feat(ingest/s3): Partition support

metadata-ingestion/docs/sources/s3/s3.md

@@ -3,16 +3,31 @@
 
 Path Specs (`path_specs`) is a list of Path Spec (`path_spec`) objects where each individual `path_spec` represents one or more datasets. Include path (`path_spec.include`) represents formatted path to the dataset. This path must end with `*.*` or `*.[ext]` to represent leaf level. If `*.[ext]` is provided then files with only specified extension type will be scanned. "`.[ext]`" can be any of [supported file types](#supported-file-types). Refer [example 1](#example-1---individual-file-as-dataset) below for more details.
 
-All folder levels need to be specified in include path. You can use `/*/` to represent a folder level and avoid specifying exact folder name. To map folder as a dataset, use `{table}` placeholder to represent folder level for which dataset is to be created. For a partitioned dataset, you can use placeholder `{partition_key[i]}` to represent name of `i`th partition and `{partition[i]}` to represent value of `i`th partition. During ingestion, `i` will be used to match partition_key to partition. Refer [example 2 and 3](#example-2---folder-of-files-as-dataset-without-partitions) below for more details.
+All folder levels need to be specified in include path. You can use `/*/` to represent a folder level and avoid specifying exact folder name. To map folder as a dataset, use `{table}` placeholder to represent folder level for which dataset is to be created. For a partitioned dataset, you can use placeholder `{partition_key[i]}` to represent name of `i`th partition and `{partition_value[i]}` to represent value of `i`th partition. During ingestion, `i` will be used to match partition_key to partition. Refer [example 2 and 3](#example-2---folder-of-files-as-dataset-without-partitions) below for more details.
 
 Exclude paths (`path_spec.exclude`) can be used to ignore paths that are not relevant to current `path_spec`. This path cannot have named variables ( `{}` ). Exclude path can have `**` to represent multiple folder levels. Refer [example 4](#example-4---folder-of-files-as-dataset-with-partitions-and-exclude-filter) below for more details.
 
 Refer [example 5](#example-5---advanced---either-individual-file-or-folder-of-files-as-dataset) if your bucket has more complex dataset representation.
 
+
 **Additional points to note**
 - Folder names should not contain {, }, *, / in their names.
 - Named variable {folder} is reserved for internal working. please do not use in named variables.
 
+#### Partitioned Dataset support
+If your dataset is partitioned by the `partition_key`=`partition_value` format, then the partition values are auto-detected.
+
+Otherwise, you can specify partitions in the following way in the path_spec:
+1. Specify partition_key and partition_value in the path like => `{partition_key[0]}={partition_value[0]}/{partition_key[1]}={partition_value[1]}/{partition_key[2]}={partition_value[2]}`
+2. Partition key can be specify using named variables in the path_spec like => `year={year}/month={month}/day={day}`
+3 if the path is in the form of /value1/value2/value3 the source infer partition value from the path and assign partition_0, partition_1, partition_2 etc
+
+Dataset creation time is determined by the creation time of earliest created file in the lowest partition while last updated time is determined by the last updated time of the latest updated file in the highest partition.
+
+How the source determines the highest/lowest partition it is based on the traversal method set in the path_spec.
+- If the traversal method is set to `MAX` then the source will try to find the latest partition by ordering the partitions each level and find the latest partiton. This traversal method won't look for earilest partition/creation time but this is the fastest.
+- If the traversal method is set to `MIN_MAX` then the source will try to find the latest and earliest partition by ordering the partitions each level and find the latest/earliest partiton. This traversal sort folders purely by name therefor it is fast but it doesn't guarantee the latest partition will have the latest created file.
+- If the traversal method is set to `ALL` then the source will try to find the latest and earliest partition by listing all the files in all the partitions and find the creation/last modification time based on the file creations. This is the slowest but for non time partitioned datasets this is the only way to find the latest/earliest partition.
 
 ### Path Specs -  Examples
 #### Example 1 - Individual file as Dataset
@@ -73,7 +88,12 @@ test-bucket
 Path specs config to ingest folders `orders` and `returns` as datasets:
 ```
 path_specs:
-    - include: s3://test-bucket/{table}/{partition_key[0]}={partition[0]}/{partition_key[1]}={partition[1]}/*.parquet
+    - include: s3://test-bucket/{table}/{partition_key[0]}={partition_value[0]}/{partition_key[1]}={partition_value[1]}/*.parquet
+```
+or with partition auto-detection:
+```
+path_specs:
+    - include: s3://test-bucket/{table}/
 ```
 
 One can also use `include: s3://test-bucket/{table}/*/*/*.parquet` here however above format is preferred as it allows declaring partitions explicitly.
@@ -99,11 +119,15 @@ test-bucket
 Path specs config to ingest folder `orders` as dataset but not folder `tmp_orders`:
 ```
 path_specs:
-    - include: s3://test-bucket/{table}/{partition_key[0]}={partition[0]}/{partition_key[1]}={partition[1]}/*.parquet
+    - include: s3://test-bucket/{table}/{partition_key[0]}={partition_value[0]}/{partition_key[1]}={partition_value[1]}/*.parquet
       exclude: 
         - **/tmp_orders/**
 ```
-
+or with partition auto-detection:
+```
+path_specs:
+    - include: s3://test-bucket/{table}/
+```
 
 #### Example 5 - Advanced - Either Individual file OR Folder of files as Dataset
 
@@ -150,6 +174,7 @@ Above config has 3 path_specs and will ingest following datasets
 s3://my-bucket/foo/tests/bar.avro # single file table   
 s3://my-bucket/foo/tests/*.* # mulitple file level tables
 s3://my-bucket/foo/tests/{table}/*.avro #table without partition
+s3://my-bucket/foo/tests/{table}/ #table with partition autodetection. Partition only can be detected if it is in the format of key=value
 s3://my-bucket/foo/tests/{table}/*/*.avro #table where partitions are not specified
 s3://my-bucket/foo/tests/{table}/*.* # table where no partitions as well as data type specified
 s3://my-bucket/{dept}/tests/{table}/*.avro # specifying keywords to be used in display name