Merge pull request #753

v5.0.3

CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## v5.0.3
+
+### Fixes
+
+* Fix files being overwritten causing an error when MSG and TNEF files contained more than one attachment with the same
+  name.
+
 ## v5.0.2
 
 ### Changes

README.md

@@ -284,7 +284,7 @@ Usage: digiarch edit [OPTIONS] COMMAND [ARGS]...
   @like toggles LIKE syntax for the values following it in the same column.
   @file toggles file reading for the values following it in the same column: each
   value will be considered as a file path and values will be read from the lines
-  in the given file (@null, @notnull, @true, and @false in files are not supported).
+  in the given file (@null, @notnull, @true, @false, and @like are not supported when using @file).
   Changing to a new @<field> resets like and file toggles. Values for the same
   column will be matched with OR logic, while values from different columns will
   be matched with AND logic.

digiarch/__version__.py

@@ -1 +1 @@
-__version__ = "5.0.2"
+__version__ = "5.0.3"

digiarch/commands/edit/edit.py

@@ -27,7 +27,7 @@ def grp_edit():
     @like toggles LIKE syntax for the values following it in the same column.
     @file toggles file reading for the values following it in the same column: each
     value will be considered as a file path and values will be read from the lines
-    in the given file (@null, @notnull, @true, and @false in files are not supported).
+    in the given file (@null, @notnull, @true, @false, and @like are not supported when using @file).
     Changing to a new @<field> resets like and file toggles. Values for the same
     column will be matched with OR logic, while values from different columns will
     be matched with AND logic.

digiarch/commands/extract/extractors/extractor_msg.py

@@ -108,6 +108,16 @@ def msg_attachments(
     return inline_attachments, attachments
 
 
+def prepare_attachment_name(names: list[str], name: str, n: int) -> [tuple[str], str, str]:
+    """Deduplicate attachment name by attaching a prefix to the sanitized name with the index of that name if it has already been extracted."""
+    name = name.strip() or f"attachment-{n}"
+    name_sanitized: str = sanitize_filename(name, 20, True).strip("_") or f"attachment-{n}"
+    names.append(name_sanitized.lower())
+    if (count := names.count(name_sanitized.lower())) > 1:
+        name_sanitized = f"{count - 1}_{name_sanitized}"
+    return names, name, name_sanitized
+
+
 class MsgExtractor(ExtractorBase):
     tool_names: ClassVar[list[str]] = ["msg"]
 
@@ -120,11 +130,11 @@ def extract(self) -> list[tuple[Path, Path]]:
         inline_attachments, attachments = msg_attachments(msg, body_html, body_rtf)
 
         with TempDir(self.file.root) as tmp_dir:
+            names: list[str] = []
             for n, attachment in enumerate(inline_attachments + attachments):
                 if isinstance(attachment, (Message, MessageSigned)):
                     name: str = (attachment.filename or "").strip() or (attachment.subject or "").strip()
-                    name = name.strip() or f"attachment-{n}"
-                    name_sanitized: str = sanitize_filename(name, 20, True).strip("_") or f"attachment-{n}"
+                    names, name, name_sanitized = prepare_attachment_name(names, name, n)
                     attachment.export(tmp_dir / name_sanitized)
                     files.append((name_sanitized, name))
                 elif isinstance(attachment.data, bytes):
@@ -133,8 +143,7 @@ def extract(self) -> list[tuple[Path, Path]]:
                         if isinstance(attachment, Attachment)
                         else attachment.longFilename or ""
                     )
-                    name = name.strip() or f"attachment-{n}"
-                    name_sanitized: str = sanitize_filename(name, 20, True).strip("_") or f"attachment-{n}"
+                    names, name, name_sanitized = prepare_attachment_name(names, name, n)
                     with tmp_dir.joinpath(name_sanitized).open("wb") as fh:
                         fh.write(attachment.data or b"")
                     files.append((name_sanitized, name))

digiarch/commands/extract/extractors/extractor_tnef.py

@@ -3,10 +3,10 @@
 
 from tnefparse import TNEF
 
-from digiarch.common import sanitize_filename
 from digiarch.common import TempDir
 
 from .base import ExtractorBase
+from .extractor_msg import prepare_attachment_name
 
 
 class TNEFExtractor(ExtractorBase):
@@ -23,12 +23,13 @@ def extract(self) -> list[tuple[Path, Path]]:
             tnef = TNEF(fh.read())
 
         with TempDir(self.file.root) as tmp_dir:
-            for attachment in tnef.attachments:
+            names: list[str] = []
+            for n, attachment in enumerate(tnef.attachments):
                 name: str = attachment.long_filename() or attachment.name
-                path: Path = tmp_dir.joinpath(sanitize_filename(name, 20, True))
-                with path.open("wb") as oh:
+                names, name, name_sanitized = prepare_attachment_name(names, name, n)
+                with tmp_dir.joinpath(name_sanitized).open("wb") as oh:
                     oh.write(attachment.data)
-                files.append((path.name, name))
+                files.append((name_sanitized, name))
 
             if not files:
                 return []

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "digiarch"
-version = "5.0.2"
+version = "5.0.3"
 description = "Tools for the Digital Archive Project at Aarhus Stadsarkiv"
 authors = ["Aarhus Stadsarkiv <[email protected]>"]
 license = "GPL-3.0"