Improve office conversion

.gitignore

@@ -21,3 +21,12 @@
 # Go workspace file
 go.work
 
+# Locally used configuration file
+/config.yaml
+/ncConverter.json
+
+# Vite build file
+/.vite
+
+# Log files
+*.log

BUILD.md

@@ -0,0 +1,10 @@
+## Ausführen
+
+set GOTMPDIR=C:\MYCOMP
+go run .\cmd\web
+npm run dev
+
+npm run build
+
+
+nodemon --watch './**/*.go' --signal SIGTERM --exec 'go' run cmd/MyProgram/main.go

README.md

@@ -1,3 +1,32 @@
 # ncDocConverter
 
-A Go program able to convert Office Documents automatically to PDF / EPUB Files via OnlyOffice
+A Go program able to convert documents automatically to PDF / EPUB Files.
+
+Currently, the following sources for documents are supported:
+
+* Nextcloud with OnlyOffice
+* Boockstack
+
+As a destination to save the converted files only **Nextcloud** is supported.
+
+
+## Setting it up
+
+For using the 
+
+
+### BookStack
+
+For converting books of BookStack you need to create an API token for the user to access the books:
+1. Login as Admin
+2. Go to *Settings → Users*
+3. Select user for API access
+4. Scroll down to `API Tokens` and click `CREATE TOKEN`
+5. Set a name and expire date. Click `save`
+6. Copy the ID and Token. The field `apiToken` will contain the combination from `id:token`
+
+Now you need also create a new role or edit an existing role.
+1. Go to *Settings → Roles*
+2. Edit and existing Role (the role which the user have) or create a new role
+3. Check the box `Access system API` and `Export content` in `System permissions`
+4. Assing View Role *(all and own)* for *Shelves, Books, Chapters and Pages* 

cmd/ncDocConverth/main.go

@@ -0,0 +1,73 @@
+package main
+
+import (
+	"crypto/tls"
+	"net/http"
+	"time"
+
+	"rpjosh.de/ncDocConverter/internal/models"
+	"rpjosh.de/ncDocConverter/internal/ncworker"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+)
+
+type WebApplication struct {
+	logger *logger.Logger
+	config *models.WebConfig
+}
+
+func main() {
+	defer logger.CloseFile()
+
+	config, err := models.SetConfig()
+	if err != nil {
+		logger.Error(err.Error())
+	}
+
+	tlsConfig := &tls.Config{
+		CurvePreferences: []tls.CurveID{tls.X25519, tls.CurveP256},
+		MinVersion:       tls.VersionTLS12,
+		CipherSuites: []uint16{
+			tls.TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384,
+			tls.TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384,
+			tls.TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305,
+			tls.TLS_ECDHE_RSA_WITH_CHACHA20_POLY1305,
+			tls.TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256,
+			tls.TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256,
+		},
+	}
+
+	webApp := WebApplication{
+		logger: logger.GetGlobalLogger(),
+		config: config,
+	}
+
+	srv := &http.Server{
+		Addr:         config.Server.Address,
+		ErrorLog:     nil,
+		Handler:      webApp.routes(),
+		TLSConfig:    tlsConfig,
+		IdleTimeout:  time.Minute,
+		ReadTimeout:  5 * time.Second,
+		WriteTimeout: 10 * time.Second,
+	}
+
+	ncConvertUsers, err := models.ParseConvertUsers("./ncConverter.json")
+	if err != nil {
+		logger.Error("Unable to parse the file %s: %s", "dd", err)
+	}
+	ncworker.NewScheduler(ncConvertUsers, config)
+
+	if 1 == 1 {
+		return
+	}
+
+	logger.Info("Server started on %s", config.Server.Address)
+	var errw error
+	if config.Server.Certificate == "" {
+		errw = srv.ListenAndServe()
+	} else {
+		errw = srv.ListenAndServeTLS(config.Server.Certificate+"cert.pem", config.Server.Certificate+"key.pem")
+	}
+
+	logger.Error("Failed to run the HTTP Server: %s", errw)
+}

cmd/ncDocConverth/middleware.go

@@ -0,0 +1,68 @@
+package main
+
+import (
+	"fmt"
+	"net/http"
+	"runtime/debug"
+
+	"github.com/justinas/nosurf"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+)
+
+func secureHeaders(next http.Handler) http.Handler {
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		// Note: This is split across multiple lines for readability. You don't
+		// need to do this in your own code.
+		w.Header().Set("Content-Security-Policy",
+			//"default-src 'self' localhost:*; style-src 'self' fonts.googleapis.com localhost:*; font-src fonts.gstatic.com")
+			"default-src * 'unsafe-inline' 'unsafe-eval'; script-src * 'unsafe-inline' 'unsafe-eval'; connect-src * 'unsafe-inline'; img-src * data: blob: 'unsafe-inline'; frame-src *; style-src * 'unsafe-inline';")
+		w.Header().Set("Referrer-Policy", "origin-when-cross-origin")
+		w.Header().Set("X-Content-Type-Options", "nosniff")
+		w.Header().Set("X-Frame-Options", "deny")
+		w.Header().Set("X-XSS-Protection", "0")
+
+		next.ServeHTTP(w, r)
+	})
+}
+
+func (app *WebApplication) logRequest(next http.Handler) http.Handler {
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		logger.Info("%s - %s %s %s", r.RemoteAddr, r.Proto, r.Method, r.URL.RequestURI())
+
+		next.ServeHTTP(w, r)
+	})
+}
+
+func (app *WebApplication) recoverPanic(next http.Handler) http.Handler {
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		// Create a deferred function (which will always be run in the event
+		// of a panic as Go unwinds the stack).
+		defer func() {
+			// Use the builtin recover function to check if there has been a
+			// panic or not. If there has...
+			if err := recover(); err != nil {
+				// Set a "Connection: close" header on the response.
+				w.Header().Set("Connection", "close")
+				// Call the app.serverError helper method to return a 500
+				// Internal Server response.
+				trace := fmt.Sprintf("%s\n%s", fmt.Errorf("%s", err).Error(), debug.Stack())
+				logger.Error(trace)
+			}
+		}()
+
+		next.ServeHTTP(w, r)
+	})
+}
+
+// Create a NoSurf middleware function which uses a customized CSRF cookie with
+// the Secure, Path and HttpOnly attributes set.
+func noSurf(next http.Handler) http.Handler {
+	csrfHandler := nosurf.New(next)
+	csrfHandler.SetBaseCookie(http.Cookie{
+		HttpOnly: true,
+		Path:     "/",
+		Secure:   true,
+	})
+
+	return csrfHandler
+}

cmd/ncDocConverth/routes.go

@@ -0,0 +1,20 @@
+package main
+
+import (
+	"net/http"
+
+	"github.com/go-chi/chi/v5"
+	"github.com/go-chi/chi/v5/middleware"
+	"rpjosh.de/ncDocConverter/internal/api"
+)
+
+func (app *WebApplication) routes() http.Handler {
+	api := api.Api{Logger: app.logger, Config: app.config}
+
+	router := chi.NewRouter()
+	router.Use(middleware.RealIP, app.recoverPanic, app.logRequest, secureHeaders)
+
+	api.SetupServer(router)
+
+	return router
+}

configs/config.yaml

@@ -0,0 +1,19 @@
+server:
+  # Address to listen on
+  address: ":4000"
+
+  # Path to the folder with the certificates file (cert.pem and key.pem) for using TLS
+  certificate: "/etc/letsencrypt/live/"
+
+  # If this parameter is given, all jobs are executed immediately after starting the program.
+  # Afterward the program does exit -> The "execution" field in the jobs are going to be ignored
+  oneShot: false
+
+logging:
+  # Minimum log Level for printing to the console (debug, info, warning, error, fatal)
+  printLogLevel: info
+  # Minimum log level for writing into the log file (debug, info, warning, error, fatal)
+  writeLogLevel: warning
+
+  # File path to log (empty = disabled)
+  logFilePath: "/home/myUser/logs/ncDocConverter.live"

configs/ncConverter.hjson

@@ -0,0 +1,78 @@
+{
+    "nextcloudUsers": [
+        {
+            // Nextcloud user and instance to save the converted files
+            "nextcloudUrl": "https://cloud.rpjosh.de",
+            "username":     "exchange",
+            "password":     "Zj9cQ-eF3n6-R6DSt-8sJXf-kYseJ",
+    
+            // OnlyOffice (docx, xlsx, ...) convertion to pdf
+            "jobs": [
+                {
+                    "jobName":          "Convert my books",
+                    "sourceDir":        "api/",
+                    "destinationDir":   "ebooks/",
+
+                    // Keep folders of source
+                    // Otherwise all files will be saved in the destination dir
+                    "keepFolders":      true,
+
+                    // If the folder should be searched recursive
+                    "recursive":        true,
+                    
+                    // Execution date in the cron format
+                    "execution":        "45 23 * * 6"
+                }
+            ],
+            
+            // Conversion from bookStack to pdf/html
+            "bookStack": {
+                "url":          "https://wiki.rpjosh.de",
+                "username":     "test@rpjosh.de",
+                "apiToken":     "typCf2LoSQDHicpeeAZQCDZwAq7BvVdl:PcKMZVDrIwEJeKIKyaD7w0cf20JCjpZz",
+
+                "jobs": [
+                    {
+                        "jobName":  "Convert my favorite books",
+                        
+                        // Shelves to filter -> convert only shelves with the names "Work" and "Linux" 
+                        // Leave empty to convert books in all shelves
+                        "shelves": [ "Work", "Linux" ],
+                        // Regex to filter after the shelv name
+                        "shelveRegex": "",
+
+                        // Books to filter (see shelves for more informations)
+                        "books": [],
+                        "booksRegex": "",
+
+                        // If books which doesn't belong to an shelf should also be converted.
+                        // The will be placed in the root folder.
+                        // Note that the field "shelves" and "shelveRegex" doesn't work as expected
+                        // (Books inside this shelve won't be excluded but will be placed in the root)
+                        "includeBooksWithoutShelve": false,
+
+                        // Destination folder to save the converted documents in nextcloud
+                        "destinationDir": "ebooks/wiki/",
+
+                        // Export format (html or pdf)
+                        "format": "html",
+
+                        // If the books should be saved inside the shelves folder
+                        // Otherwise all files will be saved in the destination dir
+                        "keepStructure": true,
+
+                        // Execution date in the cron format
+                        "execution": "45 23 * * 6",
+
+                        // The fetching of books and shelves can be ressource hungry. This value specifies the number
+                        // of jobs that are executed with cached data.
+                        // Note that new or deleted books and shelves won't be converted until the cache counter
+                        // expires. Changes in existing books will still be noted.
+                        // Specify zero to disable the cache
+                        "cache": 3
+                    }
+                ]
+            }
+        }
+    ]
+}

configs/ncConverter.json

@@ -0,0 +1,21 @@
+{
+    "users": [
+        {
+            "authUser":         "Unique Username (from ",
+            "nextcloudUrl":     "https://cloud.myDomain.de",
+            "username":         "MyUsername",
+            "password":         "App Password",
+    
+            "jobs": [
+                {
+                    "jobName":          "",
+                    "sourceDir":        "Arbeit/",
+                    "destinationDir":   "Ebooks/",
+                    "keepFolders":      true,
+                    "recursive":        true,
+                    "execution":        [ "01:00", "15:00" ]
+                }
+            ]
+        }
+    ]
+}

go.mod

@@ -1,3 +1,16 @@
 module rpjosh.de/ncDocConverter
 
 go 1.18
+
+require (
+	github.com/go-chi/chi/v5 v5.0.8 // indirect
+	github.com/go-co-op/gocron v1.18.0 // indirect
+	github.com/go-yaml/yaml v2.1.0+incompatible // indirect
+	github.com/justinas/nosurf v1.1.1 // indirect
+	github.com/robfig/cron/v3 v3.0.1 // indirect
+	golang.org/x/sync v0.1.0 // indirect
+	gopkg.in/yaml.v3 v3.0.1 // indirect
+)
+
+// https://zhwt.github.io/yaml-to-go/
+// https://www.onlinetool.io/xmltogo/

go.sum

@@ -0,0 +1,21 @@
+github.com/go-chi/chi/v5 v5.0.7 h1:rDTPXLDHGATaeHvVlLcR4Qe0zftYethFucbjVQ1PxU8=
+github.com/go-chi/chi/v5 v5.0.7/go.mod h1:DslCQbL2OYiznFReuXYUmQ2hGd1aDpCnlMNITLSKoi8=
+github.com/go-chi/chi/v5 v5.0.8 h1:lD+NLqFcAi1ovnVZpsnObHGW4xb4J8lNmoYVfECH1Y0=
+github.com/go-chi/chi/v5 v5.0.8/go.mod h1:DslCQbL2OYiznFReuXYUmQ2hGd1aDpCnlMNITLSKoi8=
+github.com/go-co-op/gocron v1.18.0 h1:SxTyJ5xnSN4byCq7b10LmmszFdxQlSQJod8s3gbnXxA=
+github.com/go-co-op/gocron v1.18.0/go.mod h1:sD/a0Aadtw5CpflUJ/lpP9Vfdk979Wl1Sg33HPHg0FY=
+github.com/go-yaml/yaml v2.1.0+incompatible h1:RYi2hDdss1u4YE7GwixGzWwVo47T8UQwnTLB6vQiq+o=
+github.com/go-yaml/yaml v2.1.0+incompatible/go.mod h1:w2MrLa16VYP0jy6N7M5kHaCkaLENm+P+Tv+MfurjSw0=
+github.com/justinas/nosurf v1.1.1 h1:92Aw44hjSK4MxJeMSyDa7jwuI9GR2J/JCQiaKvXXSlk=
+github.com/justinas/nosurf v1.1.1/go.mod h1:ALpWdSbuNGy2lZWtyXdjkYv4edL23oSEgfBT1gPJ5BQ=
+github.com/robfig/cron/v3 v3.0.1 h1:WdRxkvbJztn8LMz/QEvLN5sBU+xKpSqwwUO1Pjr4qDs=
+github.com/robfig/cron/v3 v3.0.1/go.mod h1:eQICP3HwyT7UooqI/z+Ov+PtYAWygg1TEWWzGIFLtro=
+github.com/studio-b12/gowebdav v0.0.0-20220128162035-c7b1ff8a5e62 h1:b2nJXyPCa9HY7giGM+kYcnQ71m14JnGdQabMPmyt++8=
+github.com/studio-b12/gowebdav v0.0.0-20220128162035-c7b1ff8a5e62/go.mod h1:bHA7t77X/QFExdeAnDzK6vKM34kEZAcE1OX4MfiwjkE=
+golang.org/x/sync v0.0.0-20210220032951-036812b2e83c h1:5KslGYwFpkhGh+Q16bwMP3cOontH8FOep7tGV86Y7SQ=
+golang.org/x/sync v0.0.0-20210220032951-036812b2e83c/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/sync v0.1.0 h1:wsuoTGHzEhffawBOhz5CYhcrV4IdKZbEyZjBMuTp12o=
+golang.org/x/sync v0.1.0/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
+gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=

internal/api/app.go

@@ -0,0 +1,17 @@
+package api
+
+import (
+	"github.com/go-chi/chi/v5"
+
+	"rpjosh.de/ncDocConverter/internal/models"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+)
+
+type Api struct {
+	Logger 	*logger.Logger
+	Config 	*models.WebConfig
+}
+
+func (api *Api) SetupServer(router *chi.Mux) {
+	api.routes(router)
+}

internal/api/routes.go

@@ -0,0 +1,7 @@
+package api
+
+import "github.com/go-chi/chi/v5"
+
+func (api *Api) routes(router *chi.Mux) {
+
+}

internal/models/bookstack.go

@@ -0,0 +1,38 @@
+package models
+
+// BookStack details to fetch books from
+type BookStack struct {
+	URL      string `json:"url"`
+	Username string `json:"username"`
+	Token    string `json:"apiToken"`
+
+	Jobs []BookStackJob `json:"jobs"`
+}
+
+// A concrete BookStacksJob
+type BookStackJob struct {
+	JobName        string `json:"jobName"`
+	DestinationDir string `json:"destinationDir"`
+
+	Shelves      []string `json:"shelves"`
+	ShelvesRegex string   `json:"shelveRegex"`
+
+	Books      []string `json:"books"`
+	BooksRegex string   `json:"booksRegex"`
+
+	IncludeBooksWithoutShelve bool   `json:"includeBooksWithoutShelve"`
+	Format                    Format `json:"format"`
+	KeepStructure             bool   `json:"keepStructure"`
+
+	Recursive string `json:"recursive"`
+	Execution string `json:"execution"`
+
+	CacheCount int `json:"cache"`
+}
+
+type Format string
+
+const (
+	HTML Format = "html"
+	PDF  Format = "pdf"
+)

internal/models/ncconvert.go

@@ -0,0 +1,57 @@
+package models
+
+import (
+	"encoding/json"
+	"fmt"
+	"io/ioutil"
+	"os"
+)
+
+// The root nextcloud user where the files are stored
+// and the files for onlyoffice jobs are defined
+type NextcloudUser struct {
+	NextcloudBaseUrl string `json:"nextcloudUrl"`
+	Username         string `json:"username"`
+	Password         string `json:"password"`
+
+	// OnlyOffice
+	ConvertJobs []NcConvertJob `json:"jobs"`
+
+	// BookStack
+	BookStack BookStack `json:"bookStack"`
+}
+
+// A OnlyOffice docs convert job
+type NcConvertJob struct {
+	JobName        string `json:"jobName"`
+	SourceDir      string `json:"sourceDir"`
+	DestinationDir string `json:"destinationDir"`
+	KeepFolders    string `json:"keepFolders"`
+	Recursive      string `json:"recursive"`
+	Execution      string `json:"execution"`
+}
+
+type NcConvertUsers struct {
+	Users []NextcloudUser `json:"nextcloudUsers"`
+}
+
+// Parses the given file to the in memory struct
+func ParseConvertUsers(filePath string) (*NcConvertUsers, error) {
+
+	file, err := os.OpenFile(filePath, os.O_APPEND|os.O_CREATE, 0644)
+	if err != nil {
+		return nil, fmt.Errorf("failed to open the file '%s': %s", filePath, err)
+	}
+	defer file.Close()
+
+	byteValue, err := ioutil.ReadAll(file)
+	if err != nil {
+		return nil, fmt.Errorf("failed to parse 'ncConverter.json': %s", err)
+	}
+
+	var conv NcConvertUsers
+
+	json.Unmarshal(byteValue, &conv)
+
+	return &conv, nil
+}

internal/models/webconfig.go

@@ -0,0 +1,95 @@
+package models
+
+import (
+	"flag"
+	"os"
+
+	yaml "gopkg.in/yaml.v3"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+)
+
+type WebConfig struct {
+	Server  Server  `yaml:"server"`
+	Logging Logging `yaml:"logging"`
+}
+
+type Server struct {
+	Address     string `yaml:"address"`
+	Certificate string `yaml:"certificate"`
+	OneShot     bool   `yaml:"oneShot"`
+}
+
+type Logging struct {
+	PrintLogLevel string `yaml:"printLogLevel"`
+	WriteLogLevel string `yaml:"writeLogLevel"`
+	LogFilePath   string `yaml:"logFilePath"`
+}
+
+// Parses the configuration file (.yaml file) to an WebConfiguration
+func ParseWebConfig(webConfig *WebConfig, file string) (*WebConfig, error) {
+	if file == "" {
+		return webConfig, nil
+	}
+
+	dat, err := os.ReadFile(file)
+	if err != nil {
+		return nil, err
+	}
+
+	if err := yaml.Unmarshal(dat, &webConfig); err != nil {
+		return nil, err
+	}
+
+	return webConfig, nil
+}
+
+func getDefaultConfig() *WebConfig {
+	return &WebConfig{
+		Server: Server{
+			Address: ":4000",
+		},
+		Logging: Logging{
+			PrintLogLevel: "info",
+			WriteLogLevel: "warning",
+		},
+	}
+}
+
+// Applies the cli and the configuration options from the config files
+func SetConfig() (*WebConfig, error) {
+	configPath := "./config.yaml"
+	// the path of the configuration file is needed first to determine the "default" values
+	for i, arg := range os.Args {
+		if arg == "-config" || arg == "--config" && len(os.Args) > i {
+			configPath = os.Args[i+1]
+			break
+		}
+	}
+	webConfig := getDefaultConfig()
+	webConfig, err := ParseWebConfig(webConfig, configPath)
+	if err != nil {
+		logger.Error("Unable to parse the configuration file '%s': %s", configPath, err)
+		webConfig = getDefaultConfig()
+		err = nil
+	}
+
+	_ = flag.String("config", "./config.yaml", "Path to the configuration file (see configs/config.yaml) for an example")
+	address := flag.String("address", webConfig.Server.Address, "Address and port on which the api and the web server should listen to")
+	printLogLevel := flag.String("printLogLevel", webConfig.Logging.PrintLogLevel, "Minimum log level to log (debug, info, warning, error, fatal)")
+	oneShot := flag.Bool("oneShot", webConfig.Server.OneShot, "All jobs are executed immediately and the program exists afterwards")
+
+	flag.Parse()
+	webConfig.Server.Address = *address
+	webConfig.Logging.PrintLogLevel = *printLogLevel
+	webConfig.Server.OneShot = *oneShot
+
+	defaultLogger := logger.Logger{
+		PrintLevel:  logger.GetLevelByName(webConfig.Logging.PrintLogLevel),
+		LogLevel:    logger.GetLevelByName(webConfig.Logging.WriteLogLevel),
+		LogFilePath: webConfig.Logging.LogFilePath,
+		PrintSource: true,
+	}
+	logger.SetGlobalLogger(&defaultLogger)
+
+	return webConfig, err
+}

internal/ncworker/bookStack.go

@@ -0,0 +1,574 @@
+package ncworker
+
+// @TODO delete folders for shelves that doesn't exist anyore
+
+import (
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"regexp"
+	"strings"
+	"sync"
+	"time"
+
+	"rpjosh.de/ncDocConverter/internal/models"
+	"rpjosh.de/ncDocConverter/internal/nextcloud"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+	"rpjosh.de/ncDocConverter/pkg/utils"
+)
+
+type BsJob struct {
+	job    *models.BookStackJob
+	ncUser *models.NextcloudUser
+
+	cacheCount   int
+	cacheBooks   map[int]book
+	cacheShelves []shelf
+	// If the cache should be usedi n the current execution
+	useCache bool
+}
+
+type shelf struct {
+	ID   int    `json:"id"`
+	Name string `json:"name"`
+
+	// This has to be fetched extra
+	books []int
+}
+type shelfDetails struct {
+	ID    int      `json:"id"`
+	Name  string   `json:"name"`
+	Tags  []string `json:"tags"`
+	Books []struct {
+		ID   int    `json:"id"`
+		Name string `json:"name"`
+	} `json:"books"`
+}
+type shelves struct {
+	Data []shelf `json:"data"`
+}
+
+type book struct {
+	ID   int    `json:"id"`
+	Name string `json:"name"`
+
+	// This has to be calculated of the latest modify page of a page
+	lastModified time.Time
+	// If the book should be ignored to convert
+	ignore bool
+
+	// If the book has been already converted
+	converted bool
+}
+type books struct {
+	Data []book `json:"data"`
+}
+type bookDetails struct {
+	ID       int    `json:"id"`
+	Name     string `json:"name"`
+	Contents []struct {
+		ID        int       `json:"id"`
+		Name      string    `json:"name"`
+		Slug      string    `json:"slug"`
+		BookID    int       `json:"book_id"`
+		ChapterID int       `json:"chapter_id"`
+		Draft     bool      `json:"draft"`
+		Template  bool      `json:"template"`
+		UpdatedAt time.Time `json:"updated_at"`
+		URL       string    `json:"url"`
+		Type      string    `json:"type"`
+	} `json:"contents"`
+	Tags []string `json:"tags"`
+}
+
+func NewBsJob(job *models.BookStackJob, ncUser *models.NextcloudUser) *BsJob {
+	bsJob := BsJob{
+		job:    job,
+		ncUser: ncUser,
+	}
+
+	return &bsJob
+}
+
+func (job *BsJob) ExecuteJob() {
+	// Get all existing files in the destination folder
+	destination, err := nextcloud.SearchInDirectory(
+		job.ncUser, job.job.DestinationDir,
+		[]string{
+			"text/html",
+			"application/pdf",
+		},
+	)
+	if err != nil {
+		logger.Error("Failed to get files in destination directory '%s': %s", job.job.DestinationDir, err)
+		return
+	}
+
+	// Make a map with path as index
+	prefix := "/remote.php/dav/files/" + job.ncUser.Username + "/"
+	destinationMap := nextcloud.ParseSearchResult(destination, prefix, job.job.DestinationDir)
+
+	// Check for cache
+	job.cache()
+
+	// Get all shelves
+	shelves, err := job.getShelves()
+	if err != nil {
+		logger.Error("Failed to get shelves: %s", err)
+		return
+	}
+
+	// Get all books
+	books, err := job.getBooks()
+	if err != nil {
+		logger.Error("Failed to get books: %s", err)
+		return
+	}
+
+	// Index books by path
+	indexedBooks := job.getIndexedBooks(shelves, books)
+
+	// Cache data
+	if job.job.CacheCount > 0 && !job.useCache {
+		job.cacheCount = job.job.CacheCount
+
+		job.cacheShelves = *shelves
+		job.cacheBooks = utils.CopyMap(*books)
+	}
+
+	// Now finally convert the books :)
+	convertCount := 0
+	var wg sync.WaitGroup
+	for i, b := range indexedBooks {
+		// mark as converted
+		indexedBooks[i].converted = true
+		(*books)[b.ID] = *indexedBooks[i]
+
+		// check if it has to be converted again (updated) or for the first time
+		des, exists := destinationMap[i]
+
+		if (!exists || b.lastModified.After(des.LastModified)) && !b.ignore {
+			wg.Add(1)
+			convertCount++
+			go func(book book, path string) {
+				defer wg.Done()
+				job.convertBook(book, path)
+			}(*b, i)
+		} else if b.ignore {
+			logger.Debug("Duplicate book name: %s", b.Name)
+		}
+
+		// Ignore states that a book with a duplicate name exists → delete the orig also
+		if !b.ignore {
+			delete(destinationMap, i)
+		}
+	}
+	wg.Wait()
+
+	// Convert remaining books
+	if job.job.IncludeBooksWithoutShelve {
+		for _, b := range *books {
+			// check if it has to be converted again (updated) or for the first time
+			des, exists := destinationMap[b.Name]
+
+			if !b.converted && !b.ignore && (!exists || b.lastModified.After(des.LastModified)) {
+				wg.Add(1)
+				convertCount++
+				go func(book book, path string) {
+					defer wg.Done()
+					job.convertBook(book, path)
+				}(b, b.Name)
+			}
+			delete(destinationMap, b.Name)
+		}
+		wg.Wait()
+	}
+
+	// Delete the files which are not available anymore
+	for _, dest := range destinationMap {
+		err := nextcloud.DeleteFile(job.ncUser, dest.Path)
+		if err != nil {
+			logger.Error(utils.FirstCharToUppercase(err.Error()))
+		}
+	}
+
+	logger.Info("Finished BookStack job \"%s\": %d books converted", job.job.JobName, convertCount)
+}
+
+// Checks and initializes the cache
+func (job *BsJob) cache() {
+	if job.job.CacheCount > 0 {
+		job.cacheCount--
+		if job.cacheCount < 0 {
+			job.useCache = false
+		} else {
+			job.useCache = true
+		}
+	}
+}
+
+// Return the relative path of the book to save in nextcloud
+func (job *BsJob) getPath(bookName string, shelfName string) string {
+	if job.job.KeepStructure {
+		return shelfName + "/" + bookName
+	} else {
+		return bookName
+	}
+}
+
+// Gets all shelves
+func (job *BsJob) getShelves() (*[]shelf, error) {
+	if job.useCache {
+		return &job.cacheShelves, nil
+	}
+
+	client := http.Client{Timeout: 10 * time.Second}
+
+	req := job.getRequest(http.MethodGet, "shelves", nil)
+
+	// Add shelve filter
+	q := req.URL.Query()
+	for _, j := range job.job.Shelves {
+		q.Add("filter[name:eq]", j)
+	}
+	req.URL.RawQuery = q.Encode()
+
+	res, err := client.Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer res.Body.Close()
+
+	if res.StatusCode != 200 {
+		return nil, fmt.Errorf("expected status code 200, got %d", res.StatusCode)
+	}
+
+	rtc := shelves{}
+	if err = json.NewDecoder(res.Body).Decode(&rtc); err != nil {
+		return nil, fmt.Errorf("failed to decode response: %s", err)
+	}
+
+	if job.job.ShelvesRegex != "" {
+		reg, err := regexp.Compile(job.job.ShelvesRegex)
+		// This is fatal
+		logger.Fatal("Failed to parse the regex '%s': %s", job.job.ShelvesRegex, err)
+
+		rtc2 := shelves{}
+
+		for i, shelve := range rtc.Data {
+			if reg.Match([]byte(shelve.Name)) {
+				rtc2.Data = append(rtc2.Data, rtc.Data[i])
+			} else {
+				logger.Debug("Ignoring shelve %s", shelve.Name)
+			}
+		}
+
+		rtc = rtc2
+	}
+
+	return &rtc.Data, nil
+}
+
+// Returns the IDs of books which belongs to the shelf
+func (job *BsJob) getBooksInShelve(id int) ([]int, error) {
+	client := http.Client{Timeout: 10 * time.Second}
+	req := job.getRequest(http.MethodGet, "shelves/"+fmt.Sprintf("%d", id), nil)
+
+	res, err := client.Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer res.Body.Close()
+
+	if res.StatusCode != 200 {
+		return nil, fmt.Errorf("expected status code 200, got %d", res.StatusCode)
+	}
+
+	shelfDetails := shelfDetails{}
+	if err = json.NewDecoder(res.Body).Decode(&shelfDetails); err != nil {
+		return nil, fmt.Errorf("failed to decode response: %s", err)
+	}
+
+	rtc := make([]int, len(shelfDetails.Books))
+	for i, details := range shelfDetails.Books {
+		rtc[i] = details.ID
+	}
+
+	return rtc, nil
+}
+
+// Indexes the books by the relative path
+func (job *BsJob) getIndexedBooks(shelves *[]shelf, books *map[int]book) map[string]*book {
+	// Now it has to be checked which book belongs to which shelve.
+	// When cached this was already done
+	if !job.useCache {
+		var wg sync.WaitGroup
+		for i, shelv := range *shelves {
+			wg.Add(1)
+
+			go func(shelf shelf, index int) {
+				defer wg.Done()
+
+				ids, err := job.getBooksInShelve(shelf.ID)
+				if err != nil {
+					logger.Error("Failed to get shelf details: %s", err)
+				} else {
+					b := make([]int, 0)
+
+					for _, id := range ids {
+						// Check if book should be excluded → it is not contained in the book map
+						book, exists := (*books)[id]
+						if exists {
+							b = append(b, book.ID)
+						}
+					}
+
+					(*shelves)[index].books = b
+				}
+			}(shelv, i)
+		}
+		wg.Wait()
+	}
+
+	// A book can have the same name. This would lead to conflicts
+	// if they are in the same shelve / folder.
+	// In such a case the ID of the book will be appended to the name "bookName_123".
+	// Because of that a map indexed by the path will be created and AFTERWARDS the file is converted
+	indexedBooks := make(map[string]*book)
+	for _, shelf := range *shelves {
+		for _, bookId := range shelf.books {
+			b := (*books)[bookId]
+			bookPath := job.getPath(b.Name, shelf.Name)
+			existingBook, doesExists := indexedBooks[bookPath]
+
+			if doesExists {
+				// The book path will be appended
+				newBookPath := fmt.Sprintf("%s_%d", bookPath, b.ID)
+				indexedBooks[newBookPath] = &b
+
+				// Also add the other book with the ID
+				otherNewBookPath := fmt.Sprintf("%s_%d", bookPath, existingBook.ID)
+				indexedBooks[otherNewBookPath] = existingBook
+
+				// The original book won't be removed because otherwise a third book with the same
+				// name will be inserted using its real name.
+				// But because this is a pointer, a copy is needed
+				var existingBookCopy book
+				utils.Copy(existingBook, &existingBookCopy)
+				existingBookCopy.ignore = true
+				indexedBooks[bookPath] = &existingBookCopy
+			} else {
+				indexedBooks[bookPath] = &b
+			}
+		}
+
+		// If the structure should be keept, a folder for every shelve has to be created
+		if job.job.KeepStructure && !job.useCache {
+			nextcloud.CreateFoldersRecursively(job.ncUser, job.job.DestinationDir+shelf.Name+"/")
+		}
+	}
+
+	return indexedBooks
+}
+
+// Gets all books and returns a map indexed by the ID of the book
+func (job *BsJob) getBooks() (*map[int]book, error) {
+	if job.useCache {
+		books := utils.CopyMap(job.cacheBooks)
+
+		// The last Change date has to be updated even in cache
+		var wg sync.WaitGroup
+		var mut = &sync.Mutex{}
+		for i, b := range books {
+			wg.Add(1)
+
+			go func(book book, index int) {
+				defer wg.Done()
+				lastModified, err := job.getLastModifiedOfBook(book.ID)
+				if err != nil {
+					logger.Warning("Failed to get last modified date of book %s (%d) - using old date: %s", book.Name, book.ID, err)
+					return
+				}
+
+				book.lastModified = *lastModified
+
+				mut.Lock()
+				books[index] = book
+				mut.Unlock()
+			}(b, i)
+		}
+		wg.Wait()
+
+		return &books, nil
+	}
+
+	client := http.Client{Timeout: 10 * time.Second}
+	req := job.getRequest(http.MethodGet, "books", nil)
+
+	// Add shelve filter
+	q := req.URL.Query()
+	for _, j := range job.job.Books {
+		q.Add("filter[name:eq]", j)
+	}
+	req.URL.RawQuery = q.Encode()
+
+	res, err := client.Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer res.Body.Close()
+
+	if res.StatusCode != 200 {
+		return nil, fmt.Errorf("expected status code 200, got %d", res.StatusCode)
+	}
+
+	booksArray := books{}
+	if err = json.NewDecoder(res.Body).Decode(&booksArray); err != nil {
+		return nil, fmt.Errorf("failed to decode response: %s", err)
+	}
+
+	if job.job.BooksRegex != "" {
+		reg, err := regexp.Compile(job.job.BooksRegex)
+		// This is fatal
+		logger.Fatal("Failed to parse the regex '%s': %s", job.job.BooksRegex, err)
+
+		booksArray2 := books{}
+
+		for i, book := range booksArray.Data {
+			if reg.Match([]byte(book.Name)) {
+				booksArray2.Data = append(booksArray2.Data, booksArray.Data[i])
+			} else {
+				logger.Debug("Ignoring shelve %s", book.Name)
+			}
+		}
+
+		booksArray = booksArray2
+	}
+
+	// Create indexed map
+	rtc := make(map[int]book)
+	var wg sync.WaitGroup
+	var mut = &sync.Mutex{}
+	for _, b := range booksArray.Data {
+		wg.Add(1)
+
+		go func(b book) {
+			defer wg.Done()
+			lastModified, err := job.getLastModifiedOfBook(b.ID)
+			if err != nil {
+				logger.Warning("Failed to get last modified date of book %s (%d) - skipping: %s", b.Name, b.ID, err)
+				return
+			}
+
+			if lastModified.Unix() == 0 {
+				logger.Info("Skipping book %s (%d) because of no content", b.Name, b.ID)
+				return
+			}
+
+			mut.Lock()
+			rtc[b.ID] = book{
+				ID:           b.ID,
+				Name:         b.Name,
+				lastModified: *lastModified,
+			}
+			mut.Unlock()
+		}(b)
+	}
+	wg.Wait()
+
+	return &rtc, nil
+}
+
+// Returns the last modified time of a book
+func (job *BsJob) getLastModifiedOfBook(id int) (*time.Time, error) {
+	client := http.Client{Timeout: 10 * time.Second}
+	req := job.getRequest(http.MethodGet, "books/"+fmt.Sprintf("%d", id), nil)
+
+	res, err := client.Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer res.Body.Close()
+
+	if res.StatusCode != 200 {
+		return nil, fmt.Errorf("expected status code 200, got %d", res.StatusCode)
+	}
+
+	bd := bookDetails{}
+	if err = json.NewDecoder(res.Body).Decode(&bd); err != nil {
+		return nil, fmt.Errorf("failed to decode response: %s", err)
+	}
+
+	lastMod := time.Unix(0, 0)
+	for i, content := range bd.Contents {
+		if content.Template || content.Draft {
+			continue
+		}
+
+		if content.UpdatedAt.After(lastMod) {
+			lastMod = bd.Contents[i].UpdatedAt
+		}
+	}
+
+	return &lastMod, nil
+}
+
+// Returns a new request to the bookStack API.
+// The path beginning AFTER /api/ should be given (e.g.: shelves)
+func (job *BsJob) getRequest(method string, path string, body io.Reader) *http.Request {
+	req, err := http.NewRequest(method, job.ncUser.BookStack.URL+"/api/"+path, body)
+	if err != nil {
+		logger.Error("%s", err)
+	}
+	req.Header.Set("Authorization", "Token "+job.ncUser.BookStack.Token)
+
+	return req
+}
+
+// Converts the given book and uploads it to nextcloud.
+// The path is being expected relative to the root dir of the jobs directory and does
+// not contain a file extension
+func (job *BsJob) convertBook(book book, path string) {
+	fileExtension, url := job.getFileExtension()
+
+	client := http.Client{Timeout: 10 * time.Second}
+	req := job.getRequest(http.MethodGet, fmt.Sprintf("books/%d/export/%s", book.ID, url), nil)
+
+	res, err := client.Do(req)
+	if err != nil {
+		logger.Error("Failed to convert book: %s", err)
+	}
+	defer res.Body.Close()
+
+	if res.StatusCode != 200 {
+		logger.Error("Failed to convert book: expected status code 200, got %d", res.StatusCode)
+		return
+	}
+
+	err = nextcloud.UploadFile(job.ncUser, job.job.DestinationDir+path+fileExtension, res.Body)
+	if err != nil {
+		logger.Error("Failed to upload book to nextcloud: %s", err)
+	}
+}
+
+func (job *BsJob) getFileExtension() (fileExtension string, url string) {
+	switch strings.ToLower(string(job.job.Format)) {
+	case "html":
+		{
+			fileExtension = ".html"
+			url = "html"
+		}
+	case "pdf":
+		{
+			fileExtension = ".pdf"
+			url = "pdf"
+		}
+	default:
+		{
+			logger.Fatal("Invalid format given: '%s'. Expected 'html' or 'pdf'", job.job.Format)
+		}
+	}
+
+	return
+}

internal/ncworker/convertscheduler.go

@@ -0,0 +1,114 @@
+package ncworker
+
+import (
+	"bufio"
+	"fmt"
+	"os"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/go-co-op/gocron"
+	"rpjosh.de/ncDocConverter/internal/models"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+)
+
+type NcConvertScheduler struct {
+	users  *models.NcConvertUsers
+	config *models.WebConfig
+
+	scheduler *gocron.Scheduler
+}
+
+func NewScheduler(users *models.NcConvertUsers, config *models.WebConfig) *NcConvertScheduler {
+	scheduler := NcConvertScheduler{
+		users:     users,
+		config:    config,
+		scheduler: gocron.NewScheduler(time.Local),
+	}
+	// Don't reschedule a task if it's still running
+	scheduler.scheduler.SingletonMode()
+	scheduler.scheduler.StartAsync()
+
+	if config.Server.OneShot {
+		scheduler.ScheduleExecutionsOneShot()
+	} else {
+		scheduler.ScheduleExecutions()
+
+		fmt.Println("Started in schedule mode.\nType \"exit\" to leave or \"execute\" to execute all jobs")
+		// Endless loop
+		for {
+			reader := bufio.NewReader(os.Stdin)
+			text, err := reader.ReadString('\n')
+			if err != nil {
+				// No console input
+				var wg sync.WaitGroup
+				logger.Debug("No console available")
+				wg.Add(1)
+				wg.Wait()
+			}
+
+			input := strings.Trim(strings.ToLower(text), "\n")
+			if input == "exit" {
+				break
+			} else if input == "execute" {
+				scheduler.scheduler.RunAll()
+			}
+		}
+	}
+
+	return &scheduler
+}
+
+// Executes all jobs and exits the program afterwards
+func (scheduler NcConvertScheduler) ScheduleExecutionsOneShot() {
+	for _, user := range scheduler.users.Users {
+
+		// Schedule Nextcloud jobs
+		for _, job := range user.ConvertJobs {
+			convJob := NewNcJob(&job, &user)
+			convJob.ExecuteJob()
+		}
+
+		// Schedule boockstack jobs
+		if user.BookStack.URL != "" {
+			for _, job := range user.BookStack.Jobs {
+				bsJob := NewBsJob(&job, &user)
+				bsJob.ExecuteJob()
+			}
+		}
+
+	}
+}
+
+// Schedules all jobs with gocron
+func (s NcConvertScheduler) ScheduleExecutions() {
+	for ui, user := range s.users.Users {
+
+		// Schedule Nextcloud jobs
+		for i, job := range user.ConvertJobs {
+			convJob := NewNcJob(&s.users.Users[ui].ConvertJobs[i], &s.users.Users[i])
+
+			_, err := s.scheduler.Cron(job.Execution).DoWithJobDetails(s.executeJob, convJob)
+			if err != nil {
+				logger.Fatal("Failed to schedule office job '%s': %s", job.JobName, err)
+			}
+		}
+
+		// Schedule boockstack jobs
+		if user.BookStack.URL != "" {
+			for i, job := range user.BookStack.Jobs {
+				bsJob := NewBsJob(&s.users.Users[ui].BookStack.Jobs[i], &s.users.Users[i])
+
+				_, err := s.scheduler.Cron(job.Execution).DoWithJobDetails(s.executeJob, bsJob)
+				if err != nil {
+					logger.Fatal("Failed to schedule BookStack job '%s': %s", job.JobName, err)
+				}
+			}
+		}
+	}
+}
+
+func (s NcConvertScheduler) executeJob(job Job, scheduledJob gocron.Job) {
+	job.ExecuteJob()
+}

internal/ncworker/job.go

@@ -0,0 +1,5 @@
+package ncworker
+
+type Job interface {
+	ExecuteJob()
+}

internal/ncworker/office.go

@@ -0,0 +1,206 @@
+package ncworker
+
+import (
+	"fmt"
+	"io"
+	"net/http"
+	"path/filepath"
+	"strings"
+	"sync"
+	"time"
+
+	"rpjosh.de/ncDocConverter/internal/models"
+	"rpjosh.de/ncDocConverter/internal/nextcloud"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+	"rpjosh.de/ncDocConverter/pkg/utils"
+)
+
+type convertJob struct {
+	job    *models.NcConvertJob
+	ncUser *models.NextcloudUser
+}
+
+type convertQueu struct {
+	source      nextcloud.NcFile
+	destination string
+}
+
+func NewNcJob(job *models.NcConvertJob, ncUser *models.NextcloudUser) *convertJob {
+	convJob := &convertJob{
+		job:    job,
+		ncUser: ncUser,
+	}
+
+	return convJob
+}
+
+func (job *convertJob) ExecuteJob() {
+
+	// Get existing directory contents
+	sourceFolder, err := nextcloud.SearchInDirectory(
+		job.ncUser,
+		job.job.SourceDir,
+		[]string{
+			"application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+			"application/msword",
+		},
+	)
+	if err != nil {
+		logger.Error("Failed to get files in source directory '%s': %s", job.job.SourceDir, err)
+		return
+	}
+
+	destinationFolder, err := nextcloud.SearchInDirectory(
+		job.ncUser,
+		job.job.DestinationDir,
+		[]string{
+			"application/pdf",
+		},
+	)
+	if err != nil {
+		logger.Error("Failed to get files in destination directory '%s': %s", job.job.DestinationDir, err)
+		return
+	}
+
+	// Store all files in a map
+	prefix := "/remote.php/dav/files/" + job.ncUser.Username + "/"
+	sourceMap := nextcloud.ParseSearchResult(sourceFolder, prefix, job.job.SourceDir)
+	destinationMap := nextcloud.ParseSearchResult(destinationFolder, prefix, job.job.DestinationDir)
+
+	// check which files should be converted
+	var filesToConvert []convertQueu
+	var directorys []string
+
+	for index, source := range sourceMap {
+		// Check if the file exists in the destination map
+		if dest, exists := destinationMap[index]; exists {
+			// Compare timestamp and size
+			if dest.LastModified.Before(source.LastModified) {
+				filesToConvert = append(filesToConvert, convertQueu{source: source, destination: dest.Path})
+			}
+			delete(destinationMap, index)
+		} else {
+			// the directory could not be existing -> check for existance
+			destinationDir := job.getDestinationDir(source.Path)
+			appendIfNotExists(&directorys, destinationDir[0:strings.LastIndex(destinationDir, "/")+1])
+
+			filesToConvert = append(filesToConvert, convertQueu{source: source, destination: destinationDir})
+
+			delete(destinationMap, index)
+		}
+	}
+
+	var wg sync.WaitGroup
+
+	// Delete the files which are not available anymore
+	wg.Add(len(destinationMap))
+	for _, dest := range destinationMap {
+		go func(file *nextcloud.NcFile) {
+			err := nextcloud.DeleteFile(job.ncUser, dest.Path)
+			if err != nil {
+				logger.Error(utils.FirstCharToUppercase(err.Error()))
+			}
+			wg.Done()
+		}(&dest)
+	}
+	wg.Wait()
+
+	// Create required directorys
+	wg.Add(len(directorys))
+	for _, dest := range directorys {
+		go func(path string) {
+			nextcloud.CreateFoldersRecursively(job.ncUser, path)
+			wg.Done()
+		}(dest)
+	}
+	wg.Wait()
+
+	// Convert the files
+	wg.Add(len(filesToConvert))
+	for _, file := range filesToConvert {
+		logger.Info("Path: %s", file.source.Path)
+		go func(cvt convertQueu) {
+			job.convertFile(cvt.source.Path, cvt.source.Fileid, cvt.destination)
+			wg.Done()
+		}(file)
+	}
+	wg.Wait()
+
+	logger.Info("Finished Nextcloud job \"%s\": %d documents converted", job.job.JobName, len(filesToConvert))
+}
+
+// Appends the directory to the array if it isn't contained
+// by another element already
+func appendIfNotExists(dirs *[]string, directory string) {
+	directoryLength := len(directory)
+	for i, currentDir := range *dirs {
+		currentLength := len(currentDir)
+
+		// the existing directory is already referenced in the current
+		if directoryLength > currentLength && directory[0:currentLength] == currentDir {
+			(*dirs)[i] = directory
+			continue
+		} else if directoryLength <= currentLength && currentDir[0:directoryLength] == directory {
+			continue
+		}
+	}
+	*dirs = append(*dirs, directory)
+}
+
+func (job *convertJob) getDestinationDir(sourceFile string) string {
+	sourceFile = sourceFile[len(job.job.SourceDir):]
+	var extension = filepath.Ext(sourceFile)
+	var name = sourceFile[0 : len(sourceFile)-len(extension)]
+
+	return job.job.DestinationDir + name + ".pdf"
+}
+
+// Converts the source file to the destination file utilizing the onlyoffice convert api
+func (job *convertJob) convertFile(sourceFile string, sourceid int, destinationFile string) {
+	logger.Debug("Converting %s (%d) to %s", sourceFile, sourceid, destinationFile)
+
+	client := http.Client{Timeout: 10 * time.Second}
+	req, err := http.NewRequest(http.MethodGet, job.ncUser.NextcloudBaseUrl+"/apps/onlyoffice/downloadas", nil)
+	if err != nil {
+		logger.Error("%s", err)
+	}
+	req.SetBasicAuth(job.ncUser.Username, job.ncUser.Password)
+
+	q := req.URL.Query()
+	q.Add("fileId", fmt.Sprint(sourceid))
+	q.Add("toExtension", "pdf")
+	req.URL.RawQuery = q.Encode()
+
+	res, err := client.Do(req)
+	if err != nil {
+		logger.Error("Failed to access the convert api: %s", err)
+		return
+	}
+	defer res.Body.Close()
+
+	if res.StatusCode != 200 {
+		body, _ := io.ReadAll(res.Body)
+		logger.Error("Failed to access the convert api (#%d). Do you have OnlyOffice installed?: %s", res.StatusCode, body)
+		return
+	}
+
+	uploadClient := http.Client{Timeout: 10 * time.Second}
+	uploadReq, err := http.NewRequest(http.MethodPut, job.ncUser.NextcloudBaseUrl+"/remote.php/dav/files/"+job.ncUser.Username+"/"+destinationFile, res.Body)
+
+	if err != nil {
+		logger.Error("%s", err)
+	}
+	uploadReq.SetBasicAuth(job.ncUser.Username, job.ncUser.Password)
+	uploadReq.Header.Set("Content-Type", "application/binary")
+
+	res, err = uploadClient.Do(uploadReq)
+	if err != nil {
+		logger.Error("%s", err)
+	}
+
+	if res.StatusCode != 204 && res.StatusCode != 201 {
+		logger.Error("Failed to create file %s (#%d)", destinationFile, res.StatusCode)
+	}
+	// Status Code 201
+	res.Body.Close()
+}

internal/nextcloud/nextcloud.go

@@ -0,0 +1,238 @@
+package nextcloud
+
+import (
+	"bytes"
+	"encoding/xml"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"path/filepath"
+	"strconv"
+	"strings"
+	"text/template"
+	"time"
+
+	"rpjosh.de/ncDocConverter/internal/models"
+	"rpjosh.de/ncDocConverter/pkg/logger"
+	"rpjosh.de/ncDocConverter/web"
+)
+
+// The internal representation of a nextcloud file
+type NcFile struct {
+	// File extension: txt
+	Extension string
+	// Relative path of the file to the nextcloud root: /folder/file.txt
+	Path         string
+	LastModified time.Time
+	ContentType  string
+	// Size in Bytes
+	Size int
+	// The unique file ID of the nextcloud server
+	Fileid int
+	// The Webdav URL for file reference
+	WebdavURL string
+}
+
+type searchTemplateData struct {
+	Username    string
+	Directory   string
+	ContentType []string
+}
+
+type searchResult struct {
+	XMLName  xml.Name               `xml:"multistatus"`
+	Text     string                 `xml:",chardata"`
+	D        string                 `xml:"d,attr"`
+	S        string                 `xml:"s,attr"`
+	Oc       string                 `xml:"oc,attr"`
+	Nc       string                 `xml:"nc,attr"`
+	Response []searchResultResponse `xml:"response"`
+}
+type searchResultResponse struct {
+	Text     string `xml:",chardata"`
+	Href     string `xml:"href"`
+	Propstat struct {
+		Text string `xml:",chardata"`
+		Prop struct {
+			Text            string `xml:",chardata"`
+			Getcontenttype  string `xml:"getcontenttype"`
+			Getlastmodified string `xml:"getlastmodified"`
+			Size            string `xml:"size"`
+			Fileid          int    `xml:"fileid"`
+		} `xml:"prop"`
+		Status string `xml:"status"`
+	} `xml:"propstat"`
+}
+
+func (r *searchResultResponse) GetLastModified() time.Time {
+	// Time format: Fri, 23 Sep 2022 05:46:31 GMT
+	rtc, err := time.Parse("Mon, 02 Jan 2006 15:04:05 GMT", r.Propstat.Prop.Getlastmodified)
+	if err != nil {
+		logger.Warning("%s", err)
+		rtc = time.Unix(0, 1)
+	}
+
+	return rtc
+}
+
+// Returns a new request to the Nexcloud API.
+// The path beginning AFTER /dav/ should be given (e.g.: myUser/folder/file.txt)
+func getRequest(method string, path string, body io.Reader, ncUser *models.NextcloudUser) *http.Request {
+	req, err := http.NewRequest(method, ncUser.NextcloudBaseUrl+"/remote.php/dav/"+path, body)
+	if err != nil {
+		logger.Error("%s", err)
+	}
+	req.SetBasicAuth(ncUser.Username, ncUser.Password)
+
+	return req
+}
+
+// Searches for all files of the given content type starting in the given directory.
+func SearchInDirectory(ncUser *models.NextcloudUser, directory string, contentType []string) (*searchResult, error) {
+	client := http.Client{Timeout: 5 * time.Second}
+
+	template, err := template.ParseFS(web.ApiTemplateFiles, "apitemplate/ncsearch.tmpl.xml")
+	if err != nil {
+		return nil, err
+	}
+	var buf bytes.Buffer
+	templateData := searchTemplateData{
+		Username:    ncUser.Username,
+		Directory:   directory,
+		ContentType: contentType,
+	}
+	if err = template.Execute(&buf, templateData); err != nil {
+		return nil, err
+	}
+
+	// Status code 207
+	req := getRequest("SEARCH", "", &buf, ncUser)
+	req.Header.Set("Content-Type", "application/xml")
+
+	res, err := client.Do(req)
+	if err != nil {
+		return nil, err
+	}
+
+	// Decody body first before checking status code to print in error message
+	defer res.Body.Close()
+
+	resBody, err := io.ReadAll(res.Body)
+	if err != nil {
+		return nil, err
+	}
+
+	// Create folder if not existing
+	if res.StatusCode == 404 {
+		logger.Info("Creating directory '%s' because it does not exist", "/"+directory)
+		CreateFoldersRecursively(ncUser, "/"+directory+"notExisting.pdf")
+		return &searchResult{}, nil
+	}
+
+	if res.StatusCode != 207 {
+		return nil, fmt.Errorf("status code %d: %s", res.StatusCode, resBody)
+	}
+
+	var result searchResult
+	if err = xml.Unmarshal(resBody, &result); err != nil {
+		return nil, err
+	}
+
+	return &result, nil
+}
+
+// Parses the response from the given search format to an NcFile.
+// A map with the relative path based on the source Directory ("someFolder/file.txt")
+// and the mathing NcFile will be returned. Therefore, also the source Directory has to be given.
+//
+// To determine the path without the prefix "/remote.php/dav/user/" it has to be given.
+func ParseSearchResult(result *searchResult, prefix string, sourceDir string) map[string]NcFile {
+	preCount := len(prefix)
+	rtc := make(map[string]NcFile)
+
+	for _, file := range result.Response {
+		href, _ := url.QueryUnescape(file.Href)
+		path := href[preCount:]
+		var extension = filepath.Ext(path)
+		var name = path[0 : len(path)-len(extension)][len(sourceDir):]
+		time := file.GetLastModified()
+		size, err := strconv.Atoi(file.Propstat.Prop.Size)
+		if err != nil {
+			logger.Error("Failed to parse the file size '%s' to an integer: %s", file.Propstat.Prop.Size, err)
+			continue
+		}
+		rtc[name] = NcFile{
+			Extension:    extension,
+			Path:         path,
+			LastModified: time,
+			Size:         size,
+			ContentType:  file.Propstat.Prop.Getcontenttype,
+			Fileid:       file.Propstat.Prop.Fileid,
+			WebdavURL:    file.Href,
+		}
+	}
+
+	return rtc
+}
+
+// Delets a file with the given path.
+// The path has to start at the root level: Ebook/myFolder/file.txt
+func DeleteFile(ncUser *models.NextcloudUser, filePath string) error {
+	client := http.Client{Timeout: 5 * time.Second}
+
+	req := getRequest(http.MethodDelete, "files/"+ncUser.Username+"/"+filePath, nil, ncUser)
+
+	res, err := client.Do(req)
+	if err != nil {
+		logger.Error("%s", err)
+	}
+
+	if res.StatusCode != 204 {
+		return fmt.Errorf("failed to delete file %s (%d)", filePath, res.StatusCode)
+	}
+
+	return nil
+}
+
+// Creates all required directorys to create the destination file recursively.
+// The path should be relative to the root: ebook/folder1/folder2/file.txt
+func CreateFoldersRecursively(ncUser *models.NextcloudUser, destinationFile string) {
+	s := strings.Split(destinationFile, "/")
+	folderTree := ""
+
+	// Webdav doesn't have a function to create directories recursively → iterate
+	for _, folder := range s[:len(s)-1] {
+		folderTree += folder + "/"
+
+		client := http.Client{Timeout: 5 * time.Second}
+		req := getRequest("MKCOL", "files/"+ncUser.Username+"/"+folderTree, nil, ncUser)
+
+		res, err := client.Do(req)
+		if err != nil {
+			logger.Error("%s", err)
+		}
+
+		if res.StatusCode != 201 && res.StatusCode != 405 {
+			logger.Error("Failed to create directory '%s'", folderTree)
+		}
+	}
+}
+
+// Uploads a file to the nextcloud server.
+// It will be saved to the destination as a relative path to the nextcloud root (ebook/file.txt).
+func UploadFile(ncUser *models.NextcloudUser, destination string, content io.ReadCloser) error {
+	client := http.Client{Timeout: 5 * time.Second}
+	req := getRequest(http.MethodPut, "files/"+ncUser.Username+"/"+destination, content, ncUser)
+
+	res, err := client.Do(req)
+	if err != nil {
+		return err
+	}
+
+	if res.StatusCode != 201 && res.StatusCode != 204 {
+		return fmt.Errorf("expected status code 201 or 204 but got %d", res.StatusCode)
+	}
+
+	return nil
+}

main.go

@@ -1,22 +0,0 @@
-package main
-
-import (
-	"rpjosh.de/ncDocConverter/logger"
-)
-
-func init() {
-	defaultLogger := logger.Logger {
-		PrintLevel: 0,
-		LogLevel: 1,
-		LogFilePath: "log.log",
-		PrintSource: true,
-	}
-
-	logger.SetGlobalLogger(&defaultLogger)
-}
-
-func main() {
-	defer logger.CloseFile()
-
-
-}

pkg/logger/logger.go

@@ -3,15 +3,16 @@ package logger
 import (
 	"fmt"
 	"log"
+	"os"
 	"runtime"
 	"strconv"
 	"strings"
 	"time"
-	"os"
 )
 
-// define available log levels
+// Level of the log message
 type Level uint8
+
 const (
 	LevelDebug Level = iota
 	LevelInfo
@@ -35,7 +36,6 @@ type Logger struct {
 var dLogger Logger
 
 func init() {
-
 	dLogger = Logger{
 		PrintLevel:  LevelDebug,
 		LogLevel:    LevelInfo,
@@ -52,45 +52,50 @@ func (l Logger) Log(level Level, message string, parameters ...any) {
 }
 
 func (l Logger) log(level Level, message string, parameters ...any) {
-	pc, file, line, ok := runtime.Caller(2)
-	if (!ok) {
+	pc, file, line, ok := runtime.Caller(3)
+	if !ok {
 		file = "#unknown"
 		line = 0
 	}
 
 	// get the name of the level
 	var levelName string
-	switch (level) {
-		case LevelDebug:	 	levelName = "DEBUG"
-		case LevelInfo: 		levelName = "INFO "
-		case LevelWarning:		levelName = "WARN "
-		case LevelError: 		levelName = "ERROR"
-		case LevelFatal:		levelName = "FATAL"
+	switch level {
+	case LevelDebug:
+		levelName = "DEBUG"
+	case LevelInfo:
+		levelName = "INFO "
+	case LevelWarning:
+		levelName = "WARN "
+	case LevelError:
+		levelName = "ERROR"
+	case LevelFatal:
+		levelName = "FATAL"
 	}
 
-	if (levelName == "") {
+	if levelName == "" {
 		message = fmt.Sprintf("Invalid level value given: %d. Original message: ", level) + message
 		levelName = "WARN "
 		level = LevelWarning
 	}
 
-	printMessage := "[" + levelName + "] " + time.Now().UTC().Format("2006-01-02 03:04:05") + 
+	printMessage := "[" + levelName + "] " + time.Now().Local().Format("2006-01-02 15:04:05") +
 		getSourceMessage(file, line, pc, l) +
 		fmt.Sprintf(message, parameters...)
 
-	if (l.LogLevel <= level && l.fileLogger != nil) {
+	if l.LogLevel <= level && l.fileLogger != nil {
 		l.fileLogger.Println(printMessage)
 		l.logFile.Sync()
 
-		if (level == LevelFatal) {
+		if level == LevelFatal {
 			l.CloseFile()
 		}
 	}
 
-	if (l.PrintLevel <= level) {
-		if (level == LevelError) {
+	if l.PrintLevel <= level {
+		if level == LevelError {
 			l.consoleLoggerErr.Println(printMessage)
-		} else if (level == LevelFatal) {
+		} else if level == LevelFatal {
 			l.consoleLoggerErr.Fatal(printMessage)
 		} else {
 			l.consoleLogger.Println(printMessage)
@@ -99,8 +104,8 @@ func (l Logger) log(level Level, message string, parameters ...any) {
 
 }
 
-func getSourceMessage(file string, line int, pc uintptr, l Logger) (string) {
-	if (!l.PrintSource) {
+func getSourceMessage(file string, line int, pc uintptr, l Logger) string {
+	if !l.PrintSource {
 		return " - "
 	}
 
@@ -110,10 +115,11 @@ func getSourceMessage(file string, line int, pc uintptr, l Logger) (string) {
 }
 
 func (l *Logger) setup() {
+	// log.Ldate|log.Ltime|log.Lshortfile
 	l.consoleLogger = log.New(os.Stdout, "", 0)
 	l.consoleLoggerErr = log.New(os.Stderr, "", 0)
 
-	if (l.LogFilePath != "") {
+	if strings.TrimSpace(l.LogFilePath) != "" {
 		file, err := os.OpenFile(l.LogFilePath, os.O_APPEND|os.O_CREATE|os.O_WRONLY, 0644)
 		if err == nil {
 			l.fileLogger = log.New(file, "", 0)
@@ -123,7 +129,7 @@ func (l *Logger) setup() {
 		}
 	} else {
 		l.fileLogger = nil
-		if (l.logFile != nil) {
+		if l.logFile != nil {
 			l.logFile.Close()
 			l.logFile = nil
 		}
@@ -131,18 +137,20 @@ func (l *Logger) setup() {
 }
 
 func (l *Logger) CloseFile() {
-	if (dLogger.logFile != nil) {
-		dLogger.logFile.Close()
-		dLogger.logFile = nil
-		dLogger.fileLogger = nil
+	if l.logFile != nil {
+		l.logFile.Close()
+		l.logFile = nil
+		l.fileLogger = nil
 	}
 }
 
-
 func SetGlobalLogger(l *Logger) {
 	dLogger = *l
 	dLogger.setup()
 }
+func GetGlobalLogger() *Logger {
+	return &dLogger
+}
 
 func Debug(message string, parameters ...any) {
 	dLogger.Log(LevelDebug, message, parameters...)
@@ -163,3 +171,28 @@ func Fatal(message string, parameters ...any) {
 func CloseFile() {
 	dLogger.CloseFile()
 }
+
+// Tries to convert the given level name to the corresponding level code.
+// Allowed values are: 'debug', 'info', 'warn', 'warning', 'error', 'panic' and 'fatal'
+// If an incorrect level name was given an warning is logged and info will be returned
+func GetLevelByName(levelName string) Level {
+	levelName = strings.ToLower(levelName)
+	switch levelName {
+	case "debug":
+		return LevelDebug
+	case "info":
+		return LevelInfo
+	case "warn", "warning":
+		return LevelWarning
+	case "error":
+		return LevelError
+	case "panic", "fatal":
+		return LevelFatal
+
+	default:
+		{
+			Warning("Unable to parse the level name '%s'. Expected 'debug', 'info', 'warn', 'error' or 'fatal'", levelName)
+			return LevelInfo
+		}
+	}
+}

pkg/utils/utils.go

@@ -0,0 +1,38 @@
+package utils
+
+import (
+	"reflect"
+	"unicode"
+)
+
+// Makes the first character of the given string to uppercase
+func FirstCharToUppercase(text string) string {
+	a := []rune(text)
+	a[0] = unicode.ToLower(a[0])
+	return string(a)
+}
+
+// Copys a map. The structs are also cloned
+func CopyMap[T comparable, Val any](m map[T]Val) map[T]Val {
+	cp := make(map[T]Val)
+	for k, v := range m {
+		var u Val
+		Copy(&v, &u)
+		cp[k] = u
+	}
+
+	return cp
+}
+
+// Copies a struct
+func Copy(source interface{}, destin interface{}) {
+	x := reflect.ValueOf(source)
+	if reflect.ValueOf(destin).Kind() != reflect.Ptr {
+		return
+	}
+	if x.Kind() == reflect.Ptr {
+		reflect.ValueOf(destin).Elem().Set(x.Elem())
+	} else {
+		reflect.ValueOf(destin).Elem().Set(x)
+	}
+}

scripts/run.cmd

@@ -0,0 +1,15 @@
+@ECHO OFF
+
+:: Bypass the "Terminate Batch Job" prompt
+if "%~1"=="-FIXED_CTRL_C" (
+   :: Remove the -FIXED_CTRL_C parameter
+   SHIFT
+) ELSE (
+   :: Run the batch with <NUL and -FIXED_CTRL_C
+   CALL <NUL %0 -FIXED_CTRL_C %*
+   GOTO :EOF
+)
+
+SET PATH=%PATH%;C:\Windows\System32
+set GOTMPDIR=C:\MYCOMP
+nodemon --delay 1s -e go,html --ignore web/app/ --signal SIGKILL --exec go run ./cmd/ncDocConverth || exit 1

scripts/run.sh

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+nodemon --delay 1s -e go,html,yaml --ignore web/app/ --signal SIGTERM --exec 'go run ./cmd/ncDocConverth || exit 1'

web/apitemplate/ncsearch.tmpl.xml

@@ -0,0 +1,40 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<d:searchrequest xmlns:d="DAV:" xmlns:oc="http://owncloud.org/ns">
+    <d:basicsearch>
+        <d:select>
+            <d:prop>
+                <d:getcontenttype/>
+                <d:getlastmodified/>
+                <oc:size/>
+                <oc:fileid/>
+            </d:prop>
+        </d:select>
+        <d:from>
+            <d:scope>
+                <d:href>/files/{{.Username}}/{{.Directory}}</d:href>
+                <d:depth></d:depth>
+            </d:scope>
+        </d:from>
+        <d:where>
+            <d:and>
+                <d:or>
+                    {{range .ContentType}}
+                        <d:eq>
+                            <d:prop>
+                                <d:getcontenttype/>
+                            </d:prop>
+                            <d:literal>{{ . }}</d:literal>
+                        </d:eq>
+                    {{end}}
+                </d:or>
+                    <d:gt>
+                        <d:prop>
+                            <oc:size/>
+                        </d:prop>
+                        <d:literal>100</d:literal>
+                    </d:gt>
+            </d:and>
+        </d:where>
+        <d:orderby/>
+    </d:basicsearch>
+</d:searchrequest>

web/efs.go

@@ -0,0 +1,8 @@
+package web
+
+import (
+	"embed"
+)
+
+//go:embed "apitemplate"
+var ApiTemplateFiles embed.FS