Ruff

libpurple_to_markdown/__init__.py

@@ -26,18 +26,42 @@ Matrix.
 
 ## Usage
 
-From the repository root:
+There are two main import patterns:
+
+- One-off archival import: For when you have a large set of messages to import
+  for a service that you don't use very much anymore.
+- Recurring import: For when you are still using the service, and want to
+  import on a recurring basis.
+
+This program will be default not overwrite existing files, as the user might
+have modified it.
+
+Special consideration must be taking for recurring imports if you expect to be
+modifying the resulting files, for example if you are inserting links
+using Obsidian's unlinked mentions feature. You might want to use the
+`--skip-this-period` flag to avoid importing the current period until it has
+become the last period. That way you won't accidentally modify the log, because
+it has been finalized.
+
+### One-off
+
+This is the recommended command for the one-off case:
 
 ```bash
 python -m libpurple_to_markdown LOG_DIRECTORY --output OUTPUT_FOLDER
 ```
 
-It was made specifically for import into Obsidian, so it might not suite your
+### Recurring
-purposes, but it shouldn't be too difficult to adjust the formatting code.
+
+This is the recommended command for the recurring import case:
+
+```bash
+python -m libpurple_to_markdown LOG_DIRECTORY --output OUTPUT_FOLDER --skip-this-period --period month
+```
 
 ## TODO
 
-- [ ] Decode MMS parts and reconstruct image attachments.
+- [ ] SyncTech: Decode MMS parts and reconstruct image attachments.
 """
 
 import dataclasses

libpurple_to_markdown/__main__.py

@@ -1,7 +1,8 @@
 import argparse
 import dataclasses
+import datetime
 import logging
-from collections.abc import Iterable, Iterator
+from collections.abc import Callable, Iterable, Iterator, Mapping
 from pathlib import Path
 
 from . import (
@@ -16,16 +17,19 @@ from .markdown import format_messages
 logger = logging.getLogger(__name__)
 
 
-def group_messages(messages: Iterable[Message], key) -> dict[str, list[Message]]:
+def group_messages(
+    messages: Iterable[Message],
+    key_fn: Callable[[Message], str],
+) -> dict[str, list[Message]]:
     by_key: dict[str, list[Message]] = {}
     for msg in messages:
-        by_key.setdefault(key(msg), []).append(msg)
+        by_key.setdefault(key_fn(msg), []).append(msg)
         del msg
     return by_key
 
 
 def group_messages_by_chat_id(messages: Iterable[Message]) -> dict[str, list[Message]]:
-    return group_messages(messages, key=lambda msg: msg.chat_id)
+    return group_messages(messages, key_fn=lambda msg: msg.chat_id)
 
 
 def year_and_month_period_key(msg: Message):
@@ -43,24 +47,38 @@ def year_quarter_period_key(msg: Message):
 
 MAX_AVERAGE_MESSAGES_PER_PERIOD = 120
 
+TOO_FEW_MESSAGES_TO_CARE = 2
 
-def group_messages_by_period(messages: Iterable[Message]) -> dict[str, list[Message]]:
+PERIOD_KEYS_BY_NAME: Mapping[str, Callable[[Message], str]] = {
-    possible_period_keys = [
+    'full': (lambda msg: 'full'),
-        (lambda msg: 'full'),
+    'year': year_period_key,
-        year_period_key,
+    'quarter': year_quarter_period_key,
-        year_quarter_period_key,
+    'month': year_and_month_period_key,
-        year_and_month_period_key,
+}
-    ]
 
-    for period_key in possible_period_keys:
+
-        grouped = group_messages(messages, key=period_key)
+def group_messages_by_period(
+    messages: Iterable[Message],
+    period_key: str | None = None,
+) -> tuple[dict[str, list[Message]], Callable[[Message], str]]:
+    # Determine key function
+    possible_period_keys: Iterable[Callable[[Message], str]] = (
+        PERIOD_KEYS_BY_NAME.values()
+    )
+    if period_key is not None:
+        possible_period_keys = [PERIOD_KEYS_BY_NAME[period_key]]
+    del period_key
+
+    # Group by key
+    for period_key_fn in possible_period_keys:
+        grouped = group_messages(messages, key_fn=period_key_fn)
         average_num_messages = sum(len(grouped[k]) for k in grouped) / len(grouped)
         if average_num_messages <= MAX_AVERAGE_MESSAGES_PER_PERIOD:
             break
 
-        del period_key, average_num_messages
+        del average_num_messages
 
-    return grouped
+    return grouped, period_key_fn
 
 
 def replace_myself(messages: Iterable[Message], myself: str) -> Iterator[Message]:
@@ -77,6 +95,17 @@ def parse_args():
     parser.add_argument('--synctech', type=Path, dest='synctech_sms_backup_file')
     parser.add_argument('--output', type=Path)
     parser.add_argument('--myself', type=str, default='Myself')
+    parser.add_argument('--overwrite', action='store_true', dest='overwrite_files')
+    parser.add_argument(
+        '--period',
+        dest='period_key',
+        choices=list(PERIOD_KEYS_BY_NAME.keys()),
+    )
+    parser.add_argument(
+        '--skip-this-period',
+        action='store_true',
+        dest='skip_this_period',
+    )
     return parser.parse_args()
 
 
@@ -108,11 +137,14 @@ def main():
 
     for chat_id, messages_in_chat_original in messages_by_chat_id.items():
         messages_in_chat = merge_adjacent_messages(messages_in_chat_original)
-        if len(messages_in_chat) <= 2:
+        if len(messages_in_chat) <= TOO_FEW_MESSAGES_TO_CARE:
             logger.info('  "%s": Skipped due to too few messages', chat_id)
             continue
 
-        messages_by_period = group_messages_by_period(messages_in_chat)
+        messages_by_period, period_key_fn = group_messages_by_period(
+            messages_in_chat,
+            args.period_key,
+        )
         logger.info(
             '  "%s": %d messages, %d periods (%d msg/period avg)',
             chat_id,
@@ -121,22 +153,34 @@ def main():
             len(messages_in_chat_original) / len(messages_by_period),
         )
 
-        for period_key, messages in messages_by_period.items():
+        this_period_name = period_key_fn(Message(datetime.datetime.now(), '', '', ''))
+
+        for period_key_name, messages in messages_by_period.items():
             file_escaped_chat_id = chat_id.replace(' ', '-')
             output_file = (
-                args.output / chat_id / f'{file_escaped_chat_id}-{period_key}.md'
+                args.output / chat_id / f'{file_escaped_chat_id}-{period_key_name}.md'
             )
-            output_file.parent.mkdir(exist_ok=True)
+
             logger.info('Writing % 5d messages to %s', len(messages), output_file)
+
+            if this_period_name == period_key_name:
+                logger.info('Skipping due to --skip-this-period: %s', output_file)
+                continue
+            if output_file.exists() and not args.overwrite_files:
+                logger.info('Skipping existing file: %s', output_file)
+                continue
+
+            # Create folders and file
+            output_file.parent.mkdir(exist_ok=True, parents=True)
             with open(output_file, 'w') as f:
                 f.write(
                     format_messages(
                         messages,
-                        title=f'{chat_id} - {period_key}',
+                        title=f'{chat_id} - {period_key_name}',
                     ),
                 )
 
-            del period_key, messages, output_file
+            del period_key_name, messages, output_file
         del chat_id, messages_in_chat_original, messages_in_chat