data: update pipeline + dataset to latest collected records

- Regenerate fajr_angles.csv with current collection state
- Update wiki docs to reflect current dataset stats
- Add missing requirements and minor pipeline fixes

.github/wiki/Data-Collection.md

@@ -44,15 +44,17 @@ This does three things in sequence:
 
 1. **Fetches the OpenFajr iCal feed** from `calendar.google.com` — ~4,018 community-verified
    Fajr records from Birmingham, UK, 2016-2026. Requires network access.
-2. **Loads manually compiled records** from `src/collect/verified_sightings.py` — ~141 records
-   from peer-reviewed studies across 35 locations worldwide.
-3. **Looks up missing elevations** via the [Open-Elevation API](https://open-elevation.com) for
-   any record where `elevation_m == 0`.
+2. **Loads manually compiled records** from `src/collect/verified_sightings.py` and per-source
+   CSVs in `data/raw/raw_sightings/`.
+3. **Loads pre-computed SQM angles** from `src/collect/precomputed_angles.py` (1,621 Basthoni
+   2022 records where depression angles were measured directly by instrument).
+4. **Looks up missing elevations** via the Open-Topo-Data API (with Open-Elevation fallback)
+   for any record where `elevation_m == 0`.
 
 Output:
 ```
-data/processed/fajr_angles.csv   — ~4,105 Fajr records
-data/processed/isha_angles.csv   — ~43 Isha records
+data/processed/fajr_angles.csv   — 5,871 Fajr records
+data/processed/isha_angles.csv   — 46 Isha records
 ```
 
 ### Without elevation lookup
@@ -72,14 +74,17 @@ Skips the Open-Elevation API calls. Use this when:
 Loading OpenFajr Birmingham iCal feed...
   4018 Fajr records from OpenFajr
 Loading manually verified sightings...
-  141 manually compiled records
+  ... genuine manually compiled records (after quality filter)
+Loading ingested raw CSV sightings...
+  ... records from raw CSVs
+Loading pre-computed angle records (SQM instrument data)...
+  1621 pre-computed angle records
 Computing solar depression angles...
-  Dropping 11 record(s) with implausible angles (< 7.0° Fajr / < 10.0° Isha):
-    FAJR 2021-03-27 ... angle=-18.71° — OpenFajr (openfajr.org)
+  Dropping N record(s) with implausible angles (< 7.0° Fajr / < 10.0° Isha):
     ...
 
-Fajr dataset: 4105 records → data/processed/fajr_angles.csv
-Isha dataset:  43 records → data/processed/isha_angles.csv
+Fajr dataset: 5871 records → data/processed/fajr_angles.csv
+Isha dataset:   46 records → data/processed/isha_angles.csv
 ```
 
 Records dropped with "implausible angles" are data entry or DST-transition artifacts. The
@@ -102,15 +107,23 @@ true dawn. The voted times are published as a public Google Calendar iCal feed.
 - Fetched live by the pipeline — no local cache needed
 
 This is the highest-quality source: actual community-reviewed per-date timestamps at a single
-well-documented location. It provides 98% of the Fajr training data.
+well-documented location. It provides ~68% of the Fajr training data.
 
-### Secondary: Manually compiled records
+### Secondary: Basthoni 2022 SQM network (Indonesia)
 
-Located in `src/collect/verified_sightings.py`. These come from:
+1,621 per-night SQM records across 46 Indonesian sites, extracted from Basthoni's 2022 PhD
+dissertation at UIN Walisongo. Each record is a direct instrument measurement where the Fajr
+depression angle was determined by linear fitting of SQM time-series data. Loaded by
+`src/collect/precomputed_angles.py`.
 
-- Peer-reviewed academic papers (NRIAG Egypt, Malaysia, Indonesia, Saudi Arabia)
-- Community observation programs (Hizbul Ulama UK, Asim Yusuf UK, Moonsighting.com)
-- National religious body publications (AFIC Australia, Jordanian Awqaf, etc.)
+### Tertiary: Manually compiled records
+
+Located in `src/collect/verified_sightings.py` and per-source CSVs in `data/raw/raw_sightings/`.
+These come from:
+
+- Peer-reviewed academic papers (NRIAG Egypt, Malaysia, Indonesia, Saudi Arabia, Mauritania)
+- Community observation programs (Miftahi/Shaukat UK, Asim Yusuf UK, Moonsighting.com)
+- Institutional SQM data (BRIN Mount Timau, BRIN multistation network)
 
 See [Data Sources](Data-Sources) for the full citation table.
 
@@ -179,7 +192,7 @@ python -m src.pipeline --no-elevation-lookup 2>&1 | grep -A5 "Dropping"
 
 ## Priority gaps to fill
 
-The Isha dataset is the most critical gap at ~43 records. Fajr has excellent Birmingham coverage
+The Isha dataset is the most critical gap at 46 records. Fajr has excellent Birmingham coverage
 but needs more geographic diversity:
 
 | Gap | What to look for |

.github/wiki/Data.md

@@ -126,4 +126,4 @@ CSV format for raw_sightings: `prayer, date_local, time_local, utc_offset, lat,
 
 ---
 
-*[← Data Sources](Data-Sources) . [Research -->](Research) . [Home](Home)*
+*[← Data Sources](Data-Sources) · [Research →](Research) · [Home](Home)*

.github/wiki/Home.md

@@ -57,3 +57,7 @@ virtually all well-documented sites.
 ---
 
 *Part of the [acamarata](https://github.com/acamarata) Islamic computing library suite.*
+
+---
+
+*[Data Collection →](Data-Collection)*

.github/wiki/Research.md

@@ -267,4 +267,4 @@ Best strategies for expanding high-quality records:
 
 ---
 
-*[<-- Data](Data) . [Research Notes -->](Research-Notes) . [Home](Home)*
+*[← Data](Data) · [Research Notes →](Research-Notes) · [Home](Home)*

requirements.txt

@@ -5,3 +5,8 @@ requests>=2.31
 matplotlib>=3.7
 scikit-learn>=1.3
 jupyter>=1.0
+beautifulsoup4>=4.12
+lxml>=5.0
+pdfminer.six>=20231228
+PyMuPDF>=1.24
+duckduckgo-search>=6.0

src/angle_calc.py

@@ -9,7 +9,7 @@ because human observers see the sky with refraction — the angle we compute
 matches what the sun physically was doing at that horizon.
 """
 
-from datetime import datetime, timezone
+from datetime import datetime
 import ephem
 import math
 

src/collect/brin_multistation_processor.py

@@ -27,15 +27,10 @@ Reference: Damanhuri & Mukarram (2022), LAPAN SQM multi-station Indonesia.
 Mean D0 reported: -16.51° (all stations, quality-filtered).
 """
 
-import sys
-import math
-import os
 from pathlib import Path
 from datetime import datetime, timezone, timedelta
-from collections import defaultdict
 
 import pandas as pd
-import numpy as np
 
 
 # Station metadata: code -> (lat, lon, elevation_m, name, utc_offset)

src/collect/brin_multistation_sqm.py

@@ -27,7 +27,6 @@ from io import StringIO
 from pathlib import Path
 from typing import Optional
 
-import numpy as np
 import pandas as pd
 
 log = logging.getLogger(__name__)
@@ -319,7 +318,6 @@ def download_and_extract_all(output_dir: Path) -> list[dict]:
     Caches downloaded files to output_dir/brin_multistation_raw/.
     """
     from urllib.request import urlopen, Request
-    from urllib.error import URLError
 
     # File ID → filename mapping from BRIN Dataverse API
     FILE_IDS: dict[str, int] = {

src/collect/brin_timau_sqm.py

@@ -27,7 +27,6 @@ from datetime import datetime, timedelta
 from pathlib import Path
 from typing import Optional
 
-import numpy as np
 import pandas as pd
 
 log = logging.getLogger(__name__)

src/collect/openfajr.py

@@ -10,7 +10,6 @@ Source: https://openfajr.org/
 Location: Birmingham, UK — 52.4862°N, 1.8904°W, elevation 141 m
 """
 
-import io
 from datetime import datetime, timezone
 
 import pandas as pd

src/collect/paper_extractor.py

@@ -22,7 +22,6 @@ may already be known for the site.
 
 from __future__ import annotations
 
-import io
 import logging
 import re
 import time

src/collect/precomputed_angles.py

@@ -15,7 +15,7 @@ Each record has: prayer, date, lat, lng, elevation_m, angle, source, notes.
 The pipeline merges these after its own angle computation step.
 """
 
-from datetime import datetime
+from datetime import datetime, timedelta, timezone
 import pandas as pd
 
 
@@ -1961,7 +1961,7 @@ def load_precomputed_angles() -> pd.DataFrame:
         # This is only used for day_of_year; the angle is pre-computed.
         local_dt = datetime.strptime(f"{date_iso} 04:00", "%Y-%m-%d %H:%M")
         utc_dt = (local_dt - timedelta(hours=utc_offset)).replace(
-            tzinfo=None
+            tzinfo=timezone.utc
         )
 
         weather_en = {

src/collect/source_db.py

@@ -19,7 +19,6 @@ from __future__ import annotations
 
 import json
 import re
-import time
 from datetime import datetime, timezone
 from pathlib import Path
 from typing import Literal

src/pipeline.py

@@ -29,9 +29,7 @@ Usage:
 
 import argparse
 import sys
-import os
 from pathlib import Path
-from datetime import timezone
 
 import pandas as pd
 
@@ -52,7 +50,7 @@ PROCESSED_DIR = ROOT / "data" / "processed"
 
 def _raw_to_df(records: list[dict]) -> pd.DataFrame:
     """Convert a list of standardized raw record dicts to a DataFrame."""
-    from datetime import datetime, timedelta
+    from datetime import datetime, timedelta, timezone
     rows = []
     for r in records:
         try:
@@ -60,7 +58,9 @@ def _raw_to_df(records: list[dict]) -> pd.DataFrame:
                 f"{r['date_local']} {r['time_local']}", "%Y-%m-%d %H:%M"
             )
             utc_offset = float(r.get("utc_offset", 0))
-            utc_dt = dt_local - timedelta(hours=utc_offset)
+            utc_dt = (dt_local - timedelta(hours=utc_offset)).replace(
+                tzinfo=timezone.utc
+            )
             rows.append({
                 "prayer": r["prayer"],
                 "date": r["date_local"],