unit tests passing

src/stan/services/sample/fixed_param.hpp

@@ -140,7 +140,7 @@ int fixed_param(Model& model, const std::size_t num_chains,
   }
   std::vector<boost::ecuyer1988> rngs;
   std::vector<Eigen::VectorXd> cont_vectors;
-  std::vector<callbacks::structured_writer> dummy_metric_writers;
+  std::vector<stan::callbacks::structured_writer> dummy_metric_writers;
   std::vector<util::mcmc_writer> writers;
   std::vector<stan::mcmc::sample> samples;
   std::vector<stan::mcmc::fixed_param_sampler> samplers(num_chains);

src/stan/services/sample/hmc_nuts_dense_e_adapt.hpp

@@ -631,7 +631,7 @@ int hmc_nuts_dense_e_adapt(
     unit_e_metric.emplace_back(std::make_unique<stan::io::dump>(
         util::create_unit_e_dense_inv_metric(model.num_params_r())));
   }
-  std::vector<callbacks::structured_writer> dummy_metric_writer;
+  std::vector<stan::callbacks::structured_writer> dummy_metric_writer;
   dummy_metric_writer.reserve(num_chains);
   for (size_t i = 0; i < num_chains; ++i) {
     dummy_metric_writer.emplace_back(stan::callbacks::structured_writer());

src/stan/services/sample/hmc_nuts_diag_e_adapt.hpp

@@ -336,7 +336,8 @@ int hmc_nuts_diag_e_adapt(
  * @return error_codes::OK if successful
  */
 template <class Model, typename InitContextPtr, typename InitInvContextPtr,
-          typename InitWriter, typename SampleWriter, typename DiagnosticWriter>
+          typename InitWriter, typename SampleWriter, typename DiagnosticWriter,
+          typename MetricWriter>
 int hmc_nuts_diag_e_adapt(
     Model& model, size_t num_chains, const std::vector<InitContextPtr>& init,
     const std::vector<InitInvContextPtr>& init_inv_metric,
@@ -349,7 +350,7 @@ int hmc_nuts_diag_e_adapt(
     std::vector<InitWriter>& init_writer,
     std::vector<SampleWriter>& sample_writer,
     std::vector<DiagnosticWriter>& diagnostic_writer,
-    std::vector<callbacks::structured_writer>& metric_writer) {
+    std::vector<MetricWriter>& metric_writer) {
   if (num_chains == 1) {
     return hmc_nuts_diag_e_adapt(
         model, *init[0], *init_inv_metric[0], random_seed, init_chain_id,
@@ -470,7 +471,7 @@ int hmc_nuts_diag_e_adapt(
     std::vector<InitWriter>& init_writer,
     std::vector<SampleWriter>& sample_writer,
     std::vector<DiagnosticWriter>& diagnostic_writer) {
-  std::vector<callbacks::structured_writer> dummy_metric_writer;
+  std::vector<stan::callbacks::structured_writer> dummy_metric_writer;
   dummy_metric_writer.reserve(num_chains);
   for (size_t i = 0; i < num_chains; ++i) {
     dummy_metric_writer.emplace_back(stan::callbacks::structured_writer());
@@ -499,9 +500,10 @@ int hmc_nuts_diag_e_adapt(
  * @tparam Model Model class
  * @tparam InitContextPtr A pointer with underlying type derived from
  * `stan::io::var_context`
+ * @tparam InitWriter A type derived from `stan::callbacks::writer`
  * @tparam SamplerWriter A type derived from `stan::callbacks::writer`
  * @tparam DiagnosticWriter A type derived from `stan::callbacks::writer`
- * @tparam InitWriter A type derived from `stan::callbacks::writer`
+ * @tparam MetricWriter A type derived from `stan::callbacks::structured_writer`
  * @param[in] model Input model (with data already instantiated)
  * @param[in] num_chains The number of chains to run in parallel. `init`,
  * `init_writer`, `sample_writer`, and `diagnostic_writer` must be the same
@@ -540,7 +542,8 @@ int hmc_nuts_diag_e_adapt(
  * @return error_codes::OK if successful
  */
 template <class Model, typename InitContextPtr, typename InitWriter,
-          typename SampleWriter, typename DiagnosticWriter>
+          typename SampleWriter, typename DiagnosticWriter,
+          typename MetricWriter>
 int hmc_nuts_diag_e_adapt(
     Model& model, size_t num_chains, const std::vector<InitContextPtr>& init,
     unsigned int random_seed, unsigned int init_chain_id, double init_radius,
@@ -552,7 +555,7 @@ int hmc_nuts_diag_e_adapt(
     std::vector<InitWriter>& init_writer,
     std::vector<SampleWriter>& sample_writer,
     std::vector<DiagnosticWriter>& diagnostic_writer,
-    std::vector<callbacks::structured_writer>& metric_writer) {
+    std::vector<MetricWriter>& metric_writer) {
   std::vector<std::unique_ptr<stan::io::dump>> unit_e_metric;
   unit_e_metric.reserve(num_chains);
   for (size_t i = 0; i < num_chains; ++i) {
@@ -640,7 +643,7 @@ int hmc_nuts_diag_e_adapt(
     unit_e_metric.emplace_back(std::make_unique<stan::io::dump>(
         util::create_unit_e_diag_inv_metric(model.num_params_r())));
   }
-  std::vector<callbacks::structured_writer> dummy_metric_writer;
+  std::vector<stan::callbacks::structured_writer> dummy_metric_writer;
   dummy_metric_writer.reserve(num_chains);
   for (size_t i = 0; i < num_chains; ++i) {
     dummy_metric_writer.emplace_back(stan::callbacks::structured_writer());

src/stan/services/sample/hmc_nuts_unit_e_adapt.hpp

@@ -146,8 +146,7 @@ int hmc_nuts_unit_e_adapt(
  * @tparam InitWriter A type derived from `stan::callbacks::writer`
  * @tparam SamplerWriter A type derived from `stan::callbacks::writer`
  * @tparam DiagnosticWriter A type derived from `stan::callbacks::writer`
- * @tparam Stream A type with with a valid `operator<<(std::string)`
- * @tparam Deleter A class with a valid `operator()` method for deleting the
+ * @tparam MetricWriter A type derived from `stan::callbacks::structured_writer`
  * @param[in] model Input model (with data already instantiated)
  * @param[in] num_chains The number of chains to run in parallel. `init`,
  * `init_inv_metric`, `init_writer`, `sample_writer`, and `diagnostic_writer`
@@ -181,7 +180,7 @@ int hmc_nuts_unit_e_adapt(
  * @return error_codes::OK if successful
  */
 template <class Model, typename InitContextPtr, typename InitWriter,
-          typename SampleWriter, typename DiagnosticWriter>
+          typename SampleWriter, typename DiagnosticWriter, typename MetricWriter>
 int hmc_nuts_unit_e_adapt(
     Model& model, size_t num_chains, const std::vector<InitContextPtr>& init,
     unsigned int random_seed, unsigned int init_chain_id, double init_radius,
@@ -192,7 +191,7 @@ int hmc_nuts_unit_e_adapt(
     std::vector<InitWriter>& init_writer,
     std::vector<SampleWriter>& sample_writer,
     std::vector<DiagnosticWriter>& diagnostic_writer,
-    std::vector<callbacks::structured_writer>& metric_writer) {
+    std::vector<MetricWriter>& metric_writer) {
   if (num_chains == 1) {
     return hmc_nuts_unit_e_adapt(
         model, *init[0], random_seed, init_chain_id, init_radius, num_warmup,
@@ -298,7 +297,7 @@ int hmc_nuts_unit_e_adapt(
     std::vector<InitWriter>& init_writer,
     std::vector<SampleWriter>& sample_writer,
     std::vector<DiagnosticWriter>& diagnostic_writer) {
-  std::vector<callbacks::structured_writer> dummy_metric_writer;
+  std::vector<stan::callbacks::structured_writer> dummy_metric_writer;
   dummy_metric_writer.reserve(num_chains);
   for (size_t i = 0; i < num_chains; ++i) {
     dummy_metric_writer.emplace_back(stan::callbacks::structured_writer());

src/test/unit/services/sample/hmc_nuts_dense_e_adapt_parallel_match_test.cpp

@@ -1,7 +1,7 @@
 #include <stan/services/sample/hmc_nuts_dense_e_adapt.hpp>
-#include <stan/io/empty_var_context.hpp>
 #include <stan/callbacks/json_writer.hpp>
 #include <stan/callbacks/unique_stream_writer.hpp>
+#include <stan/io/empty_var_context.hpp>
 #include <test/unit/util.hpp>
 #include <src/test/unit/services/util.hpp>
 #include <test/test-models/good/optimization/rosenbrock.hpp>
@@ -44,8 +44,6 @@ class ServicesSampleHmcNutsDenseEAdaptParMatch : public testing::Test {
     for (int i = 0; i < num_chains; ++i) {
       ss_metric[i].str(std::string());
       ss_metric[i].clear();
-      metrics[i].begin_record();
-      metrics[i].end_record();
     }
   }
 
@@ -60,7 +58,8 @@ class ServicesSampleHmcNutsDenseEAdaptParMatch : public testing::Test {
       = stan::callbacks::unique_stream_writer<std::stringstream, deleter_noop>;
   std::vector<str_writer> par_parameters;
   std::vector<str_writer> seq_parameters;
-  std::vector<stan::callbacks::structured_writer> metrics;
+  std::vector<stan::callbacks::json_writer<std::stringstream, deleter_noop>>
+      metrics;
   std::vector<stan::test::unit::instrumented_writer> diagnostics;
   std::vector<std::shared_ptr<stan::io::empty_var_context>> context;
   std::unique_ptr<rosenbrock_model_namespace::rosenbrock_model> model;
@@ -125,12 +124,11 @@ TEST_F(ServicesSampleHmcNutsDenseEAdaptParMatch, single_multi_match) {
     par_res.push_back(par_mat);
 
     par_metrics.push_back(ss_metric[i].str());
-    std::cout << "metric " << par_metrics[i] << std::endl << std::flush;
-    // rapidjson::Document document;
-    // ASSERT_FALSE(document.Parse<0>(par_metrics[i].c_str()).HasParseError());
-    // EXPECT_EQ(count_matches("stepsize", par_metrics[i]), 1);
-    // EXPECT_EQ(count_matches("inv_metric", par_metrics[i]), 1);
-    // EXPECT_EQ(count_matches("[", par_metrics[i]), 3);  // list has 2 rows
+    rapidjson::Document document;
+    ASSERT_FALSE(document.Parse<0>(par_metrics[i].c_str()).HasParseError());
+    EXPECT_EQ(count_matches("stepsize", par_metrics[i]), 1);
+    EXPECT_EQ(count_matches("inv_metric", par_metrics[i]), 1);
+    EXPECT_EQ(count_matches("[", par_metrics[i]), 3);  // list has 2 rows
   }
 
   std::vector<Eigen::MatrixXd> seq_res;

src/test/unit/services/sample/hmc_nuts_diag_e_adapt_parallel_match_test.cpp

@@ -1,12 +1,12 @@
 #include <stan/services/sample/hmc_nuts_diag_e_adapt.hpp>
-#include <stan/io/empty_var_context.hpp>
 #include <stan/callbacks/json_writer.hpp>
 #include <stan/callbacks/unique_stream_writer.hpp>
 #include <stan/callbacks/structured_writer.hpp>
-#include <test/unit/util.hpp>
+#include <stan/io/empty_var_context.hpp>
 #include <src/test/unit/services/util.hpp>
 #include <test/test-models/good/optimization/rosenbrock.hpp>
 #include <test/unit/services/instrumented_callbacks.hpp>
+#include <test/unit/util.hpp>
 #include <rapidjson/document.h>
 #include <gtest/gtest.h>
 #include <iostream>

src/test/unit/services/sample/hmc_nuts_unit_e_adapt_parallel_test.cpp

@@ -1,35 +1,58 @@
 #include <stan/services/sample/hmc_nuts_unit_e_adapt.hpp>
-#include <stan/callbacks/structured_writer.hpp>
+#include <stan/callbacks/json_writer.hpp>
+#include <stan/callbacks/unique_stream_writer.hpp>
 #include <stan/io/empty_var_context.hpp>
+#include <src/test/unit/services/util.hpp>
 #include <test/test-models/good/optimization/rosenbrock.hpp>
 #include <test/unit/services/instrumented_callbacks.hpp>
-#include <iostream>
+#include <test/unit/util.hpp>
+#include <rapidjson/document.h>
 #include <gtest/gtest.h>
+#include <iostream>
 
 auto&& blah = stan::math::init_threadpool_tbb();
 
 static constexpr size_t num_chains = 4;
+
+struct deleter_noop {
+  template <typename T>
+  constexpr void operator()(T* arg) const {}
+};
+
 class ServicesSampleHmcNutsUnitEAdaptPar : public testing::Test {
  public:
-  ServicesSampleHmcNutsUnitEAdaptPar() : model(data_context, 0, &model_log) {
+  ServicesSampleHmcNutsUnitEAdaptPar()
+      : ss_metric(num_chains),
+        model(data_context, 0, &model_log) {
     for (int i = 0; i < num_chains; ++i) {
       init.push_back(stan::test::unit::instrumented_writer{});
       parameter.push_back(stan::test::unit::instrumented_writer{});
       diagnostic.push_back(stan::test::unit::instrumented_writer{});
-      metric.push_back(stan::callbacks::structured_writer(
-          std::unique_ptr<std::ofstream>(nullptr)));
+      metric.push_back(
+          stan::callbacks::json_writer<std::stringstream, deleter_noop>(
+              std::unique_ptr<std::stringstream, deleter_noop>(&ss_metric[i])));
       context.push_back(std::make_shared<stan::io::empty_var_context>());
     }
   }
+
+  void SetUp() {
+    for (int i = 0; i < num_chains; ++i) {
+      ss_metric[i].str(std::string());
+      ss_metric[i].clear();
+    }
+  }
+
   stan::io::empty_var_context data_context;
   std::stringstream model_log;
   stan::test::unit::instrumented_logger logger;
   std::vector<stan::test::unit::instrumented_writer> init;
   std::vector<stan::test::unit::instrumented_writer> parameter;
   std::vector<stan::test::unit::instrumented_writer> diagnostic;
-  std::vector<stan::callbacks::structured_writer> metric;
   std::vector<std::shared_ptr<stan::io::empty_var_context>> context;
   stan_model model;
+  std::vector<std::stringstream> ss_metric;
+  std::vector<stan::callbacks::json_writer<std::stringstream, deleter_noop>>
+  metric;
 };
 
 TEST_F(ServicesSampleHmcNutsUnitEAdaptPar, call_count) {
@@ -92,7 +115,7 @@ TEST_F(ServicesSampleHmcNutsUnitEAdaptPar, parameter_checks) {
       model, num_chains, context, random_seed, chain, init_radius, num_warmup,
       num_samples, num_thin, save_warmup, refresh, stepsize, stepsize_jitter,
       max_depth, delta, gamma, kappa, t0, interrupt, logger, init, parameter,
-      diagnostic);
+      diagnostic, metric);
 
   for (size_t i = 0; i < num_chains; ++i) {
     std::vector<std::vector<std::string>> parameter_names;
@@ -103,6 +126,14 @@ TEST_F(ServicesSampleHmcNutsUnitEAdaptPar, parameter_checks) {
     diagnostic_names = diagnostic[i].vector_string_values();
     std::vector<std::vector<double>> diagnostic_values;
     diagnostic_values = diagnostic[i].vector_double_values();
+    std::string metric = ss_metric[i].str();
+    // Adapted metric
+    rapidjson::Document document;
+    ASSERT_FALSE(document.Parse<0>(metric.c_str()).HasParseError());
+    EXPECT_EQ(count_matches("stepsize", metric), 1);
+    EXPECT_EQ(count_matches("inv_metric", metric), 1);
+    EXPECT_EQ(count_matches("[", metric), 1);         // single list
+    EXPECT_EQ(count_matches("[ 1, 1 ]", metric), 1);  // unit diagonal
 
     // Expectations of parameter parameter names.
     ASSERT_EQ(9, parameter_names[0].size());