[FEATURE] Add config::read_from and config::write_to.

seqan · Aug 23, 2023 · 8817a76 · 8817a76
1 parent 7817dc2
commit 8817a76
Show file tree

Hide file tree

Showing 5 changed files with 196 additions and 0 deletions.
diff --git a/include/hibf/config.hpp b/include/hibf/config.hpp
@@ -11,6 +11,7 @@
 #include <cstddef>    // for size_t
 #include <filesystem> // for path
 #include <functional> // for function
+#include <iosfwd>     // for ostream
 #include <iterator>   // for insert_iterator
 
 #include <hibf/contrib/robin_hood.hpp> // for unordered_flat_set
@@ -82,6 +83,9 @@ struct config
     // bool compressed{false};
     //!\}
 
+    void read_from(std::istream & stream);
+    void write_to(std::ostream & stream) const;
+
 private:
     friend class cereal::access;
 

diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
@@ -1,5 +1,6 @@
 set (HIBF_SOURCE_FILES
      hierarchical_interleaved_bloom_filter.cpp
+     config.cpp
      detail/layout/simple_binning.cpp
      detail/layout/execute.cpp
      detail/layout/compute_fpr_correction.cpp

diff --git a/src/config.cpp b/src/config.cpp
@@ -0,0 +1,60 @@
+// ---------------------------------------------------------------------------------------------------
+// Copyright (c) 2006-2023, Knut Reinert & Freie Universität Berlin
+// Copyright (c) 2016-2023, Knut Reinert & MPI für molekulare Genetik
+// This file may be used, modified and/or redistributed under the terms of the 3-clause BSD-License
+// shipped with this file and also available at: https://github.com/seqan/hibf/blob/main/LICENSE.md
+// ---------------------------------------------------------------------------------------------------
+
+#include <cassert>
+#include <charconv>
+#include <iostream>
+
+#include <hibf/config.hpp>
+#include <hibf/detail/prefixes.hpp>
+
+#include <cereal/archives/json.hpp>
+
+namespace hibf
+{
+
+void config::read_from(std::istream & stream)
+{
+    std::string line;
+    std::stringstream config_str;
+
+    while (std::getline(stream, line) && line != prefix::meta_hibf_config_start)
+        ;
+
+    assert(line == prefix::meta_hibf_config_start);
+
+    // TODO ##CONFIG: as prefix
+    while (std::getline(stream, line) && line != prefix::meta_hibf_config_end)
+    {
+        assert(line.size() >= 2);
+        assert(std::string_view{line}.substr(0, 1) == hibf::prefix::meta_header);
+        config_str << line.substr(1); // remove hibf::prefix::meta_header
+    }
+
+    assert(line == prefix::meta_hibf_config_end);
+
+    cereal::JSONInputArchive iarchive(config_str);
+    iarchive(*this);
+}
+
+void config::write_to(std::ostream & stream) const
+{
+    // write json file to temprorary string stream with cereal
+    std::stringstream config_stream{};
+    cereal::JSONOutputArchive output(config_stream); // stream to cout
+    output(cereal::make_nvp("hibf_config", *this));
+
+    // write config
+    stream << prefix::meta_hibf_config_start << '\n';
+    std::string line;
+    while (std::getline(config_stream, line, '\n'))
+        stream << prefix::meta_header << line << '\n';
+    stream << prefix::meta_header << "}\n" // last closing bracket isn't written by loop above
+           << prefix::meta_hibf_config_end << '\n';
+}
+
+} // namespace hibf
diff --git a/test/unit/hibf/CMakeLists.txt b/test/unit/hibf/CMakeLists.txt
@@ -1,4 +1,5 @@
 add_subdirectories ()
 
+hibf_test (config_test.cpp)
 hibf_test (hierarchical_interleaved_bloom_filter_test.cpp)
 hibf_test (interleaved_bloom_filter_test.cpp)
diff --git a/test/unit/hibf/config_test.cpp b/test/unit/hibf/config_test.cpp
@@ -0,0 +1,130 @@
+#include <gtest/gtest.h> // for Test, TestInfo, EXPECT_EQ, Message, TEST, TestPartResult
+
+#include <cstddef>     // for size_t
+#include <sstream>     // for operator<<, char_traits, basic_ostream, basic_stringstream, strings...
+#include <string>      // for allocator, string
+#include <string_view> // for operator<<
+#include <vector>      // for vector
+
+#include <hibf/config.hpp> // for config
+
+TEST(config_test, write_to)
+{
+    std::stringstream ss{};
+
+    hibf::config configuration;
+
+    configuration.number_of_user_bins = 123456789;
+    configuration.number_of_hash_functions = 4;
+    configuration.maximum_false_positive_rate = 0.0001;
+    configuration.threads = 31;
+    configuration.sketch_bits = 8;
+    configuration.tmax = 128;
+    configuration.alpha = 1.0;
+    configuration.max_rearrangement_ratio = 0.333;
+    configuration.disable_estimate_union = true;
+    configuration.disable_rearrangement = false;
+    configuration.disable_cutoffs = false;
+
+    configuration.write_to(ss);
+
+    std::string const expected_file{"@HIBF_CONFIG\n"
+                                    "@{\n"
+                                    "@    \"hibf_config\": {\n"
+                                    "@        \"version\": 1,\n"
+                                    "@        \"number_of_user_bins\": 123456789,\n"
+                                    "@        \"number_of_hash_functions\": 4,\n"
+                                    "@        \"maximum_false_positive_rate\": 0.0001,\n"
+                                    "@        \"threads\": 31,\n"
+                                    "@        \"sketch_bits\": 8,\n"
+                                    "@        \"tmax\": 128,\n"
+                                    "@        \"alpha\": 1.0,\n"
+                                    "@        \"max_rearrangement_ratio\": 0.333,\n"
+                                    "@        \"disable_estimate_union\": true,\n"
+                                    "@        \"disable_rearrangement\": false,\n"
+                                    "@        \"disable_cutoffs\": false\n"
+                                    "@    }\n"
+                                    "@}\n"
+                                    "@HIBF_CONFIG_END\n"};
+
+    EXPECT_EQ(ss.str(), expected_file);
+}
+
+TEST(config_test, read_from)
+{
+    std::stringstream ss{"@HIBF_CONFIG\n"
+                         "@{\n"
+                         "@    \"hibf_config\": {\n"
+                         "@        \"version\": 1,\n"
+                         "@        \"number_of_user_bins\": 123456789,\n"
+                         "@        \"number_of_hash_functions\": 4,\n"
+                         "@        \"maximum_false_positive_rate\": 0.0001,\n"
+                         "@        \"threads\": 31,\n"
+                         "@        \"sketch_bits\": 8,\n"
+                         "@        \"tmax\": 128,\n"
+                         "@        \"alpha\": 1.0,\n"
+                         "@        \"max_rearrangement_ratio\": 0.333,\n"
+                         "@        \"disable_estimate_union\": true,\n"
+                         "@        \"disable_rearrangement\": false,\n"
+                         "@        \"disable_cutoffs\": false\n"
+                         "@    }\n"
+                         "@}\n"
+                         "@HIBF_CONFIG_END\n"};
+
+    hibf::config configuration;
+    configuration.read_from(ss);
+
+    EXPECT_EQ(configuration.number_of_user_bins, 123456789);
+    EXPECT_EQ(configuration.number_of_hash_functions, 4);
+    EXPECT_EQ(configuration.maximum_false_positive_rate, 0.0001);
+    EXPECT_EQ(configuration.threads, 31);
+    EXPECT_EQ(configuration.sketch_bits, 8);
+    EXPECT_EQ(configuration.tmax, 128);
+    EXPECT_EQ(configuration.alpha, 1.0);
+    EXPECT_EQ(configuration.max_rearrangement_ratio, 0.333);
+    EXPECT_EQ(configuration.disable_estimate_union, true);
+    EXPECT_EQ(configuration.disable_rearrangement, false);
+    EXPECT_EQ(configuration.disable_cutoffs, false);
+}
+
+TEST(config_test, read_from_with_more_meta)
+{
+    std::stringstream ss{"@blah some chopper stuff\n"
+                         "@blah some chopper stuff\n"
+                         "@blah some chopper stuff\n"
+                         "@blah some chopper stuff\n"
+                         "@blah some chopper stuff\n"
+                         "@HIBF_CONFIG\n"
+                         "@{\n"
+                         "@    \"hibf_config\": {\n"
+                         "@        \"version\": 1,\n"
+                         "@        \"number_of_user_bins\": 123456789,\n"
+                         "@        \"number_of_hash_functions\": 4,\n"
+                         "@        \"maximum_false_positive_rate\": 0.0001,\n"
+                         "@        \"threads\": 31,\n"
+                         "@        \"sketch_bits\": 8,\n"
+                         "@        \"tmax\": 128,\n"
+                         "@        \"alpha\": 1.0,\n"
+                         "@        \"max_rearrangement_ratio\": 0.333,\n"
+                         "@        \"disable_estimate_union\": true,\n"
+                         "@        \"disable_rearrangement\": false,\n"
+                         "@        \"disable_cutoffs\": false\n"
+                         "@    }\n"
+                         "@}\n"
+                         "@HIBF_CONFIG_END\n"};
+
+    hibf::config configuration;
+    configuration.read_from(ss);
+
+    EXPECT_EQ(configuration.number_of_user_bins, 123456789);
+    EXPECT_EQ(configuration.number_of_hash_functions, 4);
+    EXPECT_EQ(configuration.maximum_false_positive_rate, 0.0001);
+    EXPECT_EQ(configuration.threads, 31);
+    EXPECT_EQ(configuration.sketch_bits, 8);
+    EXPECT_EQ(configuration.tmax, 128);
+    EXPECT_EQ(configuration.alpha, 1.0);
+    EXPECT_EQ(configuration.max_rearrangement_ratio, 0.333);
+    EXPECT_EQ(configuration.disable_estimate_union, true);
+    EXPECT_EQ(configuration.disable_rearrangement, false);
+    EXPECT_EQ(configuration.disable_cutoffs, false);
+}