fix #77: use processed ad dir for duplicate checking, not slug

config.go

@@ -34,7 +34,7 @@ import (
 )
 
 const (
-	VERSION    string = "0.3.3"
+	VERSION    string = "0.3.4"
 	Baseuri    string = "https://www.kleinanzeigen.de"
 	Listuri    string = "/s-bestandsliste.html"
 	Defaultdir string = "."
@@ -52,6 +52,8 @@ const (
 
 	DefaultAdNameTemplate string = "{{.Slug}}"
 
+	DefaultOutdirTemplate string = "."
+
 	// for image download throttling
 	MinThrottle int = 2
 	MaxThrottle int = 20
@@ -65,6 +67,8 @@ const (
 	WIN string = "windows"
 )
 
+var DirsVisited map[string]int
+
 const Usage string = `This is kleingebaeck, the kleinanzeigen.de backup tool.
 
 Usage: kleingebaeck [-dvVhmoclu] [<ad-listing-url>,...]
@@ -77,7 +81,7 @@ Options:
 -l --limit   <num>      Limit the ads to download to <num>, default: load all.
 -c --config  <file>     Use config file <file> (default: ~/.kleingebaeck).
    --ignoreerrors       Ignore HTTP errors, may lead to incomplete ad backup.
--f --force              Download images even if they already exist.
+-f --force              Overwrite images and ads even if the already exist.
 -m --manual             Show manual.
 -h --help               Show usage.
 -V --version            Show program version.
@@ -126,7 +130,7 @@ func InitConfig(output io.Writer) (*Config, error) {
 	// Load default values using the confmap provider.
 	if err := kloader.Load(confmap.Provider(map[string]interface{}{
 		"template":       template,
-		"outdir":         ".",
+		"outdir":         DefaultOutdirTemplate,
 		"loglevel":       "notice",
 		"userid":         0,
 		"adnametemplate": DefaultAdNameTemplate,

go.mod

@@ -14,7 +14,7 @@ require (
 	github.com/lmittmann/tint v1.0.4
 	github.com/mattn/go-isatty v0.0.20
 	github.com/spf13/pflag v1.0.5
-	github.com/tlinden/yadu v0.1.1
+	github.com/tlinden/yadu v0.1.2
 	golang.org/x/sync v0.5.0
 )
 
@@ -33,7 +33,7 @@ require (
 	github.com/pelletier/go-toml v1.9.5 // indirect
 	github.com/pkg/errors v0.9.1 // indirect
 	golang.org/x/net v0.0.0-20220722155237-a158d28d115b // indirect
-	golang.org/x/sys v0.14.0 // indirect
+	golang.org/x/sys v0.17.0 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 
 )

go.sum

@@ -66,6 +66,8 @@ github.com/tlinden/yadu v0.1.0 h1:qtCi1jxg392qVRLFyrJ2LYu6/PiKSp1LT02EX+mNLME=
 github.com/tlinden/yadu v0.1.0/go.mod h1:l3bRmHKL9zGAR6pnBHY2HRPxBecf7L74BoBgOOpTcUA=
 github.com/tlinden/yadu v0.1.1 h1:116oEUy9b4PcMF5wLL2dCFA/sn/praYutOnao07MROw=
 github.com/tlinden/yadu v0.1.1/go.mod h1:l3bRmHKL9zGAR6pnBHY2HRPxBecf7L74BoBgOOpTcUA=
+github.com/tlinden/yadu v0.1.2 h1:TYYVnUJwziRJ9YPbIbRf9ikmDw0Q8Ifixm+J/kBQFh8=
+github.com/tlinden/yadu v0.1.2/go.mod h1:l3bRmHKL9zGAR6pnBHY2HRPxBecf7L74BoBgOOpTcUA=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/net v0.0.0-20180218175443-cbe0f9307d01/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
 golang.org/x/net v0.0.0-20181114220301-adae6a3d119a/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
@@ -81,6 +83,8 @@ golang.org/x/sys v0.0.0-20220908164124-27713097b956/go.mod h1:oPkhp1MJrh7nUepCBc
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.14.0 h1:Vz7Qs629MkJkGyHxUlRHizWJRG2j8fbQKjELVSNhy7Q=
 golang.org/x/sys v0.14.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
+golang.org/x/sys v0.17.0 h1:25cE3gD+tdBA7lp7QfhuV+rJiE9YXTcS3VG1SqssI/Y=
+golang.org/x/sys v0.17.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=

kleingebaeck.1

@@ -133,7 +133,7 @@
 .\" ========================================================================
 .\"
 .IX Title "KLEINGEBAECK 1"
-.TH KLEINGEBAECK 1 "2024-01-25" "1" "User Commands"
+.TH KLEINGEBAECK 1 "2024-02-10" "1" "User Commands"
 .\" For nroff, turn off justification.  Always turn off hyphenation; it makes
 .\" way too many mistakes in technical documents.
 .if n .ad l
@@ -152,7 +152,7 @@ kleingebaeck \- kleinanzeigen.de backup tool
 \&    \-l \-\-limit   <num>      Limit the ads to download to <num>, default: load all.
 \&    \-c \-\-config  <file>     Use config file <file> (default: ~/.kleingebaeck).
 \&       \-\-ignoreerrors       Ignore HTTP errors, may lead to incomplete ad backup.
-\&    \-f \-\-force              Download images even if they already exist.
+\&    \-f \-\-force              Overwrite images and ads even if the already exist.
 \&    \-m \-\-manual             Show manual.
 \&    \-h \-\-help               Show usage.
 \&    \-V \-\-version            Show program version.
@@ -195,7 +195,7 @@ Be careful  if you want  to change the  template.  The variable  is a
 multiline string surrounded  by three double quotes. You  can left out
 certain   fields  and   use  any   formatting  you   like.  Refer   to
 <https://pkg.go.dev/text/template>  for   details  how  to   write  a
-template.
+template. Also read the \s-1TEMPLATES\s0 section below.
 .PP
 If you're on  windows and want to customize the  output directory, put
 it into single  quotes to avoid the backslashes  interpreted as escape
@@ -204,6 +204,94 @@ chars like this:
 .Vb 1
 \&    outdir = \*(AqC:\eData\eAds\*(Aq
 .Ve
+.SH "TEMPLATES"
+.IX Header "TEMPLATES"
+Various parts  of the configuration  can be modified  using templates:
+the output directory, the ad directory and the ad listing itself.
+.SS "\s-1OUTPUT DIR TEMPLATE\s0"
+.IX Subsection "OUTPUT DIR TEMPLATE"
+The config varialbe \f(CW\*(C`outdir\*(C'\fR or the command line parameter \f(CW\*(C`\-o\*(C'\fR take a
+template which may contain:
+.ie n .IP """{{.Year}}""" 4
+.el .IP "\f(CW{{.Year}}\fR" 4
+.IX Item "{{.Year}}"
+.PD 0
+.ie n .IP """{{.Month}}""" 4
+.el .IP "\f(CW{{.Month}}\fR" 4
+.IX Item "{{.Month}}"
+.ie n .IP """{{.Day}}""" 4
+.el .IP "\f(CW{{.Day}}\fR" 4
+.IX Item "{{.Day}}"
+.PD
+.PP
+That way you can create a new output directory for every backup
+run. For example:
+.PP
+.Vb 1
+\&    outdir = "/home/backups/ads\-{{.Year}}\-{{.Month}}\-{{.Day}}"
+.Ve
+.PP
+Or using the command line flag:
+.PP
+.Vb 1
+\&    \-o "/home/backups/ads\-{{.Year}}\-{{.Month}}\-{{.Day}}"
+.Ve
+.PP
+The default value is \f(CW\*(C`.\*(C'\fR \- the current directory.
+.SS "\s-1AD DIRECTORY TEMPLATE\s0"
+.IX Subsection "AD DIRECTORY TEMPLATE"
+The ad directory name can be modified using the following ad values:
+.IP "{{.Price}}" 4
+.IX Item "{{.Price}}"
+.PD 0
+.IP "{{.ID}}" 4
+.IX Item "{{.ID}}"
+.IP "{{.Category}}" 4
+.IX Item "{{.Category}}"
+.IP "{{.Condition}}" 4
+.IX Item "{{.Condition}}"
+.IP "{{.Created}}" 4
+.IX Item "{{.Created}}"
+.IP "{{.Slug}}" 4
+.IX Item "{{.Slug}}"
+.IP "{{.Text}}" 4
+.IX Item "{{.Text}}"
+.PD
+.PP
+It can only be configured in the config file. By default only
+\&\f(CW\*(C`{{.Slug}}\*(C'\fR is being used, this is the title of the ad in url format.
+.SS "\s-1AD TEMPLATE\s0"
+.IX Subsection "AD TEMPLATE"
+The ad listing itself can be modified as well, using the same
+variables as the ad name template above.
+.PP
+This is the default template:
+.PP
+.Vb 7
+\&    Title: {{.Title}}
+\&    Price: {{.Price}}
+\&    Id: {{.ID}}
+\&    Category: {{.Category}}
+\&    Condition: {{.Condition}}
+\&    Created: {{.Created}}
+\&    Expire: {{.Expire}}
+\&    
+\&    {{.Text}}
+.Ve
+.PP
+The config parameter to modify is \f(CW\*(C`template\*(C'\fR. See example.conf in the
+source repository. Please take care, since this is a multiline
+string. This is how it shall look if you modify it:
+.PP
+.Vb 2
+\&    template="""
+\&    Title: {{.Title}}
+\&    
+\&    {{.Text}}
+\&    """
+.Ve
+.PP
+That is, the content between the two \f(CW"""\fR chars is the template.
 .SH "SETUP"
 .IX Header "SETUP"
 To setup the tool, you need to lookup your userid on

kleingebaeck.go

@@ -14,7 +14,7 @@ SYNOPSYS
         -l --limit   <num>      Limit the ads to download to <num>, default: load all.
         -c --config  <file>     Use config file <file> (default: ~/.kleingebaeck).
            --ignoreerrors       Ignore HTTP errors, may lead to incomplete ad backup.
-        -f --force              Download images even if they already exist.
+        -f --force              Overwrite images and ads even if the already exist.
         -m --manual             Show manual.
         -h --help               Show usage.
         -V --version            Show program version.
@@ -55,6 +55,7 @@ CONFIGURATION
     multiline string surrounded by three double quotes. You can left out
     certain fields and use any formatting you like. Refer to
     <https://pkg.go.dev/text/template> for details how to write a template.
+    Also read the TEMPLATES section below.
 
     If you're on windows and want to customize the output directory, put it
     into single quotes to avoid the backslashes interpreted as escape chars
@@ -62,6 +63,71 @@ CONFIGURATION
 
         outdir = 'C:\Data\Ads'
 
+TEMPLATES
+    Various parts of the configuration can be modified using templates: the
+    output directory, the ad directory and the ad listing itself.
+
+  OUTPUT DIR TEMPLATE
+    The config varialbe "outdir" or the command line parameter "-o" take a
+    template which may contain:
+
+    "{{.Year}}"
+    "{{.Month}}"
+    "{{.Day}}"
+
+    That way you can create a new output directory for every backup run. For
+    example:
+
+        outdir = "/home/backups/ads-{{.Year}}-{{.Month}}-{{.Day}}"
+
+    Or using the command line flag:
+
+        -o "/home/backups/ads-{{.Year}}-{{.Month}}-{{.Day}}"
+
+    The default value is "." - the current directory.
+
+  AD DIRECTORY TEMPLATE
+    The ad directory name can be modified using the following ad values:
+
+    {{.Price}}
+    {{.ID}}
+    {{.Category}}
+    {{.Condition}}
+    {{.Created}}
+    {{.Slug}}
+    {{.Text}}
+
+    It can only be configured in the config file. By default only
+    "{{.Slug}}" is being used, this is the title of the ad in url format.
+
+  AD TEMPLATE
+    The ad listing itself can be modified as well, using the same variables
+    as the ad name template above.
+
+    This is the default template:
+
+        Title: {{.Title}}
+        Price: {{.Price}}
+        Id: {{.ID}}
+        Category: {{.Category}}
+        Condition: {{.Condition}}
+        Created: {{.Created}}
+        Expire: {{.Expire}}
+    
+        {{.Text}}
+
+    The config parameter to modify is "template". See example.conf in the
+    source repository. Please take care, since this is a multiline string.
+    This is how it shall look if you modify it:
+
+        template="""
+        Title: {{.Title}}
+    
+        {{.Text}}
+        """
+
+    That is, the content between the two """ chars is the template.
+
 SETUP
     To setup the tool, you need to lookup your userid on kleinanzeigen.de.
     Go to your ad overview page while NOT being logged in:

kleingebaeck.pod

@@ -13,7 +13,7 @@ kleingebaeck - kleinanzeigen.de backup tool
     -l --limit   <num>      Limit the ads to download to <num>, default: load all.
     -c --config  <file>     Use config file <file> (default: ~/.kleingebaeck).
        --ignoreerrors       Ignore HTTP errors, may lead to incomplete ad backup.
-    -f --force              Download images even if they already exist.
+    -f --force              Overwrite images and ads even if the already exist.
     -m --manual             Show manual.
     -h --help               Show usage.
     -V --version            Show program version.
@@ -55,7 +55,7 @@ Be careful  if you want  to change the  template.  The variable  is a
 multiline string surrounded  by three double quotes. You  can left out
 certain   fields  and   use  any   formatting  you   like.  Refer   to
 L<https://pkg.go.dev/text/template>  for   details  how  to   write  a
-template.
+template. Also read the TEMPLATES section below.
 
 If you're on  windows and want to customize the  output directory, put
 it into single  quotes to avoid the backslashes  interpreted as escape
@@ -63,6 +63,91 @@ chars like this:
 
     outdir = 'C:\Data\Ads'
 
+=head1 TEMPLATES
+
+Various parts  of the configuration  can be modified  using templates:
+the output directory, the ad directory and the ad listing itself.
+
+=head2 OUTPUT DIR TEMPLATE
+
+The config varialbe C<outdir> or the command line parameter C<-o> take a
+template which may contain:
+
+=over
+
+=item C<{{.Year}}>
+
+=item C<{{.Month}}>
+
+=item C<{{.Day}}>
+
+=back
+
+That way you can create a new output directory for every backup
+run. For example:
+
+    outdir = "/home/backups/ads-{{.Year}}-{{.Month}}-{{.Day}}"
+
+Or using the command line flag:
+
+    -o "/home/backups/ads-{{.Year}}-{{.Month}}-{{.Day}}"
+
+The default value is C<.> - the current directory.
+
+=head2 AD DIRECTORY TEMPLATE
+
+The ad directory name can be modified using the following ad values:
+
+=over
+
+=item {{.Price}}
+
+=item {{.ID}}
+
+=item {{.Category}}
+
+=item {{.Condition}}
+
+=item {{.Created}}
+
+=item {{.Slug}}
+
+=item {{.Text}}
+
+=back
+
+It can only be configured in the config file. By default only
+C<{{.Slug}}> is being used, this is the title of the ad in url format.
+
+=head2 AD TEMPLATE
+
+The ad listing itself can be modified as well, using the same
+variables as the ad name template above.
+
+This is the default template:
+
+    Title: {{.Title}}
+    Price: {{.Price}}
+    Id: {{.ID}}
+    Category: {{.Category}}
+    Condition: {{.Condition}}
+    Created: {{.Created}}
+    Expire: {{.Expire}}
+    
+    {{.Text}}
+
+The config parameter to modify is C<template>. See example.conf in the
+source repository. Please take care, since this is a multiline
+string. This is how it shall look if you modify it:
+
+    template="""
+    Title: {{.Title}}
+    
+    {{.Text}}
+    """
+
+That is, the content between the two C<"""> chars is the template.
+
 =head1 SETUP
 
 To setup the tool, you need to lookup your userid on

main.go

@@ -112,17 +112,26 @@ func Main(output io.Writer) int {
 	slog.Debug("config", "conf", conf)
 
 	// prepare output dir
-	err = Mkdir(conf.Outdir)
+	outdir, err := OutDirName(conf)
 	if err != nil {
 		return Die(err)
 	}
 
+	err = Mkdir(outdir)
+	if err != nil {
+		return Die(err)
+	}
+	conf.Outdir = outdir
+
 	// used for all HTTP requests
 	fetch, err := NewFetcher(conf)
 	if err != nil {
 		return Die(err)
 	}
 
+	// setup ad dir registry, needed to check for duplicates
+	DirsVisited = make(map[string]int)
+
 	switch {
 	case len(conf.Adlinks) >= 1:
 		// directly backup ad listing[s]

scrape.go

@@ -126,16 +126,32 @@ func ScrapeAd(fetch *Fetcher, uri string) error {
 
 	advertisement.CalculateExpire()
 
-	// write listing
-	addir, err := WriteAd(fetch.Config, advertisement)
+	// prepare ad dir name
+	addir, err := AdDirName(fetch.Config, advertisement)
 	if err != nil {
 		return err
 	}
 
+	proceed := CheckAdVisited(fetch.Config, addir)
+	if !proceed {
+		return nil
+	}
+
+	// write listing
+	err = WriteAd(fetch.Config, advertisement, addir)
+	if err != nil {
+		return err
+	}
+
+	// tell the user
 	slog.Debug("extracted ad listing", "ad", advertisement)
 
+	// stats
 	fetch.Config.IncrAds()
 
+	// register for later checks
+	DirsVisited[addir] = 1
+
 	return ScrapeImages(fetch, advertisement, addir)
 }
 

store.go

@@ -26,8 +26,36 @@ import (
 	"runtime"
 	"strings"
 	tpl "text/template"
+	"time"
 )
 
+type OutdirData struct {
+	Year, Day, Month string
+}
+
+func OutDirName(conf *Config) (string, error) {
+	tmpl, err := tpl.New("outdir").Parse(conf.Outdir)
+	if err != nil {
+		return "", fmt.Errorf("failed to parse outdir template: %w", err)
+	}
+
+	buf := bytes.Buffer{}
+
+	now := time.Now()
+	data := OutdirData{
+		Year:  now.Format("2006"),
+		Month: now.Format("02"),
+		Day:   now.Format("01"),
+	}
+
+	err = tmpl.Execute(&buf, data)
+	if err != nil {
+		return "", fmt.Errorf("failed to execute outdir template: %w", err)
+	}
+
+	return buf.String(), nil
+}
+
 func AdDirName(conf *Config, advertisement *Ad) (string, error) {
 	tmpl, err := tpl.New("adname").Parse(conf.Adnametemplate)
 	if err != nil {
@@ -44,19 +72,13 @@ func AdDirName(conf *Config, advertisement *Ad) (string, error) {
 	return buf.String(), nil
 }
 
-func WriteAd(conf *Config, advertisement *Ad) (string, error) {
-	// prepare ad dir name
-	addir, err := AdDirName(conf, advertisement)
-	if err != nil {
-		return "", err
-	}
-
+func WriteAd(conf *Config, advertisement *Ad, addir string) error {
 	// prepare output dir
 	dir := filepath.Join(conf.Outdir, addir)
 
-	err = Mkdir(dir)
+	err := Mkdir(dir)
 	if err != nil {
-		return "", err
+		return err
 	}
 
 	// write ad file
@@ -64,7 +86,7 @@ func WriteAd(conf *Config, advertisement *Ad) (string, error) {
 
 	listingfd, err := os.Create(listingfile)
 	if err != nil {
-		return "", fmt.Errorf("failed to create Adlisting.txt: %w", err)
+		return fmt.Errorf("failed to create Adlisting.txt: %w", err)
 	}
 	defer listingfd.Close()
 
@@ -76,17 +98,17 @@ func WriteAd(conf *Config, advertisement *Ad) (string, error) {
 
 	tmpl, err := tpl.New("adlisting").Parse(conf.Template)
 	if err != nil {
-		return "", fmt.Errorf("failed to parse adlisting template: %w", err)
+		return fmt.Errorf("failed to parse adlisting template: %w", err)
 	}
 
 	err = tmpl.Execute(listingfd, advertisement)
 	if err != nil {
-		return "", fmt.Errorf("failed to execute adlisting template: %w", err)
+		return fmt.Errorf("failed to execute adlisting template: %w", err)
 	}
 
 	slog.Info("wrote ad listing", "listingfile", listingfile)
 
-	return addir, nil
+	return nil
 }
 
 func WriteImage(filename string, reader *bytes.Reader) error {
@@ -133,3 +155,21 @@ func fileExists(filename string) bool {
 
 	return !info.IsDir()
 }
+
+// check if  an addir has  already been  processed by current  run and
+// decide what to do
+func CheckAdVisited(conf *Config, adname string) bool {
+	if Exists(DirsVisited, adname) {
+		if conf.ForceDownload {
+			slog.Warn("an ad with the same name has already been downloaded, overwriting", "addir", adname)
+			return true
+		}
+
+		// don't overwrite
+		slog.Warn("an ad with the same name has already been downloaded, skipping (use -f to overwrite)", "addir", adname)
+		return false
+	}
+
+	// overwrite
+	return true
+}

t/httproot/templates/render.sh

@@ -1,5 +1,7 @@
 #!/bin/sh -x
 base="../kleinanzeigen"
+
+rm -rf $base
 mkdir -p $base
 
 echo "Generating /s-bestandsliste.html"

util.go

@@ -74,3 +74,12 @@ func IsNoTty() bool {
 func GetThrottleTime() time.Duration {
 	return time.Duration(rand.Intn(MaxThrottle-MinThrottle+1)+MinThrottle) * time.Millisecond
 }
+
+// look if a key in a map exists, generic variant
+func Exists[K comparable, V any](m map[K]V, v K) bool {
+	if _, ok := m[v]; ok {
+		return true
+	}
+
+	return false
+}