From a60d8d462908c74ad194914a3cb5b793364355ae Mon Sep 17 00:00:00 2001
From: Dirk Engling <erdgeist@erdgeist.org>
Date: Thu, 7 May 2015 16:57:52 +0200
Subject: We now officially can dissect the second version of Telefonbuch from
 1995/1996, aka version 2

---
 src/export/split_version_2.c | 178 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 178 insertions(+)
 create mode 100644 src/export/split_version_2.c

(limited to 'src')

diff --git a/src/export/split_version_2.c b/src/export/split_version_2.c
new file mode 100644
index 0000000..9f04412
--- /dev/null
+++ b/src/export/split_version_2.c
@@ -0,0 +1,178 @@
+#include <stdlib.h>
+#include <stdint.h>
+#include <ctype.h>
+
+#include "mystdlib.h"
+
+static int g_first_field_length;
+static uint8_t cp437_to_iso8859_1_table[] = {
+  0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
+  0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
+  0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f,
+  0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f,
+  0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f,
+  0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f,
+  0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f,
+  0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f,
+  0xc7, 0xfc, 0xe9, 0xe2, 0xe4, 0xe0, 0xe5, 0xe7, 0xea, 0xeb, 0xe8, 0xef, 0xee, 0xec, 0xc4, 0xc5,
+  0xc9, 0xe6, 0xc6, 0xf4, 0xf6, 0xf2, 0xfb, 0xf9, 0xff, 0xd6, 0xdc, 0xa2, 0xa3, 0xa5, 0x2e, 0x2e,
+  0xe1, 0xed, 0xf3, 0xfa, 0xf1, 0xd1, 0xaa, 0xba, 0xbf, 0x2e, 0xac, 0xbd, 0xbc, 0xa1, 0xab, 0xbb,
+  0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e,
+  0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e,
+  0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e,
+  0x2e, 0xdf, 0x2e, 0x2e, 0x2e, 0x2e, 0xb5, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e,
+  0x2e, 0xb1, 0x2e, 0x2e, 0x2e, 0x2e, 0xf7, 0x2e, 0xb0, 0x2e, 0xb7, 0x2e, 0x2e, 0xb2, 0x2e, 0xa0
+};
+static void dump_field( uint8_t **end) {
+  uint8_t *e = *end;
+  uint8_t len = *--e;
+  *end = e - len;
+  printf( "_%02d_ ", len );
+  while( len-- )
+    putchar( cp437_to_iso8859_1_table[ e[-len-1] ] );
+  putchar( 9 );
+}
+
+static uint16_t load_word( uint8_t *table_start, uint16_t base, int offset ) {
+  int dword_offset = offset >> 1;
+
+  /* If a word(subflag, t_off) would span over the end of a 0x4000
+     byte page, it is placed on the next page. All other words are
+     shifted backwards, as well. */
+  table_start += 4;
+  if( base + 4 + 4 * dword_offset > 0x1ffc )
+    table_start += 3 & ( 4 - ( base & 3 ) );
+
+  return ((uint16_t*)table_start)[offset];
+}
+
+static void dump_primary( uint8_t *end, uint32_t flags ) {
+  /* First dump type of record (should be 1) */
+  putchar( *end );
+  putchar( 9 );
+
+  /* Dump first 5 chars of zip, they're always there */
+  printf( "%c%c%c%c%c\t", end[-5], end[-4], end[-3], end[-2], end[-1] );
+  end -= 5;
+
+  if( g_first_field_length == 5 ) {
+    /* Dump first 5 chars of prefix, they're always there */
+    printf( "%c%c%c%c%c\t", end[-5], end[-4], end[-3], end[-2], end[-1] );
+    end -= 5;
+
+    /* There is another version of the zip code present, if this bit is set */
+    if( flags & 0x0080 ) {
+      printf( "%c%c%c%c%c\t", end[-5], end[-4], end[-3], end[-2], end[-1] );
+      end -= 5;
+    }
+
+    /* There is an unclear X present, if this bit is set */
+    if( flags & 0x0040 ) {
+       printf( "%c\t", end[-1] );
+       end--;
+    }
+
+    /* There is another version of the prefix present, if this bit is set */
+    if( flags & 0x0020 ) {
+      printf( "%c%c%c%c%c\t", end[-5], end[-4], end[-3], end[-2], end[-1] );
+      end -= 5;
+    }
+
+    if( flags & 0x0010 ) dump_field( &end );
+    if( flags & 0x0008 ) dump_field( &end );
+    if( flags & 0x0004 ) dump_field( &end );
+    if( flags & 0x0002 ) dump_field( &end );
+    if( flags & 0x0001 ) dump_field( &end );
+    if( flags & 0x8000 ) dump_field( &end );
+    if( flags & 0x4000 ) dump_field( &end );
+    if( flags & 0x2000 ) dump_field( &end );
+    if( flags & 0x1000 ) dump_field( &end );
+
+    dump_field( &end );
+    dump_field( &end );
+  } else {
+
+    /* There is another version of the zip code present, if this bit is set */
+    if( flags & 0x0080 ) {
+      printf( "%c%c%c%c%c\t", end[-5], end[-4], end[-3], end[-2], end[-1] );
+      end -= 5;
+    }
+
+    /* There is an unclear X present, if this bit is set */
+    if( flags & 0x0040 ) {
+       printf( "%c\t", end[-1] );
+       end--;
+    }
+
+    if( flags & 0x0010 ) dump_field( &end );
+    if( flags & 0x0008 ) dump_field( &end );
+    if( flags & 0x0004 ) dump_field( &end );
+    if( flags & 0x0002 ) dump_field( &end );
+    if( flags & 0x0001 ) dump_field( &end );
+    if( flags & 0x8000 ) dump_field( &end );
+    if( flags & 0x4000 ) dump_field( &end );
+    if( flags & 0x2000 ) dump_field( &end );
+    if( flags & 0x1000 ) dump_field( &end );
+
+    /* There is another version of the prefix present, if this bit is set */
+    if( flags & 0x0020 ) dump_field( &end );
+
+    dump_field( &end );
+    dump_field( &end );
+    dump_field( &end );
+  }
+}
+
+static void act_on_record( uint8_t * end, uint16_t base ) {
+  uint16_t *rec  = (uint16_t*)end;
+  uint8_t  *e = end;
+  uint16_t num_dwords = rec[0];
+  uint16_t flagbytes  = rec[1];
+  uint32_t flags = 0;
+  int bold = 0, i, j;
+
+  for (i=0; i<flagbytes; ++i) flags = (flags<<8) | *--end;
+
+  /*
+  putchar( 27 ) ; putchar( '[' ); putchar( '3' );
+  putchar( ( flags & test_me ) ? '1' : '2' );
+  putchar( 'm' ); */
+
+  printf( "-------- %03d: %06X\n", num_dwords, flags );
+  for (i = 0; i < num_dwords; i++ ) {
+    uint16_t subflag = load_word( e, base, 2*i );
+    uint16_t t_off   = load_word( e, base, 2*i+1 );
+    uint8_t *rec_start = end - t_off;
+
+    printf( "%04x (%04x): ", subflag, t_off );
+    if (i == 0)
+      dump_primary( rec_start, flags );
+    else {
+      uint8_t * rec_end = end - ( ( i+1 == num_dwords ) ? 0 : load_word( e, base, 2*i+3 ) );
+      while ( rec_start < rec_end )
+        putchar( cp437_to_iso8859_1_table[ *(rec_start++) ] );
+    }
+    putchar(10);
+  }
+}
+
+int main( int args, char **argv ) {
+  MAP data, index;
+  uint32_t * indoff;
+
+  data  = map_file( argv[1], 1 );
+  index = map_file( argv[2], 1 );
+
+  if( !data || !index )
+    exit( 1 );
+
+  g_first_field_length = data->addr[0x21e];
+
+  for( indoff = (uint32_t*)(index->addr + 8);
+       indoff < (uint32_t*)(index->addr + index->size) && *indoff;
+       indoff++ ) {
+//        printf( "\nActing on record at off: %08X\n", *indoff );
+//        if( indoff[0] >> 14 < indoff[1] >> 14 )
+        act_on_record( data->addr + *indoff, *indoff & 0x1fff );
+  }
+}
-- 
cgit v1.2.3