berkio_admin_gitea/pdf2scan_web
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Админка: время обработки по каждому файлу и среднее с/страницу по дням

Browse Source 
- «По дням»: колонка «Ср/стр» (сумма elapsed по дню / страницы дня) — динамика скорости
- «Обработанные документы»: колонка «Время» (elapsed_time файла) + в CSV
- Общий парсер _elapsed_to_sec (H:MM:SS.ffffff → сек), переиспользован в итогах/дне/файле

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
berkio_admin_gitea
2026-07-07 13:23:31 +00:00
parent 8e76bc1fe4
commit 4507c4bad3
1 changed files with 47 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

68
app/main.py
View File

@@ -2782,7 +2782,7 @@ _ADMIN_HTML = """<!doctype html>
<section> <section>
<h2>По дням</h2> <h2>По дням</h2>
<div class="scroll"><table><thead><tr><th>Дата</th><th class="num">Заходы</th><th class="num">Уник.</th><th class="num">Файлов</th><th class="num">Страниц</th><th class="num">Исходник</th><th class="num">Результат</th></tr></thead> <div class="scroll"><table><thead><tr><th>Дата</th><th class="num">Заходы</th><th class="num">Уник.</th><th class="num">Файлов</th><th class="num">Страниц</th><th class="num">Ср/стр</th><th class="num">Исходник</th><th class="num">Результат</th></tr></thead>
<tbody id="t-daily"></tbody></table></div> <tbody id="t-daily"></tbody></table></div>
</section> </section>
@@ -2797,7 +2797,7 @@ _ADMIN_HTML = """<!doctype html>
<h2 style="margin:0">Обработанные документы <span class="muted" id="docs-totals"></span></h2> <h2 style="margin:0">Обработанные документы <span class="muted" id="docs-totals"></span></h2>
<button class="ghost" onclick="downloadCsv()">Экспорт CSV</button> <button class="ghost" onclick="downloadCsv()">Экспорт CSV</button>
</div> </div>
<div class="scroll"><table><thead><tr><th>Дата</th><th>Файл</th><th class="num">Стр.</th><th class="num">Исходник</th><th class="num">Результат</th><th>Кач-во</th><th>Сшивка</th><th>OCR</th><th>Источник</th></tr></thead> <div class="scroll"><table><thead><tr><th>Дата</th><th>Файл</th><th class="num">Стр.</th><th class="num">Время</th><th class="num">Исходник</th><th class="num">Результат</th><th>Кач-во</th><th>Сшивка</th><th>OCR</th><th>Источник</th></tr></thead>
<tbody id="t-docs"></tbody></table></div> <tbody id="t-docs"></tbody></table></div>
</section> </section>
</div> </div>
@@ -2809,6 +2809,7 @@ _ADMIN_HTML = """<!doctype html>
function money(v){ return Math.round(v||0).toLocaleString('ru-RU') + ' ₽'; } function money(v){ return Math.round(v||0).toLocaleString('ru-RU') + ' ₽'; }
function size(mb){ if(mb==null) return '—'; mb=+mb; if(!mb) return '—'; return mb>=1024 ? (mb/1024).toFixed(2)+' ГБ' : mb.toFixed(1)+' МБ'; } function size(mb){ if(mb==null) return '—'; mb=+mb; if(!mb) return '—'; return mb>=1024 ? (mb/1024).toFixed(2)+' ГБ' : mb.toFixed(1)+' МБ'; }
function dur(sec){ sec=Math.round(sec||0); if(sec<60) return sec+' с'; var m=Math.floor(sec/60); if(m<60) return m+' мин'; var h=Math.floor(m/60); if(h<24) return h+' ч '+(m%60)+' мин'; return Math.floor(h/24)+' дн '+(h%24)+' ч'; } function dur(sec){ sec=Math.round(sec||0); if(sec<60) return sec+' с'; var m=Math.floor(sec/60); if(m<60) return m+' мин'; var h=Math.floor(m/60); if(h<24) return h+' ч '+(m%60)+' мин'; return Math.floor(h/24)+' дн '+(h%24)+' ч'; }
function psec(s){ if(s==null) return '—'; s=+s; if(s<60) return s.toFixed(1)+' с'; var m=Math.floor(s/60); return m+' м '+Math.round(s%60)+' с'; }
function dts(ts){ if(!ts) return '—'; var d=new Date(ts*1000); return d.toLocaleString('ru-RU',{day:'2-digit',month:'2-digit',year:'numeric',hour:'2-digit',minute:'2-digit'}); } function dts(ts){ if(!ts) return '—'; var d=new Date(ts*1000); return d.toLocaleString('ru-RU',{day:'2-digit',month:'2-digit',year:'numeric',hour:'2-digit',minute:'2-digit'}); }
function dOnly(ts){ if(!ts) return '—'; var d=new Date(ts*1000); return d.toLocaleDateString('ru-RU'); } function dOnly(ts){ if(!ts) return '—'; var d=new Date(ts*1000); return d.toLocaleDateString('ru-RU'); }
function daysLeft(ts){ var s=Math.max(0,(ts*1000-Date.now())/86400000); return s>=365 ? Math.round(s/365)+' г.' : Math.round(s)+' дн.'; } function daysLeft(ts){ var s=Math.max(0,(ts*1000-Date.now())/86400000); return s>=365 ? Math.round(s/365)+' г.' : Math.round(s)+' дн.'; }
@@ -2866,8 +2867,9 @@ _ADMIN_HTML = """<!doctype html>
var dd = x.day.slice(8)+'.'+x.day.slice(5,7)+'.'+x.day.slice(0,4); var dd = x.day.slice(8)+'.'+x.day.slice(5,7)+'.'+x.day.slice(0,4);
return '<tr><td>'+dd+'</td><td class="num">'+(x.visits||0)+'</td><td class="num">'+(x.uniques||0)+ return '<tr><td>'+dd+'</td><td class="num">'+(x.visits||0)+'</td><td class="num">'+(x.uniques||0)+
'</td><td class="num">'+(x.files||0)+'</td><td class="num">'+(x.pages||0).toLocaleString('ru-RU')+ '</td><td class="num">'+(x.files||0)+'</td><td class="num">'+(x.pages||0).toLocaleString('ru-RU')+
'</td><td class="num">'+(x.avg_page?psec(x.avg_page):'—')+
'</td><td class="num">'+size(x.src_mb)+'</td><td class="num">'+size(x.res_mb)+'</td></tr>'; '</td><td class="num">'+size(x.src_mb)+'</td><td class="num">'+size(x.res_mb)+'</td></tr>';
}).join('') : '<tr><td colspan="7" class="muted">Пока нет данных</td></tr>'; }).join('') : '<tr><td colspan="8" class="muted">Пока нет данных</td></tr>';
// Платежи // Платежи
var bp = d.payments.by_purpose || {}; var bp = d.payments.by_purpose || {};
@@ -2890,9 +2892,9 @@ _ADMIN_HTML = """<!doctype html>
var srcPill = web ? '<span class="pill web">сайт</span>' : '<span class="pill bot">бот</span>'; var srcPill = web ? '<span class="pill web">сайт</span>' : '<span class="pill bot">бот</span>';
var dd = (f.date||'') + (f.time ? ' '+String(f.time).slice(0,5) : ''); var dd = (f.date||'') + (f.time ? ' '+String(f.time).slice(0,5) : '');
return '<tr><td>'+esc(dd)+'</td><td class="fname" title="'+esc(f.name)+'">'+esc(f.name)+'</td><td class="num">'+(f.pages||'—')+ return '<tr><td>'+esc(dd)+'</td><td class="fname" title="'+esc(f.name)+'">'+esc(f.name)+'</td><td class="num">'+(f.pages||'—')+
'</td><td class="num">'+size(f.source_size_mb)+'</td><td class="num">'+size(f.result_size_mb)+ '</td><td class="num">'+psec(f.secs)+'</td><td class="num">'+size(f.source_size_mb)+'</td><td class="num">'+size(f.result_size_mb)+
'</td><td>'+esc(f.quality||'—')+'</td><td>'+esc(f.stitching||'—')+'</td><td>'+(f.ocr?'✓':'')+'</td><td>'+srcPill+'</td></tr>'; '</td><td>'+esc(f.quality||'—')+'</td><td>'+esc(f.stitching||'—')+'</td><td>'+(f.ocr?'✓':'')+'</td><td>'+srcPill+'</td></tr>';
}).join('') : '<tr><td colspan="9" class="muted">Пока нет документов</td></tr>'; }).join('') : '<tr><td colspan="10" class="muted">Пока нет документов</td></tr>';
} }
function showGate(err){ function showGate(err){
@@ -2967,6 +2969,19 @@ def _require_admin(request: Request) -> bool:
return False return False
def _elapsed_to_sec(et):
"""Парсит elapsed_time ('H:MM:SS.ffffff') в секунды; None если не распарсилось."""
if not et:
return None
try:
pp = str(et).split(":")
if len(pp) == 3:
return int(pp[0]) * 3600 + int(pp[1]) * 60 + float(pp[2])
except Exception:
pass
return None
def _admin_collect_stats() -> dict: def _admin_collect_stats() -> dict:
now = time.time() now = time.time()
out = {"generated_at": now, "web_user_id": WEB_USER_ID} out = {"generated_at": now, "web_user_id": WEB_USER_ID}
@@ -3009,8 +3024,10 @@ def _admin_collect_stats() -> dict:
pages_expr = "COALESCE(NULLIF(pages_total,0), num_pages, 0)" if "pages_total" in cols else "COALESCE(num_pages,0)" pages_expr = "COALESCE(NULLIF(pages_total,0), num_pages, 0)" if "pages_total" in cols else "COALESCE(num_pages,0)"
docs = [dict(r) for r in con.execute( docs = [dict(r) for r in con.execute(
f"SELECT date, time, COALESCE(NULLIF(original_name,''), name_or_f) AS name, {pages_expr} AS pages, " f"SELECT date, time, COALESCE(NULLIF(original_name,''), name_or_f) AS name, {pages_expr} AS pages, "
"source_size_mb, result_size_mb, quality, stitching, ocr, user_id, url_res_f " "source_size_mb, result_size_mb, quality, stitching, ocr, elapsed_time, user_id, url_res_f "
"FROM FILES ORDER BY rowid DESC LIMIT 150")] "FROM FILES ORDER BY rowid DESC LIMIT 150")]
for _d in docs:
_d["secs"] = _elapsed_to_sec(_d.get("elapsed_time"))
agg = con.execute( agg = con.execute(
f"SELECT COUNT(*) c, COALESCE(SUM({pages_expr}),0) p, " f"SELECT COUNT(*) c, COALESCE(SUM({pages_expr}),0) p, "
"COALESCE(SUM(source_size_mb),0) ssz, COALESCE(SUM(result_size_mb),0) rsz FROM FILES").fetchone() "COALESCE(SUM(source_size_mb),0) ssz, COALESCE(SUM(result_size_mb),0) rsz FROM FILES").fetchone()
@@ -3020,13 +3037,10 @@ def _admin_collect_stats() -> dict:
for r in con.execute( for r in con.execute(
f"SELECT elapsed_time AS et, {pages_expr} AS pg FROM FILES " f"SELECT elapsed_time AS et, {pages_expr} AS pg FROM FILES "
"WHERE elapsed_time IS NOT NULL AND TRIM(elapsed_time) NOT IN ('','-')"): "WHERE elapsed_time IS NOT NULL AND TRIM(elapsed_time) NOT IN ('','-')"):
try: s = _elapsed_to_sec(r["et"])
pp = str(r["et"]).split(":") if s is not None:
if len(pp) == 3: tot_sec += s
tot_sec += int(pp[0]) * 3600 + int(pp[1]) * 60 + float(pp[2]) pages_timed += (r["pg"] or 0)
pages_timed += (r["pg"] or 0)
except Exception:
pass
out["docs"] = {"recent": docs, "total_count": agg["c"], "total_pages": agg["p"], out["docs"] = {"recent": docs, "total_count": agg["c"], "total_pages": agg["p"],
"total_source_mb": agg["ssz"], "total_result_mb": agg["rsz"], "total_source_mb": agg["ssz"], "total_result_mb": agg["rsz"],
"total_seconds": tot_sec, "avg_page_seconds": (tot_sec / pages_timed) if pages_timed else 0} "total_seconds": tot_sec, "avg_page_seconds": (tot_sec / pages_timed) if pages_timed else 0}
@@ -3035,19 +3049,29 @@ def _admin_collect_stats() -> dict:
_cutoff = (_today - datetime.timedelta(days=13)).strftime("%Y.%m.%d") _cutoff = (_today - datetime.timedelta(days=13)).strftime("%Y.%m.%d")
files_by_day = {} files_by_day = {}
for r in con.execute( for r in con.execute(
f"SELECT date AS d, COUNT(*) c, COALESCE(SUM({pages_expr}),0) p, " f"SELECT date AS d, {pages_expr} AS pg, source_size_mb AS s, result_size_mb AS rs, elapsed_time AS et "
"COALESCE(SUM(source_size_mb),0) s, COALESCE(SUM(result_size_mb),0) rs " "FROM FILES WHERE date >= ?", (_cutoff,)):
"FROM FILES WHERE date >= ? GROUP BY date", (_cutoff,)): key = (r["d"] or "").replace(".", "-")
files_by_day[(r["d"] or "").replace(".", "-")] = { e = files_by_day.setdefault(key, {"files": 0, "pages": 0, "src_mb": 0.0,
"files": r["c"], "pages": r["p"], "src_mb": r["s"], "res_mb": r["rs"]} "res_mb": 0.0, "secs": 0.0, "pages_timed": 0})
e["files"] += 1
e["pages"] += (r["pg"] or 0)
e["src_mb"] += (r["s"] or 0)
e["res_mb"] += (r["rs"] or 0)
sec = _elapsed_to_sec(r["et"])
if sec is not None:
e["secs"] += sec
e["pages_timed"] += (r["pg"] or 0)
daily = [] daily = []
for i in range(13, -1, -1): for i in range(13, -1, -1):
dkey = (_today - datetime.timedelta(days=i)).strftime("%Y-%m-%d") dkey = (_today - datetime.timedelta(days=i)).strftime("%Y-%m-%d")
vc = next((x["count"] for x in days if x["day"] == dkey), 0) vc = next((x["count"] for x in days if x["day"] == dkey), 0)
fb = files_by_day.get(dkey, {}) fb = files_by_day.get(dkey, {})
ptimed = fb.get("pages_timed", 0)
daily.append({"day": dkey, "visits": vc, "uniques": udays.get(dkey, 0), daily.append({"day": dkey, "visits": vc, "uniques": udays.get(dkey, 0),
"files": fb.get("files", 0), "pages": fb.get("pages", 0), "files": fb.get("files", 0), "pages": fb.get("pages", 0),
"src_mb": fb.get("src_mb", 0), "res_mb": fb.get("res_mb", 0)}) "src_mb": fb.get("src_mb", 0), "res_mb": fb.get("res_mb", 0),
"avg_page": (fb.get("secs", 0) / ptimed) if ptimed else 0})
out["daily"] = daily out["daily"] = daily
return out return out
@@ -3063,7 +3087,7 @@ def _build_docs_csv() -> bytes:
import csv, io import csv, io
buf = io.StringIO() buf = io.StringIO()
w = csv.writer(buf, delimiter=";") w = csv.writer(buf, delimiter=";")
w.writerow(["Дата", "Время", "Файл", "Страниц", "Исходник, МБ", w.writerow(["Дата", "Время", "Файл", "Страниц", "Время обраб., с", "Исходник, МБ",
"Результат, МБ", "Качество", "Сшивка", "OCR", "Источник", "Статус"]) "Результат, МБ", "Качество", "Сшивка", "OCR", "Источник", "Статус"])
def _mb(v): def _mb(v):
return ("%.2f" % v) if v else "" return ("%.2f" % v) if v else ""
@@ -3073,9 +3097,11 @@ def _build_docs_csv() -> bytes:
pages_expr = "COALESCE(NULLIF(pages_total,0), num_pages, 0)" if "pages_total" in cols else "COALESCE(num_pages,0)" pages_expr = "COALESCE(NULLIF(pages_total,0), num_pages, 0)" if "pages_total" in cols else "COALESCE(num_pages,0)"
for r in con.execute( for r in con.execute(
f"SELECT date, time, COALESCE(NULLIF(original_name,''), name_or_f) AS name, {pages_expr} AS pages, " f"SELECT date, time, COALESCE(NULLIF(original_name,''), name_or_f) AS name, {pages_expr} AS pages, "
"source_size_mb, result_size_mb, quality, stitching, ocr, user_id, url_res_f FROM FILES ORDER BY rowid DESC"): "source_size_mb, result_size_mb, quality, stitching, ocr, elapsed_time, user_id, url_res_f FROM FILES ORDER BY rowid DESC"):
src = "сайт" if (r["user_id"] == WEB_USER_ID) else "бот" src = "сайт" if (r["user_id"] == WEB_USER_ID) else "бот"
_et = _elapsed_to_sec(r["elapsed_time"])
w.writerow([r["date"] or "", (r["time"] or "")[:8], r["name"] or "", r["pages"] or "", w.writerow([r["date"] or "", (r["time"] or "")[:8], r["name"] or "", r["pages"] or "",
("%.1f" % _et) if _et is not None else "",
_mb(r["source_size_mb"]), _mb(r["result_size_mb"]), r["quality"] or "", _mb(r["source_size_mb"]), _mb(r["result_size_mb"]), r["quality"] or "",
r["stitching"] or "", "да" if r["ocr"] else "", src, r["url_res_f"] or ""]) r["stitching"] or "", "да" if r["ocr"] else "", src, r["url_res_f"] or ""])
return ("" + buf.getvalue()).encode("utf-8") # BOM → кириллица в Excel return ("" + buf.getvalue()).encode("utf-8") # BOM → кириллица в Excel