1 files changed, 131 insertions, 38 deletions

diff --git a/makecolumns.sh b/makecolumns.sh
index 4381f13..3f05a61 100755
--- a/makecolumns.sh
+++ b/makecolumns.sh
@@ -104,14 +104,13 @@ handle_format_version_1() {
 
     # rename our extracted columns
     mv 01_unknown 01_Flags
+
     mv 02_unknown 02_Nachname
     mv 03_unknown 03_Vorname
-    mv 04_unknown 05_Adresszusatz
     mv 05_unknown 06_Ortszusatz
     mv 06_unknown 10_Zustellamt_PLZOst
     mv 07_unknown 07_Strasse
     mv 08_unknown 08_Hausnummer
-    mv 09_unknown 04_Namenszusatz
     mv 10_unknown 09_Verweise
     mv 11_unknown 12_Vorwahl
     mv 12_unknown 13_Rufnummer
@@ -119,10 +118,19 @@ handle_format_version_1() {
     mv 14_unknown 10_Postleitzahl
 
     printf "Normalizing zusaetze ... "
-    paste 04_Namenszusatz 05_Adresszusatz | tr '\t' ' ' | sed -E s/' +'/' '/g > 04_Zusaetze
+    sed -E -e 's:^, +:u. :' 09_unknown > 04_Namenszusatz
+    sed -E -e 's:^, +:u. :' 04_unknown > 05_Adresszusatz
+    paste 04_Namenszusatz 05_Adresszusatz | awk '{$1=$1};1' > 04_Zusaetze
     printf "done.\n"
 
-    tidy_streetnames 07_Strasse
+    # For consistency, create files with empty lines
+    tr -dC '\n' < 01_Flags > 14_Webadresse
+    cp 14_Webadresse 15_Email
+    sed $'s:.*:\t:' 01_Flags > 16_Koordinaten
+
+    tidy_columns
+
+    rm ??_unknown
 }
 
 handle_format_version_2() {
@@ -150,8 +158,6 @@ handle_format_version_2() {
     mv 01_unknown 01_Flags
     mv 16_unknown 02_Nachname
     mv 07_unknown 03_Vorname
-    mv 14_unknown 04_Namenszusatz
-    mv 11_unknown 05_Adresszusatz
     mv 12_unknown 06_Ortszusatz
     mv 08_unknown 07_Strasse
     mv 10_unknown 08_Hausnummer
@@ -162,14 +168,20 @@ handle_format_version_2() {
     mv 05_unknown 12_Vorwahl
     mv 06_unknown 13_Rufnummer
 
-    # remove entries that are for searching only
-    rm 03_unknown 04_unknown
-
     printf "Normalizing zusaetze ... "
-    paste 04_Namenszusatz 05_Adresszusatz | tr '\t' ' ' | sed -E s/' +'/' '/g > 04_Zusaetze
+    sed -E -e 's:^, +:u. :' 14_unknown > 04_Namenszusatz
+    sed -E -e 's:^, +:u. :' 11_unknown > 05_Adresszusatz
+    paste 04_Namenszusatz 05_Adresszusatz | awk '{$1=$1};1' > 04_Zusaetze
     printf "done.\n"
 
-    tidy_streetnames 07_Strasse
+    # For consistency, create files with empty lines
+    tr -dC '\n' < 01_Flags > 14_Webadresse
+    cp 14_Webadresse 15_Email
+    sed $'s:.*:\t:' 01_Flags > 16_Koordinaten
+
+    tidy_columns
+
+    rm ??_unknown
 }
 
 handle_format_version_3() {
@@ -227,11 +239,10 @@ handle_format_version_3() {
         cut -c 1  < 01_unknown > 01_Flags
         cut -c 2- < 01_unknown > 02_Nachname
     fi
-    rm 01_unknown
     printf "done.\n"
 
     printf "Splitting decompress vname chunks into their columns ... "
-    JOT "%0${filename_len}d" ${vname_file} $(( number_of_files - 1 )) 3 | xargs -n 128 cat | tr '\n\0' '\t\n' | tr -d '\377' > 03_Vorname
+    JOT "%0${filename_len}d" ${vname_file} $(( number_of_files - 1 )) 3 | xargs -n 128 cat | tr '\n\0' '\t\n' | tr -d '\377' | awk '{$1=$1};1' > 03_Vorname
     printf "done.\n"
 
     printf "Splitting decompress table file chunks into their columns ... "
@@ -256,20 +267,21 @@ handle_format_version_3() {
     printf "done.\n"
 
     # rename our columns extracted from the table file
-    mv 04_unknown 04_Namenszusatz
-    mv 05_unknown 05_Adresszusatz
     mv 06_unknown 06_Ortszusatz
     mv 08_unknown 08_Hausnummer
-    mv 09_unknown 09_Verweise
     mv 10_unknown 10_Postleitzahl
     mv 11_unknown 11_Ort
     mv 12_unknown 12_Vorwahl
     mv 13_unknown 13_Rufnummer
-    if [ -f 14_unknown -a -f 15_unknown ]; then
-      paste 15_unknown 14_unknown | sed 's/[[:space:]]$//g' > 14_15_Email_Webadresse
-      rm 14_unknown 15_unknown
+    if [ -f 14_unknown ]; then
+      tr '\\' '/' < 14_unknown | iconv -f iso-8859-15 -t utf-8 > 15_Email
     else
-      tr -dC '\n' < 01_Flags > 14_15_Email_Webadresse
+      tr -dC '\n' < 01_Flags > 15_Email
+    fi
+    if [ -f 15_unknown ]; then
+      tr '\\' '/' < 15_unknown | iconv -f iso-8859-15 -t utf-8 > 14_Webadresse
+    else
+      tr -dC '\n' < 01_Flags > 14_Webadresse
     fi
 
     printf "Normalizing flags ... "
@@ -278,9 +290,34 @@ handle_format_version_3() {
     printf "done.\n"
 
     printf "Normalizing zusaetze ... "
-    paste 04_Namenszusatz 05_Adresszusatz | tr '\t' ' ' | sed -E -e 's/ +/ /g' -e 's/^ +//g' -e 's/ +$//g' > 04_Zusaetze
+    sed -E -e 's:^, +:u. :' 04_unknown > 04_Namenszusatz
+    sed -E -e 's:^, +:u. :' 05_unknown > 05_Adresszusatz
+    paste 04_Namenszusatz 05_Adresszusatz | awk '{$1=$1};1' > 04_Zusaetze
+    printf "done.\n"
+
+    printf "Normalizing verweise ... "
+    sed -E -e 's:^\|::g;s:\|$::g;s:\|:, :g' 09_unknown | awk '{$1=$1};1' > 09_Verweise
     printf "done.\n"
 
+    # At least 2002_Q3 and 2003_Q1 are known to sport | -separated vname and nname fields
+    # those fields are redundant, as they are being made explicit in 09_-column for these entries
+    if grep -q '|' 03_Vorname; then
+
+      printf "Treating vname fields with pipe separator ... (adds 3 minutes) ... "
+      # Identify entries with | in nname and move content of 09_Verweise to 04_Zusaetze
+      paste 03_Vorname 04_Zusaetze 09_Verweise | sed -E $'s:^([^|]*)\|.*\t(.*)\t(.*):\\1\t\\2 \\3\t:;' > 03_04_09_Temp
+      cut -f 1 03_04_09_Temp | awk '{$1=$1};1' > 03_Vorname
+      cut -f 2 03_04_09_Temp | awk '{$1=$1};1' > 04_Zusaetze
+      cut -f 3 03_04_09_Temp | awk '{$1=$1};1' > 09_Verweise
+      rm 03_04_09_Temp
+
+      # Delete redundant nachnamen values
+      cut -d '|' -f 1 02_Nachname > 02_Nachname.new
+      mv 02_Nachname.new 02_Nachname
+
+      printf "done.\n"
+    fi
+
     # If street names come in an extra file, extract
     # street names first
     if [ -f "${streets}" ]; then
@@ -293,30 +330,27 @@ handle_format_version_3() {
     # instead of street names
     if [ -f 99_Strassenname ]; then
         mv 07_unknown 07_Strassenindex
-        printf "Looking up street names from indexes ... "
 
         # fix up known broken Strassennamen file
         [ `stat -f %z ${streets}` -eq 1642716 ] && printf '9. Str.\n91. Str.\n91er-Str.\n' >> 99_Strassenname
 
-        tidy_streetnames 99_Strassenname
-
+        printf "Looking up street names from indexes ... "
         cut -d ';' -f 1 07_Strassenindex | ${EL} -0x 99_Strassenname > 07_Strasse
         printf "done.\n"
     else
         mv 07_unknown 07_Strasse
-        tidy_streetnames 07_Strasse
     fi
 
     if [ -f "${karto}" ]; then
       do_processfile_version_3 "${karto}" "geo coordinates" 90_Geokoordinaten_hnr_raw
 
       printf "Looking up geo coordinates for each phonebook entry ... "
-      tr '\0' '\n' < 90_Geokoordinaten_hnr_raw | tr ';' '\t' | cut -f "1,2,3,4,6,7" | tr '\n' '\0' > 90_Geokoordinaten_hnr
+      tr '\0;' '\n\t' < 90_Geokoordinaten_hnr_raw | cut -f "1,2,3,4,6,7" | tr '\n' '\0' > 90_Geokoordinaten_hnr
       rm 90_Geokoordinaten_hnr_raw
       paste 10_Postleitzahl 11_Ort 07_Strasse 08_Hausnummer | map_coords 90_Geokoordinaten_hnr | convert_coords > 16_Koordinaten
       printf "done.\n"
     else
-      sed $'s:.*:\t' 01_Flags > 16_Koordinaten
+      sed $'s:.*:\t:' 01_Flags > 16_Koordinaten
     fi
 
     if [ -f "${braid}" ]; then
@@ -326,6 +360,10 @@ handle_format_version_3() {
       map_branches_v3 97_Branchenname < 09_Branchenindex > 09_Branchen
       printf "done.\n"
     fi
+
+    tidy_columns
+
+    rm ??_unknown
 }
 
 handle_format_version_4() {
@@ -367,15 +405,14 @@ handle_format_version_4() {
     if grep -q ^40 column_0; then
       printf "Cleanung up inverted reverse search flags ... "
       awk '{ a=substr($0,1,1); printf "%x%x\n",index("5670123cdef89ab4",a)%16 ,substr($0,2,1) }' < column_0 > 01_Flags
-      rm column_0
       printf "done\n"
     else
       mv column_0 01_Flags
     fi
+
     mv column_1 02_Nachname
     mv column_2 03_Vorname
     mv column_3 04_05_Namenszusatz_Addresszusatz
-    mv column_4 09_Verweise
     mv column_5 07_08_Strassenindex_Hausnummer
     mv column_6 12_Vorwahl
     mv column_7 10_Postleitzahl
@@ -383,8 +420,6 @@ handle_format_version_4() {
     mv column_9 13_Rufnummer
     mv column_10 14_15_Email_Webadresse
 
-    tidy_streetnames 99_Strassenname
-
     printf "Looking up street names from indexes ... "
     cut -f 1 07_08_Strassenindex_Hausnummer | ${EL} -0 99_Strassenname > 07_Strasse
     printf "done.\n"
@@ -394,7 +429,19 @@ handle_format_version_4() {
     printf "done.\n"
 
     printf "Normalizing zusaetze ... "
-    tr '\t' ' ' < 04_05_Namenszusatz_Addresszusatz | sed -E s/' +'/' '/g > 04_Zusaetze
+    sed -E -e $'s:(^|\t),: u. :g' 04_05_Namenszusatz_Addresszusatz | awk '{$1=$1};1' > 04_Zusaetze
+    printf "done.\n"
+
+    printf "Normalizing verweise ... "
+    sed -E -e $'s:^\|+::g;s:\|+$::g;s:\|:, :g' column_4 | awk '{$1=$1};1' > 09_Verweise
+    printf "done.\n"
+
+    printf "Splitting webaddress ... "
+    cut -d $'\t' -f 1 14_15_Email_Webadresse | tr '\\' '/' | iconv -f iso-8859-15 -t utf-8 > 14_Webadresse
+    printf "done.\n"
+
+    printf "Splitting email ... "
+    sed $'s:$:\t:' < 14_15_Email_Webadresse | cut -sd $'\t' -f 2 | tr '\\' '/' | iconv -f iso-8859-15 -t utf-8 > 15_Email
     printf "done.\n"
 
     if [ -f "$1/zip-streets-hn-geo.tl" ]; then
@@ -414,7 +461,7 @@ handle_format_version_4() {
       paste 10_Postleitzahl 07_Strasse | map_coords 91_Geokoordinaten_str | convert_coords > 16_Koordinaten
       printf "done.\n"
     else
-      sed $'s:.*:\t' 01_Flags > 16_Koordinaten
+      sed $'s:.*:\t:' 01_Flags > 16_Koordinaten
     fi
     rm file_*
 
@@ -429,19 +476,65 @@ handle_format_version_4() {
       map_branches_v4 97_Branchenname < 09_Verweise > 09_Branchen
       printf "done.\n"
     fi
+
+    tidy_columns
+    rm column_*
 }
 
-tidy_streetnames () {
-  streets="$1"
+tidy_columns () {
 
-  printf "Tyding up streetnames ... "
+  printf "Removing backslashes from Nachnamen ... "
+  sed -E -e 's:\\::g' 02_Nachname | awk '{$1=$1};1' | iconv -f iso-8859-15 -t utf-8 > 02_Nachname.new
+  mv 02_Nachname.new 02_Nachname
+  printf "done.\n"
+
+  printf "Unicoding Vornamen ... "
+  iconv -f iso-8859-15 -t utf-8 03_Vorname > 03_Vorname.new
+  mv 03_Vorname.new 03_Vorname
+  printf "done.\n"
 
+  printf "Unicoding Zusaetze ... "
+  iconv -f iso-8859-15 -t utf-8 04_Zusaetze > 04_Zusaetze.new
+  mv 04_Zusaetze.new 04_Zusaetze
+  printf "done.\n"
+
+  printf "Tidying up streetnames ... "
   # Replace any dots at end of line by a single one
   # finish any str abbreviation without a period with a period
-  sed -E -i.bak 's/\.+/./g;s/(S|s)tr( |:)?$/\1tr./;s/ -$/ Str./;s/-$/str./' ${streets}
-  rm "${streets}".bak
+  sed -E 's/\.+/./g;s/(S|s)tr( |:)?$/\1tr./;s/(.*)-(.*) -/\1-\2-Str./;s/ -$/ Str./;s/-$/str./' 07_Strasse | iconv -f iso-8859-15 -t utf-8 > 07_Strasse.new
+  mv 07_Strasse.new 07_Strasse
+  printf "done.\n"
 
+  printf "Normalizing house numbers ... "
+  sed -E -e 's:^([[:digit:]]+) *([A-Za-z])$:\1 \2:' -e 's: a$: A:;s: b$: B:;s: c$: C:;s: d$: D:;s: e$: E:;s: f$: F:;s: g$: G:;s: h$: H:;s: i$: I:;s: j$: J:;s: k$: K:;s: l$: L:;s: m$: M:;s: n$: N:;s: o$: O:;' 08_Hausnummer | iconv -f iso-8859-15 -t utf-8 > 08_Hausnummer.new
+  mv 08_Hausnummer.new 08_Hausnummer
   printf "done.\n"
+
+  printf "Unicoding Verweise ... "
+  iconv -f iso-8859-15 -t utf-8 09_Verweise > 09_Verweise.new
+  mv 09_Verweise.new 09_Verweise
+  printf "done.\n"
+
+  printf "Unicoding Postleitzahl ...  "
+  iconv -f iso-8859-15 -t utf-8 10_Postleitzahl > 10_Postleitzahl.new
+  mv 10_Postleitzahl.new 10_Postleitzahl
+  printf "done.\n"
+
+  printf "Removing trailing * from Ort ... "
+  sed -E -e 's:\*$::' 11_Ort | iconv -f iso-8859-15 -t utf-8 > 11_Ort.new
+  mv 11_Ort.new 11_Ort
+  printf "done.\n"
+
+  printf "Unicoding Vorwahl ...  "
+  iconv -f iso-8859-15 -t utf-8 12_Vorwahl > 12_Vorwahl.new
+  mv 12_Vorwahl.new 12_Vorwahl
+  printf "done.\n"
+
+  printf "Unicoding Rufnummer ...  "
+  iconv -f iso-8859-15 -t utf-8 13_Rufnummer > 13_Rufnummer.new
+  mv 13_Rufnummer.new 13_Rufnummer
+  printf "done.\n"
+
 }
 
 # JOT <format> <begin> <end> <step>