Drop stringi dependency

.github/copilot-instructions.md

@@ -18,7 +18,7 @@ Always reference these instructions first and fallback to search or bash command
   ```
 - Install TreeTools dependencies:
   ```bash
-  sudo R -e "install.packages(c('ape', 'bit64', 'lifecycle', 'fastmatch', 'RCurl', 'Rdpack', 'stringi', 'PlotTools'), repos='https://cran.r-project.org/', dependencies=TRUE)"
+  sudo R -e "install.packages(c('ape', 'bit64', 'lifecycle', 'fastmatch', 'RCurl', 'Rdpack', 'PlotTools'), repos='https://cran.r-project.org/', dependencies=TRUE)"
   ```
 
 ### Building and Checking

.github/workflows/ASan.yml

@@ -43,8 +43,6 @@ jobs:
       _R_CHECK_FORCE_SUGGESTS_: false
       RSPM: https://packagemanager.rstudio.com/cran/__linux__/noble/latest
       USING_ASAN: true
-      STRINGI_DISABLE_PKG_CONFIG: true
-      BIOCONDUCTOR_USE_CONTAINER_REPOSITORY: FALSE # For stringi
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
       ASAN_OPTIONS: detect_container_overflow=1:verify_asan_link_order=0
 

.github/workflows/R-CMD-check.yml

@@ -75,6 +75,7 @@ jobs:
       - name: Set up R dependencies
         uses: r-lib/actions/setup-r-dependencies@v2
         with:
+          cache-version: 2-${{ runner.arch }}
           needs: |
             check
 

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: TreeTools
 Title: Create, Modify and Analyse Phylogenetic Trees
-Version: 2.2.0
+Version: 2.2.0.9001
 Authors@R: c(
   person("Martin R.", 'Smith', role = c("aut", "cre", "cph"), 
          email = "martin.smith@durham.ac.uk",
@@ -45,7 +45,6 @@ Imports:
   methods,
   PlotTools,
   Rdpack (>= 2.6.6),
-  stringi,
 Suggests:
   RCurl,
   spelling,

NAMESPACE

@@ -508,7 +508,6 @@ importFrom(stats,median)
 importFrom(stats,na.omit)
 importFrom(stats,runif)
 importFrom(stats,setNames)
-importFrom(stringi,stri_paste)
 importFrom(utils,combn)
 importFrom(utils,globalVariables)
 importFrom(utils,head)

NEWS.md

@@ -1,3 +1,15 @@
+# TreeTools 2.2.0.9001 #
+
+## New functionality
+
+- `Paste0()` provides a fast Rcpp-backed drop-in for `paste0()` / `stri_paste()`
+  with `NA` propagation. Exported for use by downstream packages.
+
+## Performance
+
+- Drop `stringi` dependency.
+- `as.character.Splits()` reimplemented in C++; ~3× faster on 200-tip trees.
+
 # TreeTools 2.2.0 (2026-03-18) #
 
 ## New functionality

R/PhyToString.R

@@ -64,7 +64,6 @@ StringToPhydat <- StringToPhyDat
 #'
 #' @family phylogenetic matrix conversion functions
 #' @template MRS
-#' @importFrom stringi stri_paste
 #' @export
 PhyToString <- function(phy, parentheses = "{", collapse = "", ps = "",
                         useIndex = TRUE, byTaxon = TRUE, concatenate = TRUE) {
@@ -119,17 +118,17 @@ PhyToString <- function(phy, parentheses = "{", collapse = "", ps = "",
       ret <- t(ret)
     }
     if (isTRUE(concatenate)) {
-      stri_paste(c(ret, ps), collapse = "")
+      paste0(c(ret, ps), collapse = "")
     } else {
       if (isTRUE(byTaxon)) {
-        stri_paste(stri_paste(ret, ps))
+        paste0(ret, ps)
       } else {
-        stri_paste(ret, ps, collapse = "")
+        paste0(ret, ps, collapse = "")
       }
     }
   } else {
     if (isTRUE(byTaxon)) ret <- t(ret)
-    stri_paste(apply(ret, 1, stri_paste, collapse = ""), ps)
+    paste0(apply(ret, 1, paste0, collapse = ""), ps)
   }
   # Return:
   ret

R/RcppExports.R

@@ -45,6 +45,10 @@ descendant_tips <- function(parent, child, postorder) {
     .Call(`_TreeTools_descendant_tips`, parent, child, postorder)
 }
 
+splits_to_char <- function(membership, labels) {
+    .Call(`_TreeTools_splits_to_char`, membership, labels)
+}
+
 first_matching_split_pair <- function(x, table) {
     .Call(`_TreeTools_first_matching_split_pair`, x, table)
 }

R/Splits.R

@@ -452,17 +452,16 @@ names.Splits <- function(x) rownames(x)
 
 
 #' @family Splits operations
-#' @importFrom stringi stri_paste
 #' @export
 as.character.Splits <- function(x, ...) {
   tipLabels <- attr(x, "tip.label")
-  nTip <- attr(x, "nTip")
-
-  apply(as.logical(x), 1L, function(inSplit) {
-    stri_paste(stri_paste(tipLabels[inSplit], collapse = " "), " | ",
-               stri_paste(tipLabels[!inSplit], collapse = " "))
-  })
-
+  if (is.null(tipLabels)) {
+    tipLabels <- paste0("t", seq_len(attr(x, "nTip")))
+  }
+  logx <- as.logical(x)
+  ret <- splits_to_char(logx, tipLabels)
+  names(ret) <- rownames(logx)
+  ret
 }
 
 #' @family Splits operations

codemeta.json

@@ -8,13 +8,13 @@
   "codeRepository": "https://github.com/ms609/TreeTools/",
   "issueTracker": "https://github.com/ms609/TreeTools/issues/",
   "license": "https://spdx.org/licenses/GPL-3.0",
-  "version": "2.2.0",
+  "version": "2.2.0.9001",
   "programmingLanguage": {
     "@type": "ComputerLanguage",
     "name": "R",
     "url": "https://r-project.org"
   },
-  "runtimePlatform": "R version 4.5.3 (2026-03-11)",
+  "runtimePlatform": "R version 4.5.3 (2026-03-11 ucrt)",
   "provider": {
     "@id": "https://cran.r-project.org",
     "@type": "Organization",
@@ -261,21 +261,9 @@
       },
       "sameAs": "https://CRAN.R-project.org/package=Rdpack"
     },
-    "8": {
-      "@type": "SoftwareApplication",
-      "identifier": "stringi",
-      "name": "stringi",
-      "provider": {
-        "@id": "https://cran.r-project.org",
-        "@type": "Organization",
-        "name": "Comprehensive R Archive Network (CRAN)",
-        "url": "https://cran.r-project.org"
-      },
-      "sameAs": "https://CRAN.R-project.org/package=stringi"
-    },
     "SystemRequirements": "C++17"
   },
-  "fileSize": "1849.432KB",
+  "fileSize": "1871.905KB",
   "citation": [
     {
       "@type": "SoftwareSourceCode",
@@ -290,7 +278,7 @@
       ],
       "name": "TreeTools: create, modify and analyse phylogenetic trees",
       "identifier": "10.32614/CRAN.package.TreeTools",
-      "description": "R package version 2.2.0",
+      "description": "R package version 2.2.0.9001",
       "@id": "https://doi.org/10.32614/CRAN.package.TreeTools",
       "sameAs": "https://doi.org/10.32614/CRAN.package.TreeTools"
     }

src/RcppExports.cpp

@@ -149,6 +149,18 @@ BEGIN_RCPP
     return rcpp_result_gen;
 END_RCPP
 }
+// splits_to_char
+CharacterVector splits_to_char(const LogicalMatrix membership, const CharacterVector labels);
+RcppExport SEXP _TreeTools_splits_to_char(SEXP membershipSEXP, SEXP labelsSEXP) {
+BEGIN_RCPP
+    Rcpp::RObject rcpp_result_gen;
+    Rcpp::RNGScope rcpp_rngScope_gen;
+    Rcpp::traits::input_parameter< const LogicalMatrix >::type membership(membershipSEXP);
+    Rcpp::traits::input_parameter< const CharacterVector >::type labels(labelsSEXP);
+    rcpp_result_gen = Rcpp::wrap(splits_to_char(membership, labels));
+    return rcpp_result_gen;
+END_RCPP
+}
 // first_matching_split_pair
 IntegerVector first_matching_split_pair(const RawMatrix x, const RawMatrix table);
 RcppExport SEXP _TreeTools_first_matching_split_pair(SEXP xSEXP, SEXP tableSEXP) {
@@ -534,6 +546,7 @@ static const R_CallMethodDef CallEntries[] = {
     {"_TreeTools_descendant_edges", (DL_FUNC) &_TreeTools_descendant_edges, 3},
     {"_TreeTools_descendant_edges_single", (DL_FUNC) &_TreeTools_descendant_edges_single, 5},
     {"_TreeTools_descendant_tips", (DL_FUNC) &_TreeTools_descendant_tips, 3},
+    {"_TreeTools_splits_to_char", (DL_FUNC) &_TreeTools_splits_to_char, 2},
     {"_TreeTools_first_matching_split_pair", (DL_FUNC) &_TreeTools_first_matching_split_pair, 2},
     {"_TreeTools_first_matching_split_index", (DL_FUNC) &_TreeTools_first_matching_split_index, 2},
     {"_TreeTools_num_to_parent", (DL_FUNC) &_TreeTools_num_to_parent, 2},

src/fast_paste.cpp

@@ -0,0 +1,48 @@
+#include <Rcpp/Lighter>
+#include <cstring>
+#include <string>
+#include <vector>
+
+using namespace Rcpp;
+
+// Format each row of a logical split matrix as "A B | C D".
+// `membership` is nSplit x nTip (logical); `labels` is length nTip.
+// [[Rcpp::export]]
+CharacterVector splits_to_char(const LogicalMatrix membership,
+                              const CharacterVector labels) {
+  const int n_split = membership.nrow();
+  const int n_tip = membership.ncol();
+  CharacterVector out(n_split);
+
+  // Cache translated label pointers and lengths
+  std::vector<const char *> lab(n_tip);
+  std::vector<size_t> lab_len(n_tip);
+  for (int j = 0; j < n_tip; ++j) {
+    lab[j] = Rf_translateCharUTF8(STRING_ELT(labels, j));
+    lab_len[j] = std::strlen(lab[j]);
+  }
+
+  std::string buf;
+  for (int i = 0; i < n_split; ++i) {
+    buf.clear();
+    bool first_in = true;
+    for (int j = 0; j < n_tip; ++j) {
+      if (membership(i, j)) {
+        if (!first_in) buf.push_back(' ');
+        buf.append(lab[j], lab_len[j]);
+        first_in = false;
+      }
+    }
+    buf.append(" | ", 3);
+    bool first_out = true;
+    for (int j = 0; j < n_tip; ++j) {
+      if (!membership(i, j)) {
+        if (!first_out) buf.push_back(' ');
+        buf.append(lab[j], lab_len[j]);
+        first_out = false;
+      }
+    }
+    SET_STRING_ELT(out, i, Rf_mkCharLenCE(buf.c_str(), buf.size(), CE_UTF8));
+  }
+  return out;
+}