewriterAttrValue("ANCHORTAIL")
-	if catenate != "" {
-		openA.WriteString(" ")
-		openA.WriteString(catenate)
-	}
-	openA.WriteString(">")
-	return &markupData{
-		beginMarkup: openA.String(),
-		text:        data,
-		endMarkup:   "</a>",
-		rescan:      false,
-	}
-}
-
 // postLinkRewriter is the rewriter that handles links to conference posts.
 type postLinkRewriter struct{}
 

htmlcheck/url_rewriter.go

@@ -0,0 +1,110 @@
+/*
+ * Amsterdam Web Communities System
+ * Copyright (c) 2025 Erbosoft Metaverse Design Solutions, All Rights Reserved
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at https://mozilla.org/MPL/2.0/.
+ */
+// The htmlcheck package contains the HTML Checker.
+package htmlcheck
+
+import (
+	"net/url"
+	"regexp"
+	"strings"
+)
+
+// urlElement is a basic element
+type urlElement struct {
+	re     *regexp.Regexp
+	prefix string
+}
+
+// eval matches the argument against our regular expression and adds an optional prefix.
+func (e *urlElement) eval(s string) string {
+	if e.re.FindString(s) == "" {
+		return ""
+	}
+	return e.prefix + s
+}
+
+// urlSetupData contains the data needed to set up the URL rewriter elements.
+var urlSetupData = [...]string{
+	`^(?i:http)://[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+.*$`, "",
+	`^(?i:https)://[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+.*$`, "",
+	`^(?i:ftp)://[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+.*$`, "",
+	`^(?i:gopher)://[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+.*$`, "",
+	"^(?i:mailto):[A-Za-z0-9!#$%*+-/=?^_`{|}~.]+@[A-Za-z0-9_-]+(\\.[A-Za-z0-9_-]+)+$", "",
+	`^(?i:news):[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+$`, "",
+	`^(?i:nntp)://[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+.*$`, "",
+	`^(?i:telnet)://[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+.*$`, "",
+	`^(?i:tn3270)://[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)+.*$`, "",
+	`^(?i:www)\.[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)*.*$`, "http://",
+	`^(?i:ftp)\.[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)*.*$`, "ftp://",
+	`^(?i:gopher)\.[A-Za-z0-9_-]+(\.[A-Za-z0-9_-]+)*.*$`, "gopher://",
+}
+
+// The URL elements we can match against.
+var urlElements []urlElement
+
+// init builds the URL elements and registers the rewriter.
+func init() {
+	urlElements = make([]urlElement, len(urlSetupData)/2)
+	i, j := 0, 0
+	for i < len(urlSetupData) {
+		urlElements[j].re = regexp.MustCompile(urlSetupData[i])
+		urlElements[j].prefix = urlSetupData[i+1]
+		i += 2
+		j++
+	}
+	r := urlRewriter{}
+	rewriterRegistry[r.Name()] = &r
+}
+
+// urlRewriter is an implementation of Rewriter that recognizes URLs.
+type urlRewriter struct{}
+
+// Name returns the rewriter's name.
+func (rw *urlRewriter) Name() string {
+	return "url"
+}
+
+/* Rewrite rewrites the given string data and adds markup before and after if needed.
+ * Parameters:
+ *     data - The data to be rewritten.
+ *     svc - Services interface we can use.
+ * Returns:
+ *     Pointer to markup data, or nil.
+ */
+func (rw *urlRewriter) Rewrite(data string, svc rewriterServices) *markupData {
+	for _, ue := range urlElements {
+		s := ue.eval(data)
+		if s != "" {
+			ls := strings.ToLower(s)
+			if strings.HasPrefix(ls, "http:") || strings.HasPrefix(ls, "https:") {
+				url, err := url.Parse(s)
+				if err == nil {
+					svc.addExternalRef(url)
+				}
+			}
+			var openA strings.Builder
+			openA.WriteString("<a href=\"")
+			openA.WriteString(s)
+			openA.WriteString("\"")
+			catenate := svc.rewriterAttrValue("ANCHORTAIL")
+			if catenate != "" {
+				openA.WriteString(" ")
+				openA.WriteString(catenate)
+			}
+			openA.WriteString(">")
+			return &markupData{
+				beginMarkup: openA.String(),
+				text:        data,
+				endMarkup:   "</a>",
+				rescan:      false,
+			}
+		}
+	}
+	return nil
+}