From e06286d721f7538e379fb9853f36a9dfa6d7acb8 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 18 Dec 2022 21:23:04 +0300
Subject: [PATCH 01/25] kphp-yaml v0.1: parse, emit, parse_file, emit_file, c++
 tests

---
 builtin-functions/_functions.txt      |   5 +
 runtime/runtime.cmake                 |   1 +
 runtime/yaml.cpp                      | 169 +++++++++++++++++++++
 runtime/yaml.h                        |  23 +++
 tests/cpp/runtime/runtime-tests.cmake |   1 +
 tests/cpp/runtime/yaml-test.cpp       | 211 ++++++++++++++++++++++++++
 6 files changed, 410 insertions(+)
 create mode 100644 runtime/yaml.cpp
 create mode 100644 runtime/yaml.h
 create mode 100644 tests/cpp/runtime/yaml-test.cpp

diff --git a/builtin-functions/_functions.txt b/builtin-functions/_functions.txt
index f4f9cd13fe..c17c686e01 100644
--- a/builtin-functions/_functions.txt
+++ b/builtin-functions/_functions.txt
@@ -1611,3 +1611,8 @@ class DateTimeImmutable implements DateTimeInterface {
   public function getOffset(): int;
   public function getTimestamp(): int;
 }
+
+function yaml_emit_file ($filename ::: string, $data ::: mixed) ::: bool;
+function yaml_emit ($data ::: mixed) ::: string;
+function yaml_parse_file ($filename ::: string, $pos ::: int = 0) ::: mixed;
+function yaml_parse ($data ::: string, $pos ::: int = 0) ::: mixed;
diff --git a/runtime/runtime.cmake b/runtime/runtime.cmake
index bf6df3dcde..cfa9a155c3 100644
--- a/runtime/runtime.cmake
+++ b/runtime/runtime.cmake
@@ -113,6 +113,7 @@ prepend(KPHP_RUNTIME_SOURCES ${BASE_DIR}/runtime/
         vkext.cpp
         vkext_stats.cpp
         ffi.cpp
+        yaml.cpp
         zlib.cpp
         zstd.cpp)
 
diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
new file mode 100644
index 0000000000..60e37cc49c
--- /dev/null
+++ b/runtime/yaml.cpp
@@ -0,0 +1,169 @@
+#include <yaml-cpp/yaml.h>
+
+#include "runtime/optional.h"
+#include "runtime/streams.h"
+#include "runtime/critical_section.h"
+
+#include "runtime/yaml.h"
+
+// TODO: add key-arrays support
+void yaml_node_to_mixed(const YAML::Node &node, mixed &data) {
+  data.clear();
+  if (node.IsNull()) {
+    php_warning("Yaml node is null. Skipping it");
+    return;
+  }
+  if (node.IsScalar()) {
+    string string_data = string(node.as<std::string>().c_str());
+    if (string_data.is_int()) {
+      data = string_data.to_int();
+      return;
+    }
+    dl::enter_critical_section();
+    double *float_data = new double;
+    if (string_data.try_to_float(float_data)) {
+      data = *float_data;
+    }
+    else data = string_data;
+    delete float_data;
+    dl::leave_critical_section();
+    return;
+  }
+  if (node.IsSequence()) {
+    for (auto it = node.begin(); it != node.end(); ++it) {
+      mixed data_piece;
+      yaml_node_to_mixed(*it, data_piece);
+      data.push_back(data_piece);
+    }
+  }
+  else if (node.IsMap()) {
+    for (auto it : node) {
+      mixed data_piece;
+      yaml_node_to_mixed(it.second, data_piece);
+      data[string(it.first.as<std::string>().c_str())] = data_piece;
+    }
+  }
+}
+
+// TODO: rewrite using switch
+// TODO: add key-arrays support
+void mixed_to_yaml_node(const mixed &data, YAML::Node &node) {
+  if (!data.is_array()) {
+    if (data.is_null()) {
+      php_warning("Cannot convert (mixed)null into yaml node");
+      return;
+    }
+    if (data.is_string()) {
+      node = std::string(data.as_string().c_str());
+      return;
+    }
+    if (data.is_int()) {
+      node = data.as_int();
+      return;
+    }
+    if (data.is_float()) {
+      node = data.as_double();
+      return;
+    }
+    if (data.is_bool()) {
+      if (!data.as_bool()) {
+          node = false;
+      }
+      else {
+          node = true;
+      }
+      return;
+    }
+  }
+  array<mixed> data_array = data.as_array();
+  if (data_array.is_pseudo_vector()) {
+    for (auto it = data_array.begin(); it != data_array.end(); ++it) {
+      mixed data_piece = it.get_value();
+      if (data_piece.is_array()) {
+        YAML::Node node_piece;
+        mixed_to_yaml_node(data_piece, node_piece);
+        node.push_back(node_piece);
+      }
+      else if (data_piece.is_string()) {
+        node.push_back(std::string(data_piece.as_string().c_str()));
+      }
+      else if (data_piece.is_int()) {
+        node.push_back(data_piece.as_int());
+      }
+      else if (data_piece.is_float()) {
+        node.push_back(data_piece.as_double());
+      }
+      else if (data_piece.is_bool()) {
+          node.push_back(data_piece.as_bool());
+      }
+      else if (data_piece.is_null()) {
+        php_warning("Data piece is null. Skipping it");
+      }
+      else php_warning("Unknown data type. Skipping it"); // maybe this is redundant?
+    }
+  } else {
+    for (auto it = data_array.begin(); it != data_array.end(); ++it) {
+      mixed data_piece = it.get_value();
+      YAML::Node node_piece;
+      mixed_to_yaml_node(data_piece, node_piece);
+      mixed data_key = it.get_key();
+      if (data_key.is_null() || data_key.is_array()) {
+        php_warning("Key is null or an array. Cannot handle it. Pushing data as in a vector");
+        node.push_back(node_piece);
+      }
+      else if (data_key.is_string()) {
+        node[std::string(data_key.as_string().c_str())] = node_piece;
+      }
+      else if (data_key.is_int()) { // float and bool are cast to int
+        node[data_key.as_int()] = node_piece;
+      }
+      else { // maybe this is redundant?
+        php_warning("Unknown key type. Pushing data as in a vector");
+        node.push_back(node_piece);
+      }
+    }
+  }
+}
+
+bool f$yaml_emit_file(const string &filename, const mixed &data) {
+  if (filename.empty()) {
+    php_warning("Filename cannot be empty");
+    return false;
+  }
+  string emitted_data = f$yaml_emit(data);
+  Optional<int64_t> size = f$file_put_contents(filename, emitted_data);
+  if (size.is_false()) {
+    php_warning("Error while writing to file \"%s\"", filename.c_str());
+    return false;
+  }
+  return true;
+}
+
+string f$yaml_emit(const mixed &data) {
+  YAML::Node node;
+  mixed_to_yaml_node(data, node);
+  return string(YAML::Dump(node).c_str());
+}
+
+mixed f$yaml_parse_file(const string &filename, int pos) {
+  if (filename.empty()) {
+    php_warning("Filename cannot be empty");
+    return {};
+  }
+  Optional<string> data = f$file_get_contents(filename);
+  if (data.is_false()) {
+    php_warning("Error while reading file \"%s\"", filename.c_str());
+    return {};
+  }
+  return f$yaml_parse(data.ref(), pos);
+}
+
+mixed f$yaml_parse(const string &data, int pos) {
+  if (pos < 0) {
+    php_warning("Argument \"pos\" = %d. Values other than 0 are not supported yet. Setting to default (pos = 0)", pos);
+  }
+  YAML::Node node = YAML::Load(data.c_str());
+  mixed parsed_data;
+  yaml_node_to_mixed(node, parsed_data);
+  return parsed_data;
+}
diff --git a/runtime/yaml.h b/runtime/yaml.h
new file mode 100644
index 0000000000..0a694ef596
--- /dev/null
+++ b/runtime/yaml.h
@@ -0,0 +1,23 @@
+#pragma once
+
+#include "runtime/kphp_core.h"
+
+/*
+ * emit mixed into a yaml file
+ */
+bool f$yaml_emit_file(const string &filename, const mixed &data);
+
+/*
+ * emit mixed into a yaml string
+ */
+string f$yaml_emit(const mixed &data);
+
+/*
+ * parse yaml file into mixed
+ */
+mixed f$yaml_parse_file(const string &filename, int pos = 0);
+
+/*
+ * parse yaml string into mixed
+ */
+mixed f$yaml_parse(const string &data, int pos = 0);
\ No newline at end of file
diff --git a/tests/cpp/runtime/runtime-tests.cmake b/tests/cpp/runtime/runtime-tests.cmake
index 32f74e5d7a..621dbb33a1 100644
--- a/tests/cpp/runtime/runtime-tests.cmake
+++ b/tests/cpp/runtime/runtime-tests.cmake
@@ -19,6 +19,7 @@ prepend(RUNTIME_TESTS_SOURCES ${BASE_DIR}/tests/cpp/runtime/
         memory_resource/unsynchronized_pool_resource-test.cpp
         string-list-test.cpp
         string-test.cpp
+        yaml-test.cpp
         zstd-test.cpp)
 
 allow_deprecated_declarations_for_apple(${BASE_DIR}/tests/cpp/runtime/inter-process-mutex-test.cpp)
diff --git a/tests/cpp/runtime/yaml-test.cpp b/tests/cpp/runtime/yaml-test.cpp
new file mode 100644
index 0000000000..60fbc83693
--- /dev/null
+++ b/tests/cpp/runtime/yaml-test.cpp
@@ -0,0 +1,211 @@
+#include <gtest/gtest.h>
+
+#include "runtime/files.h"
+#include "runtime/yaml.h"
+
+TEST(yaml_test, test_yaml_string) {
+  mixed example = string("string");
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_string());
+  ASSERT_TRUE(example.as_string() == result.as_string());
+}
+
+TEST(yaml_test, test_yaml_int) {
+  mixed example = 13;
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_int());
+  ASSERT_EQ(example.as_int(), result.as_int());
+}
+
+TEST(yaml_test, test_yaml_float) {
+  mixed example = 3.1416;
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_float());
+  ASSERT_DOUBLE_EQ(example.as_double(), result.as_double());
+}
+
+TEST(yaml_test, test_yaml_vector) {
+  mixed example;
+  example.push_back(string("string"));
+  example.push_back(13);
+  example.push_back(3.1416);
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_array());
+  ASSERT_TRUE(result.as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[0].is_string());
+  ASSERT_TRUE(example[0].as_string() == result[0].as_string());
+  ASSERT_TRUE(result[1].is_int());
+  ASSERT_EQ(example[1].as_int(), result[1].as_int());
+  ASSERT_TRUE(result[2].is_float());
+  ASSERT_DOUBLE_EQ(example[2].as_double(), result[2].as_double());
+}
+
+TEST(yaml_test, test_yaml_vector_recursive) {
+  mixed example, copy;
+  example.push_back(string("string"));
+  example.push_back(13);
+  example.push_back(3.1416);
+  for (auto it = example.begin(); it != example.end(); ++it) {
+    copy.push_back(it.get_value());
+  }
+  example.push_back(copy);
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result[3].is_array());
+  ASSERT_TRUE(result[3].as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[3][0].is_string());
+  ASSERT_TRUE(example[3][0].as_string() == result[3][0].as_string());
+  ASSERT_TRUE(result[3][1].is_int());
+  ASSERT_EQ(example[3][1].as_int(), result[3][1].as_int());
+  ASSERT_TRUE(result[3][2].is_float());
+  ASSERT_DOUBLE_EQ(example[3][2].as_double(), result[3][2].as_double());
+}
+
+TEST(yaml_test, test_yaml_map) {
+  mixed example;
+  example[string("first")] = string("string");
+  example[string("second")] = 13;
+  example[string("third")] = 3.1416;
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_array());
+  ASSERT_FALSE(result.as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[string("first")].is_string());
+  ASSERT_TRUE(example[string("first")].as_string() == result[string("first")].as_string());
+  ASSERT_TRUE(result[string("second")].is_int());
+  ASSERT_EQ(example[string("second")].as_int(), result[string("second")].as_int());
+  ASSERT_TRUE(result[string("third")].is_float());
+  ASSERT_DOUBLE_EQ(example[string("third")].as_double(), result[string("third")].as_double());
+}
+
+TEST(yaml_test, test_yaml_map_recursive) {
+  mixed example, copy;
+  example[string("first")] = string("string");
+  example[string("second")] = 13;
+  example[string("third")] = 3.1416;
+  for (auto it = example.begin(); it != example.end(); ++it) {
+    copy[it.get_key()] = it.get_value();
+  }
+  example[string("self")] = copy;
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result[string("self")].is_array());
+  ASSERT_FALSE(result[string("self")].as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[string("self")][string("first")].is_string());
+  ASSERT_TRUE(example[string("self")][string("first")].as_string() == result[string("self")][string("first")].as_string());
+  ASSERT_TRUE(result[string("self")][string("second")].is_int());
+  ASSERT_EQ(example[string("self")][string("second")].as_int(), result[string("self")][string("second")].as_int());
+  ASSERT_TRUE(result[string("self")][string("third")].is_float());
+  ASSERT_DOUBLE_EQ(example[string("self")][string("third")].as_double(), result[string("self")][string("third")].as_double());
+}
+
+TEST(yaml_test, test_yaml_string_file) {
+  mixed example = string("string");
+  string filename("test_yaml_string");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_string());
+  ASSERT_TRUE(example.as_string() == result.as_string());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_int_file) {
+  mixed example = 13;
+  string filename("test_yaml_int");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_int());
+  ASSERT_EQ(example.as_int(), result.as_int());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_float_file) {
+  mixed example = 3.1416;
+  string filename("test_yaml_float");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_float());
+  ASSERT_DOUBLE_EQ(example.as_double(), result.as_double());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_vector_file) {
+  mixed example;
+  example.push_back(string("string"));
+  example.push_back(13);
+  example.push_back(3.1416);
+  string filename("test_yaml_vector");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_array());
+  ASSERT_TRUE(result.as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[0].is_string());
+  ASSERT_TRUE(example[0].as_string() == result[0].as_string());
+  ASSERT_TRUE(result[1].is_int());
+  ASSERT_EQ(example[1].as_int(), result[1].as_int());
+  ASSERT_TRUE(result[2].is_float());
+  ASSERT_DOUBLE_EQ(example[2].as_double(), result[2].as_double());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_vector_recursive_file) {
+  mixed example, copy;
+  example.push_back(string("string"));
+  example.push_back(13);
+  example.push_back(3.1416);
+  for (auto it = example.begin(); it != example.end(); ++it) {
+    copy.push_back(it.get_value());
+  }
+  example.push_back(copy);
+  string filename("test_yaml_vector_recursive");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result[3].is_array());
+  ASSERT_TRUE(result[3].as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[3][0].is_string());
+  ASSERT_TRUE(example[3][0].as_string() == result[3][0].as_string());
+  ASSERT_TRUE(result[3][1].is_int());
+  ASSERT_EQ(example[3][1].as_int(), result[3][1].as_int());
+  ASSERT_TRUE(result[3][2].is_float());
+  ASSERT_DOUBLE_EQ(example[3][2].as_double(), result[3][2].as_double());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_map_file) {
+  mixed example;
+  example[string("first")] = string("string");
+  example[string("second")] = 13;
+  example[string("third")] = 3.1416;
+  string filename("test_yaml_map");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_array());
+  ASSERT_FALSE(result.as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[string("first")].is_string());
+  ASSERT_TRUE(example[string("first")].as_string() == result[string("first")].as_string());
+  ASSERT_TRUE(result[string("second")].is_int());
+  ASSERT_EQ(example[string("second")].as_int(), result[string("second")].as_int());
+  ASSERT_TRUE(result[string("third")].is_float());
+  ASSERT_DOUBLE_EQ(example[string("third")].as_double(), result[string("third")].as_double());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_map_recursive_file) {
+  mixed example, copy;
+  example[string("first")] = string("string");
+  example[string("second")] = 13;
+  example[string("third")] = 3.1416;
+  for (auto it = example.begin(); it != example.end(); ++it) {
+    copy[it.get_key()] = it.get_value();
+  }
+  example[string("self")] = copy;
+  string filename("test_yaml_map_recursive");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result[string("self")].is_array());
+  ASSERT_FALSE(result[string("self")].as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[string("self")][string("first")].is_string());
+  ASSERT_TRUE(example[string("self")][string("first")].as_string() == result[string("self")][string("first")].as_string());
+  ASSERT_TRUE(result[string("self")][string("second")].is_int());
+  ASSERT_EQ(example[string("self")][string("second")].as_int(), result[string("self")][string("second")].as_int());
+  ASSERT_TRUE(result[string("self")][string("third")].is_float());
+  ASSERT_DOUBLE_EQ(example[string("self")][string("third")].as_double(), result[string("self")][string("third")].as_double());
+  ASSERT_TRUE(f$unlink(filename));
+}
\ No newline at end of file

From 6cd1ccfdc933487dedf2738f146c455758ee792d Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Fri, 23 Dec 2022 21:04:15 +0300
Subject: [PATCH 02/25] khhp-yaml: codestyle fixes & memory optimizations

---
 runtime/yaml.cpp | 114 +++++++++++++++--------------------------------
 1 file changed, 37 insertions(+), 77 deletions(-)

diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index 60e37cc49c..1469128f8b 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -6,38 +6,33 @@
 
 #include "runtime/yaml.h"
 
-// TODO: add key-arrays support
 void yaml_node_to_mixed(const YAML::Node &node, mixed &data) {
   data.clear();
   if (node.IsNull()) {
     php_warning("Yaml node is null. Skipping it");
-    return;
-  }
-  if (node.IsScalar()) {
-    string string_data = string(node.as<std::string>().c_str());
+  } else if (node.IsScalar()) {
+    const string string_data = string(node.as<std::string>().c_str());
     if (string_data.is_int()) {
       data = string_data.to_int();
-      return;
-    }
-    dl::enter_critical_section();
-    double *float_data = new double;
-    if (string_data.try_to_float(float_data)) {
-      data = *float_data;
+    } else {
+      dl::enter_critical_section();
+      auto *float_data = new double;
+      if (string_data.try_to_float(float_data)) {
+        data = *float_data;
+      } else {
+        data = string_data;
+      }
+      delete float_data;
+      dl::leave_critical_section();
     }
-    else data = string_data;
-    delete float_data;
-    dl::leave_critical_section();
-    return;
-  }
-  if (node.IsSequence()) {
+  } else if (node.IsSequence()) {
     for (auto it = node.begin(); it != node.end(); ++it) {
       mixed data_piece;
       yaml_node_to_mixed(*it, data_piece);
       data.push_back(data_piece);
     }
-  }
-  else if (node.IsMap()) {
-    for (auto it : node) {
+  } else if (node.IsMap()) {
+    for (const auto &it : node) {
       mixed data_piece;
       yaml_node_to_mixed(it.second, data_piece);
       data[string(it.first.as<std::string>().c_str())] = data_piece;
@@ -45,79 +40,43 @@ void yaml_node_to_mixed(const YAML::Node &node, mixed &data) {
   }
 }
 
-// TODO: rewrite using switch
-// TODO: add key-arrays support
 void mixed_to_yaml_node(const mixed &data, YAML::Node &node) {
   if (!data.is_array()) {
     if (data.is_null()) {
-      php_warning("Cannot convert (mixed)null into yaml node");
-      return;
-    }
-    if (data.is_string()) {
+      php_warning("Cannot convert null into yaml node");
+    } else if (data.is_string()) {
       node = std::string(data.as_string().c_str());
-      return;
-    }
-    if (data.is_int()) {
+    } else if (data.is_int()) {
       node = data.as_int();
-      return;
-    }
-    if (data.is_float()) {
+    } else if (data.is_float()) {
       node = data.as_double();
-      return;
-    }
-    if (data.is_bool()) {
-      if (!data.as_bool()) {
-          node = false;
-      }
-      else {
-          node = true;
-      }
-      return;
+    } else if (data.is_bool()) {
+      node = data.as_bool();
     }
+    return;
   }
-  array<mixed> data_array = data.as_array();
+  const array<mixed> &data_array = data.as_array();
   if (data_array.is_pseudo_vector()) {
-    for (auto it = data_array.begin(); it != data_array.end(); ++it) {
-      mixed data_piece = it.get_value();
-      if (data_piece.is_array()) {
-        YAML::Node node_piece;
-        mixed_to_yaml_node(data_piece, node_piece);
-        node.push_back(node_piece);
-      }
-      else if (data_piece.is_string()) {
-        node.push_back(std::string(data_piece.as_string().c_str()));
-      }
-      else if (data_piece.is_int()) {
-        node.push_back(data_piece.as_int());
-      }
-      else if (data_piece.is_float()) {
-        node.push_back(data_piece.as_double());
-      }
-      else if (data_piece.is_bool()) {
-          node.push_back(data_piece.as_bool());
-      }
-      else if (data_piece.is_null()) {
-        php_warning("Data piece is null. Skipping it");
-      }
-      else php_warning("Unknown data type. Skipping it"); // maybe this is redundant?
+    for (const auto &it : data_array) {
+      const mixed data_piece = it.get_value();
+      YAML::Node node_piece;
+      mixed_to_yaml_node(data_piece, node_piece);
+      node.push_back(node_piece);
     }
   } else {
-    for (auto it = data_array.begin(); it != data_array.end(); ++it) {
-      mixed data_piece = it.get_value();
+    for (const auto &it : data_array) {
+      const mixed data_key = it.get_key();
+      const mixed data_piece = it.get_value();
       YAML::Node node_piece;
       mixed_to_yaml_node(data_piece, node_piece);
-      mixed data_key = it.get_key();
       if (data_key.is_null() || data_key.is_array()) {
-        php_warning("Key is null or an array. Cannot handle it. Pushing data as in a vector");
+        php_warning("Null and array keys are prohibited. Pushing data as in a vector");
         node.push_back(node_piece);
-      }
-      else if (data_key.is_string()) {
+      } else if (data_key.is_string()) {
         node[std::string(data_key.as_string().c_str())] = node_piece;
-      }
-      else if (data_key.is_int()) { // float and bool are cast to int
+      } else if (data_key.is_int()) { // float and bool keys are cast to int
         node[data_key.as_int()] = node_piece;
-      }
-      else { // maybe this is redundant?
+      } else {
         php_warning("Unknown key type. Pushing data as in a vector");
         node.push_back(node_piece);
       }
@@ -159,7 +118,7 @@ mixed f$yaml_parse_file(const string &filename, int pos) {
 }
 
 mixed f$yaml_parse(const string &data, int pos) {
-  if (pos < 0) {
+  if (pos != 0) {
     php_warning("Argument \"pos\" = %d. Values other than 0 are not supported yet. Setting to default (pos = 0)", pos);
   }
   YAML::Node node = YAML::Load(data.c_str());
@@ -167,3 +126,4 @@ mixed f$yaml_parse(const string &data, int pos) {
   yaml_node_to_mixed(node, parsed_data);
   return parsed_data;
 }
+

From da3ee16a863d2c83719c0f3f95431c7afd9a9f4b Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 26 Feb 2023 11:27:08 +0300
Subject: [PATCH 03/25] yaml: add empty string tests and number-as-string tests

---
 tests/cpp/runtime/yaml-test.cpp | 51 +++++++++++++++++++++++++++++++++
 1 file changed, 51 insertions(+)

diff --git a/tests/cpp/runtime/yaml-test.cpp b/tests/cpp/runtime/yaml-test.cpp
index 60fbc83693..9ca273ef34 100644
--- a/tests/cpp/runtime/yaml-test.cpp
+++ b/tests/cpp/runtime/yaml-test.cpp
@@ -10,6 +10,13 @@ TEST(yaml_test, test_yaml_string) {
   ASSERT_TRUE(example.as_string() == result.as_string());
 }
 
+TEST(yaml_test, test_yaml_empty_string) {
+  mixed example = string("");
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_string());
+  ASSERT_TRUE(example.as_string() == result.as_string());
+}
+
 TEST(yaml_test, test_yaml_int) {
   mixed example = 13;
   mixed result = f$yaml_parse(f$yaml_emit(example));
@@ -24,6 +31,20 @@ TEST(yaml_test, test_yaml_float) {
   ASSERT_DOUBLE_EQ(example.as_double(), result.as_double());
 }
 
+TEST(yaml_test, test_yaml_int_as_string) {
+  mixed example = string("13");
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_string());
+  ASSERT_EQ(example.as_string(), result.as_string());
+}
+
+TEST(yaml_test, test_yaml_float_as_string) {
+  mixed example = string("3.1416");
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_string());
+  ASSERT_EQ(example.as_string(), result.as_string());
+}
+
 TEST(yaml_test, test_yaml_vector) {
   mixed example;
   example.push_back(string("string"));
@@ -106,6 +127,16 @@ TEST(yaml_test, test_yaml_string_file) {
   ASSERT_TRUE(f$unlink(filename));
 }
 
+TEST(yaml_test, test_yaml_empty_string_file) {
+  mixed example = string("");
+  string filename("test_yaml_empty_string");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_string());
+  ASSERT_TRUE(example.as_string() == result.as_string());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
 TEST(yaml_test, test_yaml_int_file) {
   mixed example = 13;
   string filename("test_yaml_int");
@@ -126,6 +157,26 @@ TEST(yaml_test, test_yaml_float_file) {
   ASSERT_TRUE(f$unlink(filename));
 }
 
+TEST(yaml_test, test_yaml_int_as_string_file) {
+  mixed example = string("13");
+  string filename("test_yaml_int_as_string");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_string());
+  ASSERT_EQ(example.as_string(), result.as_string());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_float_as_string_file) {
+  mixed example = string("3.1416");
+  string filename("test_yaml_float_as_string");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_string());
+  ASSERT_EQ(example.as_string(), result.as_string());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
 TEST(yaml_test, test_yaml_vector_file) {
   mixed example;
   example.push_back(string("string"));

From 4e9cc2ebc7f9b906e7b2d0d751a064c4d20ae8b3 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 26 Feb 2023 11:28:28 +0300
Subject: [PATCH 04/25] yaml: fix number-as-string parsing and rewrite
 mixed_to_yaml() function

---
 runtime/yaml.cpp | 104 +++++++++++++++++++++++++++++------------------
 1 file changed, 64 insertions(+), 40 deletions(-)

diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index 1469128f8b..ca158b8475 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -3,22 +3,27 @@
 #include "runtime/optional.h"
 #include "runtime/streams.h"
 #include "runtime/critical_section.h"
-
 #include "runtime/yaml.h"
 
-void yaml_node_to_mixed(const YAML::Node &node, mixed &data) {
-  data.clear();
-  if (node.IsNull()) {
-    php_warning("Yaml node is null. Skipping it");
-  } else if (node.IsScalar()) {
-    const string string_data = string(node.as<std::string>().c_str());
+void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string &source) {
+  data.clear(); // sets data to NULL
+  if (node.IsScalar()) {
+    const string string_data(node.as<std::string>().c_str());
     if (string_data.is_int()) {
-      data = string_data.to_int();
+      if (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') {
+        data = string_data;
+      } else {
+        data = string_data.to_int();
+      }
     } else {
       dl::enter_critical_section();
       auto *float_data = new double;
       if (string_data.try_to_float(float_data)) {
-        data = *float_data;
+        if (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') {
+          data = string_data;
+        } else {
+          data = *float_data;
+        }
       } else {
         data = string_data;
       }
@@ -28,58 +33,78 @@ void yaml_node_to_mixed(const YAML::Node &node, mixed &data) {
   } else if (node.IsSequence()) {
     for (auto it = node.begin(); it != node.end(); ++it) {
       mixed data_piece;
-      yaml_node_to_mixed(*it, data_piece);
+      yaml_node_to_mixed(*it, data_piece, source);
       data.push_back(data_piece);
     }
   } else if (node.IsMap()) {
     for (const auto &it : node) {
       mixed data_piece;
-      yaml_node_to_mixed(it.second, data_piece);
+      yaml_node_to_mixed(it.second, data_piece, source);
       data[string(it.first.as<std::string>().c_str())] = data_piece;
     }
   }
 }
 
-void mixed_to_yaml_node(const mixed &data, YAML::Node &node) {
+string print_tabs(uint8_t nesting_level) {
+  string tabs;
+  for (uint8_t i = 0; i < 2 * nesting_level; i++) {
+    tabs.push_back(' ');
+  }
+  return tabs;
+}
+
+string print_key(const mixed& data_key) {
+  if (data_key.is_string()) {
+    return data_key.as_string();
+  }
+  return string(data_key.as_int()); // array can not be a key; bool and float keys are cast to int
+}
+
+void mixed_to_string(const mixed& data, string& string_data, uint8_t nesting_level = 0) {
   if (!data.is_array()) {
     if (data.is_null()) {
-      php_warning("Cannot convert null into yaml node");
+      string_data.push_back('~');
     } else if (data.is_string()) {
-      node = std::string(data.as_string().c_str());
+      const string& string_data_piece = data.as_string();
+      if (string_data_piece.size() < 2
+          || (string_data_piece[0] != '"' && string_data_piece[string_data_piece.size() - 1] != '"')) {
+        string_data.push_back('"');
+        string_data.append(string_data_piece);
+        string_data.push_back('"');
+      } else {
+        string_data.append(string_data_piece);
+      }
     } else if (data.is_int()) {
-      node = data.as_int();
+      string_data.append(data.as_int());
     } else if (data.is_float()) {
-      node = data.as_double();
+      string_data.append(data.as_double());
     } else if (data.is_bool()) {
-      node = data.as_bool();
+      string_data.append(data.as_bool());
     }
+    string_data.push_back('\n');
     return;
   }
   const array<mixed> &data_array = data.as_array();
   if (data_array.is_pseudo_vector()) {
     for (const auto &it : data_array) {
-      const mixed data_piece = it.get_value();
-      YAML::Node node_piece;
-      mixed_to_yaml_node(data_piece, node_piece);
-      node.push_back(node_piece);
+      const mixed &data_piece = it.get_value();
+      string_data.append(print_tabs(nesting_level));
+      string_data.append("- ");
+      if (data_piece.is_array()) {
+        string_data.push_back('\n');
+      }
+      mixed_to_string(data_piece, string_data, nesting_level + 1);
     }
   } else {
     for (const auto &it : data_array) {
-      const mixed data_key = it.get_key();
-      const mixed data_piece = it.get_value();
-      YAML::Node node_piece;
-      mixed_to_yaml_node(data_piece, node_piece);
-      if (data_key.is_null() || data_key.is_array()) {
-        php_warning("Null and array keys are prohibited. Pushing data as in a vector");
-        node.push_back(node_piece);
-      } else if (data_key.is_string()) {
-        node[std::string(data_key.as_string().c_str())] = node_piece;
-      } else if (data_key.is_int()) { // float and bool keys are cast to int
-        node[data_key.as_int()] = node_piece;
-      } else {
-        php_warning("Unknown key type. Pushing data as in a vector");
-        node.push_back(node_piece);
+      const mixed &data_piece = it.get_value();
+      string_data.append(print_tabs(nesting_level));
+      string_data.append(print_key(it.get_key()));
+      string_data.append(": ");
+      if (data_piece.is_array()) {
+        string_data.push_back('\n');
       }
+      mixed_to_string(data_piece, string_data, nesting_level + 1);
     }
   }
 }
@@ -99,9 +124,9 @@ bool f$yaml_emit_file(const string &filename, const mixed &data) {
 }
 
 string f$yaml_emit(const mixed &data) {
-  YAML::Node node;
-  mixed_to_yaml_node(data, node);
-  return string(YAML::Dump(node).c_str());
+  string string_data;
+  mixed_to_string(data, string_data);
+  return string_data;
 }
 
 mixed f$yaml_parse_file(const string &filename, int pos) {
@@ -123,7 +148,6 @@ mixed f$yaml_parse(const string &data, int pos) {
   }
   YAML::Node node = YAML::Load(data.c_str());
   mixed parsed_data;
-  yaml_node_to_mixed(node, parsed_data);
+  yaml_node_to_mixed(node, parsed_data, data);
   return parsed_data;
 }
-

From a4422d8faa36cd9223d8e6f6a8edd7a0d3a276e8 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 26 Feb 2023 22:51:41 +0300
Subject: [PATCH 05/25] kphp-yaml: fix boolean emitting & parsing

---
 runtime/yaml.cpp                | 10 +++++++--
 tests/cpp/runtime/yaml-test.cpp | 40 +++++++++++++++++++++++++++++++++
 2 files changed, 48 insertions(+), 2 deletions(-)

diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index ca158b8475..5c477dbe12 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -9,7 +9,13 @@ void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string &sourc
   data.clear(); // sets data to NULL
   if (node.IsScalar()) {
     const string string_data(node.as<std::string>().c_str());
-    if (string_data.is_int()) {
+    if (string_data == string("true") || string_data == string("false")) {
+      if (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') {
+        data = string_data;
+      } else {
+        data = string_data == string("true");
+      }
+    } else if (string_data.is_int()) {
       if (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') {
         data = string_data;
       } else {
@@ -79,7 +85,7 @@ void mixed_to_string(const mixed& data, string& string_data, uint8_t nesting_lev
     } else if (data.is_float()) {
       string_data.append(data.as_double());
     } else if (data.is_bool()) {
-      string_data.append(data.as_bool());
+      string_data.append((data.as_bool()) ? "true" : "false");
     }
     string_data.push_back('\n');
     return;
diff --git a/tests/cpp/runtime/yaml-test.cpp b/tests/cpp/runtime/yaml-test.cpp
index 9ca273ef34..2ae8431516 100644
--- a/tests/cpp/runtime/yaml-test.cpp
+++ b/tests/cpp/runtime/yaml-test.cpp
@@ -81,6 +81,25 @@ TEST(yaml_test, test_yaml_vector_recursive) {
   ASSERT_DOUBLE_EQ(example[3][2].as_double(), result[3][2].as_double());
 }
 
+TEST(yaml_test, test_yaml_bool) {
+  mixed example;
+  example.push_back(true);
+  example.push_back(false);
+  example.push_back(string("true"));
+  example.push_back(string("false"));
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_array());
+  ASSERT_TRUE(result.as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[0].is_bool());
+  ASSERT_TRUE(example[0].as_bool() == result[0].as_bool());
+  ASSERT_TRUE(result[1].is_bool());
+  ASSERT_TRUE(example[1].as_bool() == result[1].as_bool());
+  ASSERT_TRUE(result[2].is_string());
+  ASSERT_TRUE(example[2].as_string() == result[2].as_string());
+  ASSERT_TRUE(result[3].is_string());
+  ASSERT_TRUE(example[3].as_string() == result[3].as_string());
+}
+
 TEST(yaml_test, test_yaml_map) {
   mixed example;
   example[string("first")] = string("string");
@@ -219,6 +238,27 @@ TEST(yaml_test, test_yaml_vector_recursive_file) {
   ASSERT_TRUE(f$unlink(filename));
 }
 
+TEST(yaml_test, test_yaml_bool_file) {
+  mixed example;
+  example.push_back(true);
+  example.push_back(false);
+  example.push_back(string("true"));
+  example.push_back(string("false"));
+  string filename("test_yaml_bool");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_array());
+  ASSERT_TRUE(result.as_array().is_pseudo_vector());
+  ASSERT_TRUE(result[0].is_bool());
+  ASSERT_TRUE(example[0].as_bool() == result[0].as_bool());
+  ASSERT_TRUE(result[1].is_bool());
+  ASSERT_TRUE(example[1].as_bool() == result[1].as_bool());
+  ASSERT_TRUE(result[2].is_string());
+  ASSERT_TRUE(example[2].as_string() == result[2].as_string());
+  ASSERT_TRUE(result[3].is_string());
+  ASSERT_TRUE(example[3].as_string() == result[3].as_string());
+}
+
 TEST(yaml_test, test_yaml_map_file) {
   mixed example;
   example[string("first")] = string("string");

From d7322b53553e89199448f5dfb187840899053d4f Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Mon, 27 Feb 2023 21:49:06 +0300
Subject: [PATCH 06/25] kphp-yaml: add PHP test

---
 runtime/yaml.cpp                |  3 ++-
 tests/cpp/runtime/yaml-test.cpp |  1 +
 tests/phpt/yaml/01_complex.php  | 13 +++++++++++++
 3 files changed, 16 insertions(+), 1 deletion(-)
 create mode 100644 tests/phpt/yaml/01_complex.php

diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index 5c477dbe12..99b0fddaa3 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -130,8 +130,9 @@ bool f$yaml_emit_file(const string &filename, const mixed &data) {
 }
 
 string f$yaml_emit(const mixed &data) {
-  string string_data;
+  string string_data("---\n");
   mixed_to_string(data, string_data);
+  string_data.append("...");
   return string_data;
 }
 
diff --git a/tests/cpp/runtime/yaml-test.cpp b/tests/cpp/runtime/yaml-test.cpp
index 2ae8431516..62a469155b 100644
--- a/tests/cpp/runtime/yaml-test.cpp
+++ b/tests/cpp/runtime/yaml-test.cpp
@@ -257,6 +257,7 @@ TEST(yaml_test, test_yaml_bool_file) {
   ASSERT_TRUE(example[2].as_string() == result[2].as_string());
   ASSERT_TRUE(result[3].is_string());
   ASSERT_TRUE(example[3].as_string() == result[3].as_string());
+  ASSERT_TRUE(f$unlink(filename));
 }
 
 TEST(yaml_test, test_yaml_map_file) {
diff --git a/tests/phpt/yaml/01_complex.php b/tests/phpt/yaml/01_complex.php
new file mode 100644
index 0000000000..801956bd77
--- /dev/null
+++ b/tests/phpt/yaml/01_complex.php
@@ -0,0 +1,13 @@
+@ok
+<?php
+
+function test_emit_parse_complex_mixed() {
+  /** @param mixed $in */
+  $in = array("10", 10, "9.7", 9.7, "", array("true"=>"true", true=>true, false=>false, array("false"=>"false", 10 => null)));
+  echo(serialize($in) . PHP_EOL);
+  /** @param mixed $out */
+  $out = yaml_parse(yaml_emit($in));
+  echo(serialize($out) . PHP_EOL);
+}
+
+test_emit_parse_complex_mixed();
\ No newline at end of file

From 48febd953f14ba6ae91ee64fb9785b865dec6b05 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Mon, 27 Feb 2023 21:50:01 +0300
Subject: [PATCH 07/25] kphp-yaml: add yaml.so extension to kphp_run_once.py
 for PHP tests to execute

---
 tests/python/lib/kphp_run_once.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/python/lib/kphp_run_once.py b/tests/python/lib/kphp_run_once.py
index ba078a8fe8..049c20709e 100644
--- a/tests/python/lib/kphp_run_once.py
+++ b/tests/python/lib/kphp_run_once.py
@@ -40,7 +40,7 @@ def _get_extensions(self):
                 ("extension", "tokenizer.so"),
                 ("extension", "h3.so"),
                 ("extension", "zstd.so"),
-                ("extension", "ctype.so")
+                ("extension", "yaml.so"),
             ]
 
         if self._vkext_dir:

From 51771d2a2249a4f87356ccb7b056d712427cb90c Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Mon, 27 Feb 2023 21:58:35 +0300
Subject: [PATCH 08/25] kphp yaml: return an accidentally deleted line
 (tests/python/lib/kphp_run_once.py:43

---
 tests/python/lib/kphp_run_once.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/python/lib/kphp_run_once.py b/tests/python/lib/kphp_run_once.py
index 049c20709e..599518d69e 100644
--- a/tests/python/lib/kphp_run_once.py
+++ b/tests/python/lib/kphp_run_once.py
@@ -40,7 +40,8 @@ def _get_extensions(self):
                 ("extension", "tokenizer.so"),
                 ("extension", "h3.so"),
                 ("extension", "zstd.so"),
-                ("extension", "yaml.so"),
+                ("extension", "ctype.so"),
+                ("extension", "yaml.so")
             ]
 
         if self._vkext_dir:

From 66131a78d8855422e55929bff86e5b02cbe6083b Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Tue, 28 Feb 2023 19:52:27 +0300
Subject: [PATCH 09/25] kphp-yaml: temporarily deleted changes in
 _functions.txt

---
 builtin-functions/_functions.txt | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/builtin-functions/_functions.txt b/builtin-functions/_functions.txt
index c17c686e01..f4f9cd13fe 100644
--- a/builtin-functions/_functions.txt
+++ b/builtin-functions/_functions.txt
@@ -1611,8 +1611,3 @@ class DateTimeImmutable implements DateTimeInterface {
   public function getOffset(): int;
   public function getTimestamp(): int;
 }
-
-function yaml_emit_file ($filename ::: string, $data ::: mixed) ::: bool;
-function yaml_emit ($data ::: mixed) ::: string;
-function yaml_parse_file ($filename ::: string, $pos ::: int = 0) ::: mixed;
-function yaml_parse ($data ::: string, $pos ::: int = 0) ::: mixed;

From 3c1d0ddb9f5b1a63f7b035294f9aba93e69c53ce Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Tue, 28 Feb 2023 19:55:22 +0300
Subject: [PATCH 10/25] kphp-yaml: bring back changes in _functions.txt
 [conflict resolved]

---
 builtin-functions/_functions.txt | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/builtin-functions/_functions.txt b/builtin-functions/_functions.txt
index 5c7a8f3a91..fbd8bbb003 100644
--- a/builtin-functions/_functions.txt
+++ b/builtin-functions/_functions.txt
@@ -1619,3 +1619,8 @@ class DateTimeImmutable implements DateTimeInterface {
 }
 
 function getenv(string $varname = '', bool $local_only = false): mixed;
+
+function yaml_emit_file ($filename ::: string, $data ::: mixed) ::: bool;
+function yaml_emit ($data ::: mixed) ::: string;
+function yaml_parse_file ($filename ::: string, $pos ::: int = 0) ::: mixed;
+function yaml_parse ($data ::: string, $pos ::: int = 0) ::: mixed;

From c45668264db4475cfdd7ca5eb2956fffe051b2fd Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 12 Mar 2023 00:27:27 +0300
Subject: [PATCH 11/25] yaml: delete obvious comments in runtime/yaml.h

---
 runtime/yaml.h | 14 +-------------
 1 file changed, 1 insertion(+), 13 deletions(-)

diff --git a/runtime/yaml.h b/runtime/yaml.h
index 0a694ef596..cd03cdffbb 100644
--- a/runtime/yaml.h
+++ b/runtime/yaml.h
@@ -2,22 +2,10 @@
 
 #include "runtime/kphp_core.h"
 
-/*
- * emit mixed into a yaml file
- */
 bool f$yaml_emit_file(const string &filename, const mixed &data);
 
-/*
- * emit mixed into a yaml string
- */
 string f$yaml_emit(const mixed &data);
 
-/*
- * parse yaml file into mixed
- */
 mixed f$yaml_parse_file(const string &filename, int pos = 0);
 
-/*
- * parse yaml string into mixed
- */
-mixed f$yaml_parse(const string &data, int pos = 0);
\ No newline at end of file
+mixed f$yaml_parse(const string &data, int pos = 0);

From a5adf3f1d0453520ba7bf8f7c6f79816bef87c9e Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 12 Mar 2023 00:28:06 +0300
Subject: [PATCH 12/25] yaml: support for empty arrays, new comments, codestyle
 fixes

---
 runtime/yaml.cpp | 131 +++++++++++++++++++++++++----------------------
 1 file changed, 69 insertions(+), 62 deletions(-)

diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index 99b0fddaa3..c40fdf593e 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -5,37 +5,36 @@
 #include "runtime/critical_section.h"
 #include "runtime/yaml.h"
 
-void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string &source) {
+/*
+ * convert YAML::Node to mixed after parsing a YAML document into YAML::Node
+ */
+static void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string &source) noexcept {
   data.clear(); // sets data to NULL
   if (node.IsScalar()) {
     const string string_data(node.as<std::string>().c_str());
-    if (string_data == string("true") || string_data == string("false")) {
-      if (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') {
-        data = string_data;
-      } else {
-        data = string_data == string("true");
-      }
+    // check whether the primitive is put in quotes in the source YAML
+    const bool string_data_has_quotes = (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"');
+    // if so, it is a string
+    if (string_data_has_quotes) {
+      data = string_data;
+    } else if (string_data == string("true")) {
+      data = true; // "true" without quotes is boolean(1)
+    } else if (string_data == string("false")) {
+      data = false; // "false" without quotes is boolean(0)
     } else if (string_data.is_int()) {
-      if (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') {
-        data = string_data;
-      } else {
-        data = string_data.to_int();
-      }
+      data = string_data.to_int();
     } else {
-      dl::enter_critical_section();
-      auto *float_data = new double;
-      if (string_data.try_to_float(float_data)) {
-        if (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') {
-          data = string_data;
-        } else {
-          data = *float_data;
-        }
+      double float_data = 0.0;
+      if (string_data.try_to_float(&float_data)) {
+        data = float_data;
       } else {
         data = string_data;
       }
-      delete float_data;
-      dl::leave_critical_section();
     }
+  } else if (node.size() == 0 && node.IsDefined() && !node.IsNull()) {
+    // if node is defined, is not null or scalar and has size 0, then it is an empty array
+    array<mixed> empty_array;
+    data = empty_array;
   } else if (node.IsSequence()) {
     for (auto it = node.begin(); it != node.end(); ++it) {
       mixed data_piece;
@@ -49,69 +48,77 @@ void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string &sourc
       data[string(it.first.as<std::string>().c_str())] = data_piece;
     }
   }
+  // else node is Null or Undefined, so data is Null
 }
 
-string print_tabs(uint8_t nesting_level) {
-  string tabs;
-  for (uint8_t i = 0; i < 2 * nesting_level; i++) {
-    tabs.push_back(' ');
-  }
-  return tabs;
+/*
+ * print tabs in quantity of nesting_level (used to print nested YAML entries)
+ */
+static string yaml_print_tabs(const uint8_t nesting_level) noexcept {
+  return string(2 * nesting_level, ' ');
 }
 
-string print_key(const mixed& data_key) {
+/*
+ * print the key of a YAML map entry
+ */
+static string yaml_print_key(const mixed& data_key) noexcept {
   if (data_key.is_string()) {
     return data_key.as_string();
   }
-  return string(data_key.as_int()); // array can not be a key; bool and float keys are cast to int
+  return string(data_key.as_int()); // key can not be an array; bool and float keys are cast to int
 }
 
-void mixed_to_string(const mixed& data, string& string_data, uint8_t nesting_level = 0) {
+/*
+ * get a YAML representation of mixed in a string variable
+ */
+static void mixed_to_string(const mixed& data, string& string_data, const uint8_t nesting_level = 0) noexcept {
+  string buffer;
   if (!data.is_array()) {
     if (data.is_null()) {
-      string_data.push_back('~');
+      buffer.push_back('~');
     } else if (data.is_string()) {
       const string& string_data_piece = data.as_string();
-      if (string_data_piece.size() < 2
-          || (string_data_piece[0] != '"' && string_data_piece[string_data_piece.size() - 1] != '"')) {
-        string_data.push_back('"');
-        string_data.append(string_data_piece);
-        string_data.push_back('"');
+      // check if a string has quotes
+      if (string_data_piece.size() < 2 || (string_data_piece[0] != '"' && string_data_piece[string_data_piece.size() - 1] != '"')) {
+        // if not, put it in quotes
+        buffer.push_back('"');
+        buffer.append(string_data_piece);
+        buffer.push_back('"');
       } else {
-        string_data.append(string_data_piece);
+        buffer = string_data_piece;
       }
     } else if (data.is_int()) {
-      string_data.append(data.as_int());
+      buffer.append(data.as_int());
     } else if (data.is_float()) {
-      string_data.append(data.as_double());
+      buffer.append(data.as_double());
     } else if (data.is_bool()) {
-      string_data.append((data.as_bool()) ? "true" : "false");
+      buffer = (data.as_bool()) ? string("true") : string("false");
     }
+    string_data.append(buffer);
     string_data.push_back('\n');
     return;
   }
   const array<mixed> &data_array = data.as_array();
-  if (data_array.is_pseudo_vector()) {
-    for (const auto &it : data_array) {
-      const mixed &data_piece = it.get_value();
-      string_data.append(print_tabs(nesting_level));
-      string_data.append("- ");
-      if (data_piece.is_array()) {
-        string_data.push_back('\n');
-      }
-      mixed_to_string(data_piece, string_data, nesting_level + 1);
+  if (data_array.empty()) {
+    string_data.append("[]\n"); // an empty array is represented as [] in YAML
+  }
+  const bool data_array_is_vector = data_array.is_pseudo_vector(); // check if an array has keys increasing by 1 starting from 0
+  for (const auto &it : data_array) {
+    const mixed &data_piece = it.get_value();
+    buffer = yaml_print_tabs(nesting_level);
+    if (data_array_is_vector) {
+      buffer.push_back('-');
+    } else {
+      buffer.append(yaml_print_key(it.get_key()));
+      buffer.push_back(':');
     }
-  } else {
-    for (const auto &it : data_array) {
-      const mixed &data_piece = it.get_value();
-      string_data.append(print_tabs(nesting_level));
-      string_data.append(print_key(it.get_key()));
-      string_data.append(": ");
-      if (data_piece.is_array()) {
-        string_data.push_back('\n');
-      }
-      mixed_to_string(data_piece, string_data, nesting_level + 1);
+    if (data_piece.is_array() && !data_piece.as_array().empty()) {
+      buffer.push_back('\n'); // if an element of an array is also a non-empty array, print it on the next line
+    } else {
+      buffer.push_back(' '); // if an element of an array is a primitive or an empty array, print it after a space
     }
+    string_data.append(buffer);
+    mixed_to_string(data_piece, string_data, nesting_level + 1); // for entries of an array, increase nesting level
   }
 }
 
@@ -130,9 +137,9 @@ bool f$yaml_emit_file(const string &filename, const mixed &data) {
 }
 
 string f$yaml_emit(const mixed &data) {
-  string string_data("---\n");
+  string string_data("---\n"); // beginning of a YAML document
   mixed_to_string(data, string_data);
-  string_data.append("...");
+  string_data.append("...\n"); // ending of a YAML document
   return string_data;
 }
 

From 47d382a1fe89daf61a4717024e8eb591432aa620 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 12 Mar 2023 00:28:42 +0300
Subject: [PATCH 13/25] yaml: add null and empty array C++ tests

---
 tests/cpp/runtime/yaml-test.cpp | 34 ++++++++++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/tests/cpp/runtime/yaml-test.cpp b/tests/cpp/runtime/yaml-test.cpp
index 62a469155b..88f78030c3 100644
--- a/tests/cpp/runtime/yaml-test.cpp
+++ b/tests/cpp/runtime/yaml-test.cpp
@@ -136,6 +136,19 @@ TEST(yaml_test, test_yaml_map_recursive) {
   ASSERT_DOUBLE_EQ(example[string("self")][string("third")].as_double(), result[string("self")][string("third")].as_double());
 }
 
+TEST(yaml_test, test_yaml_empty_array) {
+  array<mixed> empty_array;
+  mixed result = f$yaml_parse(f$yaml_emit(empty_array));
+  ASSERT_TRUE(result.is_array());
+  ASSERT_TRUE(result.as_array().empty());
+}
+
+TEST(yaml_test, test_yaml_null) {
+  mixed example;
+  mixed result = f$yaml_parse(f$yaml_emit(example));
+  ASSERT_TRUE(result.is_null());
+}
+
 TEST(yaml_test, test_yaml_string_file) {
   mixed example = string("string");
   string filename("test_yaml_string");
@@ -300,4 +313,23 @@ TEST(yaml_test, test_yaml_map_recursive_file) {
   ASSERT_TRUE(result[string("self")][string("third")].is_float());
   ASSERT_DOUBLE_EQ(example[string("self")][string("third")].as_double(), result[string("self")][string("third")].as_double());
   ASSERT_TRUE(f$unlink(filename));
-}
\ No newline at end of file
+}
+
+TEST(yaml_test, test_yaml_empty_array_file) {
+  array<mixed> empty_array;
+  string filename("test_yaml_empty_array");
+  ASSERT_TRUE(f$yaml_emit_file(filename, empty_array));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_array());
+  ASSERT_TRUE(result.as_array().empty());
+  ASSERT_TRUE(f$unlink(filename));
+}
+
+TEST(yaml_test, test_yaml_null_file) {
+  mixed example;
+  string filename("test_yaml_null");
+  ASSERT_TRUE(f$yaml_emit_file(filename, example));
+  mixed result = f$yaml_parse_file(filename);
+  ASSERT_TRUE(result.is_null());
+  ASSERT_TRUE(f$unlink(filename));
+}

From 4d88e37be38d76433ab2fdf8174710e2c5e1a375 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 12 Mar 2023 00:29:30 +0300
Subject: [PATCH 14/25] yaml: add PHP test for docs/_config.yml file

---
 tests/phpt/yaml/01_complex.php          | 13 -------------
 tests/phpt/yaml/01_yaml_complex.php     | 13 +++++++++++++
 tests/phpt/yaml/02_yaml_config_file.php | 15 +++++++++++++++
 3 files changed, 28 insertions(+), 13 deletions(-)
 delete mode 100644 tests/phpt/yaml/01_complex.php
 create mode 100644 tests/phpt/yaml/01_yaml_complex.php
 create mode 100644 tests/phpt/yaml/02_yaml_config_file.php

diff --git a/tests/phpt/yaml/01_complex.php b/tests/phpt/yaml/01_complex.php
deleted file mode 100644
index 801956bd77..0000000000
--- a/tests/phpt/yaml/01_complex.php
+++ /dev/null
@@ -1,13 +0,0 @@
-@ok
-<?php
-
-function test_emit_parse_complex_mixed() {
-  /** @param mixed $in */
-  $in = array("10", 10, "9.7", 9.7, "", array("true"=>"true", true=>true, false=>false, array("false"=>"false", 10 => null)));
-  echo(serialize($in) . PHP_EOL);
-  /** @param mixed $out */
-  $out = yaml_parse(yaml_emit($in));
-  echo(serialize($out) . PHP_EOL);
-}
-
-test_emit_parse_complex_mixed();
\ No newline at end of file
diff --git a/tests/phpt/yaml/01_yaml_complex.php b/tests/phpt/yaml/01_yaml_complex.php
new file mode 100644
index 0000000000..4481ff7d34
--- /dev/null
+++ b/tests/phpt/yaml/01_yaml_complex.php
@@ -0,0 +1,13 @@
+@ok
+<?php
+
+function test_yaml_complex() {
+  /** @param mixed $in */
+  $in = array("10", 10, "9.7", 9.7, "", array(), array("true"=>"true", true=>true, false=>false, array("false"=>"false", 10 => null)));
+  echo(serialize($in) . PHP_EOL);
+  /** @param mixed $out */
+  $out = yaml_parse(yaml_emit($in));
+  echo(serialize($out) . PHP_EOL);
+}
+
+test_yaml_complex();
diff --git a/tests/phpt/yaml/02_yaml_config_file.php b/tests/phpt/yaml/02_yaml_config_file.php
new file mode 100644
index 0000000000..30ee53c33b
--- /dev/null
+++ b/tests/phpt/yaml/02_yaml_config_file.php
@@ -0,0 +1,15 @@
+@ok
+<?php
+
+function test_yaml_parse_config_file() {
+  /** @param string $filename */
+  $filename = dirname(__FILE__) . '/../../../docs/_config.yml';
+  /** @param mixed $in */
+  $in = yaml_parse_file($filename);
+  echo(serialize($in) . PHP_EOL);
+  /** @param mixed $out */
+  $out = yaml_parse(yaml_emit($in));
+  echo(serialize($out) . PHP_EOL);
+}
+
+test_yaml_parse_config_file();

From a2ac35de9842f03593484af834bd84563537e0b2 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 12 Mar 2023 00:36:10 +0300
Subject: [PATCH 15/25] yaml: add comma in tests/python/lib/kphp_run_once.py:44

---
 tests/python/lib/kphp_run_once.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/python/lib/kphp_run_once.py b/tests/python/lib/kphp_run_once.py
index 599518d69e..75855c0262 100644
--- a/tests/python/lib/kphp_run_once.py
+++ b/tests/python/lib/kphp_run_once.py
@@ -41,7 +41,7 @@ def _get_extensions(self):
                 ("extension", "h3.so"),
                 ("extension", "zstd.so"),
                 ("extension", "ctype.so"),
-                ("extension", "yaml.so")
+                ("extension", "yaml.so"),
             ]
 
         if self._vkext_dir:

From 252cb5829dcbf487b58bf8d3cafd44183e02cdb5 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 23 Apr 2023 19:37:00 +0300
Subject: [PATCH 16/25] yaml: minor codestyle fixes according to @unserialize
 review

---
 runtime/yaml.cpp | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index c40fdf593e..bdf86989ce 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -45,7 +45,7 @@ static void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string
     for (const auto &it : node) {
       mixed data_piece;
       yaml_node_to_mixed(it.second, data_piece, source);
-      data[string(it.first.as<std::string>().c_str())] = data_piece;
+      data.set_value(string(it.first.as<std::string>().c_str()), data_piece);
     }
   }
   // else node is Null or Undefined, so data is Null
@@ -61,7 +61,7 @@ static string yaml_print_tabs(const uint8_t nesting_level) noexcept {
 /*
  * print the key of a YAML map entry
  */
-static string yaml_print_key(const mixed& data_key) noexcept {
+static string yaml_print_key(const mixed &data_key) noexcept {
   if (data_key.is_string()) {
     return data_key.as_string();
   }
@@ -71,13 +71,13 @@ static string yaml_print_key(const mixed& data_key) noexcept {
 /*
  * get a YAML representation of mixed in a string variable
  */
-static void mixed_to_string(const mixed& data, string& string_data, const uint8_t nesting_level = 0) noexcept {
+static void mixed_to_string(const mixed &data, string &string_data, const uint8_t nesting_level = 0) noexcept {
   string buffer;
   if (!data.is_array()) {
     if (data.is_null()) {
-      buffer.push_back('~');
+      buffer.push_back('~'); // tilda is a YAML representation of NULL
     } else if (data.is_string()) {
-      const string& string_data_piece = data.as_string();
+      const string &string_data_piece = data.as_string();
       // check if a string has quotes
       if (string_data_piece.size() < 2 || (string_data_piece[0] != '"' && string_data_piece[string_data_piece.size() - 1] != '"')) {
         // if not, put it in quotes
@@ -101,6 +101,7 @@ static void mixed_to_string(const mixed& data, string& string_data, const uint8_
   const array<mixed> &data_array = data.as_array();
   if (data_array.empty()) {
     string_data.append("[]\n"); // an empty array is represented as [] in YAML
+    return;
   }
   const bool data_array_is_vector = data_array.is_pseudo_vector(); // check if an array has keys increasing by 1 starting from 0
   for (const auto &it : data_array) {

From 17b3d3e670a80c456453e11dd379e57f44f0bd21 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Mon, 22 May 2023 12:13:24 +0300
Subject: [PATCH 17/25] YAML: implement special characters escaping

---
 runtime/yaml.cpp | 44 ++++++++++++++++++++++++++++++++++----------
 1 file changed, 34 insertions(+), 10 deletions(-)

diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index bdf86989ce..ef5b8335cd 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -68,6 +68,37 @@ static string yaml_print_key(const mixed &data_key) noexcept {
   return string(data_key.as_int()); // key can not be an array; bool and float keys are cast to int
 }
 
+/*
+ * escape special characters in a string entry
+ */
+static string yaml_escape(const string &data) noexcept {
+  string escaped_data;
+  for (size_t i = 0; i < data.size(); ++i) {
+    if (data[i] == 10) { // line feed - code 10
+      escaped_data.push_back(92); // backslash
+      escaped_data.push_back('n');
+    } else if (data[i] == 8) { // backspace - code 8
+      escaped_data.push_back(92); // backslash
+      escaped_data.push_back('b');
+    } else if (data[i] == 9) { // horizontal tab - code 9
+      escaped_data.push_back(92); // backslash
+      escaped_data.push_back('t');
+    } else if (data[i] == 11) { // vertical tab - code 11
+      escaped_data.push_back(92); // backslash
+      escaped_data.push_back('v');
+    } else if (data[i] == 34) { // double quotation mark - code 34
+      escaped_data.push_back(92); // backslash
+      escaped_data.push_back(34); // double quotation mark
+    } else if (data[i] == 92) { // backslash - code 92
+      escaped_data.push_back(92);
+      escaped_data.push_back(92); // double backslash
+    } else {
+      escaped_data.push_back(data[i]);
+    }
+  }
+  return escaped_data;
+}
+
 /*
  * get a YAML representation of mixed in a string variable
  */
@@ -77,16 +108,9 @@ static void mixed_to_string(const mixed &data, string &string_data, const uint8_
     if (data.is_null()) {
       buffer.push_back('~'); // tilda is a YAML representation of NULL
     } else if (data.is_string()) {
-      const string &string_data_piece = data.as_string();
-      // check if a string has quotes
-      if (string_data_piece.size() < 2 || (string_data_piece[0] != '"' && string_data_piece[string_data_piece.size() - 1] != '"')) {
-        // if not, put it in quotes
-        buffer.push_back('"');
-        buffer.append(string_data_piece);
-        buffer.push_back('"');
-      } else {
-        buffer = string_data_piece;
-      }
+      buffer.push_back('"'); // cover string entry in double quotes
+      buffer.append(yaml_escape(data.as_string())); // escape special characters
+      buffer.push_back('"');
     } else if (data.is_int()) {
       buffer.append(data.as_int());
     } else if (data.is_float()) {

From 3dcf68ac31bb3501c3e9cb7779d15ab94e301154 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Mon, 29 May 2023 13:52:40 +0300
Subject: [PATCH 18/25] YAML: add PHP tests for symbols that should be escaped

---
 tests/phpt/yaml/03_yaml_escaping.php | 16 ++++++++++++++++
 tests/phpt/yaml/yaml_escaping.yml    |  5 +++++
 2 files changed, 21 insertions(+)
 create mode 100644 tests/phpt/yaml/03_yaml_escaping.php
 create mode 100644 tests/phpt/yaml/yaml_escaping.yml

diff --git a/tests/phpt/yaml/03_yaml_escaping.php b/tests/phpt/yaml/03_yaml_escaping.php
new file mode 100644
index 0000000000..ee35035480
--- /dev/null
+++ b/tests/phpt/yaml/03_yaml_escaping.php
@@ -0,0 +1,16 @@
+@ok
+<?php
+
+function test_yaml_escaping() {
+  /** @param string $filename */
+  $filename = dirname(__FILE__) . '/yaml_escaping.yml';
+  /** @param mixed $in */
+  $in = yaml_parse_file($filename);
+  echo(serialize($in) . PHP_EOL);
+//   echo(yaml_emit($in) . PHP_EOL); // does not pass because PHP YAML does not use double quotes sometimes
+  /** @param mixed $out */
+  $out = yaml_parse(yaml_emit($in));
+  echo(serialize($out) . PHP_EOL);
+}
+
+test_yaml_escaping();
\ No newline at end of file
diff --git a/tests/phpt/yaml/yaml_escaping.yml b/tests/phpt/yaml/yaml_escaping.yml
new file mode 100644
index 0000000000..54e5ac4d05
--- /dev/null
+++ b/tests/phpt/yaml/yaml_escaping.yml
@@ -0,0 +1,5 @@
+---
+quote: "'single'quotes'are'not'escaped'double\"quotes\"are\"escaped\""
+newline: "1\n2\n3\n4\n5\nnewlines"
+other: "backslash\\horizontal tab\tvertical tab\vevery\b last\b symbol\b is\b erased\b"
+...

From 343a9c9535aae8408106fa62ccddb7b68faf8701 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Tue, 19 Sep 2023 15:45:52 +0300
Subject: [PATCH 19/25] yaml: add a linebreak in test 03

---
 tests/phpt/yaml/03_yaml_escaping.php | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/phpt/yaml/03_yaml_escaping.php b/tests/phpt/yaml/03_yaml_escaping.php
index ee35035480..a548c41ec4 100644
--- a/tests/phpt/yaml/03_yaml_escaping.php
+++ b/tests/phpt/yaml/03_yaml_escaping.php
@@ -13,4 +13,5 @@ function test_yaml_escaping() {
   echo(serialize($out) . PHP_EOL);
 }
 
-test_yaml_escaping();
\ No newline at end of file
+test_yaml_escaping();
+

From ecf7065f21669cfde4f47480004b1bfbab4173d8 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 1 Oct 2023 15:28:04 +0300
Subject: [PATCH 20/25] yaml: make yaml an extension, turning on by CMake
 option `-DYAML=ON

---
 cmake/external-libraries.cmake                         | 3 +++
 runtime/runtime.cmake                                  | 7 ++++++-
 runtime/{ => yaml}/yaml.cpp                            | 2 +-
 runtime/{ => yaml}/yaml.h                              | 0
 tests/cpp/runtime/{ => extensions-tests}/yaml-test.cpp | 2 +-
 tests/cpp/runtime/runtime-tests.cmake                  | 7 ++++++-
 6 files changed, 17 insertions(+), 4 deletions(-)
 rename runtime/{ => yaml}/yaml.cpp (99%)
 rename runtime/{ => yaml}/yaml.h (100%)
 rename tests/cpp/runtime/{ => extensions-tests}/yaml-test.cpp (99%)

diff --git a/cmake/external-libraries.cmake b/cmake/external-libraries.cmake
index 44a6734f28..6d208de75b 100644
--- a/cmake/external-libraries.cmake
+++ b/cmake/external-libraries.cmake
@@ -74,3 +74,6 @@ if(APPLE)
     add_definitions(-DEPOLL_SHIM_LIB_DIR="${epoll_BINARY_DIR}/src")
     set(EPOLL_SHIM_LIB epoll-shim)
 endif()
+
+option(YAML "build yaml" OFF)
+cmake_print_variables(YAML)
diff --git a/runtime/runtime.cmake b/runtime/runtime.cmake
index e62ad28db0..521ca8b9b7 100644
--- a/runtime/runtime.cmake
+++ b/runtime/runtime.cmake
@@ -49,6 +49,11 @@ prepend(KPHP_RUNTIME_PDO_PGSQL_SOURCES pdo/pgsql/
         pgsql_pdo_emulated_statement.cpp)
 endif()
 
+if (YAML)
+prepend(KPHP_RUNTIME_YAML_SOURCES /yaml/
+	yaml.cpp)
+endif()
+
 prepend(KPHP_RUNTIME_SOURCES ${BASE_DIR}/runtime/
         ${KPHP_RUNTIME_DATETIME_SOURCES}
         ${KPHP_RUNTIME_MEMORY_RESOURCE_SOURCES}
@@ -58,6 +63,7 @@ prepend(KPHP_RUNTIME_SOURCES ${BASE_DIR}/runtime/
         ${KPHP_RUNTIME_PDO_SOURCES}
         ${KPHP_RUNTIME_PDO_MYSQL_SOURCES}
         ${KPHP_RUNTIME_PDO_PGSQL_SOURCES}
+        ${KPHP_RUNTIME_YAML_SOURCES}
         allocator.cpp
         array_functions.cpp
         bcmath.cpp
@@ -121,7 +127,6 @@ prepend(KPHP_RUNTIME_SOURCES ${BASE_DIR}/runtime/
         vkext.cpp
         vkext_stats.cpp
         ffi.cpp
-        yaml.cpp
         zlib.cpp
         zstd.cpp)
 
diff --git a/runtime/yaml.cpp b/runtime/yaml/yaml.cpp
similarity index 99%
rename from runtime/yaml.cpp
rename to runtime/yaml/yaml.cpp
index ef5b8335cd..8407546f79 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml/yaml.cpp
@@ -3,7 +3,7 @@
 #include "runtime/optional.h"
 #include "runtime/streams.h"
 #include "runtime/critical_section.h"
-#include "runtime/yaml.h"
+#include "yaml.h"
 
 /*
  * convert YAML::Node to mixed after parsing a YAML document into YAML::Node
diff --git a/runtime/yaml.h b/runtime/yaml/yaml.h
similarity index 100%
rename from runtime/yaml.h
rename to runtime/yaml/yaml.h
diff --git a/tests/cpp/runtime/yaml-test.cpp b/tests/cpp/runtime/extensions-tests/yaml-test.cpp
similarity index 99%
rename from tests/cpp/runtime/yaml-test.cpp
rename to tests/cpp/runtime/extensions-tests/yaml-test.cpp
index 88f78030c3..7baaf3eb40 100644
--- a/tests/cpp/runtime/yaml-test.cpp
+++ b/tests/cpp/runtime/extensions-tests/yaml-test.cpp
@@ -1,7 +1,7 @@
 #include <gtest/gtest.h>
 
 #include "runtime/files.h"
-#include "runtime/yaml.h"
+#include "runtime/yaml/yaml.h"
 
 TEST(yaml_test, test_yaml_string) {
   mixed example = string("string");
diff --git a/tests/cpp/runtime/runtime-tests.cmake b/tests/cpp/runtime/runtime-tests.cmake
index daefbed534..83d7322efe 100644
--- a/tests/cpp/runtime/runtime-tests.cmake
+++ b/tests/cpp/runtime/runtime-tests.cmake
@@ -1,4 +1,10 @@
+if (YAML)
+prepend(RUNTIME_TESTS_YAML_SOURCES /extensions-tests/
+	yaml-test.cpp)
+endif()
+
 prepend(RUNTIME_TESTS_SOURCES ${BASE_DIR}/tests/cpp/runtime/
+        ${RUNTIME_TESTS_YAML_SOURCES}
         _runtime-tests-env.cpp
         allocator-malloc-replacement-test.cpp
         array-test.cpp
@@ -21,7 +27,6 @@ prepend(RUNTIME_TESTS_SOURCES ${BASE_DIR}/tests/cpp/runtime/
         memory_resource/unsynchronized_pool_resource-test.cpp
         string-list-test.cpp
         string-test.cpp
-        yaml-test.cpp
         zstd-test.cpp)
 
 allow_deprecated_declarations_for_apple(${BASE_DIR}/tests/cpp/runtime/inter-process-mutex-test.cpp)

From 361addabc52b3b102a3d0286bc3110763c0e49f1 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sat, 7 Oct 2023 13:03:04 +0300
Subject: [PATCH 21/25] yaml: minor changes

---
 runtime/runtime.cmake                            | 2 +-
 runtime/streams.h                                | 2 +-
 runtime/{yaml => }/yaml.cpp                      | 1 -
 runtime/{yaml => }/yaml.h                        | 0
 tests/cpp/runtime/extensions-tests/yaml-test.cpp | 2 +-
 5 files changed, 3 insertions(+), 4 deletions(-)
 rename runtime/{yaml => }/yaml.cpp (99%)
 rename runtime/{yaml => }/yaml.h (100%)

diff --git a/runtime/runtime.cmake b/runtime/runtime.cmake
index 521ca8b9b7..6ee237f8b2 100644
--- a/runtime/runtime.cmake
+++ b/runtime/runtime.cmake
@@ -50,7 +50,7 @@ prepend(KPHP_RUNTIME_PDO_PGSQL_SOURCES pdo/pgsql/
 endif()
 
 if (YAML)
-prepend(KPHP_RUNTIME_YAML_SOURCES /yaml/
+prepend(KPHP_RUNTIME_YAML_SOURCES /
 	yaml.cpp)
 endif()
 
diff --git a/runtime/streams.h b/runtime/streams.h
index b95321e5b5..7c59084ccd 100644
--- a/runtime/streams.h
+++ b/runtime/streams.h
@@ -6,7 +6,7 @@
 
 #include "runtime/kphp_core.h"
 
-using Stream =mixed;
+using Stream = mixed;
 
 
 constexpr int64_t STREAM_SET_BLOCKING_OPTION = 0;
diff --git a/runtime/yaml/yaml.cpp b/runtime/yaml.cpp
similarity index 99%
rename from runtime/yaml/yaml.cpp
rename to runtime/yaml.cpp
index 8407546f79..3952d8e40d 100644
--- a/runtime/yaml/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -2,7 +2,6 @@
 
 #include "runtime/optional.h"
 #include "runtime/streams.h"
-#include "runtime/critical_section.h"
 #include "yaml.h"
 
 /*
diff --git a/runtime/yaml/yaml.h b/runtime/yaml.h
similarity index 100%
rename from runtime/yaml/yaml.h
rename to runtime/yaml.h
diff --git a/tests/cpp/runtime/extensions-tests/yaml-test.cpp b/tests/cpp/runtime/extensions-tests/yaml-test.cpp
index 7baaf3eb40..88f78030c3 100644
--- a/tests/cpp/runtime/extensions-tests/yaml-test.cpp
+++ b/tests/cpp/runtime/extensions-tests/yaml-test.cpp
@@ -1,7 +1,7 @@
 #include <gtest/gtest.h>
 
 #include "runtime/files.h"
-#include "runtime/yaml/yaml.h"
+#include "runtime/yaml.h"
 
 TEST(yaml_test, test_yaml_string) {
   mixed example = string("string");

From 29797e4abf2fa18e7541133a938e50f56cb4bdd3 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Thu, 12 Oct 2023 13:16:48 +0300
Subject: [PATCH 22/25] yaml: move cpp tests to common dir, delete
 `extensions-tests` dir

---
 tests/cpp/runtime/runtime-tests.cmake                  | 2 +-
 tests/cpp/runtime/{extensions-tests => }/yaml-test.cpp | 0
 2 files changed, 1 insertion(+), 1 deletion(-)
 rename tests/cpp/runtime/{extensions-tests => }/yaml-test.cpp (100%)

diff --git a/tests/cpp/runtime/runtime-tests.cmake b/tests/cpp/runtime/runtime-tests.cmake
index 83d7322efe..cf1401fdf5 100644
--- a/tests/cpp/runtime/runtime-tests.cmake
+++ b/tests/cpp/runtime/runtime-tests.cmake
@@ -1,5 +1,5 @@
 if (YAML)
-prepend(RUNTIME_TESTS_YAML_SOURCES /extensions-tests/
+prepend(RUNTIME_TESTS_YAML_SOURCES /
 	yaml-test.cpp)
 endif()
 
diff --git a/tests/cpp/runtime/extensions-tests/yaml-test.cpp b/tests/cpp/runtime/yaml-test.cpp
similarity index 100%
rename from tests/cpp/runtime/extensions-tests/yaml-test.cpp
rename to tests/cpp/runtime/yaml-test.cpp

From 12b31da399ec8c243bcfd2dd11e4f84e9f2f94df Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Mon, 23 Oct 2023 02:18:36 +0300
Subject: [PATCH 23/25] yaml: fix issues after @unserialize review, mostly in
 implementation of char escaping

---
 builtin-functions/_functions.txt |  8 ++--
 runtime/yaml.cpp                 | 82 +++++++++++++++++---------------
 2 files changed, 48 insertions(+), 42 deletions(-)

diff --git a/builtin-functions/_functions.txt b/builtin-functions/_functions.txt
index b993b6788d..e25f7e3187 100644
--- a/builtin-functions/_functions.txt
+++ b/builtin-functions/_functions.txt
@@ -1654,7 +1654,7 @@ class DateTimeImmutable implements DateTimeInterface {
 
 function getenv(string $varname = '', bool $local_only = false): mixed;
 
-function yaml_emit_file ($filename ::: string, $data ::: mixed) ::: bool;
-function yaml_emit ($data ::: mixed) ::: string;
-function yaml_parse_file ($filename ::: string, $pos ::: int = 0) ::: mixed;
-function yaml_parse ($data ::: string, $pos ::: int = 0) ::: mixed;
+function yaml_emit_file(string $filename, mixed $data): bool;
+function yaml_emit(mixed $data): string;
+function yaml_parse_file(string $filename, int $pos = 0): mixed;
+function yaml_parse(string $data, int $pos = 0): mixed;
diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index 3952d8e40d..02fb225ef1 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -2,7 +2,7 @@
 
 #include "runtime/optional.h"
 #include "runtime/streams.h"
-#include "yaml.h"
+#include "runtime/yaml.h"
 
 /*
  * convert YAML::Node to mixed after parsing a YAML document into YAML::Node
@@ -12,7 +12,9 @@ static void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string
   if (node.IsScalar()) {
     const string string_data(node.as<std::string>().c_str());
     // check whether the primitive is put in quotes in the source YAML
-    const bool string_data_has_quotes = (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"');
+    const bool string_data_has_quotes =
+      (source[node.Mark().pos] == '"' && source[node.Mark().pos + string_data.size() + 1] == '"') ||
+      (source[node.Mark().pos] == '\'' && source[node.Mark().pos + string_data.size() + 1] == '\'');
     // if so, it is a string
     if (string_data_has_quotes) {
       data = string_data;
@@ -64,7 +66,8 @@ static string yaml_print_key(const mixed &data_key) noexcept {
   if (data_key.is_string()) {
     return data_key.as_string();
   }
-  return string(data_key.as_int()); // key can not be an array; bool and float keys are cast to int
+  // key can not be an array; bool and float keys are cast to int
+  return string(data_key.as_int());
 }
 
 /*
@@ -73,26 +76,27 @@ static string yaml_print_key(const mixed &data_key) noexcept {
 static string yaml_escape(const string &data) noexcept {
   string escaped_data;
   for (size_t i = 0; i < data.size(); ++i) {
-    if (data[i] == 10) { // line feed - code 10
-      escaped_data.push_back(92); // backslash
+    const char& current_char = data[i];
+    if (current_char == '\n') { // line feed
+      escaped_data.push_back('\\');
       escaped_data.push_back('n');
-    } else if (data[i] == 8) { // backspace - code 8
-      escaped_data.push_back(92); // backslash
+    } else if (current_char == '\b') { // backspace
+      escaped_data.push_back('\\');
       escaped_data.push_back('b');
-    } else if (data[i] == 9) { // horizontal tab - code 9
-      escaped_data.push_back(92); // backslash
+    } else if (current_char == '\t') { // horizontal tab
+      escaped_data.push_back('\\');
       escaped_data.push_back('t');
-    } else if (data[i] == 11) { // vertical tab - code 11
-      escaped_data.push_back(92); // backslash
+    } else if (current_char == '\v') { // vertical tab
+      escaped_data.push_back('\\');
       escaped_data.push_back('v');
-    } else if (data[i] == 34) { // double quotation mark - code 34
-      escaped_data.push_back(92); // backslash
-      escaped_data.push_back(34); // double quotation mark
-    } else if (data[i] == 92) { // backslash - code 92
-      escaped_data.push_back(92);
-      escaped_data.push_back(92); // double backslash
+    } else if (current_char == '\"') {
+      escaped_data.push_back('\\');
+      escaped_data.push_back('"');
+    } else if (current_char == '\\') {
+      escaped_data.push_back('\\');
+      escaped_data.push_back('\\');
     } else {
-      escaped_data.push_back(data[i]);
+      escaped_data.push_back(current_char);
     }
   }
   return escaped_data;
@@ -102,22 +106,21 @@ static string yaml_escape(const string &data) noexcept {
  * get a YAML representation of mixed in a string variable
  */
 static void mixed_to_string(const mixed &data, string &string_data, const uint8_t nesting_level = 0) noexcept {
-  string buffer;
   if (!data.is_array()) {
     if (data.is_null()) {
-      buffer.push_back('~'); // tilda is a YAML representation of NULL
+      string_data.push_back('~'); // tilda is a YAML representation of NULL
     } else if (data.is_string()) {
-      buffer.push_back('"'); // cover string entry in double quotes
-      buffer.append(yaml_escape(data.as_string())); // escape special characters
-      buffer.push_back('"');
+      string_data.push_back('"'); // cover string entry in double quotes
+      string_data.append(yaml_escape(data.as_string())); // escape special characters
+      string_data.push_back('"');
     } else if (data.is_int()) {
-      buffer.append(data.as_int());
+      string_data.append(data.as_int());
     } else if (data.is_float()) {
-      buffer.append(data.as_double());
+      string_data.append(data.as_double());
     } else if (data.is_bool()) {
-      buffer = (data.as_bool()) ? string("true") : string("false");
+      const string bool_repr = (data.as_bool()) ? string("true") : string("false");
+      string_data.append(bool_repr);
     }
-    string_data.append(buffer);
     string_data.push_back('\n');
     return;
   }
@@ -126,23 +129,26 @@ static void mixed_to_string(const mixed &data, string &string_data, const uint8_
     string_data.append("[]\n"); // an empty array is represented as [] in YAML
     return;
   }
-  const bool data_array_is_vector = data_array.is_pseudo_vector(); // check if an array has keys increasing by 1 starting from 0
+  // check if an array has keys increasing by 1 starting from 0
+  const bool data_array_is_vector = data_array.is_pseudo_vector();
   for (const auto &it : data_array) {
     const mixed &data_piece = it.get_value();
-    buffer = yaml_print_tabs(nesting_level);
+    string_data.append(yaml_print_tabs(nesting_level));
     if (data_array_is_vector) {
-      buffer.push_back('-');
+      string_data.push_back('-');
     } else {
-      buffer.append(yaml_print_key(it.get_key()));
-      buffer.push_back(':');
+      string_data.append(yaml_print_key(it.get_key()));
+      string_data.push_back(':');
     }
     if (data_piece.is_array() && !data_piece.as_array().empty()) {
-      buffer.push_back('\n'); // if an element of an array is also a non-empty array, print it on the next line
+      // if an element of an array is also a non-empty array, print it on the next line
+      string_data.push_back('\n');
     } else {
-      buffer.push_back(' '); // if an element of an array is a primitive or an empty array, print it after a space
+      // if an element of an array is a primitive or an empty array, print it after a space
+      string_data.push_back(' ');
     }
-    string_data.append(buffer);
-    mixed_to_string(data_piece, string_data, nesting_level + 1); // for entries of an array, increase nesting level
+    // for entries of an array, increase nesting level
+    mixed_to_string(data_piece, string_data, nesting_level + 1);
   }
 }
 
@@ -170,12 +176,12 @@ string f$yaml_emit(const mixed &data) {
 mixed f$yaml_parse_file(const string &filename, int pos) {
   if (filename.empty()) {
     php_warning("Filename cannot be empty");
-    return {};
+    return false;
   }
   Optional<string> data = f$file_get_contents(filename);
   if (data.is_false()) {
     php_warning("Error while reading file \"%s\"", filename.c_str());
-    return {};
+    return false;
   }
   return f$yaml_parse(data.ref(), pos);
 }

From 97474f64c6563f9fbab69834e54d8c9de5a626bf Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Sun, 29 Oct 2023 15:45:09 +0300
Subject: [PATCH 24/25] yaml: little codestyle fixes

---
 runtime/streams.h | 2 +-
 runtime/yaml.cpp  | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/runtime/streams.h b/runtime/streams.h
index 7c59084ccd..b95321e5b5 100644
--- a/runtime/streams.h
+++ b/runtime/streams.h
@@ -6,7 +6,7 @@
 
 #include "runtime/kphp_core.h"
 
-using Stream = mixed;
+using Stream =mixed;
 
 
 constexpr int64_t STREAM_SET_BLOCKING_OPTION = 0;
diff --git a/runtime/yaml.cpp b/runtime/yaml.cpp
index 02fb225ef1..0d63378250 100644
--- a/runtime/yaml.cpp
+++ b/runtime/yaml.cpp
@@ -55,7 +55,7 @@ static void yaml_node_to_mixed(const YAML::Node &node, mixed &data, const string
 /*
  * print tabs in quantity of nesting_level (used to print nested YAML entries)
  */
-static string yaml_print_tabs(const uint8_t nesting_level) noexcept {
+static string yaml_print_tabs(const int nesting_level) noexcept {
   return string(2 * nesting_level, ' ');
 }
 
@@ -105,7 +105,7 @@ static string yaml_escape(const string &data) noexcept {
 /*
  * get a YAML representation of mixed in a string variable
  */
-static void mixed_to_string(const mixed &data, string &string_data, const uint8_t nesting_level = 0) noexcept {
+static void mixed_to_string(const mixed &data, string &string_data, const int nesting_level = 0) noexcept {
   if (!data.is_array()) {
     if (data.is_null()) {
       string_data.push_back('~'); // tilda is a YAML representation of NULL

From c07c1aa05a4f1423abdadaecd5c6628a924111a5 Mon Sep 17 00:00:00 2001
From: Nikolay Osintsev <niosintsev@miem.hse.ru>
Date: Wed, 15 Nov 2023 00:42:26 +0300
Subject: [PATCH 25/25] yaml: add test for strings in single quotes

---
 tests/phpt/yaml/yaml_escaping.yml | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/tests/phpt/yaml/yaml_escaping.yml b/tests/phpt/yaml/yaml_escaping.yml
index 54e5ac4d05..486810f595 100644
--- a/tests/phpt/yaml/yaml_escaping.yml
+++ b/tests/phpt/yaml/yaml_escaping.yml
@@ -1,5 +1,8 @@
 ---
-quote: "'single'quotes'are'not'escaped'double\"quotes\"are\"escaped\""
-newline: "1\n2\n3\n4\n5\nnewlines"
-other: "backslash\\horizontal tab\tvertical tab\vevery\b last\b symbol\b is\b erased\b"
+string_in_double_quotes: "'single'quotes'are'not'escaped'double\"quotes\"are\"escaped\""
+string_in_single_quotes: 'string inside single quotes'
+string_without_quotes: this is a string
+number_in_single_quotes_i_string: '12345'
+newline_symbols: "1\n2\n3\n4\n5\nnewlines"
+string_with_escaped_symbols: "backslash\\horizontal tab\tvertical tab\vevery\b last\b symbol\b is\b erased\b"
 ...