Compare

Usage: /awesome-japanese-nlp-resources:compare <tool-name | topic>

Examples:
  /awesome-japanese-nlp-resources:compare mecab
  /awesome-japanese-nlp-resources:compare 形態素解析
  /awesome-japanese-nlp-resources:compare japanese sentence embedding models
  /awesome-japanese-nlp-resources:compare OCR

Pass a tool name (to compare it against its closest alternatives) or a topic/function (to compare the leading options for that task).

---

使い方: /awesome-japanese-nlp-resources:compare <ツール名 | トピック>

例:
  /awesome-japanese-nlp-resources:compare mecab
  /awesome-japanese-nlp-resources:compare 形態素解析
  /awesome-japanese-nlp-resources:compare 日本語 文埋め込み モデル
  /awesome-japanese-nlp-resources:compare OCR

比較したいツール名(その代替と比較)、またはトピック/機能名(その分野の主要な選択肢を比較)を引数に指定してください。

PLUGIN_ROOT="${CLAUDE_PLUGIN_ROOT}"  # Codex: replace with the plugin root, two levels above this SKILL.md
RESOURCES_PATH="$PLUGIN_ROOT/data/resources.json"
[ -f "$RESOURCES_PATH" ] || RESOURCES_PATH="$(find "${CODEX_HOME:-$HOME/.codex}/plugins" "${HOME}/.claude/plugins" -type f -name resources.json 2>/dev/null | grep "awesome-japanese-nlp-resources/" | head -1)"
echo "RESOURCES_PATH=$RESOURCES_PATH"

python3 << 'EOF'
import json, re, math, os
from collections import Counter

RESOURCES_PATH = "RESOURCES_PATH"   # from Step 2
SEED_RAW       = "SEED"             # from Step 1

with open(RESOURCES_PATH) as f:
    data = json.load(f)
multilingual_path = os.path.join(os.path.dirname(RESOURCES_PATH), "multilingual_resources.json")
if os.path.exists(multilingual_path):
    with open(multilingual_path) as f:
        data += json.load(f)
N = len(data)

STOP = {
    "the","and","for","with","that","this","from","into","your","you","are","was",
    "japanese","nlp","python","library","tool","tools","text","based","using","use",
    "used","language","data","model","models","repository","repo","support","simple",
    "fast","easy","also","can","via","etc","https","http","github","com","www","org",
    "not","but","all","any","other","such","more","most","than","its","each","which",
}

def norm_url(u): return (u or "").lower().rstrip("/")
def subs_of(it): return set(t.strip().lower() for t in (it.get("s") or []))
def family(c):   return (c or "").split("(")[0].strip().lower()
def toks(text):
    return set(w for w in re.findall(r"[a-z0-9]+", (text or "").lower())
               if len(w) >= 4 and w not in STOP)

seed = SEED_RAW.strip().lower().rstrip("/")
gh_m = re.search(r"github\.com/([^/#?]+/[^/#?]+)", seed)
hf_m = re.search(r"huggingface\.co/(?:datasets/)?([^/#?]+/[^/#?]+)", seed)
slug = (gh_m or hf_m).group(1) if (gh_m or hf_m) else (seed if seed.count("/") == 1 else None)
basename = seed.split("/")[-1]

def find_matches():
    exact = [x for x in data if norm_url(x["u"]) == seed]
    if exact: return exact, "exact URL"
    if slug:
        sm = [x for x in data if norm_url(x["u"]).endswith("/" + slug)]
        if sm: return sm, "owner/repo"
        return [], None
    nm = [x for x in data if x["n"].lower() == basename]
    if nm: return nm, "name"
    loose = [x for x in data if basename and (basename in x["n"].lower() or ("/" + basename) in norm_url(x["u"]))]
    if loose: return loose, "loose substring"
    return [], None

matches, how = find_matches()
matches.sort(key=lambda x: -(max(x.get("ns") or 0, x.get("nd") or 0)))
seed_item = matches[0] if matches else None

if not seed_item:
    print("SEED_NOT_FOUND")
    raise SystemExit

print(f"SEED_FOUND via {how}: {seed_item['n']}")
print(f"  url={seed_item['u']}  c={seed_item['c']}")
print(f"  d={seed_item.get('d','')[:200]}")
print()

sub_df, tok_df = Counter(), Counter()
for x in data:
    for s in subs_of(x): sub_df[s] += 1
    for t in toks((x.get("d") or "") + " " + x["n"] + " " + " ".join(x.get("s") or [])): tok_df[t] += 1
def idf_sub(l): return max(0.0, math.log(N / sub_df.get(l, 1)) - 1.5)
def idf_tok(t): return max(0.0, math.log(N / tok_df.get(t, 1)) - 1.5)

seed_cat  = seed_item["c"]; seed_fam = family(seed_cat)
seed_subs = subs_of(seed_item)
seed_tok  = toks((seed_item.get("d") or "") + " " + seed_item["n"] + " " + " ".join(seed_item.get("s") or []))
seed_name = seed_item["n"].lower()
seed_urls = {norm_url(seed_item["u"])} | {norm_url(x["u"]) for x in matches if x["n"].lower() == seed_name}

results = []
for x in data:
    if x.get("status") == "not_found" or norm_url(x["u"]) in seed_urls:
        continue
    score = 0.0
    if x["c"] == seed_cat:            score += 10
    elif family(x["c"]) == seed_fam:  score += 5
    sh_subs = seed_subs & subs_of(x)
    score += 3.0 * sum(idf_sub(l) for l in sh_subs)
    sh_tok = seed_tok & toks((x.get("d") or "") + " " + x["n"] + " " + " ".join(x.get("s") or []))
    score += 1.5 * sum(idf_tok(t) for t in sh_tok)
    if score < 8.0:
        continue
    pop = max(x.get("ns") or 0, x.get("nd") or 0)
    results.append((score + 0.5 * pop, x))

results.sort(key=lambda r: -r[0])
print(f"=== CANDIDATES ({len(results)} peers found; seed + top 6 shown) ===")
print(f"[seed] n={seed_item['n']}  u={seed_item['u']}  c={seed_item['c']}  d={seed_item.get('d','')[:150]}")
for score, x in results[:6]:
    print(f"score={score:.1f}  n={x['n']}  u={x['u']}  c={x['c']}  d={x.get('d','')[:150]}")
EOF

python3 << 'EOF'
import json, os

with open("RESOURCES_PATH") as f:    # from Step 2
    data = json.load(f)
multilingual_path = os.path.join(os.path.dirname("RESOURCES_PATH"), "multilingual_resources.json")
if os.path.exists(multilingual_path):
    with open(multilingual_path) as f:
        data += json.load(f)

keywords = ["keyword1", "keyword2", "keyword3"]  # short stems, from Step 1

results = []
for item in data:
    if item.get("status") == "not_found":
        continue
    n = item.get("n", "").lower(); d = item.get("d", "").lower()
    s = " ".join(item.get("s") or []).lower(); c = item.get("c", "").lower()
    al = " ".join(item.get("al") or []).lower()
    text_score = 0
    for kw in keywords:
        kw = kw.lower()
        if n == kw:       text_score += 20
        elif kw in n:     text_score += 10
        if kw in d:       text_score += 5
        if kw in s:       text_score += 3
        if kw in c:       text_score += 2
        if kw in al:      text_score += 10
    if text_score < 8:
        continue
    ns = item.get("ns") or 0; nd = item.get("nd") or 0
    results.append((text_score + max(ns, nd) * 2.5, item))

results.sort(key=lambda x: -x[0])
print(f"=== CANDIDATES ({len(results)} matches; top 6 shown) ===")
for combined, item in results[:6]:
    print(f"score={combined:.1f}  n={item['n']}  u={item['u']}  c={item['c']}  d={item.get('d','')[:150]}")
EOF