From 5d9b3a6070d8885a803202385c5850aabc759dc4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ebbe=20Ba=C3=9F?= <ebbe.bass@bechtle.com>
Date: Wed, 9 Sep 2026 11:40:35 +0200
Subject: [PATCH] changed method of fetching crontab schedule

---
 api/README.md                                 |  18 ++++-
 .../scrape_schedule.cpython-313.pyc           | Bin 0 -> 5559 bytes
 api/main.py                                   |  12 ++-
 api/scrape_schedule.py                        |  69 ++++++++++++++++--
 4 files changed, 87 insertions(+), 12 deletions(-)
 create mode 100644 api/__pycache__/scrape_schedule.cpython-313.pyc

diff --git a/api/README.md b/api/README.md
index 8e4da39..0bf8991 100644
--- a/api/README.md
+++ b/api/README.md
@@ -31,8 +31,17 @@ DATA_FILE=/path/to/events.json ARCHIVE_DATA_FILE=/path/to/archive.json uvicorn m
 ### Scrape schedule reporting
 
 `/health` reports when each scraper output file was last written (its mtime)
-and, if you tell it the cron schedule, when it's next due. This doesn't read
-your crontab — set the same expression(s) you put there as env vars:
+and, if it can determine the cron schedule, when it's next due. The schedule
+is sourced in order:
+
+1. `SCRAPE_CRON` / `ARCHIVE_SCRAPE_CRON` env vars, if set — always wins, and
+   the only option that works when the API doesn't run as the same user/host
+   as the cron job.
+2. Otherwise, the API's own OS user's crontab (`crontab -l`), looked up for a
+   line invoking `scraper.py` (with vs. without `--archive` picks archive vs.
+   plain). Only works when the API process runs as the same user whose
+   personal crontab holds the scrape job — not a system crontab/cron.d entry,
+   not a job scheduled under a different user or host.
 
 ```bash
 SCRAPE_CRON="0 * * * *" ARCHIVE_SCRAPE_CRON="0 4 * * *" uvicorn main:app
@@ -40,8 +49,9 @@ SCRAPE_CRON="0 * * * *" ARCHIVE_SCRAPE_CRON="0 4 * * *" uvicorn main:app
 
 `SCRAPE_CRON` covers the plain scrape (`fundi-scraped-output.json`);
 `ARCHIVE_SCRAPE_CRON` covers `--archive` (`fundi-archive-output.json`) and
-falls back to `SCRAPE_CRON` if unset — set it separately only if the archive
-scrape runs on its own cron line. Leave both unset to just get the
+falls back to `SCRAPE_CRON` (env or crontab-discovered) if unset — set it
+separately only if the archive scrape runs on its own cron line. If neither
+an env var nor a matching crontab line is found, you just get the
 last-scrape info with `next_scrape_at` / `seconds_until_next_scrape` as
 `null`.
 
diff --git a/api/__pycache__/scrape_schedule.cpython-313.pyc b/api/__pycache__/scrape_schedule.cpython-313.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..fac45a5624cd4c17c4970cafe6178dfbce04c780
GIT binary patch
literal 5559
zcmb7ITWl2989uW+d-2ZNyXy-Ot_Nd`H?`M5Od#fJ2nY<A#EVlJoUX^av-ZGvW_xDV
z#+EAF<e?^1!l{CAq*QWLsRgMIJf>1xg}1!4^2SkjGjh^OZ6oov2r8jG_5aW8tZi^E
zoz>2q`Okm;bIyPL&OENI4H3Amw|#V_DN4w{@Ws6XmB!<z(6~*Kgps6>^qdu#08j5(
zPp^l0DAL5U-d>T3yzM*d>-95#Z-51QgDl8t{<Af`HLM2Ofk97~CmGyGu81rIZ~j5a
z2hxF7!)f7UXix%rU@$l+4N6@e?ByI{QsNx7$uMx#4%TrlXS6?A$NM75$e_sSd|a29
zjB<K38G|wPgPs+A{$vC1i}5irn32l@j`}WtvXS>SjCkTr^4~m!^eZ`oStgwrRdgy-
zGtK0jLN(d6D3gJrXPmAyGjuwkY8us*YtWb1scI!cq2_ZH)@G`jlqqtCW|=WYtx<)J
zS?ZWVjVyn0B(q&6HS$(2Z_%u(DYS!HEU$!^G9t5#rkK#1rK83~CtnW=sjO&1HmRJp
z^0KB)QdJ)}t}2-XZSJcqrWux{WK5hX#FuKON0m%oQ`)I%<XKwDz#wX{jKXk+98<Ci
zQ}ndb9SZG9&>o#C*K(SgRxPUN<8)kRv@Lm}zvs1+sT2Jd`e-NZ=|6G$%o`_DE2wc;
zzN}5ilb}9TH`_y$(g}lIg_Y`)umPqdLVFYRBJ3iaXE2>bFC=N+R8R}JN=qK5Z9_vD
z)l3`X3QHj^1q)$ALnzzOVXuxrR0oQ;MnTItW~5P7U<8<oZ<?Ais2W9sxeQaGIfa>q
zE^902ai{E1Kt4)K;}T%yD-1-h7{l${?5O;qp~{${Av&z;G6UVil;g6h$-|nGpnZmg
zg3yPMK!=q{16;-$l`YzqQL|jj7GGF9)eXKHFv+B8DP!)^I&qFf2Iat7EOa3-k-WSG
zLkBZf)(YFACRoY>Z(s%}0d{Md?KF`{#1o;17}<Ss!476*OF{R8w)mQCjmEvUcuqAf
zJ9r^ysi+y-cQK#SkRgZ*P%VYoex_J?rl(PidqI#8FZ>^$hTCm&*`<?y?0M#DCi^Os
zehltsF1N3R+!OlZUWQ_AuWn4(A&4Sa?G$YF9(jzz?4u{T2M55Q=Ae=omQ8c;gppD8
zk-=<U&!`=haO#kA>LC7|_mu1QiQJ?eN`d5*W~60}#X#Z)_?zqDHcghKhFgc;J#_O(
zQEDkjEsIj?g4oI;(6?&qm?-(yKZC|?vh{m6k7=~ilk~z*O!{6CbuvKugE07+E9vi^
zAVdZQc9It4;mLY6Q-AI1cYIq`;C6CZbo*DB#P8CUpIx2+k--=g<}y9oV7@Fl?A=DP
zLOd{ayq_DK2LR71py4%wR8w^YoXlYK@>^8R(!9PL19qSiBU7FZ?czf;J2+ciTF!wh
zDJdu0*`fuR$_}8Bfx^&r;sHBk=7%fMY<nS`?ZAcPNyZFjN0QYZhX6T}yQYy*Y`<w`
zKne`UNZ^YDHwNc5^O|J}Ae%BMdc<x>VMgY81>!TsIT?kX&G0u3xJ{F?81&UI#Wvm9
z`s=Nw+I4f{LU>DQ)3#Eqsnpo?sK!?xn(>#TBv@ArY$yaa%=Ih=wv-~VnVOOmyLI@+
z;i9yoAnjO`b}ooJpFXJ}wT<t;S#0VoG<7aWyB`z5R}b9p)|9=_`o=^@eEWr-0C^DD
z+|%lPuv37$E#k`3xN!GEHFDQG2h`i>A}f0<T&_Eyo3&MF@;ZXtIQE$<DeSArsPe-5
zi050lz0YySx#TtCnmfCo3zuDFK?bUGRb1VfNk|58_y8aae0F)b^RFc_2;uW>5^H$y
zaEZxKf6X)7bmRjT5sAA4m@nwg7ni0^BzXdbT;rx-iu#A_YZ<i7O(B?a$Ho8v#1BH6
z$L=g-aLBt@A`-T+qt6zOJw%OS3LTTLac6vpp3U8xHA3ql2Ba;}xX<>P7E^QJvz82i
zV?wn?frJaOy_%xi;uS;H;~ush2l#TxW1=itb!DKL$|_%GyMdt{{^oDsHckF1MQ(N9
z=)QSq`qi(b$ZXf0qwgJ^f2r88tI)8k6lo|0q+eeD#r0XU7}!(@Y$`<??}Xk9&G*cP
z7NV^)z9p${CV#VgE_U;1sbPI7(m4D2jcapFb3glF!$M^HLU4O29KF?hqjz4I>0Jmn
zPoG>0gl6`=Gd27AJJ(UVRCnvZjRUhM=bGpC%$`}4w$A(J<&T03Qrm*q#*?cZuBbFM
z3c9ufGE0xD<I9eZ%EYY39e*4eP;QXMFW1={$mt&uqI)h^U3|df8d~=bc+;S@^9|r+
z-aNE&ColO9f(*>H=eH7)^bGh8g1iC2&A{i;twqyAAu~#Zk`W>9l>;#3lnRwxrem@N
ztAw;zL1JKi#8L<kTvW3f(f}$1+~D{Q-zb0wL=6;@9r^%>QZoRC)a)d-QBXy6We{ys
z5+jLFbEU>XEo%lJ92%;q3&iErG@vXT+1$}lAvZftJZ>g<M@|D|0}u)@n3Lnpsu%O6
zSKwD#)^}L|TO8;L2_8TS4+g$%#Y}|yrfRBlfPYd@9{0PrpY6f+Uc7CGo9*Wwnz4g!
z2tjFZ)L>#50KS3CNzC=nV?MSYcn-ne?1S4hc~VDWEq{;hoDr8|E%O%_V=vD5A?4mU
z@<q6@w7Izud|@V7lGa1*MnZV~-2Otid46J1O5l5B{kxGz4J6iDZY1HT<9%WdfY7Qe
zTXnV%Vagv;;2~G8k*!?U>|dk&x~{oao^r>v5GdzW2ms|8T$_?@WTaZEnFI?2jY9{q
zYw?7El@wCRSk4yV0l3bzQk=*TH`uXLaJ&F?dddK}#4BiAU<aUQWqFLH@DSn*s{o}V
z!15Zh9fcbdM~~OnxD;7mjO_d@vU5K3d-Wsrw^xg;-G$cf#mK>8<WwPYYB6$pA$a=f
zlC&AiUE>##1}t`s-<Sxf-adGQ{3X;Qc^`xXxZ5>|7C6YaeR38+C&p{_sv6{XKth8I
z2x}F-BpgnmJgmq8mpuC_a@I1?<*H%+e{)<8yQ||!>N(TIJqWn{)|9Ge4Zt)3ra<ha
z6^6La4vr{R3XhE3bdGI8Z^&9+abQjGtTO4C$IDRV>==%cbQ$tuitEQjn^yy;jtg^6
zN&uMHY3Teh{LS-lLwkw7#?rb?cgEfuyW2H8wy<vZj0E<Zt67Y679$4=kpq9)uoyXB
zjGQk-&M!tTECeq+T?()J5YqJawXD~DnEW}~6ZJlb2ykCzE8HKh?eQ!m%e6AK?immc
z3oh8mfPL_z!CKJ74TR^oJpd}I0XI-B=#b~Da%BjOW#DnC-t~hEfW>ujKp614E06&b
z4tT#S5OK^3{!Jhm1CL*qc6309eitfteTz#GSIsJZ@2t!GTUO0_{D;n(^d@{QP~V(q
zT%7vD@w}?xS=9NS(TQGIIXBT^I2K){S;+T@Y$}Cj<(CPkh`FKaR4{(3b$BZWc>HPr
zsM~}CrBk$M^KSa0^MjfdLfq5|=L%U<GqmFf|8~-W-%hfsqQU11nJHAyYg!^?!?D$8
z*Yf00EgyCG9R|*H_)R7SIqn5hfn#vSOqnWtx=5|)wd-(qYq92`&}vT5ID+B<ei~#i
zf>jx&J03%JjcOX84=8)Kk73fZYc0c)wNz!!xR0k#CrB|Ruyc6BT4kb#LWH`-Erv1g
zZX$(gfhB<>g{=4B2DL3rq^6tg(<e%?jd!-q4Hu)k3ejDQ(ReYs=kCSnGyiUCowpX7
z+Nb-z+JEf*pU$88yy>M+GShu!KiRnBs{^Oz62*;s3mf--zOn1mSHC`d_E#roUjmF*
ztbe&s|MFt}fzQLQEWAGO^_q-FRx&PzUwIS@h5cm`^5H3`UwA`UCh)QuZ?prcRK`fB
zQVfsKP7MlRE!EUvh8Y{O-((K33`0dQ%(V<p0z3n9W!qi=KFiR-7KaT(;{|_(1;mV+
zwz#HP9gfE8;h=@!p)mtLW*uQ~0u8baG5-oTs4{}^$WtqbkGB$`{!6muOS0h~q^&^O
z{zJBu{erM-_Vg11k8-e9Xq~$_zvF|SJR#65(^?@umo7s)K0kUle?R&0j{BGHT`#mA
z#r}Ai#)Rm6vP|G{Kf@m%U;EVi$DcnzO1VuG8s}au6L`#Dx;t9fx&H~a%XL1WgH(9j
oZR3yo*FFgr_Mdrzq;hnf;KyO`xO?e-`rcc-`LXq1r10(j4**mvn*aa+

literal 0
HcmV?d00001

diff --git a/api/main.py b/api/main.py
index 4cb50fa..6e959bd 100644
--- a/api/main.py
+++ b/api/main.py
@@ -17,7 +17,7 @@ from fastapi.responses import PlainTextResponse
 from calendar_feed import build_ics
 from datasource import EventStore, iter_artist_names, parse_event_date
 from models import Event, EventList
-from scrape_schedule import get_schedule_info
+from scrape_schedule import discover_cron_expr, get_schedule_info
 
 app = FastAPI(
     title="Fundi Scraper API",
@@ -59,8 +59,14 @@ def _matches(
 
 @app.get("/health")
 def health():
-    scrape_cron = os.environ.get("SCRAPE_CRON")
-    archive_scrape_cron = os.environ.get("ARCHIVE_SCRAPE_CRON") or scrape_cron
+    scrape_cron = os.environ.get("SCRAPE_CRON") or discover_cron_expr(
+        "scraper.py", archive=False
+    )
+    archive_scrape_cron = (
+        os.environ.get("ARCHIVE_SCRAPE_CRON")
+        or discover_cron_expr("scraper.py", archive=True)
+        or scrape_cron
+    )
     return {
         "status": "ok",
         "events_loaded": len(store.all()),
diff --git a/api/scrape_schedule.py b/api/scrape_schedule.py
index d2e876d..e584219 100644
--- a/api/scrape_schedule.py
+++ b/api/scrape_schedule.py
@@ -1,15 +1,22 @@
 """Reports when a scrape last ran and when cron will next run it.
 
 "Last ran" is read from the mtime of the scraper's output file - true
-regardless of how/when the scrape was actually invoked. "Next run" requires
-telling this module the cron schedule via an env var (there's no reliable,
-portable way to introspect another process's crontab from here), given as
-a standard 5-field cron expression, e.g. "0 * * * *".
+regardless of how/when the scrape was actually invoked. "Next run" needs the
+cron schedule, sourced in order of preference:
+
+1. An explicit env var (SCRAPE_CRON / ARCHIVE_SCRAPE_CRON) - always wins,
+   works anywhere.
+2. The current OS user's crontab (``discover_cron_expr``) - only works when
+   this process runs as the same user whose personal crontab (``crontab -e``)
+   contains the scrape job, and the ``crontab`` binary is available. Not
+   portable beyond that (different user, no cron, system crontab/cron.d
+   instead of a personal one, containers split across hosts, ...).
 """
 
+import subprocess
 from datetime import datetime
 from pathlib import Path
-from typing import Optional
+from typing import List, Optional, Tuple
 
 from croniter import croniter
 
@@ -18,6 +25,58 @@ def _now_local() -> datetime:
     return datetime.now().astimezone()
 
 
+def _read_crontab_lines() -> List[str]:
+    """Return the invoking OS user's crontab lines, or [] if unavailable."""
+    try:
+        result = subprocess.run(
+            ["crontab", "-l"],
+            capture_output=True,
+            text=True,
+            timeout=5,
+        )
+    except (OSError, subprocess.SubprocessError):
+        return []
+    if result.returncode != 0:
+        return []
+    return result.stdout.splitlines()
+
+
+def _parse_crontab_line(line: str) -> Optional[Tuple[str, str]]:
+    """Split one crontab line into (schedule, command); None if not a job."""
+    line = line.strip()
+    if not line or line.startswith("#"):
+        return None
+    if line.startswith("@"):
+        parts = line.split(maxsplit=1)
+        if len(parts) != 2:
+            return None
+        return parts[0], parts[1]
+    parts = line.split(maxsplit=5)
+    if len(parts) != 6:
+        return None
+    return " ".join(parts[:5]), parts[5]
+
+
+def discover_cron_expr(command_hint: str, *, archive: bool) -> Optional[str]:
+    """Find a matching scrape job in the current user's crontab.
+
+    ``command_hint`` is a substring identifying the scrape command (e.g.
+    "scraper.py"); ``archive`` picks the "--archive" line vs. the plain one.
+    Returns None if the crontab can't be read or no line matches.
+    """
+    for line in _read_crontab_lines():
+        parsed = _parse_crontab_line(line)
+        if parsed is None:
+            continue
+        schedule, command = parsed
+        if command_hint not in command:
+            continue
+        if ("--archive" in command) != archive:
+            continue
+        return schedule
+    return None
+
+
 def _file_mtime_local(path: Path) -> Optional[datetime]:
     try:
         return datetime.fromtimestamp(path.stat().st_mtime).astimezone()