Use std::uint64_t for row index. (#10120)

- Use std::uint64_t instead of size_t to avoid implementation-defined type.
- Rename to bst_idx_t, to account for other types of indexing.
- Small cleanup to the base header.

include/xgboost/base.h

@@ -1,20 +1,18 @@
 /**
- * Copyright 2015-2023 by XGBoost Contributors
+ * Copyright 2015-2024, XGBoost Contributors
  * \file base.h
  * \brief Defines configuration macros and basic types for xgboost.
  */
 #ifndef XGBOOST_BASE_H_
 #define XGBOOST_BASE_H_
 
-#include <dmlc/base.h>
-#include <dmlc/omp.h>
+#include <dmlc/omp.h>  // for omp_uint, omp_ulong
 
-#include <cmath>
-#include <cstdint>
-#include <iostream>
-#include <string>
-#include <utility>
-#include <vector>
+#include <cstdint>  // for int32_t, uint64_t, int16_t
+#include <ostream>  // for ostream
+#include <string>   // for string
+#include <utility>  // for pair
+#include <vector>   // for vector
 
 /*!
  * \brief string flag for R library, to leave hooks when needed.
@@ -86,34 +84,31 @@
 
 #endif  // !defined(XGBOOST_MM_PREFETCH_PRESENT) && !defined()
 
-/*! \brief namespace of xgboost*/
 namespace xgboost {
-
 /*! \brief unsigned integer type used for feature index. */
-using bst_uint = uint32_t;  // NOLINT
+using bst_uint = std::uint32_t;  // NOLINT
 /*! \brief unsigned long integers */
-using bst_ulong = uint64_t;  // NOLINT
+using bst_ulong = std::uint64_t;  // NOLINT
 /*! \brief float type, used for storing statistics */
 using bst_float = float;  // NOLINT
 /*! \brief Categorical value type. */
-using bst_cat_t = int32_t;  // NOLINT
+using bst_cat_t = std::int32_t;  // NOLINT
 /*! \brief Type for data column (feature) index. */
-using bst_feature_t = uint32_t;  // NOLINT
-/*! \brief Type for histogram bin index. */
-using bst_bin_t = int32_t;  // NOLINT
-/*! \brief Type for data row index.
- *
- * Be careful `std::size_t' is implementation-defined.  Meaning that the binary
- * representation of DMatrix might not be portable across platform.  Booster model should
- * be portable as parameters are floating points.
+using bst_feature_t = std::uint32_t;  // NOLINT
+/**
+ * @brief Type for histogram bin index.  We sometimes use -1 to indicate invalid bin.
  */
-using bst_row_t = std::size_t;   // NOLINT
+using bst_bin_t = std::int32_t;  // NOLINT
+/**
+ * @brief Type for data row index (sample).
+ */
+using bst_idx_t = std::uint64_t;  // NOLINT
 /*! \brief Type for tree node index. */
 using bst_node_t = std::int32_t;      // NOLINT
 /*! \brief Type for ranking group index. */
 using bst_group_t = std::uint32_t;  // NOLINT
 /**
- * \brief Type for indexing into output targets.
+ * @brief Type for indexing into output targets.
  */
 using bst_target_t = std::uint32_t;  // NOLINT
 /**
@@ -306,8 +301,7 @@ class GradientPairInt64 {
   XGBOOST_DEVICE bool operator==(const GradientPairInt64 &rhs) const {
     return grad_ == rhs.grad_ && hess_ == rhs.hess_;
   }
-  friend std::ostream &operator<<(std::ostream &os,
-                                  const GradientPairInt64 &g) {
+  friend std::ostream &operator<<(std::ostream &os, const GradientPairInt64 &g) {
     os << g.GetQuantisedGrad() << "/" << g.GetQuantisedHess();
     return os;
   }
@@ -323,7 +317,7 @@ using omp_ulong = dmlc::omp_ulong;  // NOLINT
 /*! \brief define unsigned int for openmp loop */
 using bst_omp_uint = dmlc::omp_uint;  // NOLINT
 /*! \brief Type used for representing version number in binary form.*/
-using XGBoostVersionT = int32_t;
+using XGBoostVersionT = std::int32_t;
 }  // namespace xgboost
 
 #endif  // XGBOOST_BASE_H_

include/xgboost/data.h

@@ -315,7 +315,7 @@ struct BatchParam {
 struct HostSparsePageView {
   using Inst = common::Span<Entry const>;
 
-  common::Span<bst_row_t const> offset;
+  common::Span<bst_idx_t const> offset;
   common::Span<Entry const> data;
 
   Inst operator[](size_t i) const {
@@ -333,7 +333,7 @@ struct HostSparsePageView {
 class SparsePage {
  public:
   // Offset for each row.
-  HostDeviceVector<bst_row_t> offset;
+  HostDeviceVector<bst_idx_t> offset;
   /*! \brief the data of the segments */
   HostDeviceVector<Entry> data;
 

include/xgboost/json.h

@@ -60,9 +60,7 @@ class Value {
   virtual Json& operator[](int ind);
 
   virtual bool operator==(Value const& rhs) const = 0;
-#if !defined(__APPLE__)
   virtual Value& operator=(Value const& rhs) = delete;
-#endif  // !defined(__APPLE__)
 
   std::string TypeStr() const;
 
@@ -105,6 +103,7 @@ class JsonString : public Value {
   std::string&       GetString()       & { return str_; }
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   static bool IsClassOf(Value const* value) {
     return value->Type() == ValueKind::kString;
@@ -134,6 +133,7 @@ class JsonArray : public Value {
   std::vector<Json>&       GetArray()       & { return vec_; }
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   static bool IsClassOf(Value const* value) {
     return value->Type() == ValueKind::kArray;
@@ -158,6 +158,7 @@ class JsonTypedArray : public Value {
   JsonTypedArray(JsonTypedArray&& that) noexcept : Value{kind}, vec_{std::move(that.vec_)} {}
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   void Set(size_t i, T v) { vec_[i] = v; }
   size_t Size() const { return vec_.size(); }
@@ -216,6 +217,7 @@ class JsonObject : public Value {
   Map& GetObject() & { return object_; }
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   static bool IsClassOf(Value const* value) { return value->Type() == ValueKind::kObject; }
   ~JsonObject() override = default;
@@ -249,6 +251,7 @@ class JsonNumber : public Value {
   Float&       GetNumber()       & { return number_; }
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   static bool IsClassOf(Value const* value) {
     return value->Type() == ValueKind::kNumber;
@@ -287,6 +290,7 @@ class JsonInteger : public Value {
       : Value{ValueKind::kInteger}, integer_{that.integer_} {}
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   Int const& GetInteger() &&      { return integer_; }
   Int const& GetInteger() const & { return integer_; }
@@ -307,6 +311,7 @@ class JsonNull : public Value {
   void Save(JsonWriter* writer) const override;
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   static bool IsClassOf(Value const* value) {
     return value->Type() == ValueKind::kNull;
@@ -336,6 +341,7 @@ class JsonBoolean : public Value {
   bool&       GetBoolean()       & { return boolean_; }
 
   bool operator==(Value const& rhs) const override;
+  Value& operator=(Value const& rhs) override = delete;
 
   static bool IsClassOf(Value const* value) {
     return value->Type() == ValueKind::kBoolean;