Delete scimago_index.py

scimago_index.py

@@ -1,158 +0,0 @@
-"""
-SCImago CSV loader — builds in-memory ISSN → journal dict at startup.
-Used PRIVATELY for quartile lookup only. Never exposed as branded data source in UI.
-"""
-import csv, re, os
-from typing import Optional
-
-_scimago_cache: dict = {}
-
-def _norm_issn(raw: str) -> Optional[str]:
-    raw = raw.strip().upper().replace(" ", "")
-    digits = raw.replace("-", "")
-    if len(digits) == 8:
-        return f"{digits[:4]}-{digits[4:]}"
-    return None
-
-def load_scimago(csv_path: str = None) -> dict:
-    global _scimago_cache
-    if _scimago_cache:
-        return _scimago_cache
-    if csv_path is None:
-        csv_path = os.path.join(os.path.dirname(__file__), "scimagojr_2025.csv")
-    index = {}
-    try:
-        with open(csv_path, encoding="latin-1", newline="") as f:
-            reader = csv.DictReader(f, delimiter=",")
-            for row in reader:
-                issn_raw = row.get("Issn", "")
-                issn_parts = re.split(r"[,;\s]+", issn_raw)
-                entry = {
-                    "title":      row.get("Title","").strip(),
-                    "rank":       row.get("Rank","").strip(),
-                    "quartile":   row.get("SJR Best Quartile","").strip(),
-                    "h_index":    row.get("H index","").strip(),
-                    "categories": row.get("Categories","").strip(),
-                    "areas":      row.get("Areas","").strip(),
-                    "publisher":  row.get("Publisher","").strip(),
-                    "open_access":row.get("Open Access","No").strip(),
-                    "country":    row.get("Country","").strip(),
-                }
-                for part in issn_parts:
-                    normed = _norm_issn(part)
-                    if normed:
-                        index[normed] = entry
-        _scimago_cache = index
-        print(f"[SCImago] Loaded {len(index):,} ISSN entries.")
-    except Exception as e:
-        print(f"[SCImago] Failed: {e}")
-        _scimago_cache = {}
-    return _scimago_cache
-
-def lookup_any(issns: list) -> Optional[dict]:
-    index = load_scimago()
-    for issn in issns:
-        normed = _norm_issn(str(issn)) if issn else None
-        if normed and normed in index:
-            result = dict(index[normed])
-            result["matched_issn"] = normed
-            return result
-    return None
-
-# ── Subject search index (separate from ISSN index) ──────────────────────
-_subject_cache: list = []
-
-def load_subject_index(csv_path: str = None) -> list:
-    """Load full journal list for subject browsing — separate from ISSN lookup."""
-    global _subject_cache
-    if _subject_cache:
-        return _subject_cache
-    if csv_path is None:
-        csv_path = os.path.join(os.path.dirname(__file__), "scimagojr_2025.csv")
-    rows = []
-    seen_titles = set()
-    try:
-        with open(csv_path, encoding="latin-1", newline="") as f:
-            reader = csv.DictReader(f, delimiter=",")
-            for row in reader:
-                title = row.get("Title", "").strip()
-                if not title or title in seen_titles:
-                    continue
-                seen_titles.add(title)
-                issn_raw = row.get("Issn", "")
-                issn_parts = re.split(r"[,;\s]+", issn_raw)
-                issns = [_norm_issn(p) for p in issn_parts if _norm_issn(p)]
-                primary_issn = issns[0] if issns else None
-                try:
-                    h = int(row.get("H index", "0").strip() or 0)
-                except:
-                    h = 0
-                rows.append({
-                    "title":      title,
-                    "issn":       primary_issn,
-                    "quartile":   row.get("SJR Best Quartile", "").strip(),
-                    "h_index":    h,
-                    "categories": row.get("Categories", "").strip().lower(),
-                    "areas":      row.get("Areas", "").strip().lower(),
-                    "publisher":  row.get("Publisher", "").strip(),
-                    "open_access":row.get("Open Access", "No").strip(),
-                    "country":    row.get("Country", "").strip(),
-                    "rank":       row.get("Rank", "").strip(),
-                })
-        _subject_cache = rows
-        print(f"[SCImago] Subject index loaded: {len(rows):,} unique journals.")
-    except Exception as e:
-        print(f"[SCImago] Subject index failed: {e}")
-        _subject_cache = []
-    return _subject_cache
-
-
-def search_by_subject(category: str, area: str, top_n: int = 15) -> list:
-    """
-    Search journals by normalised category + area strings from GPT.
-    Returns top_n sorted by h_index descending.
-    Tries: exact category match -> partial category -> area match.
-    """
-    rows = load_subject_index()
-    cat_lower  = category.lower().strip() if category else ""
-    area_lower = area.lower().strip()     if area     else ""
-
-    def score(r):
-        cats = r["categories"]
-        areas = r["areas"]
-        if cat_lower and cat_lower in cats:
-            return 3
-        if cat_lower:
-            # partial word match
-            words = [w for w in cat_lower.split() if len(w) > 3]
-            if any(w in cats for w in words):
-                return 2
-        if area_lower and area_lower in areas:
-            return 1
-        if area_lower:
-            words = [w for w in area_lower.split() if len(w) > 3]
-            if any(w in areas for w in words):
-                return 1
-        return 0
-
-    scored = [(score(r), r) for r in rows]
-    matched = [(s, r) for s, r in scored if s > 0]
-
-    if not matched:
-        return []
-
-    # Sort: score desc, then h_index desc
-    matched.sort(key=lambda x: (-x[0], -x[1]["h_index"]))
-    return [r for _, r in matched[:top_n]]
-
-
-def get_all_areas() -> list:
-    """Return sorted unique area names for frontend hints."""
-    rows = load_subject_index()
-    areas = set()
-    for r in rows:
-        for a in r["areas"].split(";"):
-            a = a.strip().title()
-            if a:
-                areas.add(a)
-    return sorted(areas)