Merge pull request #69 from ropensci-review-tools/src

separate two sub-fns in src/bm25.cpp

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: pkgmatch
 Title:  Find R Packages Matching Either Descriptions or Other R Packages
-Version: 0.4.1.043
+Version: 0.4.1.044
 Authors@R: c(
     person("Mark", "Padgham", , "[email protected]", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-2172-5265")),

codemeta.json

@@ -8,7 +8,7 @@
   "codeRepository": "https://github.com/ropensci-review-tools/pkgmatch",
   "issueTracker": "https://github.com/ropensci-review-tools/pkgmatch/issues",
   "license": "https://spdx.org/licenses/MIT",
-  "version": "0.4.1.043",
+  "version": "0.4.1.044",
   "programmingLanguage": {
     "@type": "ComputerLanguage",
     "name": "R",

src/bm25.cpp

@@ -1,34 +1,46 @@
 #include "bm25.h"
 
-// [[Rcpp::export]]
-Rcpp::NumericVector rcpp_bm25 (const Rcpp::DataFrame &idfs, const Rcpp::List &tokensList, Rcpp::DataFrame &these_tokens, const double ntoks_avg) {
-
-    // Fixed parameters used in the BM25 function. See wikipedia reference for
-    // these values.
-    const double k = 1.2;
-    const double b = 0.75;
+void make_idf_map (
+    const Rcpp::DataFrame &idfs,
+    std::unordered_map <std::string, double> &idf_map) {
 
-    // Set up primary 'idf_map' to map all tokens to their IDFs over whole corpus:
-    std::unordered_map <std::string, double> idf_map;
     const Rcpp::CharacterVector idf_tokens = idfs ["token"];
     const Rcpp::NumericVector idf_idf = idfs ["idf"];
     for (int i = 0; i < idfs.nrow (); i++) {
         std::string this_tok = static_cast<std::string> (idf_tokens [i]);
         idf_map.emplace (this_tok, idf_idf [i]);
     }
+}
 
-    const int ndocs = tokensList.size();
+void make_these_tokens_map (
+    const Rcpp::DataFrame &these_tokens,
+    std::unordered_map <std::string, int> &these_tokens_map) {
 
-    // Then make a map of the input tokens and counts:
-    std::unordered_map <std::string, int> these_tokens_map;
     const Rcpp::CharacterVector these_tokens_str = these_tokens ["token"];
     const Rcpp::IntegerVector these_tokens_n = these_tokens ["np"];
 
     for (int i = 0; i < these_tokens.nrow (); i++) {
         const std::string this_string = static_cast <std::string> (these_tokens_str [i]);
         these_tokens_map.emplace (this_string, these_tokens_n [i]);
     }
+}
 
+// [[Rcpp::export]]
+Rcpp::NumericVector rcpp_bm25 (const Rcpp::DataFrame &idfs, const Rcpp::List &tokensList, Rcpp::DataFrame &these_tokens, const double ntoks_avg) {
+
+    // Fixed parameters used in the BM25 function. See wikipedia reference for
+    // these values.
+    const double k = 1.2;
+    const double b = 0.75;
+
+    // Set up primary 'idf_map' to map all tokens to their IDFs over whole corpus:
+    std::unordered_map <std::string, double> idf_map;
+    make_idf_map (idfs, idf_map);
+
+    std::unordered_map <std::string, int> these_tokens_map;
+    make_these_tokens_map (these_tokens, these_tokens_map);
+
+    const int ndocs = tokensList.size();
     Rcpp::NumericVector bm25 (ndocs, 0.0);
 
     for (int i = 0; i < ndocs; i++) {

src/bm25.h

@@ -5,4 +5,11 @@
 
 #include <Rcpp.h>
 
+void make_idf_map (
+    const Rcpp::DataFrame &idfs,
+    std::unordered_map <std::string, double> &idf_map);
+void make_these_tokens_map (
+    const Rcpp::DataFrame &these_tokens,
+    std::unordered_map <std::string, double> &these_tokens_map);
+
 Rcpp::NumericVector rcpp_bm25 (const Rcpp::DataFrame &idfs, const Rcpp::List &tokensList, Rcpp::DataFrame &these_tokens, const double ntoks_avg);