Get into working state

- Make compatible with at least go 1.19
- Fix paths
- Add more log output
- Update README.md
- Fix runner
- Update systemd service

README.md

@@ -12,7 +12,7 @@ The idea is to let it watch a directory any scanner will scan into, and then thi
 
 1. Install the project somewhere.
 2. Edit [main.go](main.go) to use the correct paths to your scanner and paperless consumption directories.
-3. Copy the [ocrmypdf-runner.service](service/linux/systemd/ocrmypdf-runner.service) into your paperless systemd services directory (`%HOME/.config/systemd/user/ocrmypdf-runner.service`).
-4. `systemctl --user daemon-reload`
-5. `systemctl --user enable ocrmypdf-runner.service`
-6. `systemctl --user start ocrmypdf-runner.service`
+3. Copy the [ocrmypdf-runner.service](service/linux/systemd/ocrmypdf-runner.service) into your systemd services directory (`etc/systemd/system/ocrmypdf-runner.service`). You may have to change the user in the service file.
+4. `systemctl daemon-reload`
+5. `systemctl enable ocrmypdf-runner.service`
+6. `systemctl start ocrmypdf-runner.service`

go.mod

@@ -1,3 +1,3 @@
 module ocrmypdf-runner
 
-go 1.23.0
+go 1.19

init.go

@@ -7,8 +7,8 @@ import (
 
 func init() {
 	runner := Runner{
-		InputPatterns: []string{filepath.Join(".", "input", "*.pdf")},
-		OutputPath:    filepath.Join("."),
+		InputPatterns: []string{filepath.Join("/", "media", "vault", "Paperless", "Consume", "input", "*.pdf")},
+		OutputPath:    filepath.Join("/", "media", "vault", "Paperless", "Consume", "processed"),
 		Interval:      5 * time.Second,
 	}
 	runner.Run()

main.go

@@ -1,6 +1,10 @@
 package main
 
+import "log"
+
 func main() {
+	log.Printf("Starting OCRmyPDF-runner.")
+
 	// The runners will be started in some init.go file.
 
 	// Wait forever.

runner.go

@@ -1,6 +1,7 @@
 package main
 
 import (
+	"errors"
 	"log"
 	"os"
 	"os/exec"
@@ -23,33 +24,54 @@ func (r Runner) Run() {
 	}
 
 	go func() {
+		log.Printf("Starting runner polling %q every %s.", r.InputPatterns, r.Interval)
+
 		ticker := time.NewTicker(r.Interval)
 		for range ticker.C {
 
 			for _, inputPattern := range r.InputPatterns {
 				filenames, err := filepath.Glob(inputPattern)
 				if err != nil {
-					log.Panicf("Failed to get input files: %v", err)
+					log.Panicf("Failed to get input files: %v.", err)
 				}
 
 				for _, filename := range filenames {
 
-					outputPath := filepath.Join(r.OutputPath, filepath.Base(filename))
+					base := filepath.Base(filename)
+					outputPath := filepath.Join(r.OutputPath, base)
+
+					log.Printf("Starting to process %q into %q.", base, outputPath)
 
 					args := []string{
-						"-l deu+eng",
+						"-l", "deu+eng",
 						"--rotate-pages",
-						"--optimize 1",
+						"--optimize", "1",
+						"--skip-text",
 					}
 					args = append(args, filename, outputPath)
 					cmd := exec.Command(OCRMyPDFExecutable, args...)
 
-					if err := cmd.Run(); err != nil {
-						log.Panicf("Failed to run OCRmyPDF: %v", err)
+					if output, err := cmd.CombinedOutput(); err != nil {
+						log.Printf("OCRmyPDF stdout + stderr: %v.", string(output))
+						var exitErr *exec.ExitError
+						if errors.As(err, &exitErr) {
+							log.Printf("OCRmyPDF failed with ExitCode: %v.", exitErr)
+							log.Printf("We will move the document into the output directory instead.")
+							if err := os.Rename(filename, outputPath); err != nil {
+								log.Panicf("Failed to move %q to %q.", filename, outputPath)
+							}
+							log.Printf("Exec: %q", cmd)
+						} else {
+							log.Panicf("Failed to run OCRmyPDF: %v.", err)
+						}
+					} else {
+						// Only delete the PDF if the previous steps did succeed.
+						if err := os.Remove(filename); err != nil {
+							log.Printf("Failed to remove original document: %v.", err)
+						}
 					}
 
-					// Only delete the PDF if the previous steps did succeed.
-					os.Remove(filename)
+					log.Printf("Finished processing of %q.", base)
 				}
 			}
 

service/linux/systemd/ocrmypdf-runner.service

@@ -2,8 +2,12 @@
 Description=A runner that will watch directories and runs OCRmyPDF on files in them.
 
 [Service]
+Restart=on-failure
+RestartSec=60s
 WorkingDirectory=/home/paperless/ocrmypdf-runner/
 ExecStart=go run .
+User=paperless
+Group=paperless
 
 [Install]
-WantedBy=default.target
+WantedBy=multi-user.target