Update documentation for COPY command

docs/source/user-guide/sql/dml.md

@@ -35,8 +35,22 @@ TO '<i><b>file_name</i></b>'
 [ OPTIONS( <i><b>option</i></b> [, ... ] ) ]
 </pre>
 
+`STORED AS` specifies the file format the `COPY` command will write. If this
+clause is not specified, it will be inferred from the file extension if possible.
+
+`PARTITIONED BY` specifies the columns to use for partitioning the output files into
+separate hive-style directories.
+
+The output format is determined by the first match of the following rules:
+
+1. Value of `STORED AS`
+2. Value of the `OPTION (FORMAT ..)`
+3. Filename extension (e.g. `foo.parquet` implies `PARQUET` format)
+
 For a detailed list of valid OPTIONS, see [Write Options](write_options).
 
+### Examples
+
 Copy the contents of `source_table` to `file_name.json` in JSON format:
 
 ```sql
@@ -72,6 +86,23 @@ of hive-style partitioned parquet files:
 +-------+
 ```
 
+If the the data contains values of `x` and `y` in column1 and only `a` in
+column2, output files will appear in the following directory structure:
+
+```
+dir_name/
+  column1=x/
+    column2=a/
+      <file>.parquet
+      <file>.parquet
+      ...
+  column1=y/
+    column2=a/
+      <file>.parquet
+      <file>.parquet
+      ...
+```
+
 Run the query `SELECT * from source ORDER BY time` and write the
 results (maintaining the order) to a parquet file named
 `output.parquet` with a maximum parquet row group size of 10MB:
@@ -85,14 +116,10 @@ results (maintaining the order) to a parquet file named
 +-------+
 ```
 
-The output format is determined by the first match of the following rules:
-
-1. Value of `STORED AS`
-2. Value of the `OPTION (FORMAT ..)`
-3. Filename extension (e.g. `foo.parquet` implies `PARQUET` format)
-
 ## INSERT
 
+### Examples
+
 Insert values into a table.
 
 <pre>

docs/source/user-guide/sql/write_options.md

@@ -35,15 +35,15 @@ If inserting to an external table, table specific write options can be specified
 
 ```sql
 CREATE EXTERNAL TABLE
-my_table(a bigint, b bigint)
-STORED AS csv
-COMPRESSION TYPE gzip
-WITH HEADER ROW
-DELIMITER ';'
-LOCATION '/test/location/my_csv_table/'
-OPTIONS(
-NULL_VALUE 'NAN'
-);
+  my_table(a bigint, b bigint)
+  STORED AS csv
+  COMPRESSION TYPE gzip
+  WITH HEADER ROW
+  DELIMITER ';'
+  LOCATION '/test/location/my_csv_table/'
+  OPTIONS(
+    NULL_VALUE 'NAN'
+  )
 ```
 
 When running `INSERT INTO my_table ...`, the options from the `CREATE TABLE` will be respected (gzip compression, special delimiter, and header row included). There will be a single output file if the output path doesn't have folder format, i.e. ending with a `\`. Note that compression, header, and delimiter settings can also be specified within the `OPTIONS` tuple list. Dedicated syntax within the SQL statement always takes precedence over arbitrary option tuples, so if both are specified the `OPTIONS` setting will be ignored. NULL_VALUE is a CSV format specific option that determines how null values should be encoded within the CSV file.
@@ -53,26 +53,18 @@ Finally, options can be passed when running a `COPY` command.
 ```sql
 COPY source_table
 TO 'test/table_with_options'
-(format parquet,
-compression snappy,
-'compression::col1' 'zstd(5)',
-partition_by 'column3, column4'
+OPTIONS (
+  format parquet,
+  compression snappy,
+  'compression::col1' 'zstd(5)',
+  partition_by 'column3, column4'
 )
 ```
 
 In this example, we write the entirety of `source_table` out to a folder of parquet files. One parquet file will be written in parallel to the folder for each partition in the query. The next option `compression` set to `snappy` indicates that unless otherwise specified all columns should use the snappy compression codec. The option `compression::col1` sets an override, so that the column `col1` in the parquet file will use `ZSTD` compression codec with compression level `5`. In general, parquet options which support column specific settings can be specified with the syntax `OPTION::COLUMN.NESTED.PATH`.
 
 ## Available Options
 
-### COPY Specific Options
-
-The following special options are specific to the `COPY` command.
-
-| Option       | Description                                                                                                                                                                         | Default Value |
-| ------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------- |
-| FORMAT       | Specifies the file format COPY query will write out. If there're more than one output file or the format cannot be inferred from the file extension, then FORMAT must be specified. | N/A           |
-| PARTITION_BY | Specifies the columns that the output files should be partitioned by into separate hive-style directories. Value should be a comma separated string literal, e.g. 'col1,col2'       | N/A           |
-
 ### JSON Format Specific Options
 
 The following options are available when writing JSON files. Note: If any unsupported option is specified, an error will be raised and the query will fail.