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 8a97d081 deploy: 17e4351a4a4d52c8d33f7f791384bae467186cd7
8a97d081 is described below
commit 8a97d081210b79d6b48c32654d6dcb6e46ba871e
Author: liurenjie1024 <[email protected]>
AuthorDate: Fri Aug 8 10:13:13 2025 +0000
deploy: 17e4351a4a4d52c8d33f7f791384bae467186cd7
---
api/iceberg_datafusion/physical_plan/index.html | 2 +-
api/src-files.js | 4 +-
.../physical_plan/commit.rs.html | 506 +++++++++++++++++++++
.../iceberg_datafusion/physical_plan/mod.rs.html | 12 +-
4 files changed, 517 insertions(+), 7 deletions(-)
diff --git a/api/iceberg_datafusion/physical_plan/index.html
b/api/iceberg_datafusion/physical_plan/index.html
index 012ccbf0..92c33976 100644
--- a/api/iceberg_datafusion/physical_plan/index.html
+++ b/api/iceberg_datafusion/physical_plan/index.html
@@ -1,2 +1,2 @@
-<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta
name="viewport" content="width=device-width, initial-scale=1.0"><meta
name="generator" content="rustdoc"><meta name="description" content="API
documentation for the Rust `physical_plan` mod in crate
`iceberg_datafusion`."><title>iceberg_datafusion::physical_plan -
Rust</title><script>if(window.location.protocol!=="file:")document.head.insertAdjacentHTML("beforeend","SourceSerif4-Regular-6b053e98.ttf.woff2,FiraSans-Italic-81
[...]
+<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta
name="viewport" content="width=device-width, initial-scale=1.0"><meta
name="generator" content="rustdoc"><meta name="description" content="API
documentation for the Rust `physical_plan` mod in crate
`iceberg_datafusion`."><title>iceberg_datafusion::physical_plan -
Rust</title><script>if(window.location.protocol!=="file:")document.head.insertAdjacentHTML("beforeend","SourceSerif4-Regular-6b053e98.ttf.woff2,FiraSans-Italic-81
[...]
necessary details and computed properties required for execution
planning.</dd></dl></section></div></main></body></html>
\ No newline at end of file
diff --git a/api/src-files.js b/api/src-files.js
index a340c85b..62df00ae 100644
--- a/api/src-files.js
+++ b/api/src-files.js
@@ -1,3 +1,3 @@
-var srcIndex = new
Map(JSON.parse('[["iceberg",["",[["arrow",[],["caching_delete_file_loader.rs","delete_file_loader.rs","delete_filter.rs","mod.rs","nan_val_cnt_visitor.rs","reader.rs","record_batch_projector.rs","record_batch_transformer.rs","schema.rs","value.rs"]],["avro",[],["mod.rs","schema.rs"]],["catalog",[["memory",[],["catalog.rs","mod.rs","namespace_state.rs"]]],["metadata_location.rs","mod.rs"]],["expr",[["visitors",[],["bound_predicate_visitor.rs","expression_evaluator.rs","
[...]
+var srcIndex = new
Map(JSON.parse('[["iceberg",["",[["arrow",[],["caching_delete_file_loader.rs","delete_file_loader.rs","delete_filter.rs","mod.rs","nan_val_cnt_visitor.rs","reader.rs","record_batch_projector.rs","record_batch_transformer.rs","schema.rs","value.rs"]],["avro",[],["mod.rs","schema.rs"]],["catalog",[["memory",[],["catalog.rs","mod.rs","namespace_state.rs"]]],["metadata_location.rs","mod.rs"]],["expr",[["visitors",[],["bound_predicate_visitor.rs","expression_evaluator.rs","
[...]
createSrcSidebar();
-//{"start":36,"fragment_lengths":[2283,42,91,90,80,72,67,233,49,55,95,63]}
\ No newline at end of file
+//{"start":36,"fragment_lengths":[2283,42,91,90,80,72,67,245,49,55,95,63]}
\ No newline at end of file
diff --git a/api/src/iceberg_datafusion/physical_plan/commit.rs.html
b/api/src/iceberg_datafusion/physical_plan/commit.rs.html
new file mode 100644
index 00000000..f50f0382
--- /dev/null
+++ b/api/src/iceberg_datafusion/physical_plan/commit.rs.html
@@ -0,0 +1,506 @@
+<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta
name="viewport" content="width=device-width, initial-scale=1.0"><meta
name="generator" content="rustdoc"><meta name="description" content="Source of
the Rust file
`crates/integrations/datafusion/src/physical_plan/commit.rs`."><title>commit.rs
-
source</title><script>if(window.location.protocol!=="file:")document.head.insertAdjacentHTML("beforeend","SourceSerif4-Regular-6b053e98.ttf.woff2,FiraSans-Italic-81dc35de.woff2,FiraS
[...]
+<a href=#2 id=2 data-nosnippet>2</a>// or more contributor license agreements.
See the NOTICE file
+<a href=#3 id=3 data-nosnippet>3</a>// distributed with this work for
additional information
+<a href=#4 id=4 data-nosnippet>4</a>// regarding copyright ownership. The ASF
licenses this file
+<a href=#5 id=5 data-nosnippet>5</a>// to you under the Apache License,
Version 2.0 (the
+<a href=#6 id=6 data-nosnippet>6</a>// "License"); you may not use this file
except in compliance
+<a href=#7 id=7 data-nosnippet>7</a>// with the License. You may obtain a
copy of the License at
+<a href=#8 id=8 data-nosnippet>8</a>//
+<a href=#9 id=9 data-nosnippet>9</a>//
http://www.apache.org/licenses/LICENSE-2.0
+<a href=#10 id=10 data-nosnippet>10</a>//
+<a href=#11 id=11 data-nosnippet>11</a>// Unless required by applicable law or
agreed to in writing,
+<a href=#12 id=12 data-nosnippet>12</a>// software distributed under the
License is distributed on an
+<a href=#13 id=13 data-nosnippet>13</a>// "AS IS" BASIS, WITHOUT WARRANTIES OR
CONDITIONS OF ANY
+<a href=#14 id=14 data-nosnippet>14</a>// KIND, either express or implied.
See the License for the
+<a href=#15 id=15 data-nosnippet>15</a>// specific language governing
permissions and limitations
+<a href=#16 id=16 data-nosnippet>16</a>// under the License.
+<a href=#17 id=17 data-nosnippet>17</a>
+<a href=#18 id=18 data-nosnippet>18</a></span><span class="kw">use
</span>std::any::Any;
+<a href=#19 id=19 data-nosnippet>19</a><span class="kw">use
</span>std::fmt::{Debug, Formatter};
+<a href=#20 id=20 data-nosnippet>20</a><span class="kw">use
</span>std::sync::Arc;
+<a href=#21 id=21 data-nosnippet>21</a>
+<a href=#22 id=22 data-nosnippet>22</a><span class="kw">use
</span>datafusion::arrow::array::{Array, ArrayRef, RecordBatch, StringArray,
UInt64Array};
+<a href=#23 id=23 data-nosnippet>23</a><span class="kw">use
</span>datafusion::arrow::datatypes::{
+<a href=#24 id=24 data-nosnippet>24</a> DataType, Field, Schema <span
class="kw">as </span>ArrowSchema, SchemaRef <span class="kw">as
</span>ArrowSchemaRef,
+<a href=#25 id=25 data-nosnippet>25</a>};
+<a href=#26 id=26 data-nosnippet>26</a><span class="kw">use
</span>datafusion::common::{DataFusionError, <span
class="prelude-ty">Result</span> <span class="kw">as </span>DFResult};
+<a href=#27 id=27 data-nosnippet>27</a><span class="kw">use
</span>datafusion::execution::{SendableRecordBatchStream, TaskContext};
+<a href=#28 id=28 data-nosnippet>28</a><span class="kw">use
</span>datafusion::physical_expr::{EquivalenceProperties, Partitioning};
+<a href=#29 id=29 data-nosnippet>29</a><span class="kw">use
</span>datafusion::physical_plan::execution_plan::{Boundedness, EmissionType};
+<a href=#30 id=30 data-nosnippet>30</a><span class="kw">use
</span>datafusion::physical_plan::stream::RecordBatchStreamAdapter;
+<a href=#31 id=31 data-nosnippet>31</a><span class="kw">use
</span>datafusion::physical_plan::{DisplayAs, DisplayFormatType, ExecutionPlan,
PlanProperties};
+<a href=#32 id=32 data-nosnippet>32</a><span class="kw">use
</span>futures::StreamExt;
+<a href=#33 id=33 data-nosnippet>33</a><span class="kw">use
</span>iceberg::Catalog;
+<a href=#34 id=34 data-nosnippet>34</a><span class="kw">use
</span>iceberg::spec::{DataFile, deserialize_data_file_from_json};
+<a href=#35 id=35 data-nosnippet>35</a><span class="kw">use
</span>iceberg::table::Table;
+<a href=#36 id=36 data-nosnippet>36</a><span class="kw">use
</span>iceberg::transaction::{ApplyTransactionAction, Transaction};
+<a href=#37 id=37 data-nosnippet>37</a>
+<a href=#38 id=38 data-nosnippet>38</a><span class="kw">use </span><span
class="kw">crate</span>::physical_plan::DATA_FILES_COL_NAME;
+<a href=#39 id=39 data-nosnippet>39</a><span class="kw">use </span><span
class="kw">crate</span>::to_datafusion_error;
+<a href=#40 id=40 data-nosnippet>40</a>
+<a href=#41 id=41 data-nosnippet>41</a><span class="doccomment">///
IcebergCommitExec is responsible for collecting the files written and use
+<a href=#42 id=42 data-nosnippet>42</a>/// [`Transaction::fast_append`] to
commit the data files written.
+<a href=#43 id=43 data-nosnippet>43</a></span><span
class="attr">#[derive(Debug)]
+<a href=#44 id=44 data-nosnippet>44</a></span><span
class="kw">pub</span>(<span class="kw">crate</span>) <span class="kw">struct
</span>IcebergCommitExec {
+<a href=#45 id=45 data-nosnippet>45</a> table: Table,
+<a href=#46 id=46 data-nosnippet>46</a> catalog: Arc<<span
class="kw">dyn </span>Catalog>,
+<a href=#47 id=47 data-nosnippet>47</a> input: Arc<<span class="kw">dyn
</span>ExecutionPlan>,
+<a href=#48 id=48 data-nosnippet>48</a> schema: ArrowSchemaRef,
+<a href=#49 id=49 data-nosnippet>49</a> count_schema: ArrowSchemaRef,
+<a href=#50 id=50 data-nosnippet>50</a> plan_properties: PlanProperties,
+<a href=#51 id=51 data-nosnippet>51</a>}
+<a href=#52 id=52 data-nosnippet>52</a>
+<a href=#53 id=53 data-nosnippet>53</a><span class="kw">impl
</span>IcebergCommitExec {
+<a href=#54 id=54 data-nosnippet>54</a> <span class="kw">pub fn </span>new(
+<a href=#55 id=55 data-nosnippet>55</a> table: Table,
+<a href=#56 id=56 data-nosnippet>56</a> catalog: Arc<<span
class="kw">dyn </span>Catalog>,
+<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=#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>with_new_children(
+<a href=#140 id=140 data-nosnippet>140</a> <span
class="self">self</span>: Arc<<span class="self">Self</span>>,
+<a href=#141 id=141 data-nosnippet>141</a> children:
Vec<Arc<<span class="kw">dyn </span>ExecutionPlan>>,
+<a href=#142 id=142 data-nosnippet>142</a> ) -> DFResult<Arc<<span
class="kw">dyn </span>ExecutionPlan>> {
+<a href=#143 id=143 data-nosnippet>143</a> <span class="kw">if
</span>children.len() != <span class="number">1 </span>{
+<a href=#144 id=144 data-nosnippet>144</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(DataFusionError::Internal(<span
class="macro">format!</span>(
+<a href=#145 id=145 data-nosnippet>145</a> <span
class="string">"IcebergCommitExec expects exactly one child, but provided
{}"</span>,
+<a href=#146 id=146 data-nosnippet>146</a> children.len()
+<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>
+<a href=#150 id=150 data-nosnippet>150</a> <span
class="prelude-val">Ok</span>(Arc::new(IcebergCommitExec::new(
+<a href=#151 id=151 data-nosnippet>151</a> <span
class="self">self</span>.table.clone(),
+<a href=#152 id=152 data-nosnippet>152</a> <span
class="self">self</span>.catalog.clone(),
+<a href=#153 id=153 data-nosnippet>153</a> children[<span
class="number">0</span>].clone(),
+<a href=#154 id=154 data-nosnippet>154</a> <span
class="self">self</span>.schema.clone(),
+<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="kw">fn
</span>execute(
+<a href=#159 id=159 data-nosnippet>159</a> <span
class="kw-2">&</span><span class="self">self</span>,
+<a href=#160 id=160 data-nosnippet>160</a> partition: usize,
+<a href=#161 id=161 data-nosnippet>161</a> context:
Arc<TaskContext>,
+<a href=#162 id=162 data-nosnippet>162</a> ) ->
DFResult<SendableRecordBatchStream> {
+<a href=#163 id=163 data-nosnippet>163</a> <span class="comment">//
IcebergCommitExec only has one partition (partition 0)
+<a href=#164 id=164 data-nosnippet>164</a> </span><span class="kw">if
</span>partition != <span class="number">0 </span>{
+<a href=#165 id=165 data-nosnippet>165</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(DataFusionError::Internal(<span
class="macro">format!</span>(
+<a href=#166 id=166 data-nosnippet>166</a> <span
class="string">"IcebergCommitExec only has one partition, but got partition
{}"</span>,
+<a href=#167 id=167 data-nosnippet>167</a> partition
+<a href=#168 id=168 data-nosnippet>168</a> )));
+<a href=#169 id=169 data-nosnippet>169</a> }
+<a href=#170 id=170 data-nosnippet>170</a>
+<a href=#171 id=171 data-nosnippet>171</a> <span class="kw">let
</span>table = <span class="self">self</span>.table.clone();
+<a href=#172 id=172 data-nosnippet>172</a> <span class="kw">let
</span>input_plan = <span class="self">self</span>.input.clone();
+<a href=#173 id=173 data-nosnippet>173</a> <span class="kw">let
</span>count_schema = Arc::clone(<span class="kw-2">&</span><span
class="self">self</span>.count_schema);
+<a href=#174 id=174 data-nosnippet>174</a>
+<a href=#175 id=175 data-nosnippet>175</a> <span class="comment">//
todo revisit this
+<a href=#176 id=176 data-nosnippet>176</a> </span><span class="kw">let
</span>spec_id = <span
class="self">self</span>.table.metadata().default_partition_spec_id();
+<a href=#177 id=177 data-nosnippet>177</a> <span class="kw">let
</span>partition_type = <span
class="self">self</span>.table.metadata().default_partition_type().clone();
+<a href=#178 id=178 data-nosnippet>178</a> <span class="kw">let
</span>current_schema = <span
class="self">self</span>.table.metadata().current_schema().clone();
+<a href=#179 id=179 data-nosnippet>179</a>
+<a href=#180 id=180 data-nosnippet>180</a> <span class="kw">let
</span>catalog = Arc::clone(<span class="kw-2">&</span><span
class="self">self</span>.catalog);
+<a href=#181 id=181 data-nosnippet>181</a>
+<a href=#182 id=182 data-nosnippet>182</a> <span class="comment">//
Process the input streams from all partitions and commit the data files
+<a href=#183 id=183 data-nosnippet>183</a> </span><span class="kw">let
</span>stream = futures::stream::once(<span class="kw">async move </span>{
+<a href=#184 id=184 data-nosnippet>184</a> <span class="kw">let
</span><span class="kw-2">mut </span>data_files: Vec<DataFile> =
Vec::new();
+<a href=#185 id=185 data-nosnippet>185</a> <span class="kw">let
</span><span class="kw-2">mut </span>total_record_count: u64 = <span
class="number">0</span>;
+<a href=#186 id=186 data-nosnippet>186</a>
+<a href=#187 id=187 data-nosnippet>187</a> <span class="comment">//
Execute and collect results from the input coalesced plan
+<a href=#188 id=188 data-nosnippet>188</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=#189 id=189 data-nosnippet>189</a>
+<a href=#190 id=190 data-nosnippet>190</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=#191 id=191 data-nosnippet>191</a> <span class="kw">let
</span>batch = batch_result<span class="question-mark">?</span>;
+<a href=#192 id=192 data-nosnippet>192</a>
+<a href=#193 id=193 data-nosnippet>193</a> <span class="kw">let
</span>files_array = batch
+<a href=#194 id=194 data-nosnippet>194</a>
.column_by_name(DATA_FILES_COL_NAME)
+<a href=#195 id=195 data-nosnippet>195</a> .ok_or_else(|| {
+<a href=#196 id=196 data-nosnippet>196</a>
DataFusionError::Internal(
+<a href=#197 id=197 data-nosnippet>197</a> <span
class="string">"Expected 'data_files' column in input batch"</span>.to_string(),
+<a href=#198 id=198 data-nosnippet>198</a> )
+<a href=#199 id=199 data-nosnippet>199</a> })<span
class="question-mark">?
+<a href=#200 id=200 data-nosnippet>200</a> </span>.as_any()
+<a href=#201 id=201 data-nosnippet>201</a>
.downcast_ref::<StringArray>()
+<a href=#202 id=202 data-nosnippet>202</a> .ok_or_else(|| {
+<a href=#203 id=203 data-nosnippet>203</a>
DataFusionError::Internal(
+<a href=#204 id=204 data-nosnippet>204</a> <span
class="string">"Expected 'data_files' column to be
StringArray"</span>.to_string(),
+<a href=#205 id=205 data-nosnippet>205</a> )
+<a href=#206 id=206 data-nosnippet>206</a> })<span
class="question-mark">?</span>;
+<a href=#207 id=207 data-nosnippet>207</a>
+<a href=#208 id=208 data-nosnippet>208</a> <span
class="comment">// Deserialize all data files from the StringArray
+<a href=#209 id=209 data-nosnippet>209</a> </span><span
class="kw">let </span>batch_files: Vec<DataFile> = files_array
+<a href=#210 id=210 data-nosnippet>210</a> .into_iter()
+<a href=#211 id=211 data-nosnippet>211</a> .flatten()
+<a href=#212 id=212 data-nosnippet>212</a> .map(|f| ->
DFResult<DataFile> {
+<a href=#213 id=213 data-nosnippet>213</a> <span
class="comment">// Parse JSON to DataFileSerde and convert to DataFile
+<a href=#214 id=214 data-nosnippet>214</a>
</span>deserialize_data_file_from_json(
+<a href=#215 id=215 data-nosnippet>215</a> f,
+<a href=#216 id=216 data-nosnippet>216</a> spec_id,
+<a href=#217 id=217 data-nosnippet>217</a> <span
class="kw-2">&</span>partition_type,
+<a href=#218 id=218 data-nosnippet>218</a> <span
class="kw-2">&</span>current_schema,
+<a href=#219 id=219 data-nosnippet>219</a> )
+<a href=#220 id=220 data-nosnippet>220</a>
.map_err(to_datafusion_error)
+<a href=#221 id=221 data-nosnippet>221</a> })
+<a href=#222 id=222 data-nosnippet>222</a>
.collect::<datafusion::common::Result<<span
class="kw">_</span>>>()<span class="question-mark">?</span>;
+<a href=#223 id=223 data-nosnippet>223</a>
+<a href=#224 id=224 data-nosnippet>224</a> <span
class="comment">// add record_counts from the current batch to total record
count
+<a href=#225 id=225 data-nosnippet>225</a>
</span>total_record_count += batch_files.iter().map(|f|
f.record_count()).sum::<u64>();
+<a href=#226 id=226 data-nosnippet>226</a>
+<a href=#227 id=227 data-nosnippet>227</a> <span
class="comment">// Add all deserialized files to our collection
+<a href=#228 id=228 data-nosnippet>228</a>
</span>data_files.extend(batch_files);
+<a href=#229 id=229 data-nosnippet>229</a> }
+<a href=#230 id=230 data-nosnippet>230</a>
+<a href=#231 id=231 data-nosnippet>231</a> <span class="comment">//
If no data files were collected, return an empty result
+<a href=#232 id=232 data-nosnippet>232</a> </span><span
class="kw">if </span>data_files.is_empty() {
+<a href=#233 id=233 data-nosnippet>233</a> <span
class="kw">return </span><span
class="prelude-val">Ok</span>(RecordBatch::new_empty(count_schema));
+<a href=#234 id=234 data-nosnippet>234</a> }
+<a href=#235 id=235 data-nosnippet>235</a>
+<a href=#236 id=236 data-nosnippet>236</a> <span class="comment">//
Create a transaction and commit the data files
+<a href=#237 id=237 data-nosnippet>237</a> </span><span
class="kw">let </span>tx = Transaction::new(<span
class="kw-2">&</span>table);
+<a href=#238 id=238 data-nosnippet>238</a> <span class="kw">let
</span>action = tx.fast_append().add_data_files(data_files);
+<a href=#239 id=239 data-nosnippet>239</a>
+<a href=#240 id=240 data-nosnippet>240</a> <span class="comment">//
Apply the action and commit the transaction
+<a href=#241 id=241 data-nosnippet>241</a> </span><span
class="kw">let </span>_updated_table = action
+<a href=#242 id=242 data-nosnippet>242</a> .apply(tx)
+<a href=#243 id=243 data-nosnippet>243</a>
.map_err(to_datafusion_error)<span class="question-mark">?
+<a href=#244 id=244 data-nosnippet>244</a>
</span>.commit(catalog.as_ref())
+<a href=#245 id=245 data-nosnippet>245</a> .<span
class="kw">await
+<a href=#246 id=246 data-nosnippet>246</a>
</span>.map_err(to_datafusion_error)<span class="question-mark">?</span>;
+<a href=#247 id=247 data-nosnippet>247</a>
+<a href=#248 id=248 data-nosnippet>248</a> <span
class="self">Self</span>::make_count_batch(total_record_count)
+<a href=#249 id=249 data-nosnippet>249</a> })
+<a href=#250 id=250 data-nosnippet>250</a> .boxed();
+<a href=#251 id=251 data-nosnippet>251</a>
+<a href=#252 id=252 data-nosnippet>252</a> <span
class="prelude-val">Ok</span>(Box::pin(RecordBatchStreamAdapter::new(
+<a href=#253 id=253 data-nosnippet>253</a> Arc::clone(<span
class="kw-2">&</span><span class="self">self</span>.count_schema),
+<a href=#254 id=254 data-nosnippet>254</a> stream,
+<a href=#255 id=255 data-nosnippet>255</a> )))
+<a href=#256 id=256 data-nosnippet>256</a> }
+<a href=#257 id=257 data-nosnippet>257</a>}
+<a href=#258 id=258 data-nosnippet>258</a>
+<a href=#259 id=259 data-nosnippet>259</a><span class="attr">#[cfg(test)]
+<a href=#260 id=260 data-nosnippet>260</a></span><span class="kw">mod
</span>tests {
+<a href=#261 id=261 data-nosnippet>261</a> <span class="kw">use
</span>std::collections::HashMap;
+<a href=#262 id=262 data-nosnippet>262</a> <span class="kw">use
</span>std::fmt;
+<a href=#263 id=263 data-nosnippet>263</a> <span class="kw">use
</span>std::sync::Arc;
+<a href=#264 id=264 data-nosnippet>264</a>
+<a href=#265 id=265 data-nosnippet>265</a> <span class="kw">use
</span>datafusion::arrow::array::{ArrayRef, RecordBatch, StringArray};
+<a href=#266 id=266 data-nosnippet>266</a> <span class="kw">use
</span>datafusion::arrow::datatypes::{DataType, Field, Schema <span
class="kw">as </span>ArrowSchema};
+<a href=#267 id=267 data-nosnippet>267</a> <span class="kw">use
</span>datafusion::execution::context::TaskContext;
+<a href=#268 id=268 data-nosnippet>268</a> <span class="kw">use
</span>datafusion::physical_expr::{EquivalenceProperties, Partitioning};
+<a href=#269 id=269 data-nosnippet>269</a> <span class="kw">use
</span>datafusion::physical_plan::common::collect;
+<a href=#270 id=270 data-nosnippet>270</a> <span class="kw">use
</span>datafusion::physical_plan::execution_plan::Boundedness;
+<a href=#271 id=271 data-nosnippet>271</a> <span class="kw">use
</span>datafusion::physical_plan::stream::RecordBatchStreamAdapter;
+<a href=#272 id=272 data-nosnippet>272</a> <span class="kw">use
</span>datafusion::physical_plan::{DisplayAs, DisplayFormatType, ExecutionPlan,
PlanProperties};
+<a href=#273 id=273 data-nosnippet>273</a> <span class="kw">use
</span>futures::StreamExt;
+<a href=#274 id=274 data-nosnippet>274</a> <span class="kw">use
</span>iceberg::io::FileIOBuilder;
+<a href=#275 id=275 data-nosnippet>275</a> <span class="kw">use
</span>iceberg::spec::{
+<a href=#276 id=276 data-nosnippet>276</a> DataContentType,
DataFileBuilder, DataFileFormat, NestedField, PrimitiveType, Schema,
+<a href=#277 id=277 data-nosnippet>277</a> Struct, Type,
+<a href=#278 id=278 data-nosnippet>278</a> };
+<a href=#279 id=279 data-nosnippet>279</a> <span class="kw">use
</span>iceberg::{Catalog, MemoryCatalog, NamespaceIdent, TableCreation,
TableIdent};
+<a href=#280 id=280 data-nosnippet>280</a>
+<a href=#281 id=281 data-nosnippet>281</a> <span class="kw">use
super</span>::<span class="kw-2">*</span>;
+<a href=#282 id=282 data-nosnippet>282</a> <span class="kw">use
</span><span class="kw">crate</span>::physical_plan::DATA_FILES_COL_NAME;
+<a href=#283 id=283 data-nosnippet>283</a>
+<a href=#284 id=284 data-nosnippet>284</a> <span class="comment">// A mock
execution plan that returns record batches with serialized data files
+<a href=#285 id=285 data-nosnippet>285</a> </span><span
class="attr">#[derive(Debug)]
+<a href=#286 id=286 data-nosnippet>286</a> </span><span class="kw">struct
</span>MockWriteExec {
+<a href=#287 id=287 data-nosnippet>287</a> schema:
Arc<ArrowSchema>,
+<a href=#288 id=288 data-nosnippet>288</a> data_files_json:
Vec<String>,
+<a href=#289 id=289 data-nosnippet>289</a> plan_properties:
PlanProperties,
+<a href=#290 id=290 data-nosnippet>290</a> }
+<a href=#291 id=291 data-nosnippet>291</a>
+<a href=#292 id=292 data-nosnippet>292</a> <span class="kw">impl
</span>MockWriteExec {
+<a href=#293 id=293 data-nosnippet>293</a> <span class="kw">fn
</span>new(data_files_json: Vec<String>) -> <span class="self">Self
</span>{
+<a href=#294 id=294 data-nosnippet>294</a> <span class="kw">let
</span>schema = Arc::new(ArrowSchema::new(<span
class="macro">vec!</span>[Field::new(
+<a href=#295 id=295 data-nosnippet>295</a> DATA_FILES_COL_NAME,
+<a href=#296 id=296 data-nosnippet>296</a> DataType::Utf8,
+<a href=#297 id=297 data-nosnippet>297</a> <span
class="bool-val">false</span>,
+<a href=#298 id=298 data-nosnippet>298</a> )]));
+<a href=#299 id=299 data-nosnippet>299</a>
+<a href=#300 id=300 data-nosnippet>300</a> <span class="kw">let
</span>plan_properties = PlanProperties::new(
+<a href=#301 id=301 data-nosnippet>301</a>
EquivalenceProperties::new(schema.clone()),
+<a href=#302 id=302 data-nosnippet>302</a>
Partitioning::UnknownPartitioning(<span class="number">1</span>),
+<a href=#303 id=303 data-nosnippet>303</a> EmissionType::Final,
+<a href=#304 id=304 data-nosnippet>304</a> Boundedness::Bounded,
+<a href=#305 id=305 data-nosnippet>305</a> );
+<a href=#306 id=306 data-nosnippet>306</a>
+<a href=#307 id=307 data-nosnippet>307</a> <span class="self">Self
</span>{
+<a href=#308 id=308 data-nosnippet>308</a> schema,
+<a href=#309 id=309 data-nosnippet>309</a> data_files_json,
+<a href=#310 id=310 data-nosnippet>310</a> plan_properties,
+<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> }
+<a href=#314 id=314 data-nosnippet>314</a>
+<a href=#315 id=315 data-nosnippet>315</a> <span class="kw">impl
</span>ExecutionPlan <span class="kw">for </span>MockWriteExec {
+<a href=#316 id=316 data-nosnippet>316</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=#317 id=317 data-nosnippet>317</a> <span
class="string">"MockWriteExec"
+<a href=#318 id=318 data-nosnippet>318</a> </span>}
+<a href=#319 id=319 data-nosnippet>319</a>
+<a href=#320 id=320 data-nosnippet>320</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=#321 id=321 data-nosnippet>321</a> <span class="self">self
+<a href=#322 id=322 data-nosnippet>322</a> </span>}
+<a href=#323 id=323 data-nosnippet>323</a>
+<a href=#324 id=324 data-nosnippet>324</a> <span class="kw">fn
</span>schema(<span class="kw-2">&</span><span class="self">self</span>)
-> Arc<ArrowSchema> {
+<a href=#325 id=325 data-nosnippet>325</a> <span
class="self">self</span>.schema.clone()
+<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">fn
</span>properties(<span class="kw-2">&</span><span
class="self">self</span>) -> <span class="kw-2">&</span>PlanProperties {
+<a href=#329 id=329 data-nosnippet>329</a> <span
class="kw-2">&</span><span class="self">self</span>.plan_properties
+<a href=#330 id=330 data-nosnippet>330</a> }
+<a href=#331 id=331 data-nosnippet>331</a>
+<a href=#332 id=332 data-nosnippet>332</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=#333 id=333 data-nosnippet>333</a> <span
class="macro">vec!</span>[]
+<a href=#334 id=334 data-nosnippet>334</a> }
+<a href=#335 id=335 data-nosnippet>335</a>
+<a href=#336 id=336 data-nosnippet>336</a> <span class="kw">fn
</span>with_new_children(
+<a href=#337 id=337 data-nosnippet>337</a> <span
class="self">self</span>: Arc<<span class="self">Self</span>>,
+<a href=#338 id=338 data-nosnippet>338</a> _children:
Vec<Arc<<span class="kw">dyn </span>ExecutionPlan>>,
+<a href=#339 id=339 data-nosnippet>339</a> ) ->
datafusion::common::Result<Arc<<span class="kw">dyn
</span>ExecutionPlan>> {
+<a href=#340 id=340 data-nosnippet>340</a> <span
class="prelude-val">Ok</span>(<span class="self">self</span>)
+<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>execute(
+<a href=#344 id=344 data-nosnippet>344</a> <span
class="kw-2">&</span><span class="self">self</span>,
+<a href=#345 id=345 data-nosnippet>345</a> _partition: usize,
+<a href=#346 id=346 data-nosnippet>346</a> _context:
Arc<TaskContext>,
+<a href=#347 id=347 data-nosnippet>347</a> ) ->
datafusion::common::Result<SendableRecordBatchStream> {
+<a href=#348 id=348 data-nosnippet>348</a> <span class="comment">//
Create a record batch with the serialized data files
+<a href=#349 id=349 data-nosnippet>349</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=#350 id=350 data-nosnippet>350</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=#351 id=351 data-nosnippet>351</a>
+<a href=#352 id=352 data-nosnippet>352</a> <span class="comment">//
Create a stream that returns this batch
+<a href=#353 id=353 data-nosnippet>353</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=#354 id=354 data-nosnippet>354</a> <span
class="prelude-val">Ok</span>(Box::pin(RecordBatchStreamAdapter::new(
+<a href=#355 id=355 data-nosnippet>355</a> <span
class="self">self</span>.schema(),
+<a href=#356 id=356 data-nosnippet>356</a> stream,
+<a href=#357 id=357 data-nosnippet>357</a> )))
+<a href=#358 id=358 data-nosnippet>358</a> }
+<a href=#359 id=359 data-nosnippet>359</a> }
+<a href=#360 id=360 data-nosnippet>360</a>
+<a href=#361 id=361 data-nosnippet>361</a> <span class="comment">//
Implement DisplayAs for MockDataFilesExec
+<a href=#362 id=362 data-nosnippet>362</a> </span><span class="kw">impl
</span>DisplayAs <span class="kw">for </span>MockWriteExec {
+<a href=#363 id=363 data-nosnippet>363</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=#364 id=364 data-nosnippet>364</a> <span class="kw">match
</span>t {
+<a href=#365 id=365 data-nosnippet>365</a>
DisplayFormatType::Default
+<a href=#366 id=366 data-nosnippet>366</a> |
DisplayFormatType::Verbose
+<a href=#367 id=367 data-nosnippet>367</a> |
DisplayFormatType::TreeRender => {
+<a href=#368 id=368 data-nosnippet>368</a> <span
class="macro">write!</span>(f, <span class="string">"MockDataFilesExec:
files={}"</span>, <span class="self">self</span>.data_files_json.len())
+<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> }
+<a href=#373 id=373 data-nosnippet>373</a>
+<a href=#374 id=374 data-nosnippet>374</a> <span class="attr">#[tokio::test]
+<a href=#375 id=375 data-nosnippet>375</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=#376 id=376 data-nosnippet>376</a> <span class="comment">//
Create a memory catalog with in-memory file IO
+<a href=#377 id=377 data-nosnippet>377</a> </span><span class="kw">let
</span>file_io = FileIOBuilder::new(<span
class="string">"memory"</span>).build()<span class="question-mark">?</span>;
+<a href=#378 id=378 data-nosnippet>378</a> <span class="kw">let
</span>catalog = Arc::new(MemoryCatalog::new(
+<a href=#379 id=379 data-nosnippet>379</a> file_io,
+<a href=#380 id=380 data-nosnippet>380</a> <span
class="prelude-val">Some</span>(<span
class="string">"memory://root"</span>.to_string()),
+<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> <span class="comment">//
Create a namespace
+<a href=#384 id=384 data-nosnippet>384</a> </span><span class="kw">let
</span>namespace = NamespaceIdent::new(<span
class="string">"test_namespace"</span>.to_string());
+<a href=#385 id=385 data-nosnippet>385</a>
catalog.create_namespace(<span class="kw-2">&</span>namespace,
HashMap::new()).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#386 id=386 data-nosnippet>386</a>
+<a href=#387 id=387 data-nosnippet>387</a> <span class="comment">//
Create a schema for the table
+<a href=#388 id=388 data-nosnippet>388</a> </span><span class="kw">let
</span>schema = Schema::builder()
+<a href=#389 id=389 data-nosnippet>389</a> .with_schema_id(<span
class="number">1</span>)
+<a href=#390 id=390 data-nosnippet>390</a> .with_fields(<span
class="macro">vec!</span>[
+<a href=#391 id=391 data-nosnippet>391</a>
NestedField::required(<span class="number">1</span>, <span
class="string">"id"</span>, Type::Primitive(PrimitiveType::Int)).into(),
+<a href=#392 id=392 data-nosnippet>392</a>
NestedField::required(<span class="number">2</span>, <span
class="string">"name"</span>, Type::Primitive(PrimitiveType::String)).into(),
+<a href=#393 id=393 data-nosnippet>393</a> ])
+<a href=#394 id=394 data-nosnippet>394</a> .build()<span
class="question-mark">?</span>;
+<a href=#395 id=395 data-nosnippet>395</a>
+<a href=#396 id=396 data-nosnippet>396</a> <span class="comment">//
Create a table
+<a href=#397 id=397 data-nosnippet>397</a> </span><span class="kw">let
</span>table_creation = TableCreation::builder()
+<a href=#398 id=398 data-nosnippet>398</a> .name(<span
class="string">"test_table"</span>.to_string())
+<a href=#399 id=399 data-nosnippet>399</a> .schema(schema)
+<a href=#400 id=400 data-nosnippet>400</a> .location(<span
class="string">"memory://root/test_table"</span>.to_string())
+<a href=#401 id=401 data-nosnippet>401</a>
.properties(HashMap::new())
+<a href=#402 id=402 data-nosnippet>402</a> .build();
+<a href=#403 id=403 data-nosnippet>403</a>
+<a href=#404 id=404 data-nosnippet>404</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=#405 id=405 data-nosnippet>405</a>
+<a href=#406 id=406 data-nosnippet>406</a> <span class="comment">//
Create data files
+<a href=#407 id=407 data-nosnippet>407</a> </span><span class="kw">let
</span>data_file1 = DataFileBuilder::default()
+<a href=#408 id=408 data-nosnippet>408</a>
.content(DataContentType::Data)
+<a href=#409 id=409 data-nosnippet>409</a> .file_path(<span
class="string">"path/to/file1.parquet"</span>.to_string())
+<a href=#410 id=410 data-nosnippet>410</a>
.file_format(DataFileFormat::Parquet)
+<a href=#411 id=411 data-nosnippet>411</a>
.file_size_in_bytes(<span class="number">1024</span>)
+<a href=#412 id=412 data-nosnippet>412</a> .record_count(<span
class="number">100</span>)
+<a href=#413 id=413 data-nosnippet>413</a>
.partition_spec_id(table.metadata().default_partition_spec_id())
+<a href=#414 id=414 data-nosnippet>414</a>
.partition(Struct::empty())
+<a href=#415 id=415 data-nosnippet>415</a> .build()<span
class="question-mark">?</span>;
+<a href=#416 id=416 data-nosnippet>416</a>
+<a href=#417 id=417 data-nosnippet>417</a> <span class="kw">let
</span>data_file2 = DataFileBuilder::default()
+<a href=#418 id=418 data-nosnippet>418</a>
.content(DataContentType::Data)
+<a href=#419 id=419 data-nosnippet>419</a> .file_path(<span
class="string">"path/to/file2.parquet"</span>.to_string())
+<a href=#420 id=420 data-nosnippet>420</a>
.file_format(DataFileFormat::Parquet)
+<a href=#421 id=421 data-nosnippet>421</a>
.file_size_in_bytes(<span class="number">2048</span>)
+<a href=#422 id=422 data-nosnippet>422</a> .record_count(<span
class="number">200</span>)
+<a href=#423 id=423 data-nosnippet>423</a>
.partition_spec_id(table.metadata().default_partition_spec_id())
+<a href=#424 id=424 data-nosnippet>424</a>
.partition(Struct::empty())
+<a href=#425 id=425 data-nosnippet>425</a> .build()<span
class="question-mark">?</span>;
+<a href=#426 id=426 data-nosnippet>426</a>
+<a href=#427 id=427 data-nosnippet>427</a> <span class="comment">//
Serialize data files to JSON
+<a href=#428 id=428 data-nosnippet>428</a> </span><span class="kw">let
</span>partition_type = table.metadata().default_partition_type().clone();
+<a href=#429 id=429 data-nosnippet>429</a> <span class="kw">let
</span>data_file1_json = iceberg::spec::serialize_data_file_to_json(
+<a href=#430 id=430 data-nosnippet>430</a> data_file1.clone(),
+<a href=#431 id=431 data-nosnippet>431</a> <span
class="kw-2">&</span>partition_type,
+<a href=#432 id=432 data-nosnippet>432</a>
table.metadata().format_version(),
+<a href=#433 id=433 data-nosnippet>433</a> )<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_json = iceberg::spec::serialize_data_file_to_json(
+<a href=#436 id=436 data-nosnippet>436</a> data_file2.clone(),
+<a href=#437 id=437 data-nosnippet>437</a> <span
class="kw-2">&</span>partition_type,
+<a href=#438 id=438 data-nosnippet>438</a>
table.metadata().format_version(),
+<a href=#439 id=439 data-nosnippet>439</a> )<span
class="question-mark">?</span>;
+<a href=#440 id=440 data-nosnippet>440</a>
+<a href=#441 id=441 data-nosnippet>441</a> <span class="comment">//
Create a mock execution plan that returns the serialized data files
+<a href=#442 id=442 data-nosnippet>442</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=#443 id=443 data-nosnippet>443</a>
+<a href=#444 id=444 data-nosnippet>444</a> <span class="comment">//
Create the IcebergCommitExec
+<a href=#445 id=445 data-nosnippet>445</a> </span><span class="kw">let
</span>arrow_schema = Arc::new(ArrowSchema::new(<span
class="macro">vec!</span>[Field::new(
+<a href=#446 id=446 data-nosnippet>446</a> DATA_FILES_COL_NAME,
+<a href=#447 id=447 data-nosnippet>447</a> DataType::Utf8,
+<a href=#448 id=448 data-nosnippet>448</a> <span
class="bool-val">false</span>,
+<a href=#449 id=449 data-nosnippet>449</a> )]));
+<a href=#450 id=450 data-nosnippet>450</a>
+<a href=#451 id=451 data-nosnippet>451</a> <span class="kw">let
</span>commit_exec =
+<a href=#452 id=452 data-nosnippet>452</a>
IcebergCommitExec::new(table.clone(), catalog.clone(), input_exec,
arrow_schema);
+<a href=#453 id=453 data-nosnippet>453</a>
+<a href=#454 id=454 data-nosnippet>454</a> <span class="comment">//
Execute the commit exec
+<a href=#455 id=455 data-nosnippet>455</a> </span><span class="kw">let
</span>task_ctx = Arc::new(TaskContext::default());
+<a href=#456 id=456 data-nosnippet>456</a> <span class="kw">let
</span>stream = commit_exec.execute(<span class="number">0</span>,
task_ctx)<span class="question-mark">?</span>;
+<a href=#457 id=457 data-nosnippet>457</a> <span class="kw">let
</span>batches = collect(stream).<span class="kw">await</span><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">//
Verify the results
+<a href=#460 id=460 data-nosnippet>460</a> </span><span
class="macro">assert_eq!</span>(batches.len(), <span class="number">1</span>);
+<a href=#461 id=461 data-nosnippet>461</a> <span class="kw">let
</span>batch = <span class="kw-2">&</span>batches[<span
class="number">0</span>];
+<a href=#462 id=462 data-nosnippet>462</a> <span
class="macro">assert_eq!</span>(batch.num_columns(), <span
class="number">1</span>);
+<a href=#463 id=463 data-nosnippet>463</a> <span
class="macro">assert_eq!</span>(batch.num_rows(), <span
class="number">1</span>);
+<a href=#464 id=464 data-nosnippet>464</a>
+<a href=#465 id=465 data-nosnippet>465</a> <span class="comment">// The
output should be a record batch with a single column "count" and a single row
+<a href=#466 id=466 data-nosnippet>466</a> // with the total record
count (100 + 200 = 300)
+<a href=#467 id=467 data-nosnippet>467</a> </span><span class="kw">let
</span>count_array = batch.column(<span class="number">0</span>);
+<a href=#468 id=468 data-nosnippet>468</a> <span
class="macro">assert_eq!</span>(count_array.len(), <span
class="number">1</span>);
+<a href=#469 id=469 data-nosnippet>469</a> <span
class="macro">assert_eq!</span>(count_array.data_type(), <span
class="kw-2">&</span>DataType::UInt64);
+<a href=#470 id=470 data-nosnippet>470</a>
+<a href=#471 id=471 data-nosnippet>471</a> <span class="comment">//
Verify that the count is correct
+<a href=#472 id=472 data-nosnippet>472</a> </span><span class="kw">let
</span>count =
count_array.as_any().downcast_ref::<UInt64Array>().unwrap();
+<a href=#473 id=473 data-nosnippet>473</a> <span
class="macro">assert_eq!</span>(count.value(<span class="number">0</span>),
<span class="number">300</span>);
+<a href=#474 id=474 data-nosnippet>474</a>
+<a href=#475 id=475 data-nosnippet>475</a> <span class="comment">//
Verify that the table has been updated with the new files
+<a href=#476 id=476 data-nosnippet>476</a> </span><span class="kw">let
</span>updated_table = catalog
+<a href=#477 id=477 data-nosnippet>477</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=#478 id=478 data-nosnippet>478</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#479 id=479 data-nosnippet>479</a> <span class="kw">let
</span>current_snapshot = updated_table.metadata().current_snapshot().unwrap();
+<a href=#480 id=480 data-nosnippet>480</a>
+<a href=#481 id=481 data-nosnippet>481</a> <span class="comment">//
Load the manifest list to verify the data files were added
+<a href=#482 id=482 data-nosnippet>482</a> </span><span class="kw">let
</span>manifest_list = current_snapshot
+<a href=#483 id=483 data-nosnippet>483</a>
.load_manifest_list(updated_table.file_io(), updated_table.metadata())
+<a href=#484 id=484 data-nosnippet>484</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#485 id=485 data-nosnippet>485</a>
+<a href=#486 id=486 data-nosnippet>486</a> <span class="comment">//
There should be at least one manifest
+<a href=#487 id=487 data-nosnippet>487</a> </span><span
class="macro">assert!</span>(!manifest_list.entries().is_empty());
+<a href=#488 id=488 data-nosnippet>488</a>
+<a href=#489 id=489 data-nosnippet>489</a> <span class="comment">//
Load the first manifest and verify it contains our data files
+<a href=#490 id=490 data-nosnippet>490</a> </span><span class="kw">let
</span>manifest = manifest_list.entries()[<span class="number">0</span>]
+<a href=#491 id=491 data-nosnippet>491</a>
.load_manifest(updated_table.file_io())
+<a href=#492 id=492 data-nosnippet>492</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#493 id=493 data-nosnippet>493</a>
+<a href=#494 id=494 data-nosnippet>494</a> <span class="comment">//
Verify that the manifest contains our data files
+<a href=#495 id=495 data-nosnippet>495</a> </span><span class="kw">let
</span>manifest_files: Vec<String> = manifest
+<a href=#496 id=496 data-nosnippet>496</a> .entries()
+<a href=#497 id=497 data-nosnippet>497</a> .iter()
+<a href=#498 id=498 data-nosnippet>498</a> .map(|entry|
entry.data_file().file_path().to_string())
+<a href=#499 id=499 data-nosnippet>499</a> .collect();
+<a href=#500 id=500 data-nosnippet>500</a>
+<a href=#501 id=501 data-nosnippet>501</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=#502 id=502 data-nosnippet>502</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=#503 id=503 data-nosnippet>503</a>
+<a href=#504 id=504 data-nosnippet>504</a> <span
class="prelude-val">Ok</span>(())
+<a href=#505 id=505 data-nosnippet>505</a> }
+<a href=#506 id=506
data-nosnippet>506</a>}</code></pre></div></section></main></body></html>
\ No newline at end of file
diff --git a/api/src/iceberg_datafusion/physical_plan/mod.rs.html
b/api/src/iceberg_datafusion/physical_plan/mod.rs.html
index f39a08a0..81d4762b 100644
--- a/api/src/iceberg_datafusion/physical_plan/mod.rs.html
+++ b/api/src/iceberg_datafusion/physical_plan/mod.rs.html
@@ -15,7 +15,11 @@
<a href=#15 id=15 data-nosnippet>15</a>// specific language governing
permissions and limitations
<a href=#16 id=16 data-nosnippet>16</a>// under the License.
<a href=#17 id=17 data-nosnippet>17</a>
-<a href=#18 id=18 data-nosnippet>18</a></span><span
class="kw">pub</span>(<span class="kw">crate</span>) <span class="kw">mod
</span>expr_to_predicate;
-<a href=#19 id=19 data-nosnippet>19</a><span class="kw">pub</span>(<span
class="kw">crate</span>) <span class="kw">mod </span>metadata_scan;
-<a href=#20 id=20 data-nosnippet>20</a><span class="kw">pub</span>(<span
class="kw">crate</span>) <span class="kw">mod </span>scan;
-<a href=#21 id=21 data-nosnippet>21</a><span class="kw">pub use
</span>scan::IcebergTableScan;</code></pre></div></section></main></body></html>
\ No newline at end of file
+<a href=#18 id=18 data-nosnippet>18</a></span><span
class="kw">pub</span>(<span class="kw">crate</span>) <span class="kw">mod
</span>commit;
+<a href=#19 id=19 data-nosnippet>19</a><span class="kw">pub</span>(<span
class="kw">crate</span>) <span class="kw">mod </span>expr_to_predicate;
+<a href=#20 id=20 data-nosnippet>20</a><span class="kw">pub</span>(<span
class="kw">crate</span>) <span class="kw">mod </span>metadata_scan;
+<a href=#21 id=21 data-nosnippet>21</a><span class="kw">pub</span>(<span
class="kw">crate</span>) <span class="kw">mod </span>scan;
+<a href=#22 id=22 data-nosnippet>22</a>
+<a href=#23 id=23 data-nosnippet>23</a><span class="kw">pub</span>(<span
class="kw">crate</span>) <span class="kw">const </span>DATA_FILES_COL_NAME:
<span class="kw-2">&</span>str = <span class="string">"data_files"</span>;
+<a href=#24 id=24 data-nosnippet>24</a>
+<a href=#25 id=25 data-nosnippet>25</a><span class="kw">pub use
</span>scan::IcebergTableScan;</code></pre></div></section></main></body></html>
\ No newline at end of file
