From 3ec4a534f11d8445dfd7ab703f6b7022e9dea7ad Mon Sep 17 00:00:00 2001
From: jcw349 <jcw349@gmail.com>
Date: Fri, 18 Jul 2025 17:45:55 -0700
Subject: [PATCH 1/2] feat(get_json_entry): empty clade names will inherit from
 parent node

feat(write_json_from_mat):
extension: nextclade added to "meta"
can now export multiple levels of annotations
---
 src/matUtils/convert.cpp | 109 +++++++++++++++++++++------------------
 1 file changed, 59 insertions(+), 50 deletions(-)

diff --git a/src/matUtils/convert.cpp b/src/matUtils/convert.cpp
index e4df3ba7d..1ba1f9a4f 100644
--- a/src/matUtils/convert.cpp
+++ b/src/matUtils/convert.cpp
@@ -507,7 +507,7 @@ MAT::Tree load_mat_from_json(std::string json_filename) {
     return T;
 }
 
-json get_json_entry(MAT::Node* n, std::vector<std::unordered_map<std::string,std::unordered_map<std::string,std::string>>>* catmeta, size_t div = 0, bool use_clade_zero = false, bool use_clade_one = false) {
+json get_json_entry(MAT::Node* n, std::vector<std::unordered_map<std::string,std::unordered_map<std::string,std::string>>>* catmeta, size_t div = 0, std::vector<bool> use_clade_n = {false}, std::vector<std::map<std::string, std::string>> parent_cav = {} ) {
     //each node has 3 constituent attributes
     //node_attrs, branch_attrs, and children. If its a leaf,
     //it also has a simple name attribute.
@@ -532,35 +532,36 @@ json get_json_entry(MAT::Node* n, std::vector<std::unordered_map<std::string,std
     //note: the below is pretty much sars-cov-2 specific. but so is all json-related things.
     //need to declare maps to get nlohmann to interpret these as key pairs.
     div += mutids.size();
-    std::string c1av = "";
-    std::string c2av = "";
-    if (n->clade_annotations.size() >= 1) {
-        c1av = n->clade_annotations[0];
-    }
-    if (n->clade_annotations.size() >= 2) {
-        //json output supports two simultaneous clade annotations
-        //being nextstrain and pangolin, since this is very specific
-        //to sars-cov-2 phylogenomics. Additional fields are ignored
-        //at this point.
-        c1av = n->clade_annotations[1];
-    }
-    std::unordered_map<std::string,std::string> c1a {{"value",c1av}};
-    std::unordered_map<std::string,std::string> c2a {{"value",c2av}};
+    
+    int annot_size = n->clade_annotations.size();
+    std::vector<std::map<std::string, std::string>> cav(annot_size);
+    for ( int c=0; c < annot_size; c++) {
+         
+        if (use_clade_n[c]) {
+            
+            if (n->clade_annotations[c] == ""){
+                cav[c] = parent_cav[c];
+            } else {
+                cav[c] = {{"value",n->clade_annotations[c]}};
+            }
+        }
+    }
+    
     std::string country = n->identifier.substr(0, n->identifier.find("/"));
     std::string date = n->identifier.substr( n->identifier.find_last_of("|")+1, n->identifier.size() );
     std::unordered_map<std::string,std::string> com {{"value",country}};
     std::unordered_map<std::string,std::string> dam {{"value",date}};
     if ((n->is_leaf()) && (country.length() != n->identifier.size()) && (date.length() != n->identifier.size()) ) {
-        sj["node_attrs"] = { {"country",com}, {"date",dam},{"div", div}, {"MAT_Clade_0", c1a}, {"MAT_Clade_1", c2a} };
+        sj["node_attrs"] = { {"country",com}, {"date",dam},{"div", div}};
     } else {
         sj["node_attrs"]["div"] = div;
-        if (use_clade_zero) {
-            sj["node_attrs"]["MAT_Clade_0"] = c1a;
-        }
-        if (use_clade_one) {
-            sj["node_attrs"]["MAT_Clade_1"] = c2a;
-        }
     }
+    
+    for (int c=0; c < annot_size; c++){
+        if (use_clade_n[c])
+            sj["node_attrs"]["MAT_Clade_"+std::to_string(c)] = cav[c];
+    }
+    
     for (const auto& cmet: *catmeta) {
         for (const auto& cmi: cmet) {
             if (cmi.second.find(n->identifier) != cmi.second.end()) {
@@ -573,7 +574,7 @@ json get_json_entry(MAT::Node* n, std::vector<std::unordered_map<std::string,std
     sj["name"] = n->identifier;
     std::vector<json> child_json;
     for (auto cn: n->children) {
-        json cj = get_json_entry(cn, catmeta, div, use_clade_zero, use_clade_one);
+        json cj = get_json_entry(cn, catmeta, div, use_clade_n, cav );
         child_json.push_back(cj);
         sj["children"] = child_json;
     }
@@ -585,6 +586,26 @@ void write_json_from_mat(MAT::Tree* T, std::string output_filename, std::vector<
     json nj;
     std::string desc = "JSON generated by matUtils. If you have metadata you wish to display, you can now drag on a CSV/TSV file and it will be added into this view, [see here](https://docs.nextstrain.org/projects/auspice/en/latest/advanced-functionality/drag-drop-csv-tsv.html) for more info.";
     std::unordered_map<std::string,std::string> lm = {{"branch_label", "none"}};
+    int annot_size = T->root->clade_annotations.size();
+    
+    // check clade annotations that should be written
+    std::vector<bool> use_clades(annot_size,{false});
+    
+    for (auto n: T->depth_first_expansion()) {
+        int idx = 0;
+        for (const auto& c: n->clade_annotations) {
+            ++idx;
+            
+            if (!c.empty()) {
+                use_clades[idx] = true;
+            }
+        }
+        if (std::all_of(use_clades.begin(), use_clades.end(), [](bool b) { return b;})) {
+            break;
+        }
+    }
+    
+    // set Nextclade extension configurations
     nj = {
         {"version","v2"},
         {
@@ -593,8 +614,8 @@ void write_json_from_mat(MAT::Tree* T, std::string output_filename, std::vector<
                 {"filters",json::array({"country","userOrOld"})},
                 {"panels",json::array({"tree"})},
                 {"colorings",{ {{"key","country"},{"title","Country"},{"type","categorical"}} }},
-                {"display_defaults",lm},
-                {"description",desc}
+                {"display_defaults", lm},
+                {"description", desc}
             }
         },
         {
@@ -604,6 +625,7 @@ void write_json_from_mat(MAT::Tree* T, std::string output_filename, std::vector<
             }
         }
     };
+    
     //add metadata to the header colorings if any exist
     if (catmeta->size()>0) {
         for (const auto& cmet: *catmeta) {
@@ -619,33 +641,19 @@ void write_json_from_mat(MAT::Tree* T, std::string output_filename, std::vector<
             }
         }
     }
-    //check whether each of the mat clade annotation fields are used by any sample.
-    bool uses_clade_0 = false;
-    bool uses_clade_1 = false;
-    for (auto n: T->depth_first_expansion()) {
-        if (n->clade_annotations.size() >= 1) {
-            if (n->clade_annotations[0] != "") {
-                uses_clade_0 = true;
-            }
-            if (n->clade_annotations.size() >= 2) {
-                if (n->clade_annotations[1] != "") {
-                    uses_clade_1 = true;
-                }
-            }
+    
+    int idx = 0;
+    for (const bool& c: use_clades) {
+        if (c) {
+            nj["meta"]["extensions"]["nextclade"]["clade_node_attrs"].push_back({{"name","MAT_Clade_" + std::to_string(idx)},{"displayName","MAT_Clade_"+ std::to_string(idx+1)},{"description","MAT_Clade_" + std::to_string(idx+1) + "as inferred or proposed by UShER, matUtils, or Autolin."},{"hideInWeb",false},{"skipAsReference",true}} );
+            
+            std::unordered_map<std::string,std::string> cmap {{"key","MAT_Clade_" + std::to_string(idx)},{"title","MAT_Clade_" + std::to_string(idx+1)},{"type","categorical"}};
+            nj["meta"]["colorings"].push_back(cmap);
         }
-        if ((uses_clade_0) && (uses_clade_1)) {
-            break;
-        }
-    }
-    if (uses_clade_0) {
-        std::unordered_map<std::string,std::string> c1map {{"key","MAT_Clade_0"},{"title","MAT_Clade_1"},{"type","categorical"}};
-        nj["meta"]["colorings"].push_back(c1map);
+        ++idx;
     }
-    if (uses_clade_1) {
-        std::unordered_map<std::string,std::string> c2map {{"key","MAT_Clade_1"},{"title","MAT_Clade_2"},{"type","categorical"}};
-        nj["meta"]["colorings"].push_back(c2map);
-    }
-    auto treestuff = get_json_entry(T->root, catmeta, 0, uses_clade_0, uses_clade_1);
+    
+    auto treestuff = get_json_entry(T->root, catmeta, 0, use_clades);
     nj["tree"]["children"] = json::array({treestuff});
     std::ofstream out(output_filename);
     // out << std::setw(4) << nj << std::endl;
@@ -653,6 +661,7 @@ void write_json_from_mat(MAT::Tree* T, std::string output_filename, std::vector<
     out.close();
 }
 
+
 void get_minimum_subtrees(MAT::Tree* T, std::vector<std::string> samples, size_t nearest_subtree_size, std::string output_dir, std::vector<std::unordered_map<std::string,std::unordered_map<std::string,std::string>>>* catmeta, std::string json_n, std::string newick_n, bool retain_original_branch_len) {
     //get the minimum set of subtrees of the indicated size which cover all input samples
     //and write them to the indicated output directory, with the indicated prefix, along with a tsv indicating which trees contain the relevant samples.

From 6d39dff5c4eb0e9d95ea2bf018bea7a7584a3fa4 Mon Sep 17 00:00:00 2001
From: jcw349 <jcw349@gmail.com>
Date: Fri, 18 Jul 2025 17:52:48 -0700
Subject: [PATCH 2/2] style(write_json_from_mat): removed white spaces

---
 src/matUtils/convert.cpp | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/src/matUtils/convert.cpp b/src/matUtils/convert.cpp
index 1ba1f9a4f..440f1d7b4 100644
--- a/src/matUtils/convert.cpp
+++ b/src/matUtils/convert.cpp
@@ -614,8 +614,8 @@ void write_json_from_mat(MAT::Tree* T, std::string output_filename, std::vector<
                 {"filters",json::array({"country","userOrOld"})},
                 {"panels",json::array({"tree"})},
                 {"colorings",{ {{"key","country"},{"title","Country"},{"type","categorical"}} }},
-                {"display_defaults", lm},
-                {"description", desc}
+                {"display_defaults",lm},
+                {"description",desc}
             }
         },
         {
@@ -661,7 +661,6 @@ void write_json_from_mat(MAT::Tree* T, std::string output_filename, std::vector<
     out.close();
 }
 
-
 void get_minimum_subtrees(MAT::Tree* T, std::vector<std::string> samples, size_t nearest_subtree_size, std::string output_dir, std::vector<std::unordered_map<std::string,std::unordered_map<std::string,std::string>>>* catmeta, std::string json_n, std::string newick_n, bool retain_original_branch_len) {
     //get the minimum set of subtrees of the indicated size which cover all input samples
     //and write them to the indicated output directory, with the indicated prefix, along with a tsv indicating which trees contain the relevant samples.