added script to create a markdown table for size overheads

scripts/create_results_table_sizes.sh

@@ -0,0 +1,114 @@
+#! /usr/bin/env bash
+
+FOLDER=results
+
+function main() {
+  PrintTableHeader
+  ParseFile books_200M_uint32_results.txt
+  ParseFile fb_200M_uint32_results.txt
+  ParseFile lognormal_200M_uint32_results.txt
+  ParseFile normal_200M_uint32_results.txt
+  ParseFile uniform_dense_200M_uint32_results.txt
+  ParseFile uniform_sparse_200M_uint32_results.txt
+  ParseFile books_200M_uint64_results.txt
+  ParseFile fb_200M_uint64_results.txt
+  ParseFile lognormal_200M_uint64_results.txt
+  ParseFile normal_200M_uint64_results.txt
+  ParseFile osm_cellids_200M_uint64_results.txt
+  ParseFile uniform_dense_200M_uint64_results.txt
+  ParseFile uniform_sparse_200M_uint64_results.txt
+  ParseFile wiki_ts_200M_uint64_results.txt
+}
+
+function PrintTableHeader() {
+  echo "|               | ART       | B-tree    | BS        | FAST      | IS        | RBS       | RMI       | RS        | TIP       |"
+  echo "| ------------- | ---------:| ---------:| ---------:| ---------:| ---------:| ---------:| ---------:| ---------:| ---------:|"
+}
+
+function ParseFile() {
+  file_name=$1
+
+  cat $FOLDER/$file_name | grep -v Repeating | grep -v read | grep -v "data contains duplicates" | \
+  awk -v file_name=$file_name -F'[, ]' '
+  # Simple insertion sort (awk does not have a sort, only gawk)
+  function my_sort(arr, new_arr) {
+    new_arr[0] = arr[0];
+    new_len = 1;
+    for(i=1; i<repeat_count; i++) {
+      for(j=0; j<new_len; j++) {
+        if(arr[i] < new_arr[j]) {
+          tmp = arr[i];
+          arr[i] = new_arr[j];
+          new_arr[j] = tmp;
+        }
+      }
+      new_arr[new_len] = arr[i];
+      new_len++;
+    }
+  }
+
+  BEGIN {
+    art = "n/a"
+    btree = "n/a"
+    bs = "n/a"
+    fast = "n/a"
+    is = "n/a"
+    rbs = "n/a"
+    rmi = "n/a"
+    rs = "n/a"
+    tip = "n/a"
+
+    data_set = "unknown"
+    if(file_name == "books_200M_uint32_results.txt") { data_set = "amzn32"; bit_count = 32; }
+    if(file_name == "fb_200M_uint32_results.txt") { data_set = "face32"; bit_count = 32; }
+    if(file_name == "lognormal_200M_uint32_results.txt") { data_set = "logn32"; bit_count = 32; }
+    if(file_name == "normal_200M_uint32_results.txt") { data_set = "norm32"; bit_count = 32; }
+    if(file_name == "uniform_dense_200M_uint32_results.txt") { data_set = "uden32"; bit_count = 32; }
+    if(file_name == "uniform_sparse_200M_uint32_results.txt") { data_set = "uspr32"; bit_count = 32; }
+
+    if(file_name == "books_200M_uint64_results.txt") { data_set = "amzn64"; bit_count = 64; }
+    if(file_name == "fb_200M_uint64_results.txt") { data_set = "face64"; bit_count = 64; }
+    if(file_name == "lognormal_200M_uint64_results.txt") { data_set = "logn64"; bit_count = 64; }
+    if(file_name == "normal_200M_uint64_results.txt") { data_set = "norm64"; bit_count = 64; }
+    if(file_name == "osm_cellids_200M_uint64_results.txt") { data_set = "osmc64"; bit_count = 64; }
+    if(file_name == "uniform_dense_200M_uint64_results.txt") { data_set = "uden64"; bit_count = 64; }
+    if(file_name == "uniform_sparse_200M_uint64_results.txt") { data_set = "uspr64"; bit_count = 64; }
+    if(file_name == "wiki_ts_200M_uint64_results.txt") { data_set = "wiki64"; bit_count = 64; }
+
+    if(data_set == "unknown") {
+      print "Unknown data set, please extend the script."
+      exit
+    }
+  }
+
+  /RESULT/ {
+    # Calculate size overhead
+    ds_size = $NF
+    byte_per_tuple = (bit_count + 64) / 8
+    data_size = byte_per_tuple * 200 * 1000 * 1000
+    overhead = (ds_size - data_size) / data_size
+    if(overhead * 100 >= 10 || overhead * 100 <= -10) {
+      result = sprintf("%.3f%%", overhead * 100)
+    } else {
+      result = sprintf("%.3f%%", overhead * 100)
+    }
+
+    # Remember in variable for respective data structure
+    if($2 == "ART") art = result
+    if($2 == "stx::btree_multimap") btree = result
+    if($2 == "BinarySearch") bs = result
+    if($2 == "FAST") fast = result
+    if($2 == "InterpolationSearch") is = result
+    if($2 == "RadixBinarySearch18") rbs = result
+    if($2 == "rmi") rmi = result
+    if($2 == "RadixSpline") rs = result
+    if($2 == "TIP") tip = result
+  }
+
+  END {
+    printf("| %-13s | %9s | %9s | %9s | %9s | %9s | %9s | %9s | %9s | %9s |\n", data_set, art, btree, bs, fast, is, rbs, rmi, rs, tip);
+  }
+  '
+}
+
+main