Improve ada_url_parse performance (fix #73)

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: adaR
 Title: A Fast 'WHATWG' Compliant URL Parser
-Version: 0.3.3
+Version: 0.3.3.9999
 Authors@R: 
     c(person("David", "Schoch", , "[email protected]", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-2952-4812")),

NEWS.md

@@ -1,3 +1,7 @@
+# adaR 0.3.3.9999
+
+* improved ada_url_parse performance(#73)
+
 # adaR 0.3.3
 
 * * bumped ada-url to 2.9.0

README.Rmd

@@ -94,7 +94,7 @@ practical circumstances.
 bench::mark(
     ada = ada_url_parse("https://user_1:[email protected]:8080/dir/../api?q=1#frag", decode = FALSE),
     urltools = urltools::url_parse("https://user_1:[email protected]:8080/dir/../api?q=1#frag"),
-    iterations = 1, check = FALSE
+    check = FALSE
 )
 ```
 

README.md

@@ -116,13 +116,13 @@ practical circumstances.
 bench::mark(
     ada = ada_url_parse("https://user_1:[email protected]:8080/dir/../api?q=1#frag", decode = FALSE),
     urltools = urltools::url_parse("https://user_1:[email protected]:8080/dir/../api?q=1#frag"),
-    iterations = 1, check = FALSE
+    check = FALSE
 )
 #> # A tibble: 2 × 6
 #>   expression      min   median `itr/sec` mem_alloc `gc/sec`
 #>   <bch:expr> <bch:tm> <bch:tm>     <dbl> <bch:byt>    <dbl>
-#> 1 ada          1.24ms   1.24ms      806.        0B        0
-#> 2 urltools    812.7µs  812.7µs     1230.        0B        0
+#> 1 ada          2.83µs   3.24µs   286849.        0B     28.7
+#> 2 urltools   123.41µs 134.36µs     7277.        0B     36.8
 ```
 
 For further benchmark results, see `benchmark.md` in `data_raw`.

src/RcppExports.cpp

@@ -11,7 +11,7 @@ Rcpp::Rostream<false>& Rcpp::Rcerr = Rcpp::Rcpp_cerr_get();
 #endif
 
 // Rcpp_ada_parse
-DataFrame Rcpp_ada_parse(const CharacterVector& input_vec, bool decode);
+List Rcpp_ada_parse(const CharacterVector& input_vec, bool decode);
 RcppExport SEXP _adaR_Rcpp_ada_parse(SEXP input_vecSEXP, SEXP decodeSEXP) {
 BEGIN_RCPP
     Rcpp::RObject rcpp_result_gen;

src/adaR.cpp

@@ -20,7 +20,7 @@ std::string charsub(const ada_string stringi, bool to_unicode = true) {
 }
 
 // [[Rcpp::export]]
-DataFrame Rcpp_ada_parse(const CharacterVector& input_vec, bool decode) {
+List Rcpp_ada_parse(const CharacterVector& input_vec, bool decode) {
   unsigned int n = input_vec.length();
   CharacterVector href(n);
   CharacterVector protocol(n);
@@ -32,6 +32,7 @@ DataFrame Rcpp_ada_parse(const CharacterVector& input_vec, bool decode) {
   CharacterVector pathname(n);
   CharacterVector search(n);
   CharacterVector hash(n);
+  Rcpp::IntegerVector row_name(n);
   for (unsigned int i = 0; i < n; i++) {
     String s = input_vec[i];
     std::string_view input(s.get_cstring());
@@ -60,6 +61,7 @@ DataFrame Rcpp_ada_parse(const CharacterVector& input_vec, bool decode) {
       hash[i] = NA_STRING;
     }
     ada_free(url);
+    row_name[i] = i + 1;
   }
   if (decode) {
     href = Rcpp_url_decode2(href);
@@ -73,13 +75,20 @@ DataFrame Rcpp_ada_parse(const CharacterVector& input_vec, bool decode) {
     search = Rcpp_url_decode2(search);
     hash = Rcpp_url_decode2(hash);
   }
-  return (DataFrame::create(Named("href") = href, _["protocol"] = protocol,
-                            _["username"] = username, _["password"] = password,
-                            _["host"] = host, _["hostname"] = hostname,
-                            _["port"] = port, _["pathname"] = pathname,
-                            _["search"] = search, _["hash"] = hash));
+
+  List result = List::create(Named("href") = href, _["protocol"] = protocol,
+                        _["username"] = username, _["password"] = password,
+                        _["host"] = host, _["hostname"] = hostname,
+                        _["port"] = port, _["pathname"] = pathname,
+                        _["search"] = search, _["hash"] = hash);
+  // as data.frame is expensive - create from the list
+  result.attr("row.names") = row_name;
+  result.attr("class") = "data.frame";
+
+  return result;
 }
 
+
 // higher-order function for all Rcpp_ada_has_*
 LogicalVector Rcpp_ada_has(const CharacterVector& url_vec,
                            std::function<bool(ada_url)> func) {