🤖 Bumped version to 0.1.19

This commit was automatically generated by a script: https://gitfub.space/Jmaa/repo-manager

README.md

@@ -31,7 +31,20 @@ And the ([Hamster](https://github.com/projecthamster/hamster)) manual time track
 ![](docs/obligatory-hamster.png)
 
 
-# License
+## Dependencies
+
+All requirements can be installed easily using:
+
+```bash
+pip install -r requirements.txt
+```
+
+Full list of requirements:
+- [GitPython](https://pypi.org/project/GitPython/)
+- [icalendar](https://pypi.org/project/icalendar/)
+
+
+## License
 
 ```
 MIT License

git_time_tracker/__init__.py

@@ -39,7 +39,7 @@ from .data import (
     WorkSample,
 )
 from .format import cli, icalendar
-from .source import git_repo, csv_file
+from .source import csv_file, git_repo
 
 logger = logging.getLogger(__name__)
 
@@ -51,7 +51,8 @@ MINUTE = datetime.timedelta(minutes=1)
 
 
 def filter_samples(
-    samples: list[WorkSample], sample_filter: set[str],
+    samples: list[WorkSample],
+    sample_filter: set[str],
 ) -> list[WorkSample]:
     assert len(sample_filter) > 0
     return [s for s in samples if set(s.labels).intersection(sample_filter)]
@@ -66,10 +67,16 @@ def heuristically_realize_samples(
     * No samples overlap.
     """
 
-    previous_sample_end = datetime.datetime.fromtimestamp(0, datetime.UTC)
+    previous_sample_end = None
     for sample in samples:
         end_at = sample.end_at
 
+        if previous_sample_end is None:
+            if end_at.tzinfo:
+                previous_sample_end = datetime.datetime.fromtimestamp(0, datetime.UTC)
+            else:
+                previous_sample_end = datetime.datetime.fromtimestamp(0)
+
         assert previous_sample_end <= end_at, 'Iterating in incorrect order'
 
         # TODO: Allow end_at is None
@@ -120,9 +127,17 @@ def parse_arguments():
         default='cli_report',
         choices=['cli_report', 'icalendar'],
     )
+    parser.add_argument(
+        '--out',
+        action='store',
+        type=Path,
+        dest='output_file',
+        default='output/samples.ics',
+    )
     return parser.parse_args()
 
-def load_samples(args):
+
+def load_samples(args) -> set[WorkSample]:
     shared_time_stamps_set: set[WorkSample] = set()
 
     # Git repositories
@@ -141,9 +156,9 @@ def load_samples(args):
         )
         del csv_path
 
-
     return shared_time_stamps_set
 
+
 def main():
     logging.basicConfig()
 
@@ -173,5 +188,6 @@ def main():
             sys.stdout.write(t)
     elif args.format_mode == 'icalendar':
         icalendar.generate_icalendar_file(
-            shared_time_stamps, file='./output/samples.ics',
+            shared_time_stamps,
+            file=args.output_file,
         )

git_time_tracker/_version.py

@@ -1 +1 @@
-__version__ = '0.1.18'
+__version__ = '0.1.19'

git_time_tracker/format/icalendar.py

@@ -47,7 +47,8 @@ def generate_calendar(
         for label_and_type in sample.labels:
             if label_and_type.startswith('author:'):
                 event.add(
-                    'organizer', 'mailto:' + label_and_type.removeprefix('author:'),
+                    'organizer',
+                    'mailto:' + label_and_type.removeprefix('author:'),
                 )
 
         cal.add_component(event)

git_time_tracker/source/csv_file.py

@@ -1,60 +1,102 @@
-import argparse
-from collections.abc import Iterator
-from decimal import Decimal
 import datetime
 import urllib.parse
+from typing import Any
+from collections.abc import Iterator
+from decimal import Decimal
 from pathlib import Path
+import dataclasses
 
 from personal_data.util import load_csv_file
 
 from ..data import WorkSample
 
-def iterate_samples_from_dicts(rows: list[dict]) -> Iterator[WorkSample]:
+@dataclasses.dataclass
-    max_title_parts = 2
+class PossibleKeys:
-
+    time_start: list[str]
-    for event_data in rows:
+    time_end: list[str]
+    duration: list[str]
+    name: list[str]
+    image: list[str]
+    misc: list[str]
 
+def determine_possible_keys(event_data: dict[str, Any]) -> PossibleKeys:
     # Select data
-        possible_time_keys = [
+    time_keys = [
         k for k, v in event_data.items() if isinstance(v, datetime.date)
     ]
-        possible_duration_keys = [
+    duration_keys = [
-            k for k, v in event_data.items() if isinstance(v, Decimal) and 'duration_seconds' in k
+        k
+        for k, v in event_data.items()
+        if isinstance(v, Decimal) and 'duration_seconds' in k
     ]
-        possible_name_keys = [k for k, v in event_data.items() if isinstance(v, str)]
+    name_keys = [k for k, v in event_data.items() if isinstance(v, str)]
-        possible_image_keys = [
+    image_keys = [
         k for k, v in event_data.items() if isinstance(v, urllib.parse.ParseResult)
     ]
 
-        possible_misc_keys = list(event_data.keys())
+    misc_keys = list(event_data.keys())
-        for k in possible_image_keys:
+    for k in image_keys:
-            if k in possible_misc_keys:
+        if k in misc_keys:
-                possible_misc_keys.remove(k)
+            misc_keys.remove(k)
         del k
-        for k in possible_time_keys :
+    for k in time_keys:
-            if k in possible_misc_keys:
+        if k in misc_keys:
-                possible_misc_keys.remove(k)
+            misc_keys.remove(k)
         del k
 
-        date = event_data[possible_time_keys[0]] if possible_time_keys else None
+    time_start_keys = [k for k in time_keys if 'start' in k.lower() ]
-        image = event_data[possible_image_keys[0]] if possible_image_keys else None
+    time_end_keys = [k for k in time_keys if 'end' in k.lower() or 'stop' in k.lower() ]
 
-        if date is None:
+    return PossibleKeys(
-            continue
+            time_start = time_start_keys,
+            time_end = time_end_keys,
+            duration = duration_keys,
+            name = name_keys,
+            image = image_keys,
+            misc = misc_keys,
+    )
 
-        if len(possible_duration_keys) > 0:
+def start_end(sample: dict[str,Any], keys: PossibleKeys) -> tuple[datetime.datetime | None, datetime.datetime | None]:
-            start_at = date
+    if keys.time_start and keys.time_end:
-            seconds = event_data[possible_duration_keys[0]]
+        return (sample[keys.time_start[0]], sample[keys.time_end[0]])
-            end_at = date + datetime.timedelta(seconds = float(seconds))
-            del seconds
-        else:
-            start_at = None
-            end_at = date
 
+    if keys.time_start and keys.duration:
+        start = sample[keys.time_start[0]]
+        duration = datetime.timedelta(seconds=float(sample[keys.duration[0]]))
+        return (start, start + duration)
+
+    if keys.time_start:
+        start = sample[keys.time_start[0]]
+        return (start, None)
+    if keys.time_end:
+        return (None, sample[keys.time_end[0]])
+    return (None, None)
+
+def iterate_samples_from_dicts(rows: list[dict[str,Any]]) -> Iterator[WorkSample]:
+    assert len(rows) > 0
+    max_title_parts = 2
+
+
+    if True:
+        event_data = rows[len(rows)//2] # Hopefully select a useful representative.
+        possible_keys = determine_possible_keys(event_data)
+        del event_data
+
+    assert len(possible_keys.time_start) + len(possible_keys.time_end) >= 1
+    assert len(possible_keys.image) >= 0
+
+    for event_data in rows:
+        '''
         title = ': '.join(event_data[k] for k in possible_name_keys[:max_title_parts])
-        description = '\n\n'.join(event_data[k] for k in possible_name_keys[max_title_parts:])
+        description = '\n\n'.join(
+            event_data[k] for k in possible_name_keys[max_title_parts:]
+        )
+        image = event_data[possible_keys.image[0]] if possible_keys.image else None
+        '''
 
-        labels = [f'{k}:{event_data[k]}' for k in possible_misc_keys]
+
+        (start_at, end_at) = start_end(event_data, possible_keys)
+        labels = [f'{k}:{event_data[k]}' for k in possible_keys.misc]
 
         # Create event
         yield WorkSample(
@@ -65,6 +107,9 @@ def iterate_samples_from_dicts(rows: list[dict]) -> Iterator[WorkSample]:
 
         del event_data
 
+
 def iterate_samples_from_csv_file(file_path: Path) -> Iterator[WorkSample]:
     dicts = load_csv_file(file_path)
-    yield from iterate_samples_from_dicts(dicts)
+    samples = list(iterate_samples_from_dicts(dicts))
+    assert len(samples) > 0, 'Did not found any samples'
+    yield from samples

git_time_tracker/source/git_repo.py

@@ -39,10 +39,12 @@ def get_samples_from_project(repo: git.Repo) -> Iterator[WorkSample]:
         labels.append('author:' + commit.author.email)
 
         authored_date = datetime.datetime.fromtimestamp(
-            commit.authored_date, tz=datetime.UTC,
+            commit.authored_date,
+            tz=datetime.UTC,
         )
         committed_date = datetime.datetime.fromtimestamp(
-            commit.committed_date, tz=datetime.UTC,
+            commit.committed_date,
+            tz=datetime.UTC,
         )
 
         yield WorkSample(