feat: attachment processing for emails (#855)

* process attachments for email

* add attachment processing to msg

* fix up metadata for attachments

* add test for processing email attachments

* added test for processing msg attachments

* update docs

* tests for error conditions

* version and changelog

CHANGELOG.md

@@ -2,6 +2,9 @@
 
 ### Enhancements
 
+* `partition_email` and `partition_msg` will now process attachments if `process_attachments=True`
+  and a attachment partitioning functions is passed through with `attachment_partitioner=partition`.
+
 ### Features
 
 ### Fixes

README.md

@@ -98,8 +98,8 @@ about the library.
 | Document Type | Partition Function | Strategies | Table Support | Options |
 | --- | --- | --- | --- | --- |
 | CSV Files (`.csv`) | `partition_csv` | N/A | Yes | None |
-| E-mails (`.eml`) | `partition_eml` | N/A | No | Encoding; Max Partition |
+| E-mails (`.eml`) | `partition_eml` | N/A | No | Encoding; Max Partition; Process Attachments |
-| E-mails (`.msg`) | `partition_msg` | N/A | No | Encoding; Max Partition |
+| E-mails (`.msg`) | `partition_msg` | N/A | No | Encoding; Max Partition; Process Attachments |
 | EPubs (`.epub`) | `partition_epub` | N/A | Yes | Include Page Breaks |
 | Excel Documents (`.xlsx`/`.xls`) | `partition_xlsx` | N/A | Yes | None |
 | HTML Pages (`.html`) | `partition_html` | N/A | No | Encoding; Include Page Breaks |

docs/source/bricks.rst

@@ -273,6 +273,22 @@ the average character length for a paragraph.
 You can disable ``max_partition`` by setting it to ``None``.
 
 
+You can optionally partition e-mail attachments by setting ``process_attachments=True``.
+If you set ``process_attachments=True``, you'll also need to pass in a partitioning
+function to ``attachment_partitioner``. The following is an example of what the
+workflow looks like:
+
+.. code:: python
+
+  from unstructured.partition.auto import partition
+  from unstructured.partition.email import partition_email
+
+  filename = "example-docs/eml/fake-email-attachment.eml"
+  elements = partition_email(
+    filename=filename, process_attachments=True, attachment_partitioner=partition
+  )
+
+
 ``partition_epub``
 ---------------------
 
@@ -439,6 +455,22 @@ the average character length for a paragraph.
 You can disable ``max_partition`` by setting it to ``None``.
 
 
+You can optionally partition e-mail attachments by setting ``process_attachments=True``.
+If you set ``process_attachments=True``, you'll also need to pass in a partitioning
+function to ``attachment_partitioner``. The following is an example of what the
+workflow looks like:
+
+.. code:: python
+
+  from unstructured.partition.auto import partition
+  from unstructured.partition.msg import partition_msg
+
+  filename = "example-docs/fake-email-attachment.msg"
+  elements = partition_msg(
+    filename=filename, process_attachments=True, attachment_partitioner=partition
+  )
+
+
 ``partition_multiple_via_api``
 ------------------------------
 

test_unstructured/partition/test_email.py

@@ -25,6 +25,7 @@ from unstructured.partition.email import (
     partition_email,
     partition_email_header,
 )
+from unstructured.partition.text import partition_text
 
 FILE_DIRECTORY = pathlib.Path(__file__).parent.resolve()
 EXAMPLE_DOCS_DIRECTORY = os.path.join(FILE_DIRECTORY, "..", "..", "example-docs", "eml")
@@ -326,3 +327,42 @@ def test_partition_email_still_works_with_no_content():
     filename = os.path.join(EXAMPLE_DOCS_DIRECTORY, "email-no-html-content-1.eml")
     elements = partition_email(filename=filename)
     assert elements == []
+
+
+def test_partition_email_can_process_attachments(
+    tmpdir,
+    filename="example-docs/eml/fake-email-attachment.eml",
+):
+    with open(filename) as f:
+        msg = email.message_from_file(f)
+    extract_attachment_info(msg, output_dir=tmpdir.dirname)
+    attachment_filename = os.path.join(tmpdir.dirname, ATTACH_EXPECTED_OUTPUT[0]["filename"])
+    attachment_elements = partition_text(
+        filename=attachment_filename,
+        metadata_filename=attachment_filename,
+    )
+    expected_metadata = attachment_elements[0].metadata
+    expected_metadata.file_directory = None
+    expected_metadata.attached_to_filename = filename
+
+    elements = partition_email(
+        filename=filename,
+        attachment_partitioner=partition_text,
+        process_attachments=True,
+    )
+
+    assert elements[0].text.startswith("Hello!")
+
+    for element in elements[:-1]:
+        assert element.metadata.filename == "fake-email-attachment.eml"
+        assert element.metadata.subject == "Fake email with attachment"
+
+    assert elements[-1].text == "Hey this is a fake attachment!"
+    assert elements[-1].metadata == expected_metadata
+
+
+def test_partition_msg_raises_with_no_partitioner(
+    filename="example-docs/eml/fake-email-attachment.eml",
+):
+    with pytest.raises(ValueError):
+        partition_email(filename=filename, process_attachments=True)

test_unstructured/partition/test_msg.py

@@ -11,6 +11,7 @@ from unstructured.documents.elements import (
     Title,
 )
 from unstructured.partition.msg import extract_msg_attachment_info, partition_msg
+from unstructured.partition.text import partition_text
 
 DIRECTORY = pathlib.Path(__file__).parent.resolve()
 EXAMPLE_DOCS_DIRECTORY = os.path.join(DIRECTORY, "..", "..", "example-docs")
@@ -94,3 +95,40 @@ def test_partition_msg_raises_with_both_specified():
 def test_partition_msg_raises_with_neither():
     with pytest.raises(ValueError):
         partition_msg()
+
+
+def test_partition_msg_can_process_attachments(
+    tmpdir,
+    filename="example-docs/fake-email-attachment.msg",
+):
+    extract_msg_attachment_info(filename=filename, output_dir=tmpdir.dirname)
+    attachment_filename = os.path.join(tmpdir.dirname, ATTACH_EXPECTED_OUTPUT[0]["filename"])
+    attachment_elements = partition_text(
+        filename=attachment_filename,
+        metadata_filename=attachment_filename,
+    )
+    expected_metadata = attachment_elements[0].metadata
+    expected_metadata.file_directory = None
+    expected_metadata.attached_to_filename = filename
+
+    elements = partition_msg(
+        filename=filename,
+        attachment_partitioner=partition_text,
+        process_attachments=True,
+    )
+
+    assert elements[0].text.startswith("Hello!")
+
+    for element in elements[:-1]:
+        assert element.metadata.filename == "fake-email-attachment.msg"
+        assert element.metadata.subject == "Fake email with attachment"
+
+    assert elements[-1].text == "Hey this is a fake attachment!"
+    assert elements[-1].metadata == expected_metadata
+
+
+def test_partition_msg_raises_with_no_partitioner(
+    filename="example-docs/fake-email-attachment.msg",
+):
+    with pytest.raises(ValueError):
+        partition_msg(filename=filename, process_attachments=True)

unstructured/documents/elements.py

@@ -50,6 +50,7 @@ class ElementMetadata:
     file_directory: Optional[str] = None
     date: Optional[str] = None
     filetype: Optional[str] = None
+    attached_to_filename: Optional[str] = None
 
     # Page numbers currenlty supported for PDF, HTML and PPT documents
     page_number: Optional[int] = None

unstructured/file_utils/filetype.py

@@ -516,11 +516,14 @@ def add_metadata_with_filetype(filetype: FileType):
                     kwarg: params.get(kwarg) for kwarg in ("filename", "url", "text_as_html")
                 }
                 for element in elements:
-                    _add_element_metadata(
+                    # NOTE(robinson) - Attached files have already run through this logic
-                        element,
+                    # in their own partitioning function
-                        filetype=FILETYPE_TO_MIMETYPE[filetype],
+                    if element.metadata.attached_to_filename is None:
-                        **metadata_kwargs,  # type: ignore
+                        _add_element_metadata(
-                    )
+                            element,
+                            filetype=FILETYPE_TO_MIMETYPE[filetype],
+                            **metadata_kwargs,  # type: ignore
+                        )
 
                 return elements
             else:

unstructured/partition/email.py

@@ -1,11 +1,12 @@
 import datetime
 import email
+import os
 import re
 import sys
 from email.message import Message
 from functools import partial
-from tempfile import SpooledTemporaryFile
+from tempfile import SpooledTemporaryFile, TemporaryDirectory
-from typing import IO, Dict, List, Optional, Tuple, Union
+from typing import IO, Callable, Dict, List, Optional, Tuple, Union
 
 from unstructured.file_utils.encoding import (
     COMMON_ENCODINGS,
@@ -226,6 +227,9 @@ def partition_email(
     encoding: Optional[str] = None,
     include_headers: bool = False,
     max_partition: Optional[int] = 1500,
+    metadata_filename: Optional[str] = None,
+    process_attachments: bool = False,
+    attachment_partitioner: Optional[Callable] = None,
     **kwargs,
 ) -> List[Element]:
     """Partitions an .eml documents into its constituent elements.
@@ -245,6 +249,13 @@ def partition_email(
     max_partition
         The maximum number of characters to include in a partition. If None is passed,
         no maximum is applied. Only applies if processing the text/plain content.
+    metadata_filename
+        The filename to use for the metadata.
+    process_attachments
+        If True, partition_email will process email attachments in addition to
+        processing the content of the email itself.
+    attachment_partitioner
+        The partitioning function to use to process attachments.
     """
     if content_source not in VALID_CONTENT_SOURCES:
         raise ValueError(
@@ -258,6 +269,8 @@ def partition_email(
     # Verify that only one of the arguments was provided
     exactly_one(filename=filename, file=file, text=text)
 
+    metadata_filename = metadata_filename or filename
+
     detected_encoding = "utf-8"
     if filename is not None:
         extracted_encoding, msg = parse_email(filename=filename)
@@ -341,7 +354,25 @@ def partition_email(
         header = partition_email_header(msg)
     all_elements = header + elements
 
-    metadata = build_email_metadata(msg, filename=filename)
+    metadata = build_email_metadata(msg, filename=metadata_filename)
     for element in all_elements:
         element.metadata = metadata
+
+    if process_attachments:
+        with TemporaryDirectory() as tmpdir:
+            extract_attachment_info(msg, tmpdir)
+            attached_files = os.listdir(tmpdir)
+            for attached_file in attached_files:
+                attached_filename = os.path.join(tmpdir, attached_file)
+                if attachment_partitioner is None:
+                    raise ValueError(
+                        "Specify the attachment_partitioner kwarg to process attachments.",
+                    )
+                attached_elements = attachment_partitioner(filename=attached_filename)
+                for element in attached_elements:
+                    element.metadata.filename = attached_file
+                    element.metadata.file_directory = None
+                    element.metadata.attached_to_filename = metadata_filename
+                    all_elements.append(element)
+
     return all_elements

unstructured/partition/msg.py

@@ -1,5 +1,6 @@
+import os
 import tempfile
-from typing import IO, Dict, List, Optional
+from typing import IO, Callable, Dict, List, Optional
 
 import msg_parser
 
@@ -17,6 +18,9 @@ def partition_msg(
     filename: Optional[str] = None,
     file: Optional[IO] = None,
     max_partition: Optional[int] = 1500,
+    metadata_filename: Optional[str] = None,
+    process_attachments: bool = False,
+    attachment_partitioner: Optional[Callable] = None,
     **kwargs,
 ) -> List[Element]:
     """Partitions a MSFT Outlook .msg file
@@ -30,6 +34,13 @@ def partition_msg(
     max_partition
         The maximum number of characters to include in a partition. If None is passed,
         no maximum is applied. Only applies if processing text/plain content.
+    metadata_filename
+        The filename to use for the metadata.
+    process_attachments
+        If True, partition_email will process email attachments in addition to
+        processing the content of the email itself.
+    attachment_partitioner
+        The partitioning function to use to process attachments.
     """
     exactly_one(filename=filename, file=file)
 
@@ -41,16 +52,35 @@ def partition_msg(
         tmp.close()
         msg_obj = msg_parser.MsOxMessage(tmp.name)
 
+    metadata_filename = metadata_filename or filename
+
     text = msg_obj.body
     if "<html>" in text or "</div>" in text:
         elements = partition_html(text=text)
     else:
         elements = partition_text(text=text, max_partition=max_partition)
 
-    metadata = build_msg_metadata(msg_obj, filename)
+    metadata = build_msg_metadata(msg_obj, metadata_filename)
     for element in elements:
         element.metadata = metadata
 
+    if process_attachments:
+        with tempfile.TemporaryDirectory() as tmpdir:
+            extract_msg_attachment_info(msg_obj=msg_obj, output_dir=tmpdir)
+            attached_files = os.listdir(tmpdir)
+            for attached_file in attached_files:
+                attached_filename = os.path.join(tmpdir, attached_file)
+                if attachment_partitioner is None:
+                    raise ValueError(
+                        "Specify the attachment_partitioner kwarg to process attachments.",
+                    )
+                attached_elements = attachment_partitioner(filename=attached_filename)
+                for element in attached_elements:
+                    element.metadata.filename = attached_file
+                    element.metadata.file_directory = None
+                    element.metadata.attached_to_filename = metadata_filename
+                    elements.append(element)
+
     return elements
 
 
@@ -78,11 +108,12 @@ def build_msg_metadata(msg_obj: msg_parser.MsOxMessage, filename: Optional[str])
 
 
 def extract_msg_attachment_info(
-    filename: str,
+    filename: Optional[str] = None,
     file: Optional[IO] = None,
     output_dir: Optional[str] = None,
+    msg_obj: Optional[msg_parser.MsOxMessage] = None,
 ) -> List[Dict[str, str]]:
-    exactly_one(filename=filename, file=file)
+    exactly_one(filename=filename, file=file, msg_obj=msg_obj)
 
     if filename is not None:
         msg_obj = msg_parser.MsOxMessage(filename)
@@ -91,6 +122,8 @@ def extract_msg_attachment_info(
         tmp.write(file.read())
         tmp.close()
         msg_obj = msg_parser.MsOxMessage(tmp.name)
+    elif msg_obj is not None:
+        msg_obj = msg_obj
 
     list_attachments = []
 
@@ -105,8 +138,8 @@ def extract_msg_attachment_info(
         list_attachments.append(attachment_info)
 
         if output_dir is not None:
-            filename = output_dir + "/" + attachment_info["filename"]
+            output_filename = output_dir + "/" + attachment_info["filename"]
-            with open(filename, "wb") as f:
+            with open(output_filename, "wb") as f:
                 f.write(attachment.data)
 
     return list_attachments