add

users-events/app/boto3clients.py

@@ -0,0 +1,14 @@
+import os
+
+import boto3
+
+
+def get_dynamodb_client():
+    if os.getenv('AWS_LAMBDA_FUNCTION_NAME'):
+        return boto3.client('dynamodb')
+
+    return boto3.client('dynamodb', endpoint_url='http://127.0.0.1:8000')
+
+
+dynamodb_client = get_dynamodb_client()
+s3_client = boto3.client('s3')

users-events/app/config.py

@@ -0,0 +1,4 @@
+import os
+
+USER_TABLE: str = os.getenv('USER_TABLE')  # type: ignore
+CHUNK_SIZE = 50

users-events/app/csv_utils.py

@@ -0,0 +1,83 @@
+import csv
+from typing import TextIO
+
+from smart_open import open
+
+
+def byte_ranges(
+    csvfile: str,
+    chunk_size: int = 100,
+    **kwargs,
+) -> list[tuple[int, int]]:
+    """Compute byte ranges for reading a CSV file in fixed-size line chunks.
+
+    Returns pairs (start_byte, end_byte) for each fixed-size group of lines.
+
+    Parameters
+    ----------
+    csvfile : str
+        Path to the CSV file, opened in binary mode internally.
+    chunk_size : int, optional
+        Number of lines per chunk. Default is 100.
+    **kwargs :
+        Extra options passed to `open()`, e.g., buffering.
+
+    Returns
+    -------
+    list of tuple[int, int]
+        Byte ranges covering each chunk of lines.
+
+    Example
+    -------
+    >>> byte_ranges("users.csv", chunk_size=500)
+    [(0, 3125), (3126, 6150), (6151, 9124)]
+    """
+    line_offsets = [0]
+
+    with open(csvfile, 'rb', **kwargs) as fp:
+        while True:
+            if not fp.readline():
+                break
+            line_offsets.append(fp.tell())
+
+    total_lines = len(line_offsets) - 1
+    byte_ranges = []
+
+    for start_line in range(1, total_lines + 1, chunk_size):
+        # Calculate the end line index, bounded by total lines
+        end_line = min(start_line + chunk_size - 1, total_lines)
+        # Get byte range for this chunk
+        start_byte = line_offsets[start_line - 1]
+        end_byte = line_offsets[end_line] - 1
+
+        byte_ranges.append((start_byte, end_byte))
+
+    return byte_ranges
+
+
+def detect_delimiter(sample: TextIO) -> str:
+    """Detect the delimiter character used in a CSV file.
+
+    Parameters
+    ----------
+    sample : TextIO
+        A file-like object opened in text mode (e.g., from `open('file.csv')`).
+        Must be readable and at position 0.
+
+    Returns
+    -------
+    str
+        The detected delimiter character (e.g., ',', ';', '\\t').
+
+    Raises
+    ------
+    csv.Error
+        If the file cannot be parsed as CSV or delimiter detection fails.
+    ValueError
+        If the file is empty or contains no detectable delimiter.
+    """
+    sniffer = csv.Sniffer()
+    dialect = sniffer.sniff(sample.read())
+    sample.seek(0)
+
+    return dialect.delimiter

users-events/app/events/batch/csv_into_chunks.py

@@ -0,0 +1,20 @@
+from aws_lambda_powertools.utilities.data_classes import (
+    EventBridgeEvent,
+    event_source,
+)
+from aws_lambda_powertools.utilities.typing import LambdaContext
+
+from boto3clients import s3_client
+from config import CHUNK_SIZE
+from csv_utils import byte_ranges
+
+transport_params = {'client': s3_client}
+
+
+@event_source(data_class=EventBridgeEvent)
+def lambda_handler(event: EventBridgeEvent, context: LambdaContext) -> bool:
+    new_image = event.detail['new_image']
+    csvfile = new_image['s3uri']
+    pairs = byte_ranges(csvfile, CHUNK_SIZE, transport_params=transport_params)
+
+    return True

users-events/app/events/batch/excel_to_csv.py

@@ -0,0 +1,14 @@
+from aws_lambda_powertools.utilities.data_classes import (
+    EventBridgeEvent,
+    event_source,
+)
+from aws_lambda_powertools.utilities.typing import LambdaContext
+
+from boto3clients import s3_client
+
+transport_params = {'client': s3_client}
+
+
+@event_source(data_class=EventBridgeEvent)
+def lambda_handler(event: EventBridgeEvent, context: LambdaContext) -> bool:
+    return True

users-events/app/events/batch/read_csv_chunk.py

@@ -0,0 +1,55 @@
+import csv
+from io import StringIO
+from typing import TYPE_CHECKING
+
+from aws_lambda_powertools.utilities.data_classes import (
+    EventBridgeEvent,
+    event_source,
+)
+from aws_lambda_powertools.utilities.typing import LambdaContext
+
+from boto3clients import s3_client
+
+if TYPE_CHECKING:
+    from mypy_boto3_s3.client import S3Client
+else:
+    S3Client = object
+
+transport_params = {'client': s3_client}
+
+
+@event_source(data_class=EventBridgeEvent)
+def lambda_handler(event: EventBridgeEvent, context: LambdaContext) -> bool:
+    new_image = event.detail['new_image']
+    csvfile = new_image['s3_uri']
+
+    data = _get_s3_object_range(
+        csvfile,
+        start_byte=new_image['start_byte'],
+        end_byte=new_image['end_byte'],
+        s3_client=s3_client,
+    )
+    reader = csv.reader(data)
+
+    for x in reader:
+        print(x)
+
+    return True
+
+
+def _get_s3_object_range(
+    s3_uri: str,
+    *,
+    start_byte: int,
+    end_byte: int,
+    s3_client: S3Client,
+) -> StringIO:
+    bucket, key = s3_uri.replace('s3://', '').split('/', 1)
+
+    response = s3_client.get_object(
+        Bucket=bucket,
+        Key=key,
+        Range=f'bytes={start_byte}-{end_byte}',
+    )
+
+    return StringIO(response['Body'].read().decode('utf-8'))

users-events/app/events/email_receiving.py

@@ -0,0 +1,40 @@
+import urllib.parse as urllib_parse
+from email.utils import parseaddr
+
+from aws_lambda_powertools import Logger
+from aws_lambda_powertools.utilities.data_classes import SESEvent, event_source
+from aws_lambda_powertools.utilities.typing import LambdaContext
+from layercake.dynamodb import DynamoDBPersistenceLayer, KeyPair, SortKey
+
+from boto3clients import dynamodb_client
+from config import USER_TABLE
+from ses_utils import get_header_value
+
+logger = Logger(__name__)
+user_layer = DynamoDBPersistenceLayer(USER_TABLE, dynamodb_client)
+
+
+@logger.inject_lambda_context
+@event_source(data_class=SESEvent)
+def lambda_handler(event: SESEvent, context: LambdaContext) -> dict:
+    ses = event.record.ses
+    to = urllib_parse.unquote(ses.receipt.recipients[0]).lower()
+    name, email_from = parseaddr(get_header_value(ses.mail.headers, 'from'))
+
+    org_id = user_layer.collection.get_item(
+        KeyPair('email', SortKey(to, path_spec='user_id')),
+        raise_on_error=False,
+        default={},
+    )
+
+    if not org_id:
+        return {'disposition': 'STOP_RULE_SET'}
+
+    print(
+        {
+            'id': f'mailbox#{org_id}',
+            'sk': ses.mail.message_id,
+        }
+    )
+
+    return {'disposition': 'CONTINUE'}

users-events/app/ses_utils.py

@@ -0,0 +1,20 @@
+from typing import Any, Iterator
+
+from aws_lambda_powertools.utilities.data_classes.ses_event import SESMailHeader
+
+
+def get_header_value(
+    headers: Iterator[SESMailHeader],
+    header_name: str,
+    *,
+    default: Any = None,
+    raise_on_missing: bool = True,
+) -> str:
+    for header in headers:
+        if header.name.lower() == header_name:
+            return header.value
+
+    if raise_on_missing:
+        raise ValueError(f'{header_name} not found.')
+
+    return default