This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a commit to branch gh-pages
in repository https://gitbox.apache.org/repos/asf/iceberg-rust.git
The following commit(s) were added to refs/heads/gh-pages by this push:
new 9f26d349 deploy: ec545f2a9f50dcbe694b2196e04b32de24f7a411
9f26d349 is described below
commit 9f26d3498cac4c250a3997c60f43f7779559a8a0
Author: liurenjie1024 <[email protected]>
AuthorDate: Tue Oct 14 10:14:56 2025 +0000
deploy: ec545f2a9f50dcbe694b2196e04b32de24f7a411
---
.../physical_plan/commit.rs.html | 1113 ++++++++++----------
1 file changed, 559 insertions(+), 554 deletions(-)
diff --git a/api/src/iceberg_datafusion/physical_plan/commit.rs.html
b/api/src/iceberg_datafusion/physical_plan/commit.rs.html
index 05c5435f..391df9e5 100644
--- a/api/src/iceberg_datafusion/physical_plan/commit.rs.html
+++ b/api/src/iceberg_datafusion/physical_plan/commit.rs.html
@@ -57,566 +57,571 @@
<a href=#57 id=57 data-nosnippet>57</a> input: Arc<<span
class="kw">dyn </span>ExecutionPlan>,
<a href=#58 id=58 data-nosnippet>58</a> schema: ArrowSchemaRef,
<a href=#59 id=59 data-nosnippet>59</a> ) -> <span class="self">Self
</span>{
-<a href=#60 id=60 data-nosnippet>60</a> <span class="kw">let
</span>plan_properties = <span
class="self">Self</span>::compute_properties(schema.clone());
+<a href=#60 id=60 data-nosnippet>60</a> <span class="kw">let
</span>count_schema = <span class="self">Self</span>::make_count_schema();
<a href=#61 id=61 data-nosnippet>61</a>
-<a href=#62 id=62 data-nosnippet>62</a> <span class="self">Self </span>{
-<a href=#63 id=63 data-nosnippet>63</a> table,
-<a href=#64 id=64 data-nosnippet>64</a> catalog,
-<a href=#65 id=65 data-nosnippet>65</a> input,
-<a href=#66 id=66 data-nosnippet>66</a> schema,
-<a href=#67 id=67 data-nosnippet>67</a> count_schema: <span
class="self">Self</span>::make_count_schema(),
-<a href=#68 id=68 data-nosnippet>68</a> plan_properties,
-<a href=#69 id=69 data-nosnippet>69</a> }
-<a href=#70 id=70 data-nosnippet>70</a> }
-<a href=#71 id=71 data-nosnippet>71</a>
-<a href=#72 id=72 data-nosnippet>72</a> <span class="comment">// Compute
the plan properties for this execution plan
-<a href=#73 id=73 data-nosnippet>73</a> </span><span class="kw">fn
</span>compute_properties(schema: ArrowSchemaRef) -> PlanProperties {
-<a href=#74 id=74 data-nosnippet>74</a> PlanProperties::new(
-<a href=#75 id=75 data-nosnippet>75</a>
EquivalenceProperties::new(schema),
-<a href=#76 id=76 data-nosnippet>76</a>
Partitioning::UnknownPartitioning(<span class="number">1</span>),
-<a href=#77 id=77 data-nosnippet>77</a> EmissionType::Final,
-<a href=#78 id=78 data-nosnippet>78</a> Boundedness::Bounded,
-<a href=#79 id=79 data-nosnippet>79</a> )
-<a href=#80 id=80 data-nosnippet>80</a> }
-<a href=#81 id=81 data-nosnippet>81</a>
-<a href=#82 id=82 data-nosnippet>82</a> <span class="comment">// Create a
record batch with just the count of rows written
-<a href=#83 id=83 data-nosnippet>83</a> </span><span class="kw">fn
</span>make_count_batch(count: u64) -> DFResult<RecordBatch> {
-<a href=#84 id=84 data-nosnippet>84</a> <span class="kw">let
</span>count_array = Arc::new(UInt64Array::from(<span
class="macro">vec!</span>[count])) <span class="kw">as </span>ArrayRef;
-<a href=#85 id=85 data-nosnippet>85</a>
-<a href=#86 id=86 data-nosnippet>86</a>
RecordBatch::try_from_iter_with_nullable(<span class="macro">vec!</span>[(<span
class="string">"count"</span>, count_array, <span
class="bool-val">false</span>)]).map_err(|e| {
-<a href=#87 id=87 data-nosnippet>87</a>
DataFusionError::ArrowError(e, <span class="prelude-val">Some</span>(<span
class="string">"Failed to make count batch!"</span>.to_string()))
-<a href=#88 id=88 data-nosnippet>88</a> })
-<a href=#89 id=89 data-nosnippet>89</a> }
-<a href=#90 id=90 data-nosnippet>90</a>
-<a href=#91 id=91 data-nosnippet>91</a> <span class="kw">fn
</span>make_count_schema() -> ArrowSchemaRef {
-<a href=#92 id=92 data-nosnippet>92</a> <span class="comment">// Define
a schema.
-<a href=#93 id=93 data-nosnippet>93</a>
</span>Arc::new(ArrowSchema::new(<span class="macro">vec!</span>[Field::new(
-<a href=#94 id=94 data-nosnippet>94</a> <span
class="string">"count"</span>,
-<a href=#95 id=95 data-nosnippet>95</a> DataType::UInt64,
-<a href=#96 id=96 data-nosnippet>96</a> <span
class="bool-val">false</span>,
-<a href=#97 id=97 data-nosnippet>97</a> )]))
-<a href=#98 id=98 data-nosnippet>98</a> }
-<a href=#99 id=99 data-nosnippet>99</a>}
-<a href=#100 id=100 data-nosnippet>100</a>
-<a href=#101 id=101 data-nosnippet>101</a><span class="kw">impl
</span>DisplayAs <span class="kw">for </span>IcebergCommitExec {
-<a href=#102 id=102 data-nosnippet>102</a> <span class="kw">fn
</span>fmt_as(<span class="kw-2">&</span><span class="self">self</span>, t:
DisplayFormatType, f: <span class="kw-2">&mut </span>Formatter) ->
std::fmt::Result {
-<a href=#103 id=103 data-nosnippet>103</a> <span class="kw">match
</span>t {
-<a href=#104 id=104 data-nosnippet>104</a>
DisplayFormatType::Default => {
-<a href=#105 id=105 data-nosnippet>105</a> <span
class="macro">write!</span>(f, <span class="string">"IcebergCommitExec:
table={}"</span>, <span class="self">self</span>.table.identifier())
-<a href=#106 id=106 data-nosnippet>106</a> }
-<a href=#107 id=107 data-nosnippet>107</a>
DisplayFormatType::Verbose => {
-<a href=#108 id=108 data-nosnippet>108</a> <span
class="macro">write!</span>(
-<a href=#109 id=109 data-nosnippet>109</a> f,
-<a href=#110 id=110 data-nosnippet>110</a> <span
class="string">"IcebergCommitExec: table={}, schema={:?}"</span>,
-<a href=#111 id=111 data-nosnippet>111</a> <span
class="self">self</span>.table.identifier(),
-<a href=#112 id=112 data-nosnippet>112</a> <span
class="self">self</span>.schema
-<a href=#113 id=113 data-nosnippet>113</a> )
-<a href=#114 id=114 data-nosnippet>114</a> }
-<a href=#115 id=115 data-nosnippet>115</a>
DisplayFormatType::TreeRender => {
-<a href=#116 id=116 data-nosnippet>116</a> <span
class="macro">write!</span>(f, <span class="string">"IcebergCommitExec:
table={}"</span>, <span class="self">self</span>.table.identifier())
-<a href=#117 id=117 data-nosnippet>117</a> }
-<a href=#118 id=118 data-nosnippet>118</a> }
-<a href=#119 id=119 data-nosnippet>119</a> }
-<a href=#120 id=120 data-nosnippet>120</a>}
-<a href=#121 id=121 data-nosnippet>121</a>
-<a href=#122 id=122 data-nosnippet>122</a><span class="kw">impl
</span>ExecutionPlan <span class="kw">for </span>IcebergCommitExec {
-<a href=#123 id=123 data-nosnippet>123</a> <span class="kw">fn
</span>name(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span>str {
-<a href=#124 id=124 data-nosnippet>124</a> <span
class="string">"IcebergCommitExec"
-<a href=#125 id=125 data-nosnippet>125</a> </span>}
-<a href=#126 id=126 data-nosnippet>126</a>
-<a href=#127 id=127 data-nosnippet>127</a> <span class="kw">fn
</span>as_any(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span><span class="kw">dyn </span>Any {
-<a href=#128 id=128 data-nosnippet>128</a> <span class="self">self
-<a href=#129 id=129 data-nosnippet>129</a> </span>}
-<a href=#130 id=130 data-nosnippet>130</a>
-<a href=#131 id=131 data-nosnippet>131</a> <span class="kw">fn
</span>properties(<span class="kw-2">&</span><span
class="self">self</span>) -> <span class="kw-2">&</span>PlanProperties {
-<a href=#132 id=132 data-nosnippet>132</a> <span
class="kw-2">&</span><span class="self">self</span>.plan_properties
-<a href=#133 id=133 data-nosnippet>133</a> }
-<a href=#134 id=134 data-nosnippet>134</a>
-<a href=#135 id=135 data-nosnippet>135</a> <span class="kw">fn
</span>children(<span class="kw-2">&</span><span class="self">self</span>)
-> Vec<<span class="kw-2">&</span>Arc<<span class="kw">dyn
</span>ExecutionPlan>> {
-<a href=#136 id=136 data-nosnippet>136</a> <span
class="macro">vec!</span>[<span class="kw-2">&</span><span
class="self">self</span>.input]
-<a href=#137 id=137 data-nosnippet>137</a> }
-<a href=#138 id=138 data-nosnippet>138</a>
-<a href=#139 id=139 data-nosnippet>139</a> <span class="kw">fn
</span>required_input_distribution(<span class="kw-2">&</span><span
class="self">self</span>) ->
Vec<datafusion::physical_plan::Distribution> {
-<a href=#140 id=140 data-nosnippet>140</a> <span
class="macro">vec!</span>[datafusion::physical_plan::Distribution::SinglePartition;
<span class="self">self</span>.children().len()]
-<a href=#141 id=141 data-nosnippet>141</a> }
-<a href=#142 id=142 data-nosnippet>142</a>
-<a href=#143 id=143 data-nosnippet>143</a> <span class="kw">fn
</span>benefits_from_input_partitioning(<span class="kw-2">&</span><span
class="self">self</span>) -> Vec<bool> {
-<a href=#144 id=144 data-nosnippet>144</a> <span
class="macro">vec!</span>[<span class="bool-val">false</span>]
-<a href=#145 id=145 data-nosnippet>145</a> }
-<a href=#146 id=146 data-nosnippet>146</a>
-<a href=#147 id=147 data-nosnippet>147</a> <span class="kw">fn
</span>with_new_children(
-<a href=#148 id=148 data-nosnippet>148</a> <span
class="self">self</span>: Arc<<span class="self">Self</span>>,
-<a href=#149 id=149 data-nosnippet>149</a> children:
Vec<Arc<<span class="kw">dyn </span>ExecutionPlan>>,
-<a href=#150 id=150 data-nosnippet>150</a> ) -> DFResult<Arc<<span
class="kw">dyn </span>ExecutionPlan>> {
-<a href=#151 id=151 data-nosnippet>151</a> <span class="kw">if
</span>children.len() != <span class="number">1 </span>{
-<a href=#152 id=152 data-nosnippet>152</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(DataFusionError::Internal(<span
class="macro">format!</span>(
-<a href=#153 id=153 data-nosnippet>153</a> <span
class="string">"IcebergCommitExec expects exactly one child, but provided
{}"</span>,
-<a href=#154 id=154 data-nosnippet>154</a> children.len()
-<a href=#155 id=155 data-nosnippet>155</a> )));
-<a href=#156 id=156 data-nosnippet>156</a> }
-<a href=#157 id=157 data-nosnippet>157</a>
-<a href=#158 id=158 data-nosnippet>158</a> <span
class="prelude-val">Ok</span>(Arc::new(IcebergCommitExec::new(
-<a href=#159 id=159 data-nosnippet>159</a> <span
class="self">self</span>.table.clone(),
-<a href=#160 id=160 data-nosnippet>160</a> <span
class="self">self</span>.catalog.clone(),
-<a href=#161 id=161 data-nosnippet>161</a> children[<span
class="number">0</span>].clone(),
-<a href=#162 id=162 data-nosnippet>162</a> <span
class="self">self</span>.schema.clone(),
-<a href=#163 id=163 data-nosnippet>163</a> )))
-<a href=#164 id=164 data-nosnippet>164</a> }
-<a href=#165 id=165 data-nosnippet>165</a>
-<a href=#166 id=166 data-nosnippet>166</a> <span class="kw">fn
</span>execute(
-<a href=#167 id=167 data-nosnippet>167</a> <span
class="kw-2">&</span><span class="self">self</span>,
-<a href=#168 id=168 data-nosnippet>168</a> partition: usize,
-<a href=#169 id=169 data-nosnippet>169</a> context:
Arc<TaskContext>,
-<a href=#170 id=170 data-nosnippet>170</a> ) ->
DFResult<SendableRecordBatchStream> {
-<a href=#171 id=171 data-nosnippet>171</a> <span class="comment">//
IcebergCommitExec only has one partition (partition 0)
-<a href=#172 id=172 data-nosnippet>172</a> </span><span class="kw">if
</span>partition != <span class="number">0 </span>{
-<a href=#173 id=173 data-nosnippet>173</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(DataFusionError::Internal(<span
class="macro">format!</span>(
-<a href=#174 id=174 data-nosnippet>174</a> <span
class="string">"IcebergCommitExec only has one partition, but got partition
{}"</span>,
-<a href=#175 id=175 data-nosnippet>175</a> partition
-<a href=#176 id=176 data-nosnippet>176</a> )));
-<a href=#177 id=177 data-nosnippet>177</a> }
-<a href=#178 id=178 data-nosnippet>178</a>
-<a href=#179 id=179 data-nosnippet>179</a> <span class="kw">let
</span>table = <span class="self">self</span>.table.clone();
-<a href=#180 id=180 data-nosnippet>180</a> <span class="kw">let
</span>input_plan = <span class="self">self</span>.input.clone();
-<a href=#181 id=181 data-nosnippet>181</a> <span class="kw">let
</span>count_schema = Arc::clone(<span class="kw-2">&</span><span
class="self">self</span>.count_schema);
-<a href=#182 id=182 data-nosnippet>182</a>
-<a href=#183 id=183 data-nosnippet>183</a> <span class="comment">//
todo revisit this
-<a href=#184 id=184 data-nosnippet>184</a> </span><span class="kw">let
</span>spec_id = <span
class="self">self</span>.table.metadata().default_partition_spec_id();
-<a href=#185 id=185 data-nosnippet>185</a> <span class="kw">let
</span>partition_type = <span
class="self">self</span>.table.metadata().default_partition_type().clone();
-<a href=#186 id=186 data-nosnippet>186</a> <span class="kw">let
</span>current_schema = <span
class="self">self</span>.table.metadata().current_schema().clone();
-<a href=#187 id=187 data-nosnippet>187</a>
-<a href=#188 id=188 data-nosnippet>188</a> <span class="kw">let
</span>catalog = Arc::clone(<span class="kw-2">&</span><span
class="self">self</span>.catalog);
+<a href=#62 id=62 data-nosnippet>62</a> <span class="kw">let
</span>plan_properties = <span
class="self">Self</span>::compute_properties(Arc::clone(<span
class="kw-2">&</span>count_schema));
+<a href=#63 id=63 data-nosnippet>63</a>
+<a href=#64 id=64 data-nosnippet>64</a> <span class="self">Self </span>{
+<a href=#65 id=65 data-nosnippet>65</a> table,
+<a href=#66 id=66 data-nosnippet>66</a> catalog,
+<a href=#67 id=67 data-nosnippet>67</a> input,
+<a href=#68 id=68 data-nosnippet>68</a> schema,
+<a href=#69 id=69 data-nosnippet>69</a> count_schema,
+<a href=#70 id=70 data-nosnippet>70</a> plan_properties,
+<a href=#71 id=71 data-nosnippet>71</a> }
+<a href=#72 id=72 data-nosnippet>72</a> }
+<a href=#73 id=73 data-nosnippet>73</a>
+<a href=#74 id=74 data-nosnippet>74</a> <span class="comment">// Compute
the plan properties for this execution plan
+<a href=#75 id=75 data-nosnippet>75</a> </span><span class="kw">fn
</span>compute_properties(schema: ArrowSchemaRef) -> PlanProperties {
+<a href=#76 id=76 data-nosnippet>76</a> PlanProperties::new(
+<a href=#77 id=77 data-nosnippet>77</a>
EquivalenceProperties::new(schema),
+<a href=#78 id=78 data-nosnippet>78</a>
Partitioning::UnknownPartitioning(<span class="number">1</span>),
+<a href=#79 id=79 data-nosnippet>79</a> EmissionType::Final,
+<a href=#80 id=80 data-nosnippet>80</a> Boundedness::Bounded,
+<a href=#81 id=81 data-nosnippet>81</a> )
+<a href=#82 id=82 data-nosnippet>82</a> }
+<a href=#83 id=83 data-nosnippet>83</a>
+<a href=#84 id=84 data-nosnippet>84</a> <span class="comment">// Create a
record batch with just the count of rows written
+<a href=#85 id=85 data-nosnippet>85</a> </span><span class="kw">fn
</span>make_count_batch(count: u64) -> DFResult<RecordBatch> {
+<a href=#86 id=86 data-nosnippet>86</a> <span class="kw">let
</span>count_array = Arc::new(UInt64Array::from(<span
class="macro">vec!</span>[count])) <span class="kw">as </span>ArrayRef;
+<a href=#87 id=87 data-nosnippet>87</a>
+<a href=#88 id=88 data-nosnippet>88</a>
RecordBatch::try_from_iter_with_nullable(<span class="macro">vec!</span>[(<span
class="string">"count"</span>, count_array, <span
class="bool-val">false</span>)]).map_err(|e| {
+<a href=#89 id=89 data-nosnippet>89</a>
DataFusionError::ArrowError(e, <span class="prelude-val">Some</span>(<span
class="string">"Failed to make count batch!"</span>.to_string()))
+<a href=#90 id=90 data-nosnippet>90</a> })
+<a href=#91 id=91 data-nosnippet>91</a> }
+<a href=#92 id=92 data-nosnippet>92</a>
+<a href=#93 id=93 data-nosnippet>93</a> <span class="kw">fn
</span>make_count_schema() -> ArrowSchemaRef {
+<a href=#94 id=94 data-nosnippet>94</a> <span class="comment">// Define
a schema.
+<a href=#95 id=95 data-nosnippet>95</a>
</span>Arc::new(ArrowSchema::new(<span class="macro">vec!</span>[Field::new(
+<a href=#96 id=96 data-nosnippet>96</a> <span
class="string">"count"</span>,
+<a href=#97 id=97 data-nosnippet>97</a> DataType::UInt64,
+<a href=#98 id=98 data-nosnippet>98</a> <span
class="bool-val">false</span>,
+<a href=#99 id=99 data-nosnippet>99</a> )]))
+<a href=#100 id=100 data-nosnippet>100</a> }
+<a href=#101 id=101 data-nosnippet>101</a>}
+<a href=#102 id=102 data-nosnippet>102</a>
+<a href=#103 id=103 data-nosnippet>103</a><span class="kw">impl
</span>DisplayAs <span class="kw">for </span>IcebergCommitExec {
+<a href=#104 id=104 data-nosnippet>104</a> <span class="kw">fn
</span>fmt_as(<span class="kw-2">&</span><span class="self">self</span>, t:
DisplayFormatType, f: <span class="kw-2">&mut </span>Formatter) ->
std::fmt::Result {
+<a href=#105 id=105 data-nosnippet>105</a> <span class="kw">match
</span>t {
+<a href=#106 id=106 data-nosnippet>106</a>
DisplayFormatType::Default => {
+<a href=#107 id=107 data-nosnippet>107</a> <span
class="macro">write!</span>(f, <span class="string">"IcebergCommitExec:
table={}"</span>, <span class="self">self</span>.table.identifier())
+<a href=#108 id=108 data-nosnippet>108</a> }
+<a href=#109 id=109 data-nosnippet>109</a>
DisplayFormatType::Verbose => {
+<a href=#110 id=110 data-nosnippet>110</a> <span
class="macro">write!</span>(
+<a href=#111 id=111 data-nosnippet>111</a> f,
+<a href=#112 id=112 data-nosnippet>112</a> <span
class="string">"IcebergCommitExec: table={}, schema={:?}"</span>,
+<a href=#113 id=113 data-nosnippet>113</a> <span
class="self">self</span>.table.identifier(),
+<a href=#114 id=114 data-nosnippet>114</a> <span
class="self">self</span>.schema
+<a href=#115 id=115 data-nosnippet>115</a> )
+<a href=#116 id=116 data-nosnippet>116</a> }
+<a href=#117 id=117 data-nosnippet>117</a>
DisplayFormatType::TreeRender => {
+<a href=#118 id=118 data-nosnippet>118</a> <span
class="macro">write!</span>(f, <span class="string">"IcebergCommitExec:
table={}"</span>, <span class="self">self</span>.table.identifier())
+<a href=#119 id=119 data-nosnippet>119</a> }
+<a href=#120 id=120 data-nosnippet>120</a> }
+<a href=#121 id=121 data-nosnippet>121</a> }
+<a href=#122 id=122 data-nosnippet>122</a>}
+<a href=#123 id=123 data-nosnippet>123</a>
+<a href=#124 id=124 data-nosnippet>124</a><span class="kw">impl
</span>ExecutionPlan <span class="kw">for </span>IcebergCommitExec {
+<a href=#125 id=125 data-nosnippet>125</a> <span class="kw">fn
</span>name(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span>str {
+<a href=#126 id=126 data-nosnippet>126</a> <span
class="string">"IcebergCommitExec"
+<a href=#127 id=127 data-nosnippet>127</a> </span>}
+<a href=#128 id=128 data-nosnippet>128</a>
+<a href=#129 id=129 data-nosnippet>129</a> <span class="kw">fn
</span>as_any(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span><span class="kw">dyn </span>Any {
+<a href=#130 id=130 data-nosnippet>130</a> <span class="self">self
+<a href=#131 id=131 data-nosnippet>131</a> </span>}
+<a href=#132 id=132 data-nosnippet>132</a>
+<a href=#133 id=133 data-nosnippet>133</a> <span class="kw">fn
</span>properties(<span class="kw-2">&</span><span
class="self">self</span>) -> <span class="kw-2">&</span>PlanProperties {
+<a href=#134 id=134 data-nosnippet>134</a> <span
class="kw-2">&</span><span class="self">self</span>.plan_properties
+<a href=#135 id=135 data-nosnippet>135</a> }
+<a href=#136 id=136 data-nosnippet>136</a>
+<a href=#137 id=137 data-nosnippet>137</a> <span class="kw">fn
</span>children(<span class="kw-2">&</span><span class="self">self</span>)
-> Vec<<span class="kw-2">&</span>Arc<<span class="kw">dyn
</span>ExecutionPlan>> {
+<a href=#138 id=138 data-nosnippet>138</a> <span
class="macro">vec!</span>[<span class="kw-2">&</span><span
class="self">self</span>.input]
+<a href=#139 id=139 data-nosnippet>139</a> }
+<a href=#140 id=140 data-nosnippet>140</a>
+<a href=#141 id=141 data-nosnippet>141</a> <span class="kw">fn
</span>required_input_distribution(<span class="kw-2">&</span><span
class="self">self</span>) ->
Vec<datafusion::physical_plan::Distribution> {
+<a href=#142 id=142 data-nosnippet>142</a> <span
class="macro">vec!</span>[datafusion::physical_plan::Distribution::SinglePartition;
<span class="self">self</span>.children().len()]
+<a href=#143 id=143 data-nosnippet>143</a> }
+<a href=#144 id=144 data-nosnippet>144</a>
+<a href=#145 id=145 data-nosnippet>145</a> <span class="kw">fn
</span>benefits_from_input_partitioning(<span class="kw-2">&</span><span
class="self">self</span>) -> Vec<bool> {
+<a href=#146 id=146 data-nosnippet>146</a> <span
class="macro">vec!</span>[<span class="bool-val">false</span>]
+<a href=#147 id=147 data-nosnippet>147</a> }
+<a href=#148 id=148 data-nosnippet>148</a>
+<a href=#149 id=149 data-nosnippet>149</a> <span class="kw">fn
</span>with_new_children(
+<a href=#150 id=150 data-nosnippet>150</a> <span
class="self">self</span>: Arc<<span class="self">Self</span>>,
+<a href=#151 id=151 data-nosnippet>151</a> children:
Vec<Arc<<span class="kw">dyn </span>ExecutionPlan>>,
+<a href=#152 id=152 data-nosnippet>152</a> ) -> DFResult<Arc<<span
class="kw">dyn </span>ExecutionPlan>> {
+<a href=#153 id=153 data-nosnippet>153</a> <span class="kw">if
</span>children.len() != <span class="number">1 </span>{
+<a href=#154 id=154 data-nosnippet>154</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(DataFusionError::Internal(<span
class="macro">format!</span>(
+<a href=#155 id=155 data-nosnippet>155</a> <span
class="string">"IcebergCommitExec expects exactly one child, but provided
{}"</span>,
+<a href=#156 id=156 data-nosnippet>156</a> children.len()
+<a href=#157 id=157 data-nosnippet>157</a> )));
+<a href=#158 id=158 data-nosnippet>158</a> }
+<a href=#159 id=159 data-nosnippet>159</a>
+<a href=#160 id=160 data-nosnippet>160</a> <span
class="prelude-val">Ok</span>(Arc::new(IcebergCommitExec::new(
+<a href=#161 id=161 data-nosnippet>161</a> <span
class="self">self</span>.table.clone(),
+<a href=#162 id=162 data-nosnippet>162</a> <span
class="self">self</span>.catalog.clone(),
+<a href=#163 id=163 data-nosnippet>163</a> children[<span
class="number">0</span>].clone(),
+<a href=#164 id=164 data-nosnippet>164</a> <span
class="self">self</span>.schema.clone(),
+<a href=#165 id=165 data-nosnippet>165</a> )))
+<a href=#166 id=166 data-nosnippet>166</a> }
+<a href=#167 id=167 data-nosnippet>167</a>
+<a href=#168 id=168 data-nosnippet>168</a> <span class="kw">fn
</span>execute(
+<a href=#169 id=169 data-nosnippet>169</a> <span
class="kw-2">&</span><span class="self">self</span>,
+<a href=#170 id=170 data-nosnippet>170</a> partition: usize,
+<a href=#171 id=171 data-nosnippet>171</a> context:
Arc<TaskContext>,
+<a href=#172 id=172 data-nosnippet>172</a> ) ->
DFResult<SendableRecordBatchStream> {
+<a href=#173 id=173 data-nosnippet>173</a> <span class="comment">//
IcebergCommitExec only has one partition (partition 0)
+<a href=#174 id=174 data-nosnippet>174</a> </span><span class="kw">if
</span>partition != <span class="number">0 </span>{
+<a href=#175 id=175 data-nosnippet>175</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(DataFusionError::Internal(<span
class="macro">format!</span>(
+<a href=#176 id=176 data-nosnippet>176</a> <span
class="string">"IcebergCommitExec only has one partition, but got partition
{}"</span>,
+<a href=#177 id=177 data-nosnippet>177</a> partition
+<a href=#178 id=178 data-nosnippet>178</a> )));
+<a href=#179 id=179 data-nosnippet>179</a> }
+<a href=#180 id=180 data-nosnippet>180</a>
+<a href=#181 id=181 data-nosnippet>181</a> <span class="kw">let
</span>table = <span class="self">self</span>.table.clone();
+<a href=#182 id=182 data-nosnippet>182</a> <span class="kw">let
</span>input_plan = <span class="self">self</span>.input.clone();
+<a href=#183 id=183 data-nosnippet>183</a> <span class="kw">let
</span>count_schema = Arc::clone(<span class="kw-2">&</span><span
class="self">self</span>.count_schema);
+<a href=#184 id=184 data-nosnippet>184</a>
+<a href=#185 id=185 data-nosnippet>185</a> <span class="comment">//
todo revisit this
+<a href=#186 id=186 data-nosnippet>186</a> </span><span class="kw">let
</span>spec_id = <span
class="self">self</span>.table.metadata().default_partition_spec_id();
+<a href=#187 id=187 data-nosnippet>187</a> <span class="kw">let
</span>partition_type = <span
class="self">self</span>.table.metadata().default_partition_type().clone();
+<a href=#188 id=188 data-nosnippet>188</a> <span class="kw">let
</span>current_schema = <span
class="self">self</span>.table.metadata().current_schema().clone();
<a href=#189 id=189 data-nosnippet>189</a>
-<a href=#190 id=190 data-nosnippet>190</a> <span class="comment">//
Process the input streams from all partitions and commit the data files
-<a href=#191 id=191 data-nosnippet>191</a> </span><span class="kw">let
</span>stream = futures::stream::once(<span class="kw">async move </span>{
-<a href=#192 id=192 data-nosnippet>192</a> <span class="kw">let
</span><span class="kw-2">mut </span>data_files: Vec<DataFile> =
Vec::new();
-<a href=#193 id=193 data-nosnippet>193</a> <span class="kw">let
</span><span class="kw-2">mut </span>total_record_count: u64 = <span
class="number">0</span>;
-<a href=#194 id=194 data-nosnippet>194</a>
-<a href=#195 id=195 data-nosnippet>195</a> <span class="comment">//
Execute and collect results from the input coalesced plan
-<a href=#196 id=196 data-nosnippet>196</a> </span><span
class="kw">let </span><span class="kw-2">mut </span>batch_stream =
input_plan.execute(<span class="number">0</span>, context)<span
class="question-mark">?</span>;
-<a href=#197 id=197 data-nosnippet>197</a>
-<a href=#198 id=198 data-nosnippet>198</a> <span class="kw">while
let </span><span class="prelude-val">Some</span>(batch_result) =
batch_stream.next().<span class="kw">await </span>{
-<a href=#199 id=199 data-nosnippet>199</a> <span class="kw">let
</span>batch = batch_result<span class="question-mark">?</span>;
-<a href=#200 id=200 data-nosnippet>200</a>
-<a href=#201 id=201 data-nosnippet>201</a> <span class="kw">let
</span>files_array = batch
-<a href=#202 id=202 data-nosnippet>202</a>
.column_by_name(DATA_FILES_COL_NAME)
-<a href=#203 id=203 data-nosnippet>203</a> .ok_or_else(|| {
-<a href=#204 id=204 data-nosnippet>204</a>
DataFusionError::Internal(
-<a href=#205 id=205 data-nosnippet>205</a> <span
class="string">"Expected 'data_files' column in input batch"</span>.to_string(),
-<a href=#206 id=206 data-nosnippet>206</a> )
-<a href=#207 id=207 data-nosnippet>207</a> })<span
class="question-mark">?
-<a href=#208 id=208 data-nosnippet>208</a> </span>.as_any()
-<a href=#209 id=209 data-nosnippet>209</a>
.downcast_ref::<StringArray>()
-<a href=#210 id=210 data-nosnippet>210</a> .ok_or_else(|| {
-<a href=#211 id=211 data-nosnippet>211</a>
DataFusionError::Internal(
-<a href=#212 id=212 data-nosnippet>212</a> <span
class="string">"Expected 'data_files' column to be
StringArray"</span>.to_string(),
-<a href=#213 id=213 data-nosnippet>213</a> )
-<a href=#214 id=214 data-nosnippet>214</a> })<span
class="question-mark">?</span>;
-<a href=#215 id=215 data-nosnippet>215</a>
-<a href=#216 id=216 data-nosnippet>216</a> <span
class="comment">// Deserialize all data files from the StringArray
-<a href=#217 id=217 data-nosnippet>217</a> </span><span
class="kw">let </span>batch_files: Vec<DataFile> = files_array
-<a href=#218 id=218 data-nosnippet>218</a> .into_iter()
-<a href=#219 id=219 data-nosnippet>219</a> .flatten()
-<a href=#220 id=220 data-nosnippet>220</a> .map(|f| ->
DFResult<DataFile> {
-<a href=#221 id=221 data-nosnippet>221</a> <span
class="comment">// Parse JSON to DataFileSerde and convert to DataFile
-<a href=#222 id=222 data-nosnippet>222</a>
</span>deserialize_data_file_from_json(
-<a href=#223 id=223 data-nosnippet>223</a> f,
-<a href=#224 id=224 data-nosnippet>224</a> spec_id,
-<a href=#225 id=225 data-nosnippet>225</a> <span
class="kw-2">&</span>partition_type,
-<a href=#226 id=226 data-nosnippet>226</a> <span
class="kw-2">&</span>current_schema,
-<a href=#227 id=227 data-nosnippet>227</a> )
-<a href=#228 id=228 data-nosnippet>228</a>
.map_err(to_datafusion_error)
-<a href=#229 id=229 data-nosnippet>229</a> })
-<a href=#230 id=230 data-nosnippet>230</a>
.collect::<datafusion::common::Result<<span
class="kw">_</span>>>()<span class="question-mark">?</span>;
-<a href=#231 id=231 data-nosnippet>231</a>
-<a href=#232 id=232 data-nosnippet>232</a> <span
class="comment">// add record_counts from the current batch to total record
count
-<a href=#233 id=233 data-nosnippet>233</a>
</span>total_record_count += batch_files.iter().map(|f|
f.record_count()).sum::<u64>();
-<a href=#234 id=234 data-nosnippet>234</a>
-<a href=#235 id=235 data-nosnippet>235</a> <span
class="comment">// Add all deserialized files to our collection
-<a href=#236 id=236 data-nosnippet>236</a>
</span>data_files.extend(batch_files);
-<a href=#237 id=237 data-nosnippet>237</a> }
-<a href=#238 id=238 data-nosnippet>238</a>
-<a href=#239 id=239 data-nosnippet>239</a> <span class="comment">//
If no data files were collected, return an empty result
-<a href=#240 id=240 data-nosnippet>240</a> </span><span
class="kw">if </span>data_files.is_empty() {
-<a href=#241 id=241 data-nosnippet>241</a> <span
class="kw">return </span><span
class="prelude-val">Ok</span>(RecordBatch::new_empty(count_schema));
-<a href=#242 id=242 data-nosnippet>242</a> }
-<a href=#243 id=243 data-nosnippet>243</a>
-<a href=#244 id=244 data-nosnippet>244</a> <span class="comment">//
Create a transaction and commit the data files
-<a href=#245 id=245 data-nosnippet>245</a> </span><span
class="kw">let </span>tx = Transaction::new(<span
class="kw-2">&</span>table);
-<a href=#246 id=246 data-nosnippet>246</a> <span class="kw">let
</span>action = tx.fast_append().add_data_files(data_files);
-<a href=#247 id=247 data-nosnippet>247</a>
-<a href=#248 id=248 data-nosnippet>248</a> <span class="comment">//
Apply the action and commit the transaction
-<a href=#249 id=249 data-nosnippet>249</a> </span><span
class="kw">let </span>_updated_table = action
-<a href=#250 id=250 data-nosnippet>250</a> .apply(tx)
-<a href=#251 id=251 data-nosnippet>251</a>
.map_err(to_datafusion_error)<span class="question-mark">?
-<a href=#252 id=252 data-nosnippet>252</a>
</span>.commit(catalog.as_ref())
-<a href=#253 id=253 data-nosnippet>253</a> .<span
class="kw">await
-<a href=#254 id=254 data-nosnippet>254</a>
</span>.map_err(to_datafusion_error)<span class="question-mark">?</span>;
-<a href=#255 id=255 data-nosnippet>255</a>
-<a href=#256 id=256 data-nosnippet>256</a> <span
class="self">Self</span>::make_count_batch(total_record_count)
-<a href=#257 id=257 data-nosnippet>257</a> })
-<a href=#258 id=258 data-nosnippet>258</a> .boxed();
-<a href=#259 id=259 data-nosnippet>259</a>
-<a href=#260 id=260 data-nosnippet>260</a> <span
class="prelude-val">Ok</span>(Box::pin(RecordBatchStreamAdapter::new(
-<a href=#261 id=261 data-nosnippet>261</a> Arc::clone(<span
class="kw-2">&</span><span class="self">self</span>.count_schema),
-<a href=#262 id=262 data-nosnippet>262</a> stream,
-<a href=#263 id=263 data-nosnippet>263</a> )))
-<a href=#264 id=264 data-nosnippet>264</a> }
-<a href=#265 id=265 data-nosnippet>265</a>}
-<a href=#266 id=266 data-nosnippet>266</a>
-<a href=#267 id=267 data-nosnippet>267</a><span class="attr">#[cfg(test)]
-<a href=#268 id=268 data-nosnippet>268</a></span><span class="kw">mod
</span>tests {
-<a href=#269 id=269 data-nosnippet>269</a> <span class="kw">use
</span>std::collections::HashMap;
-<a href=#270 id=270 data-nosnippet>270</a> <span class="kw">use
</span>std::fmt;
-<a href=#271 id=271 data-nosnippet>271</a> <span class="kw">use
</span>std::sync::Arc;
-<a href=#272 id=272 data-nosnippet>272</a>
-<a href=#273 id=273 data-nosnippet>273</a> <span class="kw">use
</span>datafusion::arrow::array::{ArrayRef, Int32Array, RecordBatch,
StringArray, UInt64Array};
-<a href=#274 id=274 data-nosnippet>274</a> <span class="kw">use
</span>datafusion::arrow::datatypes::{DataType, Field, Schema <span
class="kw">as </span>ArrowSchema};
-<a href=#275 id=275 data-nosnippet>275</a> <span class="kw">use
</span>datafusion::datasource::MemTable;
-<a href=#276 id=276 data-nosnippet>276</a> <span class="kw">use
</span>datafusion::execution::context::TaskContext;
-<a href=#277 id=277 data-nosnippet>277</a> <span class="kw">use
</span>datafusion::physical_expr::{EquivalenceProperties, Partitioning};
-<a href=#278 id=278 data-nosnippet>278</a> <span class="kw">use
</span>datafusion::physical_plan::common::collect;
-<a href=#279 id=279 data-nosnippet>279</a> <span class="kw">use
</span>datafusion::physical_plan::execution_plan::Boundedness;
-<a href=#280 id=280 data-nosnippet>280</a> <span class="kw">use
</span>datafusion::physical_plan::stream::RecordBatchStreamAdapter;
-<a href=#281 id=281 data-nosnippet>281</a> <span class="kw">use
</span>datafusion::physical_plan::{DisplayAs, DisplayFormatType, ExecutionPlan,
PlanProperties};
-<a href=#282 id=282 data-nosnippet>282</a> <span class="kw">use
</span>datafusion::prelude::<span class="kw-2">*</span>;
-<a href=#283 id=283 data-nosnippet>283</a> <span class="kw">use
</span>futures::StreamExt;
-<a href=#284 id=284 data-nosnippet>284</a> <span class="kw">use
</span>iceberg::memory::{MEMORY_CATALOG_WAREHOUSE, MemoryCatalogBuilder};
-<a href=#285 id=285 data-nosnippet>285</a> <span class="kw">use
</span>iceberg::spec::{
-<a href=#286 id=286 data-nosnippet>286</a> DataContentType,
DataFileBuilder, DataFileFormat, NestedField, PrimitiveType, Schema,
-<a href=#287 id=287 data-nosnippet>287</a> Struct, Type,
-<a href=#288 id=288 data-nosnippet>288</a> };
-<a href=#289 id=289 data-nosnippet>289</a> <span class="kw">use
</span>iceberg::{Catalog, CatalogBuilder, NamespaceIdent, TableCreation,
TableIdent};
-<a href=#290 id=290 data-nosnippet>290</a>
-<a href=#291 id=291 data-nosnippet>291</a> <span class="kw">use
super</span>::<span class="kw-2">*</span>;
-<a href=#292 id=292 data-nosnippet>292</a> <span class="kw">use
</span><span class="kw">crate</span>::physical_plan::DATA_FILES_COL_NAME;
-<a href=#293 id=293 data-nosnippet>293</a> <span class="kw">use
</span><span class="kw">crate</span>::table::IcebergTableProvider;
-<a href=#294 id=294 data-nosnippet>294</a>
-<a href=#295 id=295 data-nosnippet>295</a> <span class="comment">// A mock
execution plan that returns record batches with serialized data files
-<a href=#296 id=296 data-nosnippet>296</a> </span><span
class="attr">#[derive(Debug)]
-<a href=#297 id=297 data-nosnippet>297</a> </span><span class="kw">struct
</span>MockWriteExec {
-<a href=#298 id=298 data-nosnippet>298</a> schema:
Arc<ArrowSchema>,
-<a href=#299 id=299 data-nosnippet>299</a> data_files_json:
Vec<String>,
-<a href=#300 id=300 data-nosnippet>300</a> plan_properties:
PlanProperties,
-<a href=#301 id=301 data-nosnippet>301</a> }
-<a href=#302 id=302 data-nosnippet>302</a>
-<a href=#303 id=303 data-nosnippet>303</a> <span class="kw">impl
</span>MockWriteExec {
-<a href=#304 id=304 data-nosnippet>304</a> <span class="kw">fn
</span>new(data_files_json: Vec<String>) -> <span class="self">Self
</span>{
-<a href=#305 id=305 data-nosnippet>305</a> <span class="kw">let
</span>schema = Arc::new(ArrowSchema::new(<span
class="macro">vec!</span>[Field::new(
-<a href=#306 id=306 data-nosnippet>306</a> DATA_FILES_COL_NAME,
-<a href=#307 id=307 data-nosnippet>307</a> DataType::Utf8,
-<a href=#308 id=308 data-nosnippet>308</a> <span
class="bool-val">false</span>,
-<a href=#309 id=309 data-nosnippet>309</a> )]));
-<a href=#310 id=310 data-nosnippet>310</a>
-<a href=#311 id=311 data-nosnippet>311</a> <span class="kw">let
</span>plan_properties = PlanProperties::new(
-<a href=#312 id=312 data-nosnippet>312</a>
EquivalenceProperties::new(schema.clone()),
-<a href=#313 id=313 data-nosnippet>313</a>
Partitioning::UnknownPartitioning(<span class="number">1</span>),
-<a href=#314 id=314 data-nosnippet>314</a> EmissionType::Final,
-<a href=#315 id=315 data-nosnippet>315</a> Boundedness::Bounded,
-<a href=#316 id=316 data-nosnippet>316</a> );
-<a href=#317 id=317 data-nosnippet>317</a>
-<a href=#318 id=318 data-nosnippet>318</a> <span class="self">Self
</span>{
-<a href=#319 id=319 data-nosnippet>319</a> schema,
-<a href=#320 id=320 data-nosnippet>320</a> data_files_json,
-<a href=#321 id=321 data-nosnippet>321</a> plan_properties,
-<a href=#322 id=322 data-nosnippet>322</a> }
-<a href=#323 id=323 data-nosnippet>323</a> }
-<a href=#324 id=324 data-nosnippet>324</a> }
-<a href=#325 id=325 data-nosnippet>325</a>
-<a href=#326 id=326 data-nosnippet>326</a> <span class="kw">impl
</span>ExecutionPlan <span class="kw">for </span>MockWriteExec {
-<a href=#327 id=327 data-nosnippet>327</a> <span class="kw">fn
</span>name(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span>str {
-<a href=#328 id=328 data-nosnippet>328</a> <span
class="string">"MockWriteExec"
-<a href=#329 id=329 data-nosnippet>329</a> </span>}
-<a href=#330 id=330 data-nosnippet>330</a>
-<a href=#331 id=331 data-nosnippet>331</a> <span class="kw">fn
</span>as_any(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span><span class="kw">dyn </span>Any {
-<a href=#332 id=332 data-nosnippet>332</a> <span class="self">self
-<a href=#333 id=333 data-nosnippet>333</a> </span>}
-<a href=#334 id=334 data-nosnippet>334</a>
-<a href=#335 id=335 data-nosnippet>335</a> <span class="kw">fn
</span>schema(<span class="kw-2">&</span><span class="self">self</span>)
-> Arc<ArrowSchema> {
-<a href=#336 id=336 data-nosnippet>336</a> <span
class="self">self</span>.schema.clone()
-<a href=#337 id=337 data-nosnippet>337</a> }
-<a href=#338 id=338 data-nosnippet>338</a>
-<a href=#339 id=339 data-nosnippet>339</a> <span class="kw">fn
</span>properties(<span class="kw-2">&</span><span
class="self">self</span>) -> <span class="kw-2">&</span>PlanProperties {
-<a href=#340 id=340 data-nosnippet>340</a> <span
class="kw-2">&</span><span class="self">self</span>.plan_properties
-<a href=#341 id=341 data-nosnippet>341</a> }
-<a href=#342 id=342 data-nosnippet>342</a>
-<a href=#343 id=343 data-nosnippet>343</a> <span class="kw">fn
</span>children(<span class="kw-2">&</span><span class="self">self</span>)
-> Vec<<span class="kw-2">&</span>Arc<<span class="kw">dyn
</span>ExecutionPlan>> {
-<a href=#344 id=344 data-nosnippet>344</a> <span
class="macro">vec!</span>[]
-<a href=#345 id=345 data-nosnippet>345</a> }
-<a href=#346 id=346 data-nosnippet>346</a>
-<a href=#347 id=347 data-nosnippet>347</a> <span class="kw">fn
</span>with_new_children(
-<a href=#348 id=348 data-nosnippet>348</a> <span
class="self">self</span>: Arc<<span class="self">Self</span>>,
-<a href=#349 id=349 data-nosnippet>349</a> _children:
Vec<Arc<<span class="kw">dyn </span>ExecutionPlan>>,
-<a href=#350 id=350 data-nosnippet>350</a> ) ->
datafusion::common::Result<Arc<<span class="kw">dyn
</span>ExecutionPlan>> {
-<a href=#351 id=351 data-nosnippet>351</a> <span
class="prelude-val">Ok</span>(<span class="self">self</span>)
-<a href=#352 id=352 data-nosnippet>352</a> }
-<a href=#353 id=353 data-nosnippet>353</a>
-<a href=#354 id=354 data-nosnippet>354</a> <span class="kw">fn
</span>execute(
-<a href=#355 id=355 data-nosnippet>355</a> <span
class="kw-2">&</span><span class="self">self</span>,
-<a href=#356 id=356 data-nosnippet>356</a> _partition: usize,
-<a href=#357 id=357 data-nosnippet>357</a> _context:
Arc<TaskContext>,
-<a href=#358 id=358 data-nosnippet>358</a> ) ->
datafusion::common::Result<SendableRecordBatchStream> {
-<a href=#359 id=359 data-nosnippet>359</a> <span class="comment">//
Create a record batch with the serialized data files
-<a href=#360 id=360 data-nosnippet>360</a> </span><span
class="kw">let </span>array = Arc::new(StringArray::from(<span
class="self">self</span>.data_files_json.clone())) <span class="kw">as
</span>ArrayRef;
-<a href=#361 id=361 data-nosnippet>361</a> <span class="kw">let
</span>batch = RecordBatch::try_new(<span
class="self">self</span>.schema.clone(), <span
class="macro">vec!</span>[array])<span class="question-mark">?</span>;
-<a href=#362 id=362 data-nosnippet>362</a>
-<a href=#363 id=363 data-nosnippet>363</a> <span class="comment">//
Create a stream that returns this batch
-<a href=#364 id=364 data-nosnippet>364</a> </span><span
class="kw">let </span>stream = futures::stream::once(<span class="kw">async
move </span>{ <span class="prelude-val">Ok</span>(batch) }).boxed();
-<a href=#365 id=365 data-nosnippet>365</a> <span
class="prelude-val">Ok</span>(Box::pin(RecordBatchStreamAdapter::new(
-<a href=#366 id=366 data-nosnippet>366</a> <span
class="self">self</span>.schema(),
-<a href=#367 id=367 data-nosnippet>367</a> stream,
-<a href=#368 id=368 data-nosnippet>368</a> )))
-<a href=#369 id=369 data-nosnippet>369</a> }
-<a href=#370 id=370 data-nosnippet>370</a> }
-<a href=#371 id=371 data-nosnippet>371</a>
-<a href=#372 id=372 data-nosnippet>372</a> <span class="comment">//
Implement DisplayAs for MockDataFilesExec
-<a href=#373 id=373 data-nosnippet>373</a> </span><span class="kw">impl
</span>DisplayAs <span class="kw">for </span>MockWriteExec {
-<a href=#374 id=374 data-nosnippet>374</a> <span class="kw">fn
</span>fmt_as(<span class="kw-2">&</span><span class="self">self</span>, t:
DisplayFormatType, f: <span class="kw-2">&mut </span>Formatter) ->
fmt::Result {
-<a href=#375 id=375 data-nosnippet>375</a> <span class="kw">match
</span>t {
-<a href=#376 id=376 data-nosnippet>376</a>
DisplayFormatType::Default
-<a href=#377 id=377 data-nosnippet>377</a> |
DisplayFormatType::Verbose
-<a href=#378 id=378 data-nosnippet>378</a> |
DisplayFormatType::TreeRender => {
-<a href=#379 id=379 data-nosnippet>379</a> <span
class="macro">write!</span>(f, <span class="string">"MockDataFilesExec:
files={}"</span>, <span class="self">self</span>.data_files_json.len())
-<a href=#380 id=380 data-nosnippet>380</a> }
-<a href=#381 id=381 data-nosnippet>381</a> }
-<a href=#382 id=382 data-nosnippet>382</a> }
-<a href=#383 id=383 data-nosnippet>383</a> }
-<a href=#384 id=384 data-nosnippet>384</a>
-<a href=#385 id=385 data-nosnippet>385</a> <span class="attr">#[tokio::test]
-<a href=#386 id=386 data-nosnippet>386</a> </span><span class="kw">async fn
</span>test_iceberg_commit_exec() -> <span
class="prelude-ty">Result</span><(), Box<<span class="kw">dyn
</span>std::error::Error>> {
-<a href=#387 id=387 data-nosnippet>387</a> <span class="comment">//
Create a memory catalog with in-memory file IO
-<a href=#388 id=388 data-nosnippet>388</a> </span><span class="kw">let
</span>catalog = Arc::new(
-<a href=#389 id=389 data-nosnippet>389</a>
MemoryCatalogBuilder::default()
-<a href=#390 id=390 data-nosnippet>390</a> .load(
-<a href=#391 id=391 data-nosnippet>391</a> <span
class="string">"memory"</span>,
-<a href=#392 id=392 data-nosnippet>392</a> HashMap::from([(
-<a href=#393 id=393 data-nosnippet>393</a>
MEMORY_CATALOG_WAREHOUSE.to_string(),
-<a href=#394 id=394 data-nosnippet>394</a> <span
class="string">"memory://root"</span>.to_string(),
-<a href=#395 id=395 data-nosnippet>395</a> )]),
-<a href=#396 id=396 data-nosnippet>396</a> )
-<a href=#397 id=397 data-nosnippet>397</a> .<span
class="kw">await
-<a href=#398 id=398 data-nosnippet>398</a> </span>.unwrap(),
-<a href=#399 id=399 data-nosnippet>399</a> );
-<a href=#400 id=400 data-nosnippet>400</a>
-<a href=#401 id=401 data-nosnippet>401</a> <span class="comment">//
Create a namespace
-<a href=#402 id=402 data-nosnippet>402</a> </span><span class="kw">let
</span>namespace = NamespaceIdent::new(<span
class="string">"test_namespace"</span>.to_string());
-<a href=#403 id=403 data-nosnippet>403</a>
catalog.create_namespace(<span class="kw-2">&</span>namespace,
HashMap::new()).<span class="kw">await</span><span
class="question-mark">?</span>;
-<a href=#404 id=404 data-nosnippet>404</a>
-<a href=#405 id=405 data-nosnippet>405</a> <span class="comment">//
Create a schema for the table
-<a href=#406 id=406 data-nosnippet>406</a> </span><span class="kw">let
</span>schema = Schema::builder()
-<a href=#407 id=407 data-nosnippet>407</a> .with_schema_id(<span
class="number">1</span>)
-<a href=#408 id=408 data-nosnippet>408</a> .with_fields(<span
class="macro">vec!</span>[
-<a href=#409 id=409 data-nosnippet>409</a>
NestedField::required(<span class="number">1</span>, <span
class="string">"id"</span>, Type::Primitive(PrimitiveType::Int)).into(),
-<a href=#410 id=410 data-nosnippet>410</a>
NestedField::required(<span class="number">2</span>, <span
class="string">"name"</span>, Type::Primitive(PrimitiveType::String)).into(),
-<a href=#411 id=411 data-nosnippet>411</a> ])
-<a href=#412 id=412 data-nosnippet>412</a> .build()<span
class="question-mark">?</span>;
-<a href=#413 id=413 data-nosnippet>413</a>
-<a href=#414 id=414 data-nosnippet>414</a> <span class="comment">//
Create a table
-<a href=#415 id=415 data-nosnippet>415</a> </span><span class="kw">let
</span>table_creation = TableCreation::builder()
-<a href=#416 id=416 data-nosnippet>416</a> .name(<span
class="string">"test_table"</span>.to_string())
-<a href=#417 id=417 data-nosnippet>417</a> .schema(schema)
-<a href=#418 id=418 data-nosnippet>418</a> .location(<span
class="string">"memory://root/test_table"</span>.to_string())
-<a href=#419 id=419 data-nosnippet>419</a>
.properties(HashMap::new())
-<a href=#420 id=420 data-nosnippet>420</a> .build();
-<a href=#421 id=421 data-nosnippet>421</a>
-<a href=#422 id=422 data-nosnippet>422</a> <span class="kw">let
</span>table = catalog.create_table(<span class="kw-2">&</span>namespace,
table_creation).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#190 id=190 data-nosnippet>190</a> <span class="kw">let
</span>catalog = Arc::clone(<span class="kw-2">&</span><span
class="self">self</span>.catalog);
+<a href=#191 id=191 data-nosnippet>191</a>
+<a href=#192 id=192 data-nosnippet>192</a> <span class="comment">//
Process the input streams from all partitions and commit the data files
+<a href=#193 id=193 data-nosnippet>193</a> </span><span class="kw">let
</span>stream = futures::stream::once(<span class="kw">async move </span>{
+<a href=#194 id=194 data-nosnippet>194</a> <span class="kw">let
</span><span class="kw-2">mut </span>data_files: Vec<DataFile> =
Vec::new();
+<a href=#195 id=195 data-nosnippet>195</a> <span class="kw">let
</span><span class="kw-2">mut </span>total_record_count: u64 = <span
class="number">0</span>;
+<a href=#196 id=196 data-nosnippet>196</a>
+<a href=#197 id=197 data-nosnippet>197</a> <span class="comment">//
Execute and collect results from the input coalesced plan
+<a href=#198 id=198 data-nosnippet>198</a> </span><span
class="kw">let </span><span class="kw-2">mut </span>batch_stream =
input_plan.execute(<span class="number">0</span>, context)<span
class="question-mark">?</span>;
+<a href=#199 id=199 data-nosnippet>199</a>
+<a href=#200 id=200 data-nosnippet>200</a> <span class="kw">while
let </span><span class="prelude-val">Some</span>(batch_result) =
batch_stream.next().<span class="kw">await </span>{
+<a href=#201 id=201 data-nosnippet>201</a> <span class="kw">let
</span>batch = batch_result<span class="question-mark">?</span>;
+<a href=#202 id=202 data-nosnippet>202</a>
+<a href=#203 id=203 data-nosnippet>203</a> <span class="kw">let
</span>files_array = batch
+<a href=#204 id=204 data-nosnippet>204</a>
.column_by_name(DATA_FILES_COL_NAME)
+<a href=#205 id=205 data-nosnippet>205</a> .ok_or_else(|| {
+<a href=#206 id=206 data-nosnippet>206</a>
DataFusionError::Internal(
+<a href=#207 id=207 data-nosnippet>207</a> <span
class="string">"Expected 'data_files' column in input batch"</span>.to_string(),
+<a href=#208 id=208 data-nosnippet>208</a> )
+<a href=#209 id=209 data-nosnippet>209</a> })<span
class="question-mark">?
+<a href=#210 id=210 data-nosnippet>210</a> </span>.as_any()
+<a href=#211 id=211 data-nosnippet>211</a>
.downcast_ref::<StringArray>()
+<a href=#212 id=212 data-nosnippet>212</a> .ok_or_else(|| {
+<a href=#213 id=213 data-nosnippet>213</a>
DataFusionError::Internal(
+<a href=#214 id=214 data-nosnippet>214</a> <span
class="string">"Expected 'data_files' column to be
StringArray"</span>.to_string(),
+<a href=#215 id=215 data-nosnippet>215</a> )
+<a href=#216 id=216 data-nosnippet>216</a> })<span
class="question-mark">?</span>;
+<a href=#217 id=217 data-nosnippet>217</a>
+<a href=#218 id=218 data-nosnippet>218</a> <span
class="comment">// Deserialize all data files from the StringArray
+<a href=#219 id=219 data-nosnippet>219</a> </span><span
class="kw">let </span>batch_files: Vec<DataFile> = files_array
+<a href=#220 id=220 data-nosnippet>220</a> .into_iter()
+<a href=#221 id=221 data-nosnippet>221</a> .flatten()
+<a href=#222 id=222 data-nosnippet>222</a> .map(|f| ->
DFResult<DataFile> {
+<a href=#223 id=223 data-nosnippet>223</a> <span
class="comment">// Parse JSON to DataFileSerde and convert to DataFile
+<a href=#224 id=224 data-nosnippet>224</a>
</span>deserialize_data_file_from_json(
+<a href=#225 id=225 data-nosnippet>225</a> f,
+<a href=#226 id=226 data-nosnippet>226</a> spec_id,
+<a href=#227 id=227 data-nosnippet>227</a> <span
class="kw-2">&</span>partition_type,
+<a href=#228 id=228 data-nosnippet>228</a> <span
class="kw-2">&</span>current_schema,
+<a href=#229 id=229 data-nosnippet>229</a> )
+<a href=#230 id=230 data-nosnippet>230</a>
.map_err(to_datafusion_error)
+<a href=#231 id=231 data-nosnippet>231</a> })
+<a href=#232 id=232 data-nosnippet>232</a>
.collect::<datafusion::common::Result<<span
class="kw">_</span>>>()<span class="question-mark">?</span>;
+<a href=#233 id=233 data-nosnippet>233</a>
+<a href=#234 id=234 data-nosnippet>234</a> <span
class="comment">// add record_counts from the current batch to total record
count
+<a href=#235 id=235 data-nosnippet>235</a>
</span>total_record_count += batch_files.iter().map(|f|
f.record_count()).sum::<u64>();
+<a href=#236 id=236 data-nosnippet>236</a>
+<a href=#237 id=237 data-nosnippet>237</a> <span
class="comment">// Add all deserialized files to our collection
+<a href=#238 id=238 data-nosnippet>238</a>
</span>data_files.extend(batch_files);
+<a href=#239 id=239 data-nosnippet>239</a> }
+<a href=#240 id=240 data-nosnippet>240</a>
+<a href=#241 id=241 data-nosnippet>241</a> <span class="comment">//
If no data files were collected, return an empty result
+<a href=#242 id=242 data-nosnippet>242</a> </span><span
class="kw">if </span>data_files.is_empty() {
+<a href=#243 id=243 data-nosnippet>243</a> <span
class="kw">return </span><span
class="prelude-val">Ok</span>(RecordBatch::new_empty(count_schema));
+<a href=#244 id=244 data-nosnippet>244</a> }
+<a href=#245 id=245 data-nosnippet>245</a>
+<a href=#246 id=246 data-nosnippet>246</a> <span class="comment">//
Create a transaction and commit the data files
+<a href=#247 id=247 data-nosnippet>247</a> </span><span
class="kw">let </span>tx = Transaction::new(<span
class="kw-2">&</span>table);
+<a href=#248 id=248 data-nosnippet>248</a> <span class="kw">let
</span>action = tx.fast_append().add_data_files(data_files);
+<a href=#249 id=249 data-nosnippet>249</a>
+<a href=#250 id=250 data-nosnippet>250</a> <span class="comment">//
Apply the action and commit the transaction
+<a href=#251 id=251 data-nosnippet>251</a> </span><span
class="kw">let </span>_updated_table = action
+<a href=#252 id=252 data-nosnippet>252</a> .apply(tx)
+<a href=#253 id=253 data-nosnippet>253</a>
.map_err(to_datafusion_error)<span class="question-mark">?
+<a href=#254 id=254 data-nosnippet>254</a>
</span>.commit(catalog.as_ref())
+<a href=#255 id=255 data-nosnippet>255</a> .<span
class="kw">await
+<a href=#256 id=256 data-nosnippet>256</a>
</span>.map_err(to_datafusion_error)<span class="question-mark">?</span>;
+<a href=#257 id=257 data-nosnippet>257</a>
+<a href=#258 id=258 data-nosnippet>258</a> <span
class="self">Self</span>::make_count_batch(total_record_count)
+<a href=#259 id=259 data-nosnippet>259</a> })
+<a href=#260 id=260 data-nosnippet>260</a> .boxed();
+<a href=#261 id=261 data-nosnippet>261</a>
+<a href=#262 id=262 data-nosnippet>262</a> <span
class="prelude-val">Ok</span>(Box::pin(RecordBatchStreamAdapter::new(
+<a href=#263 id=263 data-nosnippet>263</a> Arc::clone(<span
class="kw-2">&</span><span class="self">self</span>.count_schema),
+<a href=#264 id=264 data-nosnippet>264</a> stream,
+<a href=#265 id=265 data-nosnippet>265</a> )))
+<a href=#266 id=266 data-nosnippet>266</a> }
+<a href=#267 id=267 data-nosnippet>267</a>}
+<a href=#268 id=268 data-nosnippet>268</a>
+<a href=#269 id=269 data-nosnippet>269</a><span class="attr">#[cfg(test)]
+<a href=#270 id=270 data-nosnippet>270</a></span><span class="kw">mod
</span>tests {
+<a href=#271 id=271 data-nosnippet>271</a> <span class="kw">use
</span>std::collections::HashMap;
+<a href=#272 id=272 data-nosnippet>272</a> <span class="kw">use
</span>std::fmt;
+<a href=#273 id=273 data-nosnippet>273</a> <span class="kw">use
</span>std::sync::Arc;
+<a href=#274 id=274 data-nosnippet>274</a>
+<a href=#275 id=275 data-nosnippet>275</a> <span class="kw">use
</span>datafusion::arrow::array::{ArrayRef, Int32Array, RecordBatch,
StringArray, UInt64Array};
+<a href=#276 id=276 data-nosnippet>276</a> <span class="kw">use
</span>datafusion::arrow::datatypes::{DataType, Field, Schema <span
class="kw">as </span>ArrowSchema};
+<a href=#277 id=277 data-nosnippet>277</a> <span class="kw">use
</span>datafusion::datasource::MemTable;
+<a href=#278 id=278 data-nosnippet>278</a> <span class="kw">use
</span>datafusion::execution::context::TaskContext;
+<a href=#279 id=279 data-nosnippet>279</a> <span class="kw">use
</span>datafusion::physical_expr::{EquivalenceProperties, Partitioning};
+<a href=#280 id=280 data-nosnippet>280</a> <span class="kw">use
</span>datafusion::physical_plan::common::collect;
+<a href=#281 id=281 data-nosnippet>281</a> <span class="kw">use
</span>datafusion::physical_plan::execution_plan::Boundedness;
+<a href=#282 id=282 data-nosnippet>282</a> <span class="kw">use
</span>datafusion::physical_plan::stream::RecordBatchStreamAdapter;
+<a href=#283 id=283 data-nosnippet>283</a> <span class="kw">use
</span>datafusion::physical_plan::{DisplayAs, DisplayFormatType, ExecutionPlan,
PlanProperties};
+<a href=#284 id=284 data-nosnippet>284</a> <span class="kw">use
</span>datafusion::prelude::<span class="kw-2">*</span>;
+<a href=#285 id=285 data-nosnippet>285</a> <span class="kw">use
</span>futures::StreamExt;
+<a href=#286 id=286 data-nosnippet>286</a> <span class="kw">use
</span>iceberg::memory::{MEMORY_CATALOG_WAREHOUSE, MemoryCatalogBuilder};
+<a href=#287 id=287 data-nosnippet>287</a> <span class="kw">use
</span>iceberg::spec::{
+<a href=#288 id=288 data-nosnippet>288</a> DataContentType,
DataFileBuilder, DataFileFormat, NestedField, PrimitiveType, Schema,
+<a href=#289 id=289 data-nosnippet>289</a> Struct, Type,
+<a href=#290 id=290 data-nosnippet>290</a> };
+<a href=#291 id=291 data-nosnippet>291</a> <span class="kw">use
</span>iceberg::{Catalog, CatalogBuilder, NamespaceIdent, TableCreation,
TableIdent};
+<a href=#292 id=292 data-nosnippet>292</a>
+<a href=#293 id=293 data-nosnippet>293</a> <span class="kw">use
super</span>::<span class="kw-2">*</span>;
+<a href=#294 id=294 data-nosnippet>294</a> <span class="kw">use
</span><span class="kw">crate</span>::physical_plan::DATA_FILES_COL_NAME;
+<a href=#295 id=295 data-nosnippet>295</a> <span class="kw">use
</span><span class="kw">crate</span>::table::IcebergTableProvider;
+<a href=#296 id=296 data-nosnippet>296</a>
+<a href=#297 id=297 data-nosnippet>297</a> <span class="comment">// A mock
execution plan that returns record batches with serialized data files
+<a href=#298 id=298 data-nosnippet>298</a> </span><span
class="attr">#[derive(Debug)]
+<a href=#299 id=299 data-nosnippet>299</a> </span><span class="kw">struct
</span>MockWriteExec {
+<a href=#300 id=300 data-nosnippet>300</a> schema:
Arc<ArrowSchema>,
+<a href=#301 id=301 data-nosnippet>301</a> data_files_json:
Vec<String>,
+<a href=#302 id=302 data-nosnippet>302</a> plan_properties:
PlanProperties,
+<a href=#303 id=303 data-nosnippet>303</a> }
+<a href=#304 id=304 data-nosnippet>304</a>
+<a href=#305 id=305 data-nosnippet>305</a> <span class="kw">impl
</span>MockWriteExec {
+<a href=#306 id=306 data-nosnippet>306</a> <span class="kw">fn
</span>new(data_files_json: Vec<String>) -> <span class="self">Self
</span>{
+<a href=#307 id=307 data-nosnippet>307</a> <span class="kw">let
</span>schema = Arc::new(ArrowSchema::new(<span
class="macro">vec!</span>[Field::new(
+<a href=#308 id=308 data-nosnippet>308</a> DATA_FILES_COL_NAME,
+<a href=#309 id=309 data-nosnippet>309</a> DataType::Utf8,
+<a href=#310 id=310 data-nosnippet>310</a> <span
class="bool-val">false</span>,
+<a href=#311 id=311 data-nosnippet>311</a> )]));
+<a href=#312 id=312 data-nosnippet>312</a>
+<a href=#313 id=313 data-nosnippet>313</a> <span class="kw">let
</span>plan_properties = PlanProperties::new(
+<a href=#314 id=314 data-nosnippet>314</a>
EquivalenceProperties::new(schema.clone()),
+<a href=#315 id=315 data-nosnippet>315</a>
Partitioning::UnknownPartitioning(<span class="number">1</span>),
+<a href=#316 id=316 data-nosnippet>316</a> EmissionType::Final,
+<a href=#317 id=317 data-nosnippet>317</a> Boundedness::Bounded,
+<a href=#318 id=318 data-nosnippet>318</a> );
+<a href=#319 id=319 data-nosnippet>319</a>
+<a href=#320 id=320 data-nosnippet>320</a> <span class="self">Self
</span>{
+<a href=#321 id=321 data-nosnippet>321</a> schema,
+<a href=#322 id=322 data-nosnippet>322</a> data_files_json,
+<a href=#323 id=323 data-nosnippet>323</a> plan_properties,
+<a href=#324 id=324 data-nosnippet>324</a> }
+<a href=#325 id=325 data-nosnippet>325</a> }
+<a href=#326 id=326 data-nosnippet>326</a> }
+<a href=#327 id=327 data-nosnippet>327</a>
+<a href=#328 id=328 data-nosnippet>328</a> <span class="kw">impl
</span>ExecutionPlan <span class="kw">for </span>MockWriteExec {
+<a href=#329 id=329 data-nosnippet>329</a> <span class="kw">fn
</span>name(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span>str {
+<a href=#330 id=330 data-nosnippet>330</a> <span
class="string">"MockWriteExec"
+<a href=#331 id=331 data-nosnippet>331</a> </span>}
+<a href=#332 id=332 data-nosnippet>332</a>
+<a href=#333 id=333 data-nosnippet>333</a> <span class="kw">fn
</span>as_any(<span class="kw-2">&</span><span class="self">self</span>)
-> <span class="kw-2">&</span><span class="kw">dyn </span>Any {
+<a href=#334 id=334 data-nosnippet>334</a> <span class="self">self
+<a href=#335 id=335 data-nosnippet>335</a> </span>}
+<a href=#336 id=336 data-nosnippet>336</a>
+<a href=#337 id=337 data-nosnippet>337</a> <span class="kw">fn
</span>schema(<span class="kw-2">&</span><span class="self">self</span>)
-> Arc<ArrowSchema> {
+<a href=#338 id=338 data-nosnippet>338</a> <span
class="self">self</span>.schema.clone()
+<a href=#339 id=339 data-nosnippet>339</a> }
+<a href=#340 id=340 data-nosnippet>340</a>
+<a href=#341 id=341 data-nosnippet>341</a> <span class="kw">fn
</span>properties(<span class="kw-2">&</span><span
class="self">self</span>) -> <span class="kw-2">&</span>PlanProperties {
+<a href=#342 id=342 data-nosnippet>342</a> <span
class="kw-2">&</span><span class="self">self</span>.plan_properties
+<a href=#343 id=343 data-nosnippet>343</a> }
+<a href=#344 id=344 data-nosnippet>344</a>
+<a href=#345 id=345 data-nosnippet>345</a> <span class="kw">fn
</span>children(<span class="kw-2">&</span><span class="self">self</span>)
-> Vec<<span class="kw-2">&</span>Arc<<span class="kw">dyn
</span>ExecutionPlan>> {
+<a href=#346 id=346 data-nosnippet>346</a> <span
class="macro">vec!</span>[]
+<a href=#347 id=347 data-nosnippet>347</a> }
+<a href=#348 id=348 data-nosnippet>348</a>
+<a href=#349 id=349 data-nosnippet>349</a> <span class="kw">fn
</span>with_new_children(
+<a href=#350 id=350 data-nosnippet>350</a> <span
class="self">self</span>: Arc<<span class="self">Self</span>>,
+<a href=#351 id=351 data-nosnippet>351</a> _children:
Vec<Arc<<span class="kw">dyn </span>ExecutionPlan>>,
+<a href=#352 id=352 data-nosnippet>352</a> ) ->
datafusion::common::Result<Arc<<span class="kw">dyn
</span>ExecutionPlan>> {
+<a href=#353 id=353 data-nosnippet>353</a> <span
class="prelude-val">Ok</span>(<span class="self">self</span>)
+<a href=#354 id=354 data-nosnippet>354</a> }
+<a href=#355 id=355 data-nosnippet>355</a>
+<a href=#356 id=356 data-nosnippet>356</a> <span class="kw">fn
</span>execute(
+<a href=#357 id=357 data-nosnippet>357</a> <span
class="kw-2">&</span><span class="self">self</span>,
+<a href=#358 id=358 data-nosnippet>358</a> _partition: usize,
+<a href=#359 id=359 data-nosnippet>359</a> _context:
Arc<TaskContext>,
+<a href=#360 id=360 data-nosnippet>360</a> ) ->
datafusion::common::Result<SendableRecordBatchStream> {
+<a href=#361 id=361 data-nosnippet>361</a> <span class="comment">//
Create a record batch with the serialized data files
+<a href=#362 id=362 data-nosnippet>362</a> </span><span
class="kw">let </span>array = Arc::new(StringArray::from(<span
class="self">self</span>.data_files_json.clone())) <span class="kw">as
</span>ArrayRef;
+<a href=#363 id=363 data-nosnippet>363</a> <span class="kw">let
</span>batch = RecordBatch::try_new(<span
class="self">self</span>.schema.clone(), <span
class="macro">vec!</span>[array])<span class="question-mark">?</span>;
+<a href=#364 id=364 data-nosnippet>364</a>
+<a href=#365 id=365 data-nosnippet>365</a> <span class="comment">//
Create a stream that returns this batch
+<a href=#366 id=366 data-nosnippet>366</a> </span><span
class="kw">let </span>stream = futures::stream::once(<span class="kw">async
move </span>{ <span class="prelude-val">Ok</span>(batch) }).boxed();
+<a href=#367 id=367 data-nosnippet>367</a> <span
class="prelude-val">Ok</span>(Box::pin(RecordBatchStreamAdapter::new(
+<a href=#368 id=368 data-nosnippet>368</a> <span
class="self">self</span>.schema(),
+<a href=#369 id=369 data-nosnippet>369</a> stream,
+<a href=#370 id=370 data-nosnippet>370</a> )))
+<a href=#371 id=371 data-nosnippet>371</a> }
+<a href=#372 id=372 data-nosnippet>372</a> }
+<a href=#373 id=373 data-nosnippet>373</a>
+<a href=#374 id=374 data-nosnippet>374</a> <span class="comment">//
Implement DisplayAs for MockDataFilesExec
+<a href=#375 id=375 data-nosnippet>375</a> </span><span class="kw">impl
</span>DisplayAs <span class="kw">for </span>MockWriteExec {
+<a href=#376 id=376 data-nosnippet>376</a> <span class="kw">fn
</span>fmt_as(<span class="kw-2">&</span><span class="self">self</span>, t:
DisplayFormatType, f: <span class="kw-2">&mut </span>Formatter) ->
fmt::Result {
+<a href=#377 id=377 data-nosnippet>377</a> <span class="kw">match
</span>t {
+<a href=#378 id=378 data-nosnippet>378</a>
DisplayFormatType::Default
+<a href=#379 id=379 data-nosnippet>379</a> |
DisplayFormatType::Verbose
+<a href=#380 id=380 data-nosnippet>380</a> |
DisplayFormatType::TreeRender => {
+<a href=#381 id=381 data-nosnippet>381</a> <span
class="macro">write!</span>(f, <span class="string">"MockDataFilesExec:
files={}"</span>, <span class="self">self</span>.data_files_json.len())
+<a href=#382 id=382 data-nosnippet>382</a> }
+<a href=#383 id=383 data-nosnippet>383</a> }
+<a href=#384 id=384 data-nosnippet>384</a> }
+<a href=#385 id=385 data-nosnippet>385</a> }
+<a href=#386 id=386 data-nosnippet>386</a>
+<a href=#387 id=387 data-nosnippet>387</a> <span class="attr">#[tokio::test]
+<a href=#388 id=388 data-nosnippet>388</a> </span><span class="kw">async fn
</span>test_iceberg_commit_exec() -> <span
class="prelude-ty">Result</span><(), Box<<span class="kw">dyn
</span>std::error::Error>> {
+<a href=#389 id=389 data-nosnippet>389</a> <span class="comment">//
Create a memory catalog with in-memory file IO
+<a href=#390 id=390 data-nosnippet>390</a> </span><span class="kw">let
</span>catalog = Arc::new(
+<a href=#391 id=391 data-nosnippet>391</a>
MemoryCatalogBuilder::default()
+<a href=#392 id=392 data-nosnippet>392</a> .load(
+<a href=#393 id=393 data-nosnippet>393</a> <span
class="string">"memory"</span>,
+<a href=#394 id=394 data-nosnippet>394</a> HashMap::from([(
+<a href=#395 id=395 data-nosnippet>395</a>
MEMORY_CATALOG_WAREHOUSE.to_string(),
+<a href=#396 id=396 data-nosnippet>396</a> <span
class="string">"memory://root"</span>.to_string(),
+<a href=#397 id=397 data-nosnippet>397</a> )]),
+<a href=#398 id=398 data-nosnippet>398</a> )
+<a href=#399 id=399 data-nosnippet>399</a> .<span
class="kw">await
+<a href=#400 id=400 data-nosnippet>400</a> </span>.unwrap(),
+<a href=#401 id=401 data-nosnippet>401</a> );
+<a href=#402 id=402 data-nosnippet>402</a>
+<a href=#403 id=403 data-nosnippet>403</a> <span class="comment">//
Create a namespace
+<a href=#404 id=404 data-nosnippet>404</a> </span><span class="kw">let
</span>namespace = NamespaceIdent::new(<span
class="string">"test_namespace"</span>.to_string());
+<a href=#405 id=405 data-nosnippet>405</a>
catalog.create_namespace(<span class="kw-2">&</span>namespace,
HashMap::new()).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#406 id=406 data-nosnippet>406</a>
+<a href=#407 id=407 data-nosnippet>407</a> <span class="comment">//
Create a schema for the table
+<a href=#408 id=408 data-nosnippet>408</a> </span><span class="kw">let
</span>schema = Schema::builder()
+<a href=#409 id=409 data-nosnippet>409</a> .with_schema_id(<span
class="number">1</span>)
+<a href=#410 id=410 data-nosnippet>410</a> .with_fields(<span
class="macro">vec!</span>[
+<a href=#411 id=411 data-nosnippet>411</a>
NestedField::required(<span class="number">1</span>, <span
class="string">"id"</span>, Type::Primitive(PrimitiveType::Int)).into(),
+<a href=#412 id=412 data-nosnippet>412</a>
NestedField::required(<span class="number">2</span>, <span
class="string">"name"</span>, Type::Primitive(PrimitiveType::String)).into(),
+<a href=#413 id=413 data-nosnippet>413</a> ])
+<a href=#414 id=414 data-nosnippet>414</a> .build()<span
class="question-mark">?</span>;
+<a href=#415 id=415 data-nosnippet>415</a>
+<a href=#416 id=416 data-nosnippet>416</a> <span class="comment">//
Create a table
+<a href=#417 id=417 data-nosnippet>417</a> </span><span class="kw">let
</span>table_creation = TableCreation::builder()
+<a href=#418 id=418 data-nosnippet>418</a> .name(<span
class="string">"test_table"</span>.to_string())
+<a href=#419 id=419 data-nosnippet>419</a> .schema(schema)
+<a href=#420 id=420 data-nosnippet>420</a> .location(<span
class="string">"memory://root/test_table"</span>.to_string())
+<a href=#421 id=421 data-nosnippet>421</a>
.properties(HashMap::new())
+<a href=#422 id=422 data-nosnippet>422</a> .build();
<a href=#423 id=423 data-nosnippet>423</a>
-<a href=#424 id=424 data-nosnippet>424</a> <span class="comment">//
Create data files
-<a href=#425 id=425 data-nosnippet>425</a> </span><span class="kw">let
</span>data_file1 = DataFileBuilder::default()
-<a href=#426 id=426 data-nosnippet>426</a>
.content(DataContentType::Data)
-<a href=#427 id=427 data-nosnippet>427</a> .file_path(<span
class="string">"path/to/file1.parquet"</span>.to_string())
-<a href=#428 id=428 data-nosnippet>428</a>
.file_format(DataFileFormat::Parquet)
-<a href=#429 id=429 data-nosnippet>429</a>
.file_size_in_bytes(<span class="number">1024</span>)
-<a href=#430 id=430 data-nosnippet>430</a> .record_count(<span
class="number">100</span>)
-<a href=#431 id=431 data-nosnippet>431</a>
.partition_spec_id(table.metadata().default_partition_spec_id())
-<a href=#432 id=432 data-nosnippet>432</a>
.partition(Struct::empty())
-<a href=#433 id=433 data-nosnippet>433</a> .build()<span
class="question-mark">?</span>;
-<a href=#434 id=434 data-nosnippet>434</a>
-<a href=#435 id=435 data-nosnippet>435</a> <span class="kw">let
</span>data_file2 = DataFileBuilder::default()
-<a href=#436 id=436 data-nosnippet>436</a>
.content(DataContentType::Data)
-<a href=#437 id=437 data-nosnippet>437</a> .file_path(<span
class="string">"path/to/file2.parquet"</span>.to_string())
-<a href=#438 id=438 data-nosnippet>438</a>
.file_format(DataFileFormat::Parquet)
-<a href=#439 id=439 data-nosnippet>439</a>
.file_size_in_bytes(<span class="number">2048</span>)
-<a href=#440 id=440 data-nosnippet>440</a> .record_count(<span
class="number">200</span>)
-<a href=#441 id=441 data-nosnippet>441</a>
.partition_spec_id(table.metadata().default_partition_spec_id())
-<a href=#442 id=442 data-nosnippet>442</a>
.partition(Struct::empty())
-<a href=#443 id=443 data-nosnippet>443</a> .build()<span
class="question-mark">?</span>;
-<a href=#444 id=444 data-nosnippet>444</a>
-<a href=#445 id=445 data-nosnippet>445</a> <span class="comment">//
Serialize data files to JSON
-<a href=#446 id=446 data-nosnippet>446</a> </span><span class="kw">let
</span>partition_type = table.metadata().default_partition_type().clone();
-<a href=#447 id=447 data-nosnippet>447</a> <span class="kw">let
</span>data_file1_json = iceberg::spec::serialize_data_file_to_json(
-<a href=#448 id=448 data-nosnippet>448</a> data_file1.clone(),
-<a href=#449 id=449 data-nosnippet>449</a> <span
class="kw-2">&</span>partition_type,
-<a href=#450 id=450 data-nosnippet>450</a>
table.metadata().format_version(),
-<a href=#451 id=451 data-nosnippet>451</a> )<span
class="question-mark">?</span>;
-<a href=#452 id=452 data-nosnippet>452</a>
-<a href=#453 id=453 data-nosnippet>453</a> <span class="kw">let
</span>data_file2_json = iceberg::spec::serialize_data_file_to_json(
-<a href=#454 id=454 data-nosnippet>454</a> data_file2.clone(),
-<a href=#455 id=455 data-nosnippet>455</a> <span
class="kw-2">&</span>partition_type,
-<a href=#456 id=456 data-nosnippet>456</a>
table.metadata().format_version(),
-<a href=#457 id=457 data-nosnippet>457</a> )<span
class="question-mark">?</span>;
-<a href=#458 id=458 data-nosnippet>458</a>
-<a href=#459 id=459 data-nosnippet>459</a> <span class="comment">//
Create a mock execution plan that returns the serialized data files
-<a href=#460 id=460 data-nosnippet>460</a> </span><span class="kw">let
</span>input_exec = Arc::new(MockWriteExec::new(<span
class="macro">vec!</span>[data_file1_json, data_file2_json]));
-<a href=#461 id=461 data-nosnippet>461</a>
-<a href=#462 id=462 data-nosnippet>462</a> <span class="comment">//
Create the IcebergCommitExec
-<a href=#463 id=463 data-nosnippet>463</a> </span><span class="kw">let
</span>arrow_schema = Arc::new(ArrowSchema::new(<span
class="macro">vec!</span>[Field::new(
-<a href=#464 id=464 data-nosnippet>464</a> DATA_FILES_COL_NAME,
-<a href=#465 id=465 data-nosnippet>465</a> DataType::Utf8,
-<a href=#466 id=466 data-nosnippet>466</a> <span
class="bool-val">false</span>,
-<a href=#467 id=467 data-nosnippet>467</a> )]));
-<a href=#468 id=468 data-nosnippet>468</a>
-<a href=#469 id=469 data-nosnippet>469</a> <span class="kw">let
</span>commit_exec =
-<a href=#470 id=470 data-nosnippet>470</a>
IcebergCommitExec::new(table.clone(), catalog.clone(), input_exec,
arrow_schema);
-<a href=#471 id=471 data-nosnippet>471</a>
-<a href=#472 id=472 data-nosnippet>472</a> <span class="comment">//
Execute the commit exec
-<a href=#473 id=473 data-nosnippet>473</a> </span><span class="kw">let
</span>task_ctx = Arc::new(TaskContext::default());
-<a href=#474 id=474 data-nosnippet>474</a> <span class="kw">let
</span>stream = commit_exec.execute(<span class="number">0</span>,
task_ctx)<span class="question-mark">?</span>;
-<a href=#475 id=475 data-nosnippet>475</a> <span class="kw">let
</span>batches = collect(stream).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#424 id=424 data-nosnippet>424</a> <span class="kw">let
</span>table = catalog.create_table(<span class="kw-2">&</span>namespace,
table_creation).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#425 id=425 data-nosnippet>425</a>
+<a href=#426 id=426 data-nosnippet>426</a> <span class="comment">//
Create data files
+<a href=#427 id=427 data-nosnippet>427</a> </span><span class="kw">let
</span>data_file1 = DataFileBuilder::default()
+<a href=#428 id=428 data-nosnippet>428</a>
.content(DataContentType::Data)
+<a href=#429 id=429 data-nosnippet>429</a> .file_path(<span
class="string">"path/to/file1.parquet"</span>.to_string())
+<a href=#430 id=430 data-nosnippet>430</a>
.file_format(DataFileFormat::Parquet)
+<a href=#431 id=431 data-nosnippet>431</a>
.file_size_in_bytes(<span class="number">1024</span>)
+<a href=#432 id=432 data-nosnippet>432</a> .record_count(<span
class="number">100</span>)
+<a href=#433 id=433 data-nosnippet>433</a>
.partition_spec_id(table.metadata().default_partition_spec_id())
+<a href=#434 id=434 data-nosnippet>434</a>
.partition(Struct::empty())
+<a href=#435 id=435 data-nosnippet>435</a> .build()<span
class="question-mark">?</span>;
+<a href=#436 id=436 data-nosnippet>436</a>
+<a href=#437 id=437 data-nosnippet>437</a> <span class="kw">let
</span>data_file2 = DataFileBuilder::default()
+<a href=#438 id=438 data-nosnippet>438</a>
.content(DataContentType::Data)
+<a href=#439 id=439 data-nosnippet>439</a> .file_path(<span
class="string">"path/to/file2.parquet"</span>.to_string())
+<a href=#440 id=440 data-nosnippet>440</a>
.file_format(DataFileFormat::Parquet)
+<a href=#441 id=441 data-nosnippet>441</a>
.file_size_in_bytes(<span class="number">2048</span>)
+<a href=#442 id=442 data-nosnippet>442</a> .record_count(<span
class="number">200</span>)
+<a href=#443 id=443 data-nosnippet>443</a>
.partition_spec_id(table.metadata().default_partition_spec_id())
+<a href=#444 id=444 data-nosnippet>444</a>
.partition(Struct::empty())
+<a href=#445 id=445 data-nosnippet>445</a> .build()<span
class="question-mark">?</span>;
+<a href=#446 id=446 data-nosnippet>446</a>
+<a href=#447 id=447 data-nosnippet>447</a> <span class="comment">//
Serialize data files to JSON
+<a href=#448 id=448 data-nosnippet>448</a> </span><span class="kw">let
</span>partition_type = table.metadata().default_partition_type().clone();
+<a href=#449 id=449 data-nosnippet>449</a> <span class="kw">let
</span>data_file1_json = iceberg::spec::serialize_data_file_to_json(
+<a href=#450 id=450 data-nosnippet>450</a> data_file1.clone(),
+<a href=#451 id=451 data-nosnippet>451</a> <span
class="kw-2">&</span>partition_type,
+<a href=#452 id=452 data-nosnippet>452</a>
table.metadata().format_version(),
+<a href=#453 id=453 data-nosnippet>453</a> )<span
class="question-mark">?</span>;
+<a href=#454 id=454 data-nosnippet>454</a>
+<a href=#455 id=455 data-nosnippet>455</a> <span class="kw">let
</span>data_file2_json = iceberg::spec::serialize_data_file_to_json(
+<a href=#456 id=456 data-nosnippet>456</a> data_file2.clone(),
+<a href=#457 id=457 data-nosnippet>457</a> <span
class="kw-2">&</span>partition_type,
+<a href=#458 id=458 data-nosnippet>458</a>
table.metadata().format_version(),
+<a href=#459 id=459 data-nosnippet>459</a> )<span
class="question-mark">?</span>;
+<a href=#460 id=460 data-nosnippet>460</a>
+<a href=#461 id=461 data-nosnippet>461</a> <span class="comment">//
Create a mock execution plan that returns the serialized data files
+<a href=#462 id=462 data-nosnippet>462</a> </span><span class="kw">let
</span>input_exec = Arc::new(MockWriteExec::new(<span
class="macro">vec!</span>[data_file1_json, data_file2_json]));
+<a href=#463 id=463 data-nosnippet>463</a>
+<a href=#464 id=464 data-nosnippet>464</a> <span class="comment">//
Create the IcebergCommitExec
+<a href=#465 id=465 data-nosnippet>465</a> </span><span class="kw">let
</span>arrow_schema = Arc::new(ArrowSchema::new(<span
class="macro">vec!</span>[Field::new(
+<a href=#466 id=466 data-nosnippet>466</a> DATA_FILES_COL_NAME,
+<a href=#467 id=467 data-nosnippet>467</a> DataType::Utf8,
+<a href=#468 id=468 data-nosnippet>468</a> <span
class="bool-val">false</span>,
+<a href=#469 id=469 data-nosnippet>469</a> )]));
+<a href=#470 id=470 data-nosnippet>470</a>
+<a href=#471 id=471 data-nosnippet>471</a> <span class="kw">let
</span>commit_exec =
+<a href=#472 id=472 data-nosnippet>472</a>
IcebergCommitExec::new(table.clone(), catalog.clone(), input_exec,
arrow_schema);
+<a href=#473 id=473 data-nosnippet>473</a>
+<a href=#474 id=474 data-nosnippet>474</a> <span class="comment">//
Verify Execution Plan schema matches the count schema
+<a href=#475 id=475 data-nosnippet>475</a> </span><span
class="macro">assert_eq!</span>(commit_exec.schema(),
IcebergCommitExec::make_count_schema());
<a href=#476 id=476 data-nosnippet>476</a>
-<a href=#477 id=477 data-nosnippet>477</a> <span class="comment">//
Verify the results
-<a href=#478 id=478 data-nosnippet>478</a> </span><span
class="macro">assert_eq!</span>(batches.len(), <span class="number">1</span>);
-<a href=#479 id=479 data-nosnippet>479</a> <span class="kw">let
</span>batch = <span class="kw-2">&</span>batches[<span
class="number">0</span>];
-<a href=#480 id=480 data-nosnippet>480</a> <span
class="macro">assert_eq!</span>(batch.num_columns(), <span
class="number">1</span>);
-<a href=#481 id=481 data-nosnippet>481</a> <span
class="macro">assert_eq!</span>(batch.num_rows(), <span
class="number">1</span>);
-<a href=#482 id=482 data-nosnippet>482</a>
-<a href=#483 id=483 data-nosnippet>483</a> <span class="comment">// The
output should be a record batch with a single column "count" and a single row
-<a href=#484 id=484 data-nosnippet>484</a> // with the total record
count (100 + 200 = 300)
-<a href=#485 id=485 data-nosnippet>485</a> </span><span class="kw">let
</span>count_array = batch.column(<span class="number">0</span>);
-<a href=#486 id=486 data-nosnippet>486</a> <span
class="macro">assert_eq!</span>(count_array.len(), <span
class="number">1</span>);
-<a href=#487 id=487 data-nosnippet>487</a> <span
class="macro">assert_eq!</span>(count_array.data_type(), <span
class="kw-2">&</span>DataType::UInt64);
-<a href=#488 id=488 data-nosnippet>488</a>
-<a href=#489 id=489 data-nosnippet>489</a> <span class="comment">//
Verify that the count is correct
-<a href=#490 id=490 data-nosnippet>490</a> </span><span class="kw">let
</span>count =
count_array.as_any().downcast_ref::<UInt64Array>().unwrap();
-<a href=#491 id=491 data-nosnippet>491</a> <span
class="macro">assert_eq!</span>(count.value(<span class="number">0</span>),
<span class="number">300</span>);
-<a href=#492 id=492 data-nosnippet>492</a>
-<a href=#493 id=493 data-nosnippet>493</a> <span class="comment">//
Verify that the table has been updated with the new files
-<a href=#494 id=494 data-nosnippet>494</a> </span><span class="kw">let
</span>updated_table = catalog
-<a href=#495 id=495 data-nosnippet>495</a> .load_table(<span
class="kw-2">&</span>TableIdent::from_strs([<span
class="string">"test_namespace"</span>, <span
class="string">"test_table"</span>]).unwrap())
-<a href=#496 id=496 data-nosnippet>496</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
-<a href=#497 id=497 data-nosnippet>497</a> <span class="kw">let
</span>current_snapshot = updated_table.metadata().current_snapshot().unwrap();
-<a href=#498 id=498 data-nosnippet>498</a>
-<a href=#499 id=499 data-nosnippet>499</a> <span class="comment">//
Load the manifest list to verify the data files were added
-<a href=#500 id=500 data-nosnippet>500</a> </span><span class="kw">let
</span>manifest_list = current_snapshot
-<a href=#501 id=501 data-nosnippet>501</a>
.load_manifest_list(updated_table.file_io(), updated_table.metadata())
-<a href=#502 id=502 data-nosnippet>502</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#477 id=477 data-nosnippet>477</a> <span class="comment">//
Execute the commit exec
+<a href=#478 id=478 data-nosnippet>478</a> </span><span class="kw">let
</span>task_ctx = Arc::new(TaskContext::default());
+<a href=#479 id=479 data-nosnippet>479</a> <span class="kw">let
</span>stream = commit_exec.execute(<span class="number">0</span>,
task_ctx)<span class="question-mark">?</span>;
+<a href=#480 id=480 data-nosnippet>480</a> <span class="kw">let
</span>batches = collect(stream).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#481 id=481 data-nosnippet>481</a>
+<a href=#482 id=482 data-nosnippet>482</a> <span class="comment">//
Verify the results
+<a href=#483 id=483 data-nosnippet>483</a> </span><span
class="macro">assert_eq!</span>(batches.len(), <span class="number">1</span>);
+<a href=#484 id=484 data-nosnippet>484</a> <span class="kw">let
</span>batch = <span class="kw-2">&</span>batches[<span
class="number">0</span>];
+<a href=#485 id=485 data-nosnippet>485</a> <span
class="macro">assert_eq!</span>(batch.num_columns(), <span
class="number">1</span>);
+<a href=#486 id=486 data-nosnippet>486</a> <span
class="macro">assert_eq!</span>(batch.num_rows(), <span
class="number">1</span>);
+<a href=#487 id=487 data-nosnippet>487</a>
+<a href=#488 id=488 data-nosnippet>488</a> <span class="comment">// The
output should be a record batch with a single column "count" and a single row
+<a href=#489 id=489 data-nosnippet>489</a> // with the total record
count (100 + 200 = 300)
+<a href=#490 id=490 data-nosnippet>490</a> </span><span class="kw">let
</span>count_array = batch.column(<span class="number">0</span>);
+<a href=#491 id=491 data-nosnippet>491</a> <span
class="macro">assert_eq!</span>(count_array.len(), <span
class="number">1</span>);
+<a href=#492 id=492 data-nosnippet>492</a> <span
class="macro">assert_eq!</span>(count_array.data_type(), <span
class="kw-2">&</span>DataType::UInt64);
+<a href=#493 id=493 data-nosnippet>493</a>
+<a href=#494 id=494 data-nosnippet>494</a> <span class="comment">//
Verify that the count is correct
+<a href=#495 id=495 data-nosnippet>495</a> </span><span class="kw">let
</span>count =
count_array.as_any().downcast_ref::<UInt64Array>().unwrap();
+<a href=#496 id=496 data-nosnippet>496</a> <span
class="macro">assert_eq!</span>(count.value(<span class="number">0</span>),
<span class="number">300</span>);
+<a href=#497 id=497 data-nosnippet>497</a>
+<a href=#498 id=498 data-nosnippet>498</a> <span class="comment">//
Verify that the table has been updated with the new files
+<a href=#499 id=499 data-nosnippet>499</a> </span><span class="kw">let
</span>updated_table = catalog
+<a href=#500 id=500 data-nosnippet>500</a> .load_table(<span
class="kw-2">&</span>TableIdent::from_strs([<span
class="string">"test_namespace"</span>, <span
class="string">"test_table"</span>]).unwrap())
+<a href=#501 id=501 data-nosnippet>501</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#502 id=502 data-nosnippet>502</a> <span class="kw">let
</span>current_snapshot = updated_table.metadata().current_snapshot().unwrap();
<a href=#503 id=503 data-nosnippet>503</a>
-<a href=#504 id=504 data-nosnippet>504</a> <span class="comment">//
There should be at least one manifest
-<a href=#505 id=505 data-nosnippet>505</a> </span><span
class="macro">assert!</span>(!manifest_list.entries().is_empty());
-<a href=#506 id=506 data-nosnippet>506</a>
-<a href=#507 id=507 data-nosnippet>507</a> <span class="comment">//
Load the first manifest and verify it contains our data files
-<a href=#508 id=508 data-nosnippet>508</a> </span><span class="kw">let
</span>manifest = manifest_list.entries()[<span class="number">0</span>]
-<a href=#509 id=509 data-nosnippet>509</a>
.load_manifest(updated_table.file_io())
-<a href=#510 id=510 data-nosnippet>510</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#504 id=504 data-nosnippet>504</a> <span class="comment">//
Load the manifest list to verify the data files were added
+<a href=#505 id=505 data-nosnippet>505</a> </span><span class="kw">let
</span>manifest_list = current_snapshot
+<a href=#506 id=506 data-nosnippet>506</a>
.load_manifest_list(updated_table.file_io(), updated_table.metadata())
+<a href=#507 id=507 data-nosnippet>507</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#508 id=508 data-nosnippet>508</a>
+<a href=#509 id=509 data-nosnippet>509</a> <span class="comment">//
There should be at least one manifest
+<a href=#510 id=510 data-nosnippet>510</a> </span><span
class="macro">assert!</span>(!manifest_list.entries().is_empty());
<a href=#511 id=511 data-nosnippet>511</a>
-<a href=#512 id=512 data-nosnippet>512</a> <span class="comment">//
Verify that the manifest contains our data files
-<a href=#513 id=513 data-nosnippet>513</a> </span><span class="kw">let
</span>manifest_files: Vec<String> = manifest
-<a href=#514 id=514 data-nosnippet>514</a> .entries()
-<a href=#515 id=515 data-nosnippet>515</a> .iter()
-<a href=#516 id=516 data-nosnippet>516</a> .map(|entry|
entry.data_file().file_path().to_string())
-<a href=#517 id=517 data-nosnippet>517</a> .collect();
-<a href=#518 id=518 data-nosnippet>518</a>
-<a href=#519 id=519 data-nosnippet>519</a> <span
class="macro">assert!</span>(manifest_files.contains(<span
class="kw-2">&</span><span
class="string">"path/to/file1.parquet"</span>.to_string()));
-<a href=#520 id=520 data-nosnippet>520</a> <span
class="macro">assert!</span>(manifest_files.contains(<span
class="kw-2">&</span><span
class="string">"path/to/file2.parquet"</span>.to_string()));
-<a href=#521 id=521 data-nosnippet>521</a>
-<a href=#522 id=522 data-nosnippet>522</a> <span
class="prelude-val">Ok</span>(())
-<a href=#523 id=523 data-nosnippet>523</a> }
-<a href=#524 id=524 data-nosnippet>524</a>
-<a href=#525 id=525 data-nosnippet>525</a> <span class="attr">#[tokio::test]
-<a href=#526 id=526 data-nosnippet>526</a> </span><span class="kw">async fn
</span>test_datafusion_execution_partitioned_source() -> <span
class="prelude-ty">Result</span><(), Box<<span class="kw">dyn
</span>std::error::Error>>
-<a href=#527 id=527 data-nosnippet>527</a> {
-<a href=#528 id=528 data-nosnippet>528</a> <span class="kw">let
</span>catalog = Arc::new(
-<a href=#529 id=529 data-nosnippet>529</a>
MemoryCatalogBuilder::default()
-<a href=#530 id=530 data-nosnippet>530</a> .load(
-<a href=#531 id=531 data-nosnippet>531</a> <span
class="string">"memory"</span>,
-<a href=#532 id=532 data-nosnippet>532</a> HashMap::from([(
-<a href=#533 id=533 data-nosnippet>533</a>
MEMORY_CATALOG_WAREHOUSE.to_string(),
-<a href=#534 id=534 data-nosnippet>534</a> <span
class="string">"memory://root"</span>.to_string(),
-<a href=#535 id=535 data-nosnippet>535</a> )]),
-<a href=#536 id=536 data-nosnippet>536</a> )
-<a href=#537 id=537 data-nosnippet>537</a> .<span
class="kw">await</span><span class="question-mark">?</span>,
-<a href=#538 id=538 data-nosnippet>538</a> );
-<a href=#539 id=539 data-nosnippet>539</a>
-<a href=#540 id=540 data-nosnippet>540</a> <span class="kw">let
</span>namespace = NamespaceIdent::new(<span
class="string">"test_namespace"</span>.to_string());
-<a href=#541 id=541 data-nosnippet>541</a>
catalog.create_namespace(<span class="kw-2">&</span>namespace,
HashMap::new()).<span class="kw">await</span><span
class="question-mark">?</span>;
-<a href=#542 id=542 data-nosnippet>542</a>
-<a href=#543 id=543 data-nosnippet>543</a> <span class="kw">let
</span>schema = Schema::builder()
-<a href=#544 id=544 data-nosnippet>544</a> .with_schema_id(<span
class="number">1</span>)
-<a href=#545 id=545 data-nosnippet>545</a> .with_fields(<span
class="macro">vec!</span>[
-<a href=#546 id=546 data-nosnippet>546</a>
NestedField::required(<span class="number">1</span>, <span
class="string">"id"</span>, Type::Primitive(PrimitiveType::Int)).into(),
-<a href=#547 id=547 data-nosnippet>547</a>
NestedField::required(<span class="number">2</span>, <span
class="string">"name"</span>, Type::Primitive(PrimitiveType::String)).into(),
-<a href=#548 id=548 data-nosnippet>548</a> ])
-<a href=#549 id=549 data-nosnippet>549</a> .build()<span
class="question-mark">?</span>;
-<a href=#550 id=550 data-nosnippet>550</a>
-<a href=#551 id=551 data-nosnippet>551</a> <span class="kw">let
</span>table_name = <span class="string">"test_table"</span>;
-<a href=#552 id=552 data-nosnippet>552</a> <span class="kw">let
</span>table_creation = TableCreation::builder()
-<a href=#553 id=553 data-nosnippet>553</a>
.name(table_name.to_string())
-<a href=#554 id=554 data-nosnippet>554</a> .schema(schema)
-<a href=#555 id=555 data-nosnippet>555</a> .location(<span
class="string">"memory://root/test_table"</span>.to_string())
-<a href=#556 id=556 data-nosnippet>556</a>
.properties(HashMap::new())
-<a href=#557 id=557 data-nosnippet>557</a> .build();
-<a href=#558 id=558 data-nosnippet>558</a> <span class="kw">let _
</span>= catalog.create_table(<span class="kw-2">&</span>namespace,
table_creation).<span class="kw">await</span><span
class="question-mark">?</span>;
-<a href=#559 id=559 data-nosnippet>559</a>
-<a href=#560 id=560 data-nosnippet>560</a> <span class="kw">let
</span>arrow_schema = Arc::new(ArrowSchema::new(<span class="macro">vec!</span>[
-<a href=#561 id=561 data-nosnippet>561</a> Field::new(<span
class="string">"id"</span>, DataType::Int32, <span
class="bool-val">false</span>),
-<a href=#562 id=562 data-nosnippet>562</a> Field::new(<span
class="string">"name"</span>, DataType::Utf8, <span
class="bool-val">false</span>),
-<a href=#563 id=563 data-nosnippet>563</a> ]));
+<a href=#512 id=512 data-nosnippet>512</a> <span class="comment">//
Load the first manifest and verify it contains our data files
+<a href=#513 id=513 data-nosnippet>513</a> </span><span class="kw">let
</span>manifest = manifest_list.entries()[<span class="number">0</span>]
+<a href=#514 id=514 data-nosnippet>514</a>
.load_manifest(updated_table.file_io())
+<a href=#515 id=515 data-nosnippet>515</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#516 id=516 data-nosnippet>516</a>
+<a href=#517 id=517 data-nosnippet>517</a> <span class="comment">//
Verify that the manifest contains our data files
+<a href=#518 id=518 data-nosnippet>518</a> </span><span class="kw">let
</span>manifest_files: Vec<String> = manifest
+<a href=#519 id=519 data-nosnippet>519</a> .entries()
+<a href=#520 id=520 data-nosnippet>520</a> .iter()
+<a href=#521 id=521 data-nosnippet>521</a> .map(|entry|
entry.data_file().file_path().to_string())
+<a href=#522 id=522 data-nosnippet>522</a> .collect();
+<a href=#523 id=523 data-nosnippet>523</a>
+<a href=#524 id=524 data-nosnippet>524</a> <span
class="macro">assert!</span>(manifest_files.contains(<span
class="kw-2">&</span><span
class="string">"path/to/file1.parquet"</span>.to_string()));
+<a href=#525 id=525 data-nosnippet>525</a> <span
class="macro">assert!</span>(manifest_files.contains(<span
class="kw-2">&</span><span
class="string">"path/to/file2.parquet"</span>.to_string()));
+<a href=#526 id=526 data-nosnippet>526</a>
+<a href=#527 id=527 data-nosnippet>527</a> <span
class="prelude-val">Ok</span>(())
+<a href=#528 id=528 data-nosnippet>528</a> }
+<a href=#529 id=529 data-nosnippet>529</a>
+<a href=#530 id=530 data-nosnippet>530</a> <span class="attr">#[tokio::test]
+<a href=#531 id=531 data-nosnippet>531</a> </span><span class="kw">async fn
</span>test_datafusion_execution_partitioned_source() -> <span
class="prelude-ty">Result</span><(), Box<<span class="kw">dyn
</span>std::error::Error>>
+<a href=#532 id=532 data-nosnippet>532</a> {
+<a href=#533 id=533 data-nosnippet>533</a> <span class="kw">let
</span>catalog = Arc::new(
+<a href=#534 id=534 data-nosnippet>534</a>
MemoryCatalogBuilder::default()
+<a href=#535 id=535 data-nosnippet>535</a> .load(
+<a href=#536 id=536 data-nosnippet>536</a> <span
class="string">"memory"</span>,
+<a href=#537 id=537 data-nosnippet>537</a> HashMap::from([(
+<a href=#538 id=538 data-nosnippet>538</a>
MEMORY_CATALOG_WAREHOUSE.to_string(),
+<a href=#539 id=539 data-nosnippet>539</a> <span
class="string">"memory://root"</span>.to_string(),
+<a href=#540 id=540 data-nosnippet>540</a> )]),
+<a href=#541 id=541 data-nosnippet>541</a> )
+<a href=#542 id=542 data-nosnippet>542</a> .<span
class="kw">await</span><span class="question-mark">?</span>,
+<a href=#543 id=543 data-nosnippet>543</a> );
+<a href=#544 id=544 data-nosnippet>544</a>
+<a href=#545 id=545 data-nosnippet>545</a> <span class="kw">let
</span>namespace = NamespaceIdent::new(<span
class="string">"test_namespace"</span>.to_string());
+<a href=#546 id=546 data-nosnippet>546</a>
catalog.create_namespace(<span class="kw-2">&</span>namespace,
HashMap::new()).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#547 id=547 data-nosnippet>547</a>
+<a href=#548 id=548 data-nosnippet>548</a> <span class="kw">let
</span>schema = Schema::builder()
+<a href=#549 id=549 data-nosnippet>549</a> .with_schema_id(<span
class="number">1</span>)
+<a href=#550 id=550 data-nosnippet>550</a> .with_fields(<span
class="macro">vec!</span>[
+<a href=#551 id=551 data-nosnippet>551</a>
NestedField::required(<span class="number">1</span>, <span
class="string">"id"</span>, Type::Primitive(PrimitiveType::Int)).into(),
+<a href=#552 id=552 data-nosnippet>552</a>
NestedField::required(<span class="number">2</span>, <span
class="string">"name"</span>, Type::Primitive(PrimitiveType::String)).into(),
+<a href=#553 id=553 data-nosnippet>553</a> ])
+<a href=#554 id=554 data-nosnippet>554</a> .build()<span
class="question-mark">?</span>;
+<a href=#555 id=555 data-nosnippet>555</a>
+<a href=#556 id=556 data-nosnippet>556</a> <span class="kw">let
</span>table_name = <span class="string">"test_table"</span>;
+<a href=#557 id=557 data-nosnippet>557</a> <span class="kw">let
</span>table_creation = TableCreation::builder()
+<a href=#558 id=558 data-nosnippet>558</a>
.name(table_name.to_string())
+<a href=#559 id=559 data-nosnippet>559</a> .schema(schema)
+<a href=#560 id=560 data-nosnippet>560</a> .location(<span
class="string">"memory://root/test_table"</span>.to_string())
+<a href=#561 id=561 data-nosnippet>561</a>
.properties(HashMap::new())
+<a href=#562 id=562 data-nosnippet>562</a> .build();
+<a href=#563 id=563 data-nosnippet>563</a> <span class="kw">let _
</span>= catalog.create_table(<span class="kw-2">&</span>namespace,
table_creation).<span class="kw">await</span><span
class="question-mark">?</span>;
<a href=#564 id=564 data-nosnippet>564</a>
-<a href=#565 id=565 data-nosnippet>565</a> <span class="kw">let
</span>batches: Vec<RecordBatch> = (<span class="number">1</span>..<span
class="number">4</span>)
-<a href=#566 id=566 data-nosnippet>566</a> .map(|idx| {
-<a href=#567 id=567 data-nosnippet>567</a>
RecordBatch::try_new(arrow_schema.clone(), <span class="macro">vec!</span>[
-<a href=#568 id=568 data-nosnippet>568</a>
Arc::new(Int32Array::from(<span class="macro">vec!</span>[idx])) <span
class="kw">as </span>ArrayRef,
-<a href=#569 id=569 data-nosnippet>569</a>
Arc::new(StringArray::from(<span class="macro">vec!</span>[<span
class="macro">format!</span>(<span class="string">"Name{idx}"</span>)])) <span
class="kw">as </span>ArrayRef,
-<a href=#570 id=570 data-nosnippet>570</a> ])
-<a href=#571 id=571 data-nosnippet>571</a> })
-<a href=#572 id=572 data-nosnippet>572</a> .collect::<<span
class="prelude-ty">Result</span><<span class="kw">_</span>, <span
class="kw">_</span>>>()<span class="question-mark">?</span>;
-<a href=#573 id=573 data-nosnippet>573</a>
-<a href=#574 id=574 data-nosnippet>574</a> <span class="comment">//
Create DataFusion context with specific partition configuration
-<a href=#575 id=575 data-nosnippet>575</a> </span><span class="kw">let
</span><span class="kw-2">mut </span>config = SessionConfig::new();
-<a href=#576 id=576 data-nosnippet>576</a> config =
config.set_usize(<span
class="string">"datafusion.execution.target_partitions"</span>, <span
class="number">8</span>);
-<a href=#577 id=577 data-nosnippet>577</a> <span class="kw">let
</span>ctx = SessionContext::new_with_config(config);
+<a href=#565 id=565 data-nosnippet>565</a> <span class="kw">let
</span>arrow_schema = Arc::new(ArrowSchema::new(<span class="macro">vec!</span>[
+<a href=#566 id=566 data-nosnippet>566</a> Field::new(<span
class="string">"id"</span>, DataType::Int32, <span
class="bool-val">false</span>),
+<a href=#567 id=567 data-nosnippet>567</a> Field::new(<span
class="string">"name"</span>, DataType::Utf8, <span
class="bool-val">false</span>),
+<a href=#568 id=568 data-nosnippet>568</a> ]));
+<a href=#569 id=569 data-nosnippet>569</a>
+<a href=#570 id=570 data-nosnippet>570</a> <span class="kw">let
</span>batches: Vec<RecordBatch> = (<span class="number">1</span>..<span
class="number">4</span>)
+<a href=#571 id=571 data-nosnippet>571</a> .map(|idx| {
+<a href=#572 id=572 data-nosnippet>572</a>
RecordBatch::try_new(arrow_schema.clone(), <span class="macro">vec!</span>[
+<a href=#573 id=573 data-nosnippet>573</a>
Arc::new(Int32Array::from(<span class="macro">vec!</span>[idx])) <span
class="kw">as </span>ArrayRef,
+<a href=#574 id=574 data-nosnippet>574</a>
Arc::new(StringArray::from(<span class="macro">vec!</span>[<span
class="macro">format!</span>(<span class="string">"Name{idx}"</span>)])) <span
class="kw">as </span>ArrayRef,
+<a href=#575 id=575 data-nosnippet>575</a> ])
+<a href=#576 id=576 data-nosnippet>576</a> })
+<a href=#577 id=577 data-nosnippet>577</a> .collect::<<span
class="prelude-ty">Result</span><<span class="kw">_</span>, <span
class="kw">_</span>>>()<span class="question-mark">?</span>;
<a href=#578 id=578 data-nosnippet>578</a>
-<a href=#579 id=579 data-nosnippet>579</a> <span class="comment">//
Create multiple partitions - each batch becomes a separate partition
-<a href=#580 id=580 data-nosnippet>580</a> </span><span class="kw">let
</span>partitions: Vec<Vec<RecordBatch>> =
-<a href=#581 id=581 data-nosnippet>581</a>
batches.into_iter().map(|batch| <span
class="macro">vec!</span>[batch]).collect();
-<a href=#582 id=582 data-nosnippet>582</a> <span class="kw">let
</span>source_table = Arc::new(MemTable::try_new(Arc::clone(<span
class="kw-2">&</span>arrow_schema), partitions)<span
class="question-mark">?</span>);
-<a href=#583 id=583 data-nosnippet>583</a> ctx.register_table(<span
class="string">"source_table"</span>, source_table)<span
class="question-mark">?</span>;
-<a href=#584 id=584 data-nosnippet>584</a>
-<a href=#585 id=585 data-nosnippet>585</a> <span class="kw">let
</span>iceberg_table_provider = IcebergTableProvider::try_new(
-<a href=#586 id=586 data-nosnippet>586</a> catalog.clone(),
-<a href=#587 id=587 data-nosnippet>587</a> namespace.clone(),
-<a href=#588 id=588 data-nosnippet>588</a> table_name.to_string(),
-<a href=#589 id=589 data-nosnippet>589</a> )
-<a href=#590 id=590 data-nosnippet>590</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
-<a href=#591 id=591 data-nosnippet>591</a> ctx.register_table(<span
class="string">"iceberg_table"</span>, Arc::new(iceberg_table_provider))<span
class="question-mark">?</span>;
-<a href=#592 id=592 data-nosnippet>592</a>
-<a href=#593 id=593 data-nosnippet>593</a> <span class="kw">let
</span>insert_plan = ctx
-<a href=#594 id=594 data-nosnippet>594</a> .sql(<span
class="string">"INSERT INTO iceberg_table SELECT * FROM source_table"</span>)
-<a href=#595 id=595 data-nosnippet>595</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
-<a href=#596 id=596 data-nosnippet>596</a>
-<a href=#597 id=597 data-nosnippet>597</a> <span class="kw">let
</span>physical_plan = insert_plan.create_physical_plan().<span
class="kw">await</span><span class="question-mark">?</span>;
-<a href=#598 id=598 data-nosnippet>598</a>
-<a href=#599 id=599 data-nosnippet>599</a> <span class="kw">let
</span>actual_plan = <span class="macro">format!</span>(
-<a href=#600 id=600 data-nosnippet>600</a> <span
class="string">"{}"</span>,
-<a href=#601 id=601 data-nosnippet>601</a>
datafusion::physical_plan::displayable(physical_plan.as_ref()).indent(<span
class="bool-val">false</span>)
-<a href=#602 id=602 data-nosnippet>602</a> );
+<a href=#579 id=579 data-nosnippet>579</a> <span class="comment">//
Create DataFusion context with specific partition configuration
+<a href=#580 id=580 data-nosnippet>580</a> </span><span class="kw">let
</span><span class="kw-2">mut </span>config = SessionConfig::new();
+<a href=#581 id=581 data-nosnippet>581</a> config =
config.set_usize(<span
class="string">"datafusion.execution.target_partitions"</span>, <span
class="number">8</span>);
+<a href=#582 id=582 data-nosnippet>582</a> <span class="kw">let
</span>ctx = SessionContext::new_with_config(config);
+<a href=#583 id=583 data-nosnippet>583</a>
+<a href=#584 id=584 data-nosnippet>584</a> <span class="comment">//
Create multiple partitions - each batch becomes a separate partition
+<a href=#585 id=585 data-nosnippet>585</a> </span><span class="kw">let
</span>partitions: Vec<Vec<RecordBatch>> =
+<a href=#586 id=586 data-nosnippet>586</a>
batches.into_iter().map(|batch| <span
class="macro">vec!</span>[batch]).collect();
+<a href=#587 id=587 data-nosnippet>587</a> <span class="kw">let
</span>source_table = Arc::new(MemTable::try_new(Arc::clone(<span
class="kw-2">&</span>arrow_schema), partitions)<span
class="question-mark">?</span>);
+<a href=#588 id=588 data-nosnippet>588</a> ctx.register_table(<span
class="string">"source_table"</span>, source_table)<span
class="question-mark">?</span>;
+<a href=#589 id=589 data-nosnippet>589</a>
+<a href=#590 id=590 data-nosnippet>590</a> <span class="kw">let
</span>iceberg_table_provider = IcebergTableProvider::try_new(
+<a href=#591 id=591 data-nosnippet>591</a> catalog.clone(),
+<a href=#592 id=592 data-nosnippet>592</a> namespace.clone(),
+<a href=#593 id=593 data-nosnippet>593</a> table_name.to_string(),
+<a href=#594 id=594 data-nosnippet>594</a> )
+<a href=#595 id=595 data-nosnippet>595</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#596 id=596 data-nosnippet>596</a> ctx.register_table(<span
class="string">"iceberg_table"</span>, Arc::new(iceberg_table_provider))<span
class="question-mark">?</span>;
+<a href=#597 id=597 data-nosnippet>597</a>
+<a href=#598 id=598 data-nosnippet>598</a> <span class="kw">let
</span>insert_plan = ctx
+<a href=#599 id=599 data-nosnippet>599</a> .sql(<span
class="string">"INSERT INTO iceberg_table SELECT * FROM source_table"</span>)
+<a href=#600 id=600 data-nosnippet>600</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#601 id=601 data-nosnippet>601</a>
+<a href=#602 id=602 data-nosnippet>602</a> <span class="kw">let
</span>physical_plan = insert_plan.create_physical_plan().<span
class="kw">await</span><span class="question-mark">?</span>;
<a href=#603 id=603 data-nosnippet>603</a>
-<a href=#604 id=604 data-nosnippet>604</a> <span
class="macro">println!</span>(<span class="string">"Physical
plan:\n{actual_plan}"</span>);
-<a href=#605 id=605 data-nosnippet>605</a>
-<a href=#606 id=606 data-nosnippet>606</a> <span class="kw">let
</span>expected_plan = <span class="string">"\
-<a href=#607 id=607 data-nosnippet>607</a>IcebergCommitExec:
table=test_namespace.test_table
-<a href=#608 id=608 data-nosnippet>608</a> CoalescePartitionsExec
-<a href=#609 id=609 data-nosnippet>609</a> IcebergWriteExec:
table=test_namespace.test_table
-<a href=#610 id=610 data-nosnippet>610</a> DataSourceExec: partitions=3,
partition_sizes=[1, 1, 1]"</span>;
-<a href=#611 id=611 data-nosnippet>611</a>
-<a href=#612 id=612 data-nosnippet>612</a> <span
class="macro">assert_eq!</span>(
-<a href=#613 id=613 data-nosnippet>613</a> actual_plan.trim(),
-<a href=#614 id=614 data-nosnippet>614</a> expected_plan.trim(),
-<a href=#615 id=615 data-nosnippet>615</a> <span
class="string">"Physical plan does not match
expected\n\nExpected:\n{}\n\nActual:\n{}"</span>,
-<a href=#616 id=616 data-nosnippet>616</a> expected_plan.trim(),
-<a href=#617 id=617 data-nosnippet>617</a> actual_plan.trim()
-<a href=#618 id=618 data-nosnippet>618</a> );
-<a href=#619 id=619 data-nosnippet>619</a>
-<a href=#620 id=620 data-nosnippet>620</a> <span
class="prelude-val">Ok</span>(())
-<a href=#621 id=621 data-nosnippet>621</a> }
-<a href=#622 id=622
data-nosnippet>622</a>}</code></pre></div></section></main></body></html>
\ No newline at end of file
+<a href=#604 id=604 data-nosnippet>604</a> <span class="kw">let
</span>actual_plan = <span class="macro">format!</span>(
+<a href=#605 id=605 data-nosnippet>605</a> <span
class="string">"{}"</span>,
+<a href=#606 id=606 data-nosnippet>606</a>
datafusion::physical_plan::displayable(physical_plan.as_ref()).indent(<span
class="bool-val">false</span>)
+<a href=#607 id=607 data-nosnippet>607</a> );
+<a href=#608 id=608 data-nosnippet>608</a>
+<a href=#609 id=609 data-nosnippet>609</a> <span
class="macro">println!</span>(<span class="string">"Physical
plan:\n{actual_plan}"</span>);
+<a href=#610 id=610 data-nosnippet>610</a>
+<a href=#611 id=611 data-nosnippet>611</a> <span class="kw">let
</span>expected_plan = <span class="string">"\
+<a href=#612 id=612 data-nosnippet>612</a>IcebergCommitExec:
table=test_namespace.test_table
+<a href=#613 id=613 data-nosnippet>613</a> CoalescePartitionsExec
+<a href=#614 id=614 data-nosnippet>614</a> IcebergWriteExec:
table=test_namespace.test_table
+<a href=#615 id=615 data-nosnippet>615</a> DataSourceExec: partitions=3,
partition_sizes=[1, 1, 1]"</span>;
+<a href=#616 id=616 data-nosnippet>616</a>
+<a href=#617 id=617 data-nosnippet>617</a> <span
class="macro">assert_eq!</span>(
+<a href=#618 id=618 data-nosnippet>618</a> actual_plan.trim(),
+<a href=#619 id=619 data-nosnippet>619</a> expected_plan.trim(),
+<a href=#620 id=620 data-nosnippet>620</a> <span
class="string">"Physical plan does not match
expected\n\nExpected:\n{}\n\nActual:\n{}"</span>,
+<a href=#621 id=621 data-nosnippet>621</a> expected_plan.trim(),
+<a href=#622 id=622 data-nosnippet>622</a> actual_plan.trim()
+<a href=#623 id=623 data-nosnippet>623</a> );
+<a href=#624 id=624 data-nosnippet>624</a>
+<a href=#625 id=625 data-nosnippet>625</a> <span
class="prelude-val">Ok</span>(())
+<a href=#626 id=626 data-nosnippet>626</a> }
+<a href=#627 id=627
data-nosnippet>627</a>}</code></pre></div></section></main></body></html>
\ No newline at end of file
