Ability to use external contact management software for assigning names

Currently names for addresses are assigned based on how common they are
in the users email, but this might not be always preferable. Add the
ability to override names calculated by maildir-rank-addr using an
external address lookup command.

README.md

@@ -8,6 +8,7 @@ Why? No need to manually edit an address book, yet the cached ranking is
 available extremely fast.
 
 ### Features:
+
 - scans all your emails
 - ranks based on both recency and frequency of addresses
 - collects from To, Cc, Bcc and From fields
@@ -16,13 +17,15 @@ available extremely fast.
 - uses the most frequent non-empty display name for each email
 - filters common "no reply" addresses, additional filters can be added via regexes
 - normalizes emails to lower case
-- "blazingly fast"<sup>*</sup>: crunch time for 270k emails is 7s on my machine, grepping from the output is instantaneous
+- ability to add additional email addresses from a command
+- "blazingly fast"<sup>\*</sup>: crunch time for 270k emails is 7s on my machine, grepping from the output is instantaneous
 
-<sup>*</sup>: compared to original python implementation for crunching (see Behind the scenes below) and compared to using notmuch query for address completion
+<sup>\*</sup>: compared to original python implementation for crunching (see Behind the scenes below) and compared to using notmuch query for address completion
 
 # Installation
 
 The easiest way to install is running:
+
 ```
 go install github.com/ferdinandyb/maildir-rank-addr@latest
 ```
@@ -41,6 +44,8 @@ or systemd timer).
 Supported flags:
 
 ```
+      --addr-book-cmd string   optional command to query addresses from your addressbook
+      --addr-book-add-unmatched if cmd is stated, determine wether to add unmatched addresses at the end of the file (true or false)
       --addresses strings   comma separated list of your email addresses (regex possible)
       --config string       path to config file
       --filters strings     comma separated list of regexes to filter
@@ -67,6 +72,7 @@ classification based on your explicit sends will not be possible!
 
 Uses go's `text/template` to configure output for each address (one line per address).
 Available keys:
+
 ```
 	Address
 	Name
@@ -101,6 +107,18 @@ before the @) matches any of these strings:
 	"nincsvalasz",
 ```
 
+**addr-book-cmd**
+
+Optional command to fetch email addresses and names, the output it returns must have
+an email address first, followed by a tab space and and the desired name, the name
+must end in a tab space or a newline for the command to work, this can be
+useful for integrating with command line addressbooks such as abook or khard
+
+```
+abook --mutt-query "s"
+khard email -p --remove-first-line
+```
+
 **config**
 
 Path to a config file to be loaded instead of the defaults (see below).
@@ -124,12 +142,12 @@ outputpath = "~/.mail/addressbook"
 template = "{{.Address}}\t{{.Name}}"
 ```
 
-
 ## Integration
 
 #### aerc
 
 Put something like this in your aerc config (using your favourite grep):
+
 ```
 address-book-cmd="ugrep -jP -m 100 --color=never %s /home/[myuser]/.cache/maildir-rank-addr/addressbook.tsv"
 ```
@@ -145,7 +163,6 @@ code the path without shell expansion.
 
 Ranking is actually done by first classifying and then ranking within class.
 
-
 ### Classifying addresses
 
 First we go through each email found in your maildir and for each address found
@@ -212,7 +229,6 @@ Please see [contribution guidelines](https://github.com/ferdinandyb/maildir-rank
 - [notmuch-addrlookup-c](https://github.com/aperezdc/notmuch-addrlookup-c): address lookup from notmuch
 - [addr-book-combine](https://jasoncarloscox.com/creations/addr-book-combine/): for combining generated addressbooks with hand currated ones, like [khard](https://github.com/lucc/khard)
 
-
 # Acknowledgments
 
 Some functions for parsing email was taken from [aerc](aerc-mail.org).

config.go

@@ -2,20 +2,24 @@ package main
 
 import (
 	"fmt"
-	homedir "github.com/mitchellh/go-homedir"
-	"github.com/spf13/pflag"
-	"github.com/spf13/viper"
 	"os"
+	"os/exec"
 	"regexp"
 	"strings"
 	"text/template"
+
+	homedir "github.com/mitchellh/go-homedir"
+	"github.com/spf13/pflag"
+	"github.com/spf13/viper"
 )
 
 func loadConfig() Config {
 	pflag.String("config", "", "path to config file")
 	pflag.String("maildir", "", "path to maildir folder")
 	pflag.String("outputpath", "", "path to output file")
 	pflag.String("template", "", "output template")
+	pflag.String("addr-book-cmd", "", "optional command to query addresses from your addressbook")
+	pflag.Bool("addr-book-add-unmatched", false, "flag to determine if you want unmatched addressbook contacts to be added to the output")
 	pflag.StringSlice("addresses", []string{}, "comma separated list of your email addresses (regex possible)")
 	pflag.StringSlice("filters", []string{}, "comma separated list of regexes to filter")
 	pflag.Parse()
@@ -65,6 +69,14 @@ func loadConfig() Config {
 		addresses[i] = regexp.MustCompile(filter)
 	}
 	templateString := viper.GetString("template")
+	addressbookLookupCommandString := viper.GetString("addr-book-cmd")
+	addressbookAddUnmatched := viper.GetBool("addr-book-add-unmatched")
+	var addressbookLookupCommand *exec.Cmd
+	if addressbookLookupCommandString != "" {
+		args := strings.Fields(addressbookLookupCommandString)
+		application, arguments := args[0], args[1:]
+		addressbookLookupCommand = exec.Command(application, arguments...)
+	}
 
 	if !strings.HasSuffix(templateString, "\n") {
 		templateString += "\n"
@@ -74,11 +86,13 @@ func loadConfig() Config {
 		panic(fmt.Errorf("bad template"))
 	}
 	config := Config{
-		maildir:       maildir,
-		outputpath:    outputpath,
-		addresses:     addresses,
-		template:      tmpl,
-		customFilters: customFilters,
+		maildir:                  maildir,
+		outputpath:               outputpath,
+		addresses:                addresses,
+		template:                 tmpl,
+		customFilters:            customFilters,
+		addressbookLookupCommand: addressbookLookupCommand,
+		addressbookAddUnmatched:  addressbookAddUnmatched,
 	}
 	return config
 }

data.go

@@ -1,6 +1,7 @@
 package main
 
 import (
+	"os/exec"
 	"regexp"
 	"text/template"
 )
@@ -18,9 +19,11 @@ type AddressData struct {
 }
 
 type Config struct {
-	maildir       string
-	outputpath    string
-	addresses     []*regexp.Regexp
-	template      *template.Template
-	customFilters []*regexp.Regexp
+	maildir                  string
+	outputpath               string
+	addresses                []*regexp.Regexp
+	template                 *template.Template
+	customFilters            []*regexp.Regexp
+	addressbookLookupCommand *exec.Cmd
+	addressbookAddUnmatched  bool
 }

maildir-rank-addr.go

@@ -1,10 +1,9 @@
 package main
 
-import ()
-
 func main() {
 	config := loadConfig()
-	data := walkMaildir(config.maildir, config.addresses, config.customFilters)
+	addressbook := parseAddressbook(config.addressbookLookupCommand)
+	data := walkMaildir(config.maildir, config.addresses, config.customFilters, addressbook)
 	classeddata := calculateRanks(data)
-	saveData(classeddata, config.outputpath, config.template)
+	saveData(classeddata, config.outputpath, config.template, addressbook, config.addressbookAddUnmatched)
 }

output.go

@@ -13,10 +13,11 @@ func saveData(
 	classedData map[int]map[string]AddressData,
 	path string,
 	tmpl *template.Template,
+	addressbook map[string]string,
+	addUnmatched bool,
 ) {
 	os.MkdirAll(filepath.Dir(path), os.ModePerm)
 	f, err := os.Create(path)
-
 	if err != nil {
 		log.Fatal(err)
 	}
@@ -46,5 +47,14 @@ func saveData(
 			tmpl.Execute(f, kv.Value)
 		}
 	}
+	if addUnmatched {
+		for ak, av := range addressbook {
+			aD := AddressData{}
+			aD.Address = ak
+			aD.Name = av
+			count++
+			tmpl.Execute(f, aD)
+		}
+	}
 	fmt.Println(count, " addresses written to ", path)
 }

parseAddressbook.go

@@ -0,0 +1,38 @@
+package main
+
+import (
+	"bufio"
+	"fmt"
+	"log"
+	"os/exec"
+	"strings"
+)
+
+func parseAddressbook(
+	cmd *exec.Cmd,
+) map[string]string {
+	if cmd == nil {
+		return nil
+	}
+	addressbook := make(map[string]string)
+	out, err := cmd.StdoutPipe()
+	if err != nil {
+		log.Fatal(err)
+	}
+	if err := cmd.Start(); err != nil {
+		log.Fatal(err)
+	}
+	scanner := bufio.NewScanner(out)
+	for scanner.Scan() {
+		slice := strings.Split(scanner.Text(), "\t")
+		if len(slice) < 2 {
+			fmt.Println("Couldn't parse ", scanner.Text())
+		} else {
+			addressbook[strings.ToLower(slice[0])] = slice[1]
+		}
+	}
+	if err := cmd.Wait(); err != nil {
+		log.Fatal(err)
+	}
+	return addressbook
+}

parseMail.go

@@ -95,6 +95,7 @@ func processHeaders(
 	retvalchan chan map[string]AddressData,
 	addresses []*regexp.Regexp,
 	customFilters []*regexp.Regexp,
+	addressbook map[string]string,
 ) {
 	count := 0
 	retval := make(map[string]AddressData)
@@ -124,12 +125,6 @@ func processHeaders(
 				continue
 			}
 			for _, address := range header {
-
-				dec := new(mime.WordDecoder)
-				name, err := dec.DecodeHeader(address.Name)
-				if err != nil {
-					continue
-				}
 				normaddr := strings.ToLower(address.Address)
 				if filterAddress(normaddr, customFilters) {
 					continue
@@ -140,8 +135,15 @@ func processHeaders(
 					addresses,
 				)
 				if aD, ok := retval[normaddr]; ok {
-					if normaddr != strings.ToLower(name) {
-						aD.Names = append(aD.Names, name)
+					if aD.Name == "" {
+						dec := new(mime.WordDecoder)
+						name, err := dec.DecodeHeader(address.Name)
+						if err != nil {
+							continue
+						}
+						if (strings.ToLower(name) != normaddr) && (strings.ToLower(name) != "") {
+							aD.Names = append(aD.Names, name)
+						}
 					}
 					if aD.Class < class {
 						aD.Class = class
@@ -153,11 +155,22 @@ func processHeaders(
 					retval[normaddr] = aD
 				} else {
 					aD := AddressData{}
+					addressbookname := addressbook[normaddr]
+					if addressbookname == "" {
+						dec := new(mime.WordDecoder)
+						name, err := dec.DecodeHeader(address.Name)
+						if err != nil {
+							continue
+						}
+						if (strings.ToLower(name) != normaddr) && (strings.ToLower(name) != "") {
+							aD.Names = append(aD.Names, name)
+						}
+					} else {
+						aD.Name = addressbookname
+						delete(addressbook, normaddr)
+					}
 					aD.Address = normaddr
 					aD.Class = class
-					if normaddr != strings.ToLower(name) {
-						aD.Names = append(aD.Names, name)
-					}
 					aD.ClassDate = [3]int64{0, 0, 0}
 					aD.ClassDate[class] = time.Unix()
 					aD.ClassCount = [3]int{0, 0, 0}
@@ -174,7 +187,7 @@ func processHeaders(
 	close(retvalchan)
 }
 
-func walkMaildir(path string, addresses []*regexp.Regexp, customFilters []*regexp.Regexp) map[string]AddressData {
+func walkMaildir(path string, addresses []*regexp.Regexp, customFilters []*regexp.Regexp, addressbook map[string]string) map[string]AddressData {
 	headers := make(chan *mail.Header)
 	messagePaths := make(chan string, 4096)
 
@@ -189,7 +202,7 @@ func walkMaildir(path string, addresses []*regexp.Regexp, customFilters []*regex
 	}
 
 	retvalchan := make(chan map[string]AddressData)
-	go processHeaders(headers, retvalchan, addresses, customFilters)
+	go processHeaders(headers, retvalchan, addresses, customFilters, addressbook)
 
 	filepath.Walk(path, func(path string, info os.FileInfo, err error) error {
 		if err != nil {

ranking.go

@@ -32,6 +32,9 @@ func getMostFrequent(names []string) string {
 }
 
 func normalizeAddressNames(aD AddressData) AddressData {
+	if aD.Name != "" {
+		return aD
+	}
 	aD.Name = getMostFrequent(aD.Names)
 	return aD
 }