refactor(flow-php/etl): rename entry to column

benchmarks/src/Joining/JoinEachScenario.php

@@ -24,7 +24,7 @@ public function run(): void
         data_frame(BenchmarkConfig::builder())
             ->read(from_floe(Datasets::orders($this->rows)->floe()))
             ->batchSize(1000)
-            ->withEntry('seller_id', ref('seller_id')->cast(type_string()))
+            ->withColumn('seller_id', ref('seller_id')->cast(type_string()))
             ->joinEach(new SellersDataFrameFactory(Datasets::sellers($this->rows)->parquet()), join_on([
                 'seller_id' => 'id',
             ], join_prefix: 'sellers_'))

benchmarks/src/Joining/JoinOrdersWithSellersScenario.php

@@ -27,7 +27,7 @@ public function run(): void
         data_frame(BenchmarkConfig::builder())
             ->read(from_floe(Datasets::orders($this->rows)->floe()))
             ->batchSize(1000)
-            ->withEntry('seller_id', ref('seller_id')->cast(type_string()))
+            ->withColumn('seller_id', ref('seller_id')->cast(type_string()))
             ->join(
                 data_frame()->read(from_parquet(Datasets::sellers($this->rows)->parquet())),
                 join_on(['seller_id' => 'id'], join_prefix: 'sellers_'),

benchmarks/src/Partitioning/PartitionCardinality.php

@@ -32,7 +32,7 @@ public function column(): string
      */
     public function derive(DataFrame $frame): DataFrame
     {
-        return $frame->withEntry('day', ref('created_at')->dateFormat('Y-m-d'));
+        return $frame->withColumn('day', ref('created_at')->dateFormat('Y-m-d'));
     }
 
     public function reference(): Reference

benchmarks/src/Pipeline/CanonicalPipeline.php

@@ -18,7 +18,7 @@
  * The one pipeline shape every source is measured through, so a source's number differs only by its
  * read leg. Every subject that claims to be comparable with the others must run through here.
  *
- * withEntry is the only verb that adds a column, so it forces schema widening at plan-bind time.
+ * withColumn is the only verb that adds a column, so it forces schema widening at plan-bind time.
  * filter is deliberately non-selective - created_at is never null - so the row count reaching the sink
  * cannot move for a second reason. select re-converges every source onto the same five columns, so the
  * sink does identical work whatever was read. Floe stores typed values, adding no stringification the
@@ -42,7 +42,7 @@ public function run(): void
     {
         data_frame(BenchmarkConfig::builder())
             ->read($this->extractor)
-            ->withEntry('contact', ref('customer')->concat(lit(' <'), ref('email'), lit('>')))
+            ->withColumn('contact', ref('customer')->concat(lit(' <'), ref('email'), lit('>')))
             ->filter(ref('created_at')->isNotNull())
             ->select('order_id', 'seller_id', 'created_at', 'customer', 'contact')
             ->write(to_floe($this->sink))

benchmarks/src/Pipeline/NodePipelineScenario.php

@@ -46,10 +46,10 @@ public function run(): void
                 ->read(from_xml($this->source->path($this->rows), self::XML_NODE_PATH)->withSchema(schema(xml_schema(
                     'node',
                 ))))
-                ->withEntry('body', ref('node')->cast(type_string())),
+                ->withColumn('body', ref('node')->cast(type_string())),
             NodeSource::text => data_frame(BenchmarkConfig::builder())
                 ->read(from_text($this->source->path($this->rows))->withSchema(schema(str_schema('text'))))
-                ->withEntry('body', ref('text')->concat(lit(' |'))),
+                ->withColumn('body', ref('text')->concat(lit(' |'))),
         };
 
         $frame

benchmarks/src/Pipeline/PlanDepthScenario.php

@@ -13,7 +13,7 @@
 
 /**
  * schema() answers from the plan without executing it, so this prices Planner::plan() alone (optimize, translate,
- * bind, split) rather than $depth withEntry calls per row.
+ * bind, split) rather than $depth withColumn calls per row.
  *
  * One call plans PLANS identical plans, because a single plan is far below timer resolution.
  */
@@ -33,7 +33,7 @@ public function bind(): Schema
             ->read((new SourceExtractor($this->source, SchemaMode::declared, $this->rows))->extractor());
 
         for ($step = 0; $step < $this->depth; $step++) {
-            $frame = $frame->withEntry('c' . $step, ref('customer')->concat(lit((string) $step)));
+            $frame = $frame->withColumn('c' . $step, ref('customer')->concat(lit((string) $step)));
         }
 
         return $frame->schema();

benchmarks/src/Service/Seal/SealWriteScenario.php

@@ -38,7 +38,7 @@ public function run(): void
         data_frame()
             ->read(from_floe(Datasets::orders($this->rows)->floe()))
             ->batchSize(1000)
-            ->write(to_seal_upsert($engine, $this->index())->withIdentifierEntry('order_id')->withBulkSize(1000))
+            ->write(to_seal_upsert($engine, $this->index())->withIdentifierColumn('order_id')->withBulkSize(1000))
             ->run();
     }
 

benchmarks/src/Window/MovingAverageScenario.php

@@ -29,7 +29,7 @@ public function run(): void
 
         $partitioning
             ->derive(data_frame(BenchmarkConfig::builder())->read(from_floe(Datasets::orders($this->rows)->floe())))
-            ->withEntry(
+            ->withColumn(
                 'moving_average',
                 average(ref('discount'))
                     ->over(

benchmarks/src/Window/RankScenario.php

@@ -27,7 +27,7 @@ public function run(): void
 
         $partitioning
             ->derive(data_frame(BenchmarkConfig::builder())->read(from_floe(Datasets::orders($this->rows)->floe())))
-            ->withEntry(
+            ->withColumn(
                 'rank',
                 rank()->over(window()->partitionBy($partitioning->reference())->orderBy(ref('created_at'))),
             )

benchmarks/src/Window/RowNumberScenario.php

@@ -27,7 +27,7 @@ public function run(): void
 
         $partitioning
             ->derive(data_frame(BenchmarkConfig::builder())->read(from_floe(Datasets::orders($this->rows)->floe())))
-            ->withEntry(
+            ->withColumn(
                 'row_number',
                 row_number()->over(window()->partitionBy($partitioning->reference())->orderBy(ref('created_at'))),
             )

benchmarks/src/Window/RunningTotalScenario.php

@@ -27,7 +27,7 @@ public function run(): void
 
         $partitioning
             ->derive(data_frame(BenchmarkConfig::builder())->read(from_floe(Datasets::orders($this->rows)->floe())))
-            ->withEntry(
+            ->withColumn(
                 'running_total',
                 sum(ref('discount'))
                     ->over(window()->partitionBy($partitioning->reference())->orderBy(ref('created_at'))),

benchmarks/src/Window/WholePartitionScenario.php

@@ -27,7 +27,7 @@ public function run(): void
 
         $partitioning
             ->derive(data_frame(BenchmarkConfig::builder())->read(from_floe(Datasets::orders($this->rows)->floe())))
-            ->withEntry('seller_total', sum(ref('discount'))->over(window()->partitionBy($partitioning->reference())))
+            ->withColumn('seller_total', sum(ref('discount'))->over(window()->partitionBy($partitioning->reference())))
             ->run();
     }
 }

benchmarks/suites/Pipeline/OrdersPipelineBench.php

@@ -9,7 +9,7 @@
 use PhpBench\Attributes as Bench;
 
 /**
- * Rung 4 of the ladder: read -> withEntry -> filter -> select -> write, one subject per source.
+ * Rung 4 of the ladder: read -> withColumn -> filter -> select -> write, one subject per source.
  *
  * Parquet, Floe, Doctrine and PostgreSQL have no inferred arm on purpose. They declare from metadata
  * (footer / driver describe-query) and never sample, so the inferred-vs-declared comparison is

benchmarks/tests/Flow/Benchmarks/Tests/Unit/Pipeline/OrdersSchemaTest.php

@@ -29,7 +29,7 @@ public function test_every_column_is_nullable_on_every_source(): void
     {
         foreach (Source::cases() as $source) {
             foreach (OrdersSchema::of($source)->definitions() as $definition) {
-                static::assertTrue($definition->isNullable(), $source->value . '.' . $definition->entry()->name());
+                static::assertTrue($definition->isNullable(), $source->value . '.' . $definition->reference()->name());
             }
         }
     }
@@ -40,7 +40,7 @@ public function test_the_five_projected_columns_exist_on_every_source(): void
             $names = [];
 
             foreach (OrdersSchema::of($source)->definitions() as $definition) {
-                $names[] = $definition->entry()->name();
+                $names[] = $definition->reference()->name();
             }
 
             foreach (['order_id', 'seller_id', 'created_at', 'customer', 'email'] as $projected) {

documentation/adrs/variadic-arguments-pattern.md

@@ -17,13 +17,13 @@ Several Flow PHP methods accept multiple arguments where at least one is require
 A proposal was made to simplify the API signatures from:
 
 ```php ignore
-public function partitionBy(string|Reference $entry, string|Reference ...$entries) : self
+public function partitionBy(string|Reference $column, string|Reference ...$columns) : self
 ```
 
 To:
 
 ```php ignore
-public function partitionBy(string|Reference ...$entries) : self
+public function partitionBy(string|Reference ...$columns) : self
 ```
 
 The rationale for the proposal was to reduce array operations like `array_unshift()` used internally to merge the first required argument with the variadic rest.
@@ -67,10 +67,10 @@ public function method(Type ...$args) : ReturnType
 ### 1. Pure variadic with runtime validation
 
 ```php ignore
-public function partitionBy(string|Reference ...$entries) : self
+public function partitionBy(string|Reference ...$columns) : self
 {
-    if ([] === $entries) {
-        throw new InvalidArgumentException('At least one entry is required.');
+    if ([] === $columns) {
+        throw new InvalidArgumentException('At least one column is required.');
     }
     // ...
 }

documentation/components/adapters/chartjs.md

@@ -43,7 +43,7 @@ $data = [
 
 df()
     ->read(from_array($data))
-    ->withEntry('Profit', ref('Revenue')->minus(ref('CM'))->minus(ref('Ads Spends'))->minus(ref('Storage Costs'))->minus(ref('Shipping Costs'))->round(lit(2)))
+    ->withColumn('Profit', ref('Revenue')->minus(ref('CM'))->minus(ref('Ads Spends'))->minus(ref('Storage Costs'))->minus(ref('Shipping Costs'))->round(lit(2)))
     ->write(
         to_chartjs_file(
             $chart = bar_chart(

documentation/components/adapters/excel.md

@@ -186,7 +186,7 @@ Write each row to the sheet named by its `category` value:
 <?php
 
 $loader = to_excel('path/to/output.xlsx')
-    ->withSheetNameFromEntry('category');
+    ->withSheetNameFromColumn('category');
 
 data_frame()
     ->read($extractor)

documentation/components/adapters/parquet.md

@@ -67,8 +67,8 @@ use function Flow\Types\DSL\type_json;
 
 data_frame()
     ->read(from_parquet(__DIR__ . '/orders.parquet', columns: ['id', 'payload', 'created_at']))
-    ->withEntry('payload', ref('payload')->cast(type_json()))
-    ->withEntry('created_at', to_timezone(ref('created_at'), 'Europe/Warsaw'))
+    ->withColumn('payload', ref('payload')->cast(type_json()))
+    ->withColumn('created_at', to_timezone(ref('created_at'), 'Europe/Warsaw'))
     ->write(to_output())
     ->run();
 ```

documentation/components/adapters/postgresql.md

@@ -483,7 +483,7 @@ Two helpers convert between a Flow `Schema` and a PostgreSQL table definition:
   (and related index/constraint) SQL via `toSql()`.
 - `pgsql_table_to_flow_schema()` turns a `Flow\PostgreSql\Schema\Table` back into a Flow `Schema`.
 
-Column types are resolved through the shared `EntryTypesMap` (Flow type → PostgreSQL column type), and per-column
+Column types are resolved through the shared `TypesMap` (Flow type → PostgreSQL column type), and per-column
 details - primary keys, unique constraints, indexes, length, precision/scale, defaults, identity, generated columns,
 and explicit type overrides - are driven by `PostgreSqlMetadata` entries attached to each schema definition.
 
@@ -625,11 +625,11 @@ $schema = pgsql_table_to_flow_schema($table);
 
 ### Customizing the Type Mapping
 
-Both helpers accept an optional `EntryTypesMap`. Its second constructor argument overrides the Flow type → PostgreSQL
+Both helpers accept an optional `TypesMap`. Its second constructor argument overrides the Flow type → PostgreSQL
 column type mapping (the first argument keeps overriding the value-binding types used by the loader):
 
 ```php
-use Flow\ETL\Adapter\PostgreSql\EntryTypesMap;
+use Flow\ETL\Adapter\PostgreSql\TypesMap;
 use Flow\PostgreSql\QueryBuilder\Schema\ColumnType;
 use Flow\Types\Type\Native\StringType;
 
@@ -638,7 +638,7 @@ use function Flow\ETL\Adapter\PostgreSql\to_pgsql_schema_table;
 $table = to_pgsql_schema_table(
     $schema,
     'users',
-    typesMap: new EntryTypesMap([], [
+    typesMap: new TypesMap([], [
         StringType::class => ColumnType::varchar(255),  // default strings to varchar(255) instead of text
     ]),
 );

documentation/components/adapters/seal.md

@@ -97,7 +97,7 @@ to_seal_upsert($engine, 'users')->withBulkSize(500);
 
 ### Deleting Documents
 
-`to_seal_delete()` removes documents by their identifiers, read from the `id` column. `withIdentifierEntry()` names
+`to_seal_delete()` removes documents by their identifiers, read from the `id` column. `withIdentifierColumn()` names
 another column.
 
 ```php
@@ -111,7 +111,7 @@ data_frame()
 ```
 
 ```php
-to_seal_delete($engine, 'products')->withIdentifierEntry('sku');
+to_seal_delete($engine, 'products')->withIdentifierColumn('sku');
 ```
 
 ### Data Normalization

documentation/components/adapters/xml.md

@@ -45,7 +45,7 @@ use function Flow\ETL\DSL\{data_frame, ref, to_output};
 
 data_frame()
     ->read(from_xml(__DIR__ . '/simple_items.xml')->withXMLNodePath('root/items/item'))
-    ->withEntry('id', ref('node')->xpath('id')->domElementValue())
+    ->withColumn('id', ref('node')->xpath('id')->domElementValue())
     ->write(to_output(truncate: false))
     ->run();
 ```

documentation/components/bridges/symfony-http-foundation-bridge.md

@@ -96,7 +96,7 @@ use function Flow\ETL\DSL\lit;
 new class implements Transformation {
     public function transform(DataFrame $dataFrame): DataFrame
     {
-        return $dataFrame->withEntry('time', lit(\time()));
+        return $dataFrame->withColumn('time', lit(\time()));
     }
 };
 ```

documentation/components/cli/docs.md

@@ -112,7 +112,7 @@ Options:
       --analyze[=ANALYZE]                                            Collect processing statistics and print them. [default: false]
       --config=CONFIG                                                Path to a local php file that MUST return instance of: Flow\ETL\Config
       --input-json-pointer=INPUT-JSON-POINTER                        JSON Pointer to a subtree from which schema should be extracted
-      --input-json-pointer-entry-name                                When set, JSON Pointer will be used as an entry name in the schema
+      --input-json-pointer-column-name                               When set, JSON Pointer will be used as a column name in the schema
       --output-json-date-time-format=OUTPUT-JSON-DATE-TIME-FORMAT    Date time format used in JSON output
       --output-json-rows-in-new-line                                 When set, each row will be printed in a new line
       --input-csv-header[=INPUT-CSV-HEADER]                          When set, CSV header will be used as a schema
@@ -174,7 +174,7 @@ Options:
       --output-ascii                                       Print schema as ascii list
       --config=CONFIG                                      Path to a local php file that MUST return instance of: Flow\ETL\Config
       --input-json-pointer=INPUT-JSON-POINTER              JSON Pointer to a subtree from which schema should be extracted
-      --input-json-pointer-entry-name                      When set, JSON Pointer will be used as an entry name in the schema
+      --input-json-pointer-column-name                     When set, JSON Pointer will be used as a column name in the schema
       --input-csv-header[=INPUT-CSV-HEADER]                When set, CSV header will be used as a schema
       --input-csv-empty-to-null[=INPUT-CSV-EMPTY-TO-NULL]  When set, empty CSV values will be treated as NULL values
       --input-csv-separator=INPUT-CSV-SEPARATOR            CSV separator character
@@ -243,7 +243,7 @@ Options:
       --input-file-offset=INPUT-FILE-OFFSET                Number of rows to skip before starting to read data
       --config=CONFIG                                      Path to a local php file that MUST return instance of: Flow\ETL\Config
       --input-json-pointer=INPUT-JSON-POINTER              JSON Pointer to a subtree from which schema should be extracted
-      --input-json-pointer-entry-name                      When set, JSON Pointer will be used as an entry name in the schema
+      --input-json-pointer-column-name                     When set, JSON Pointer will be used as a column name in the schema
       --input-csv-header[=INPUT-CSV-HEADER]                When set, CSV header will be used as a schema
       --input-csv-empty-to-null[=INPUT-CSV-EMPTY-TO-NULL]  When set, empty CSV values will be treated as NULL values
       --input-csv-separator=INPUT-CSV-SEPARATOR            CSV separator character
@@ -295,7 +295,7 @@ Options:
       --output-columns=OUTPUT-COLUMNS                      Columns to include in output, when not set all columns are displayed (multiple values allowed)
       --config=CONFIG                                      Path to a local php file that MUST return instance of: Flow\ETL\Config
       --input-json-pointer=INPUT-JSON-POINTER              JSON Pointer to a subtree from which schema should be extracted
-      --input-json-pointer-entry-name                      When set, JSON Pointer will be used as an entry name in the schema
+      --input-json-pointer-column-name                     When set, JSON Pointer will be used as a column name in the schema
       --input-csv-header[=INPUT-CSV-HEADER]                When set, CSV header will be used as a schema
       --input-csv-empty-to-null[=INPUT-CSV-EMPTY-TO-NULL]  When set, empty CSV values will be treated as NULL values
       --input-csv-separator=INPUT-CSV-SEPARATOR            CSV separator character
@@ -341,7 +341,7 @@ Options:
       --input-file-offset=INPUT-FILE-OFFSET                Number of rows to skip before starting to read data
       --config=CONFIG                                      Path to a local php file that MUST return instance of: Flow\ETL\Config
       --input-json-pointer=INPUT-JSON-POINTER              JSON Pointer to a subtree from which schema should be extracted
-      --input-json-pointer-entry-name                      When set, JSON Pointer will be used as an entry name in the schema
+      --input-json-pointer-column-name                     When set, JSON Pointer will be used as a column name in the schema
       --input-excel-header[=INPUT-EXCEL-HEADER]            When set, Excel header will be used as a schema
       --input-excel-sheet-name=INPUT-EXCEL-SHEET-NAME      When set, Excel sheet name will be selected for reading
       --input-excel-offset=INPUT-EXCEL-OFFSET              Offset to start reading from