naive approach

.devcontainer/devcontainer.json

@@ -1,10 +1,15 @@
 // For format details, see https://aka.ms/devcontainer.json. For config options, see the
-// README at: https://github.com/devcontainers/templates/tree/main/src/go
+// README at: https://github.com/devcontainers/templates/tree/main/src/ubuntu
 {
 	"name": "Go - Fido",
 	// Or use a Dockerfile or Docker Compose file. More info: https://containers.dev/guide/dockerfile
-	"image": "mcr.microsoft.com/devcontainers/go:1-1.21-bullseye",
-	"postCpost_create.shreateCommand": "bash .devcontainer/",
+	"image": "mcr.microsoft.com/devcontainers/base:jammy",
+	"postCreateCommand": "bash .devcontainer/",
+	"features": {
+		"ghcr.io/devcontainers/features/go:1": {
+			"version": "latest"
+		}
+	}
 
 	// Features to add to the dev container. More info: https://containers.dev/features.
 	// "features": {},
@@ -13,7 +18,7 @@
 	// "forwardPorts": [],
 
 	// Use 'postCreateCommand' to run commands after the container is created.
-	// "postCreateCommand": "go version",
+	// "postCreateCommand": "uname -a",
 
 	// Configure tool-specific properties.
 	// "customizations": {},

.devcontainer/post_create.sh

@@ -2,3 +2,6 @@
 
 go install github.com/go-task/task/v3/cmd/task@latest
 go install github.com/cosmtrek/air@latest
+
+sudo apt update && sudo apt install -y python3 python3-pip
+pip3 install linkchecker

.github/dependabot.yml

@@ -0,0 +1,12 @@
+# To get started with Dependabot version updates, you'll need to specify which
+# package ecosystems to update and where the package manifests are located.
+# Please see the documentation for more information:
+# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
+# https://containers.dev/guide/dependabot
+
+version: 2
+updates:
+ - package-ecosystem: "devcontainers"
+   directory: "/"
+   schedule:
+     interval: weekly

.gitignore

@@ -21,3 +21,6 @@
 # Go workspace file
 go.work
 
+tmp
+
+__debug_bin*

.vscode/launch.json

@@ -0,0 +1,15 @@
+{
+    // Use IntelliSense to learn about possible attributes.
+    // Hover to view descriptions of existing attributes.
+    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+    "version": "0.2.0",
+    "configurations": [
+        {
+            "name": "Launch Package",
+            "type": "go",
+            "request": "launch",
+            "mode": "auto",
+            "program": "${workspaceFolder}"
+        }
+    ]
+}

go.mod

@@ -0,0 +1,5 @@
+module codeberg.org/meadowingc/fido
+
+go 1.22.0
+
+require github.com/google/uuid v1.6.0

go.sum

@@ -0,0 +1,2 @@
+github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
+github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=

linkchecker/linkchecker.go

@@ -0,0 +1,111 @@
+package linkchecker
+
+import (
+	"bytes"
+	"errors"
+	"log"
+	"os/exec"
+	"strings"
+)
+
+type LinkCheckResult struct {
+	Summary     string
+	FoundErrors []LinkCheckError
+}
+
+type LinkCheckError struct {
+	Name           string
+	ParentURL      string
+	RealURL        string
+	CheckTime      string
+	Warning        string
+	CheckingResult string
+}
+
+func getAnalysisValueForKey(key string, line string) string {
+	components := strings.Split(line, key)
+	value := strings.Join(components[1:], " ")
+	value = strings.TrimSpace(value)
+	return value
+}
+
+func CheckLink(link string) (LinkCheckResult, error) {
+	// linkchecker https://meadow.bearblog.dev -o csv --no-status --no-warnings --timeout 1200
+	cmd := exec.Command("linkchecker", link, "-o", "text", "--no-status", "--no-warnings", "--timeout", "1200")
+
+	var out bytes.Buffer
+	cmd.Stdout = &out
+
+	err := cmd.Run()
+
+	if err != nil {
+
+		if exitError, ok := err.(*exec.ExitError); ok {
+			// if exit code is 0 then we can return early since there were no issues
+			if exitError.ExitCode() == 0 {
+				return LinkCheckResult{
+					Summary: "No issues found",
+				}, nil
+			}
+
+			if exitError.ExitCode() == 2 {
+				return LinkCheckResult{}, errors.New("there was an error while checking for links")
+			}
+
+			// exit code of 1 is expected IF there were any issues
+		} else {
+			return LinkCheckResult{}, errors.New("there was an UNKNOWN error while checking for links")
+		}
+	}
+
+	outStr := out.String()
+
+	resultsChunk := strings.Split(outStr, "Start checking at")[1]
+	resultsChunk = strings.Split(resultsChunk, "Stopped checking at")[0]
+
+	summaryLine := strings.Split(resultsChunk, "That's it. ")[1]
+	summaryLine = strings.TrimSpace(summaryLine)
+
+	analysisResultsChunk := strings.Split(resultsChunk, "Statistics:")[0]
+	analysisResults := strings.TrimSpace(analysisResultsChunk)
+
+	resultsArray := strings.Split(analysisResults, "\n\n")
+
+	allCheckResults := []LinkCheckError{}
+
+	for _, result := range resultsArray {
+		linkCheckError := LinkCheckError{}
+
+		for _, line := range strings.Split(result, "\n") {
+			if strings.HasPrefix(line, "Real URL") {
+				linkCheckError.RealURL = getAnalysisValueForKey("URL", line)
+			} else if strings.HasPrefix(line, "Name") {
+				linkCheckError.Name = getAnalysisValueForKey("Name", line)
+				linkCheckError.Name = strings.TrimLeft(linkCheckError.Name, "`")
+				linkCheckError.Name = strings.TrimRight(linkCheckError.Name, "'")
+			} else if strings.HasPrefix(line, "Parent URL") {
+				linkCheckError.ParentURL = getAnalysisValueForKey("Parent URL", line)
+				linkCheckError.ParentURL = strings.Split(linkCheckError.ParentURL, ",")[0]
+			} else if strings.HasPrefix(line, "Check time") {
+				linkCheckError.CheckTime = getAnalysisValueForKey("Check time", line)
+			} else if strings.HasPrefix(line, "Warning") {
+				linkCheckError.Warning = getAnalysisValueForKey("Warning", line)
+			} else if strings.HasPrefix(line, "          ") {
+				linkCheckError.Warning += " " + strings.TrimSpace(line)
+			} else if strings.HasPrefix(line, "Result") {
+				linkCheckError.CheckingResult = getAnalysisValueForKey("Result", line)
+			} else {
+				log.Println("Unknown line: ", line)
+			}
+		}
+
+		if linkCheckError.Name != "" {
+			allCheckResults = append(allCheckResults, linkCheckError)
+		}
+	}
+
+	return LinkCheckResult{
+		Summary:     summaryLine,
+		FoundErrors: allCheckResults,
+	}, nil
+}

linkchecker/scheduler.go

@@ -0,0 +1,70 @@
+package linkchecker
+
+import (
+	"log"
+	"sync"
+	"time"
+
+	"github.com/google/uuid"
+)
+
+type CheckLinkRequest struct {
+	UUID    string
+	PageURL string
+}
+
+type SchedulerResult struct {
+	UUID   string
+	URL    string
+	Status string
+	Result *LinkCheckResult
+}
+
+var (
+	uidToResultMap = make(map[string]*SchedulerResult)
+	mapLock        = &sync.RWMutex{}
+)
+
+func GetResultForUUID(uid string) *SchedulerResult {
+	mapLock.RLock()
+	defer mapLock.RUnlock()
+	return uidToResultMap[uid]
+}
+
+func SubmitLinkForCheck(link string) string {
+	// TODO better for this to be the sha256 of the link
+	uid := uuid.New().String()
+
+	uidToResultMap[uid] = &SchedulerResult{
+		UUID:   uid,
+		URL:    link,
+		Status: "PENDING",
+	}
+
+	log.Printf("Submitted link %s for check with UUID %s", link, uid)
+
+	go func() {
+		result, err := CheckLink(link)
+		log.Printf("Link check for %s completed with %d errors", link, len(result.FoundErrors))
+
+		mapLock.Lock()
+		if err != nil {
+			uidToResultMap[uid].Status = err.Error()
+			return
+		}
+
+		uidToResultMap[uid].Result = &result
+		uidToResultMap[uid].Status = "COMPLETED"
+		mapLock.Unlock()
+
+		// keep the results around for 24 hours and then remove from the map
+		go func() {
+			<-time.After(24 * time.Hour)
+			mapLock.Lock()
+			delete(uidToResultMap, uid)
+			mapLock.Unlock()
+		}()
+	}()
+
+	return uid
+}

main.go

@@ -0,0 +1,60 @@
+package main
+
+import (
+	"fmt"
+	"html/template"
+	"net/http"
+
+	"codeberg.org/meadowingc/fido/linkchecker"
+)
+
+func main() {
+	templates := template.Must(template.ParseGlob("templates/*.tmpl.html"))
+
+	http.HandleFunc("/", func(w http.ResponseWriter, r *http.Request) {
+		err := templates.ExecuteTemplate(w, "home.tmpl.html", nil)
+		if err != nil {
+			http.Error(w, err.Error(), http.StatusInternalServerError)
+		}
+	})
+
+	http.HandleFunc("POST /submit-eval-request", func(w http.ResponseWriter, r *http.Request) {
+		// get submitted link from request and evaluate all links on that page
+		// for broken links
+		submittedLink := r.FormValue("link")
+		if submittedLink == "" {
+			http.Error(w, "Link is required", http.StatusBadRequest)
+			return
+		}
+
+		resultUUID := linkchecker.SubmitLinkForCheck(submittedLink)
+
+		// redirect to results page
+		http.Redirect(w, r, fmt.Sprintf("/result/%s", resultUUID), http.StatusSeeOther)
+	})
+
+	http.HandleFunc("/result/{operation_id}", func(w http.ResponseWriter, r *http.Request) {
+		opId := r.PathValue("operation_id")
+
+		// get result for operation ID
+		result := linkchecker.GetResultForUUID(opId)
+		if result == nil {
+			http.Error(w, "No result found for operation ID", http.StatusNotFound)
+			return
+		}
+
+		// TODO remove this
+		templates = template.Must(template.ParseGlob("templates/*.tmpl.html"))
+		err := templates.ExecuteTemplate(w, "result.tmpl.html", result)
+		if err != nil {
+			http.Error(w, err.Error(), http.StatusInternalServerError)
+		}
+	})
+
+	http.HandleFunc("/contact", func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "Welcome to the contact page!")
+	})
+
+	fmt.Println("Server is running on http://localhost:8080")
+	http.ListenAndServe(":8080", nil)
+}

templates/footer.tmpl.html

@@ -0,0 +1,4 @@
+{{define "footer"}}
+</body>
+</html>
+{{end}}

templates/header.tmpl.html

@@ -0,0 +1,8 @@
+{{define "header"}}
+<!DOCTYPE html>
+<html>
+<head>
+    <title>Fido 🐶</title>
+</head>
+<body>
+{{end}}

templates/home.tmpl.html

@@ -0,0 +1,11 @@
+{{template "header"}}
+
+<h1>Fido 🐶!</h1>
+
+<form action="/submit-eval-request" method="post">
+    <label for="link">Link:</label><br>
+    <input type="text" id="link" name="link"><br>
+    <input type="submit" value="Submit">
+</form>
+
+{{template "footer"}}

templates/result.tmpl.html

@@ -0,0 +1,37 @@
+{{template "header"}}
+
+<h1>Results</h1>
+
+<b>Status:</b> {{.Status}}
+
+{{- if ne .Status "PENDING" -}}
+
+    {{$result := .Result}}
+    <p><b>Summary:</b> {{$result.Summary}}</p>
+
+    {{- if eq 0 (len $result.FoundErrors) -}}
+        <p>No broken links found! 🎉</p>
+    {{- end -}}
+
+    {{- range $i, $v := $result.FoundErrors -}}
+    <b>Link:</b> {{$i}}<br>
+    <ul>
+        <li><b>ParentURL:</b> <a target="_blank" href="{{$v.ParentURL}}">{{$v.ParentURL}}</a></li>
+        <li><b>Link Text:</b> <code>{{$v.Name}}</code></li>
+        <li><b>Broken URL:</b> <code>{{$v.RealURL}}</code></li>
+        <li><b>Error:</b> <code>{{$v.CheckingResult}}</code></li>
+    </ul>
+    {{- end -}}
+
+{{- end -}}
+
+{{- if eq .Status "PENDING" -}}
+<script type="text/javascript">
+    // Reload the page every 5 seconds
+    setTimeout(function(){
+        window.location.reload(1);
+    }, 5000);
+</script>
+{{- end -}}
+
+{{template "footer"}}