Initial code commit

.gitignore

@@ -0,0 +1,2 @@
+/backtone
+/backtone.exe

LICENSE

@@ -1,6 +1,8 @@
                     GNU GENERAL PUBLIC LICENSE
                        Version 3, 29 June 2007
 
+                Copyright (c) 2026 Randall Winkhart
+
  Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
  Everyone is permitted to copy and distribute verbatim copies
  of this license document, but changing it is not allowed.

README.md

@@ -0,0 +1,17 @@
+# Backtone
+Backtone is a program that converts websites to Atom RSS
+feeds and prints the results to stdout. It should be called via a
+server-side script that then writes the resulting XML to the webpage.
+
+The name is a playful nod toward converting contents back to the web of yesteryear.
+Why suffer in Web 2/3 when you could go "back to one"?
+
+# Dependencies
+To ensure Backtone always supports current ECMAScript standards, it needs a headless browser instance to execute client-side scripts. To avoid Cloudflare/similar challenge interference, we use [FlareSolverr](https://developer.chrome.com/blog/chrome-headless-shell). Its Docker image also contains the browser we use.
+
+# Usage
+All files needed for usage, including an the input XML, server-side script, and reverse proxy config, have examples in the `examples` folder in this repo. 
+
+1. Create a server-side script that calls `backtone` with the only argument being a base64-encoded XML in the format specified when running `backbone` with no arguments
+2. Serve that script via a reverse proxy
+    - Now when a feed reader fetches the feed, the feed will be created in real time

atom/rss.go

@@ -0,0 +1,79 @@
+package atom
+
+import (
+	"errors"
+	"regexp"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/gorilla/feeds"
+)
+
+// Indices indicates which capture group in the
+// regex-parsed HTML to reference for each Atom field.
+// An empty slice will use a blank/default value.
+// For the slices of strings, these will be appended between
+// captured group text.
+type IndicesT struct {
+	TitleC       []string
+	TitleI       []int8
+	LinkC        []string
+	LinkI        []int8
+	DescriptionC []string
+	DescriptionI []int8
+	AuthorNameC  []string
+	AuthorNameI  []int8
+	AuthorEmailC []string
+	AuthorEmailI []int8
+}
+
+func GetFromHTML(feed *feeds.Feed, rawHTML *string, regexString string, infoIndices IndicesT, maxFeedItems int) (*string, error) {
+	now := time.Now()
+	r := regexp.MustCompile(regexString)
+	newsItems := r.FindAllString(*rawHTML, 5)
+	for i, item := range newsItems {
+		info := r.FindStringSubmatch(item)
+		link := stitchFields(info, infoIndices.LinkC, infoIndices.LinkI)
+		feedItem := &feeds.Item{
+			Title:       stitchFields(info, infoIndices.TitleC, infoIndices.TitleI),
+			Link:        &feeds.Link{Href: link},
+			Id:          "urn:backtoneid:" + strconv.Itoa(i) + ":" + link,
+			Description: stitchFields(info, infoIndices.DescriptionC, infoIndices.DescriptionI),
+			Author: &feeds.Author{
+				Name:  stitchFields(info, infoIndices.AuthorNameC, infoIndices.AuthorNameI),
+				Email: stitchFields(info, infoIndices.AuthorEmailC, infoIndices.AuthorEmailI),
+			},
+			Created: now,
+		}
+		feed.Items = append(feed.Items, feedItem)
+		if i == maxFeedItems {
+			break
+		}
+	}
+	atom, err := feed.ToAtom()
+	if err != nil {
+		return nil, errors.New("unable to create Atom RSS feed: " + err.Error())
+	}
+	return &atom, nil
+}
+
+func stitchFields(info []string, connectors []string, indices []int8) string {
+	var output strings.Builder
+	if len(connectors) > 0 {
+		for i := range connectors {
+			output.WriteString(connectors[i])
+			if len(indices) >= i+1 {
+				output.WriteString(info[indices[i]])
+			}
+		}
+	} else {
+		for i := range indices {
+			if i > 0 {
+				output.WriteString(" | ")
+			}
+			output.WriteString(info[indices[i]])
+		}
+	}
+	return output.String()
+}

bthtml/html.go

@@ -0,0 +1,50 @@
+package bthtml
+
+import (
+	"bytes"
+	"encoding/json"
+	"errors"
+	"io"
+	"net/http"
+)
+
+type payloadT struct {
+	Cmd                    string  `json:"cmd"`
+	URL                    string  `json:"url"`
+	MaxTimeoutMilliseconds int     `json:"maxTimeout"`
+	WaitInSeconds          float32 `json:"waitInSeconds"`
+}
+
+func GetFromURL(flareSolverrURL, webPageURL string, loadSeconds float32) (string, error) {
+	client := &http.Client{}
+
+	payload := payloadT{
+		Cmd:                    "request.get",
+		URL:                    webPageURL,
+		MaxTimeoutMilliseconds: 60000,
+		WaitInSeconds:          loadSeconds,
+	}
+	payloadBytes, err := json.Marshal(payload)
+	if err != nil {
+		return "", errors.New("unable to marshal FlareSolverr payload to JSON: " + err.Error())
+	}
+
+	req, _ := http.NewRequest(
+		"POST",
+		flareSolverrURL,
+		bytes.NewBuffer(payloadBytes),
+	)
+	req.Header.Set("Content-Type", "application/json")
+
+	resp, err := client.Do(req)
+	if err != nil {
+		return "", errors.New("unable to make FlareSolverr request to " + flareSolverrURL + ": " + err.Error())
+	}
+	defer resp.Body.Close()
+
+	body, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return "", errors.New("unable to read FlareSolverr response body: " + err.Error())
+	}
+	return string(body), nil
+}

examples/Caddyfile

@@ -0,0 +1,22 @@
+backtone.example.com {
+
+    tls {
+        protocols tls1.3
+    }
+
+    log {
+        output file /tmp/caddy-backtone.log {
+            roll_size 2MiB
+            roll_keep 3
+            roll_keep_for 24h
+        }
+        level INFO
+    }
+
+    root * /path/to/feeds/directory
+    encode gzip
+    file_server
+
+    php_fastcgi 127.0.0.1:9000
+
+}

examples/feed.php

@@ -0,0 +1,8 @@
+<?php
+$command = "/usr/local/bin/backtone";
+$argument = "BASE64ENCODEDINPUTXML";
+
+$output = shell_exec(escapeshellcmd($command) . " " . escapeshellarg($argument));
+header("Content-Type: application/xml; charset=utf-8");
+echo $output;
+?>

examples/input_xmls/chorus-encore.xml

@@ -0,0 +1 @@
+<input><loadSeconds>1</loadSeconds><regex><![CDATA[<ce-search-result class=\\"card rounded-none bg-base-100 shadow-xl\\"><!----><!----><div class=\\"flex\\"><img alt=\\"Album art\\" loading=\\"lazy\\" class=\\"object-cover w-40\\" src=\\".*?\\"><!----><div class=\\"flex-grow p-3\\"><div class=\\"text-xl \[text-wrap:balance\] sm:text-2xl\\">(.*?)<\/div><!----><div><span class=\\"text-sm font-bold \[text-wrap:balance\] sm:text-base\\">Album:<\/span> (.*?)<\/div><div><span class=\\"text-sm font-bold \[text-wrap:balance\] sm:text-base\\">Genre:<\/span> .*?<\/div><div><span class=\\"text-sm font-bold \[text-wrap:balance\] sm:text-base\\">Year:<\/span> .*?<\/div><!----><\/div><\/div><ce-search-result-chart><div class=\\"collapse overflow-visible rounded-none border border-base-300 bg-base-200 collapse-close\\"><div class=\\"collapse-title flex flex-wrap items-center justify-end gap-3 rounded-none px-3 py-2\\"><div class=\\"mr-auto flex items-center gap-3\\"><div class=\\"swap swap-rotate cursor-pointer\\"><i class=\\"bi bi-chevron-down swap-off text-xl\\"><\/i><i class=\\"bi bi-chevron-up swap-on text-xl\\"><\/i><\/div><div class=\\"tooltip min-w-max\\"><img class=\\"w-16\\" .*?\\"><\/div><!----><\/div><div class=\\"flex flex-1 items-center gap-3\\" style=\\"flex-basis: 9rem;\\"><div class=\\"flex flex-wrap items-center gap-3 \[word-break:break-word\]\\"><div><div class=\\"footer-title mb-0\\">Charter:<\/div><a target=\\"_blank\\" class=\\"link-hover link\\" href=\\".*?\\">(.*?)<\/a><\/div>.*?for=\\"downloadVideos_(.*?)\\"]]></regex><flareSolverrURL>http://PLACEHOLDER:8191/v1</flareSolverrURL><maxFeedItems>10</maxFeedItems><feed><title><parent>Chorus Encore</parent><captureGroupIndex>1</captureGroupIndex></title><link><parent>https://www.enchor.us/</parent><captureGroupConnector>https://www.enchor.us/chart/</captureGroupConnector><captureGroupIndex>4</captureGroupIndex></link><description><parent>Latest YARG/CH charts published to Chorus Encore</parent><captureGroupConnector>Charter: </captureGroupConnector><captureGroupIndex>3</captureGroupIndex><captureGroupConnector> | Album: </captureGroupConnector><captureGroupIndex>2</captureGroupIndex></description><authorName><parent>Gamers</parent><captureGroupIndex>3</captureGroupIndex></authorName><authorEmail></authorEmail></feed></input>

go.mod

@@ -0,0 +1,7 @@
+module github.com/rwinkhart/backtone
+
+go 1.25.5
+
+require github.com/rwinkhart/go-boilerplate v0.1.0
+
+require github.com/gorilla/feeds v1.2.0

go.sum

@@ -0,0 +1,10 @@
+github.com/gorilla/feeds v1.2.0 h1:O6pBiXJ5JHhPvqy53NsjKOThq+dNFm8+DFrxBEdzSCc=
+github.com/gorilla/feeds v1.2.0/go.mod h1:WMib8uJP3BbY+X8Szd1rA5Pzhdfh+HCCAYT2z7Fza6Y=
+github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
+github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
+github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
+github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
+github.com/rogpeppe/go-internal v1.9.0 h1:73kH8U+JUqXU8lRuOHeVHaa/SZPifC7BkcraZVejAe8=
+github.com/rogpeppe/go-internal v1.9.0/go.mod h1:WtVeX8xhTBvf0smdhujwtBcq4Qrzq/fJaraNFVN+nFs=
+github.com/rwinkhart/go-boilerplate v0.1.0 h1:EzlVj6R7Bxtl79Nl7R5zRcg6s+Cf2FAqGIzR4giWTQg=
+github.com/rwinkhart/go-boilerplate v0.1.0/go.mod h1:cnzIF45I0FCOvE4YIB+26pLCUx2kWyY2llKYZruNaRY=

main.go

@@ -0,0 +1,108 @@
+package main
+
+import (
+	"encoding/base64"
+	"encoding/xml"
+	"fmt"
+	"os"
+
+	"github.com/gorilla/feeds"
+	"github.com/rwinkhart/backtone/atom"
+	"github.com/rwinkhart/backtone/bthtml"
+	"github.com/rwinkhart/go-boilerplate/other"
+)
+
+const cfgPath = "/etc/backtone.xml"
+
+// feedFieldInfo provides the string value for a given field in the parent feed
+// and specifies which index in the regex capture group to extract the same field
+// from per individual news item.
+// An index of -1 can be used to use the same value as the parent feed.
+// An index of -2 can be used to use a default (placeholder) value.
+type feedFieldInfoT struct {
+	Parent                string   `xml:"parent,omitempty"`
+	CaptureGroupConnector []string `xml:"captureGroupConnector,omitempty"`
+	CaptureGroupIndex     []int8   `xml:"captureGroupIndex,omitempty"`
+}
+type feedT struct {
+	Title       feedFieldInfoT `xml:"title"`
+	Link        feedFieldInfoT `xml:"link"`
+	Description feedFieldInfoT `xml:"description"`
+	AuthorName  feedFieldInfoT `xml:"authorName"`
+	AuthorEmail feedFieldInfoT `xml:"authorEmail"`
+}
+type inputT struct {
+	LoadSeconds     float32 `xml:"loadSeconds"`
+	Regex           string  `xml:"regex"`
+	FlareSolverrURL string  `xml:"flareSolverrURL"`
+	MaxFeedItems    int     `xml:"maxFeedItems"`
+	Feed            feedT   `xml:"feed"`
+}
+
+func main() {
+	// handle input argument (XML)
+	if len(os.Args) < 2 {
+		demoInputBytes, _ := xml.MarshalIndent(
+			inputT{
+				LoadSeconds:     2.5,
+				Regex:           "PLACEHOLDER",
+				FlareSolverrURL: "http://127.0.0.1:8191",
+				MaxFeedItems:    10,
+				Feed: feedT{
+					Title:       feedFieldInfoT{Parent: "Demo Title", CaptureGroupIndex: []int8{0}},
+					Link:        feedFieldInfoT{Parent: "https://example.com/", CaptureGroupIndex: []int8{1}},
+					Description: feedFieldInfoT{Parent: "Demo description", CaptureGroupConnector: []string{"Prefix - ", " - Middle - ", " - Suffix"}, CaptureGroupIndex: []int8{1, 3}},
+					AuthorName:  feedFieldInfoT{Parent: "John Doe", CaptureGroupIndex: []int8{3}},
+					AuthorEmail: feedFieldInfoT{Parent: "johndoe@example.com", CaptureGroupIndex: []int8{4}},
+				},
+			},
+			"",
+			"    ",
+		)
+		other.PrintError(os.Args[0]+" must be called with valid base64-encoded XML input using the following schema:\n\n"+string(demoInputBytes), 1)
+	}
+
+	// decode base64 input XML
+	decodedXML, err := base64.StdEncoding.DecodeString(os.Args[1])
+	if err != nil {
+		other.PrintError("Failed to decode base64 input XML: "+err.Error(), 1)
+	}
+
+	var inputInfo inputT
+	if err := xml.Unmarshal(decodedXML, &inputInfo); err != nil {
+		other.PrintError("Failed to unmarshal input XML: "+err.Error(), 1)
+	}
+
+	rawHTML, err := bthtml.GetFromURL(inputInfo.FlareSolverrURL, inputInfo.Feed.Link.Parent, inputInfo.LoadSeconds)
+	if err != nil {
+		other.PrintError("Failed to get HTML content: "+err.Error(), 1)
+	}
+
+	atomXML, err := atom.GetFromHTML(
+		&feeds.Feed{
+			Title:       inputInfo.Feed.Title.Parent,
+			Link:        &feeds.Link{Href: inputInfo.Feed.Link.Parent},
+			Description: inputInfo.Feed.Description.Parent,
+			Author:      &feeds.Author{Name: inputInfo.Feed.AuthorName.Parent, Email: inputInfo.Feed.AuthorEmail.Parent},
+		},
+		&rawHTML,
+		inputInfo.Regex,
+		atom.IndicesT{
+			TitleC:       inputInfo.Feed.Title.CaptureGroupConnector,
+			TitleI:       inputInfo.Feed.Title.CaptureGroupIndex,
+			LinkC:        inputInfo.Feed.Link.CaptureGroupConnector,
+			LinkI:        inputInfo.Feed.Link.CaptureGroupIndex,
+			DescriptionC: inputInfo.Feed.Description.CaptureGroupConnector,
+			DescriptionI: inputInfo.Feed.Description.CaptureGroupIndex,
+			AuthorNameC:  inputInfo.Feed.AuthorName.CaptureGroupConnector,
+			AuthorNameI:  inputInfo.Feed.AuthorName.CaptureGroupIndex,
+			AuthorEmailC: inputInfo.Feed.AuthorEmail.CaptureGroupConnector,
+			AuthorEmailI: inputInfo.Feed.AuthorEmail.CaptureGroupIndex,
+		},
+		inputInfo.MaxFeedItems,
+	)
+	if err != nil {
+		other.PrintError("Failed to generate Atom RSS feed from HTML:\n\n"+rawHTML+"\n\n"+err.Error(), 1)
+	}
+	fmt.Println(*atomXML)
+}