geminispace.info

commit b2dfd8f3b26f53382350fe87a088e855479c9445
parent 958574c5b641de21f5098982d905d5a34d5b4402
Author: René Wagner <rwa@clttr.info>
Date:   Sat,  5 Aug 2023 21:12:19 +0200

new excludes and README update

Diffstat:
M
README.md
 | 
5
+++++
M
gus/excludes.py
 | 
8
++++----
M
infra/update_index.sh
 | 
10
+++-------

3 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/README.md b/README.md
@@ -61,3 +61,8 @@ Now you'll have created `index.new` directory, rename it to `index`.
 ## Running the test suite
 
 Run: `poetry run pytest`
+
+## Troubleshooting
+
+* restarting gus.service fails with an error that the bus has not been found when called in a cronjob
+--> some env vars might be missing, prefix the cronjob command with `XDG_RUNTIME_DIR="/run/user/1000" DBUS_SESSION_BUS_ADDRESS="unix:path=${XDG_RUNTIME_DIR}/bus"`
diff --git a/gus/excludes.py b/gus/excludes.py
@@ -42,8 +42,8 @@ EXCLUDED_URL_PREFIXES = [
 
     # Mastodon mirror
     "gemini://vps01.rdelaage.ovh/",
-    "gemini://mastogem.picasoft.net",
-    "gemini://mastogem.remorse.us",
+    "gemini://mastogem.picasoft.net/",
+    "gemini://mastogem.remorse.us/",
 
     # various failing resources on runjimmyrunrunyoufuckerrun.com
     "gemini://runjimmyrunrunyoufuckerrun.com/fonts/",
@@ -194,9 +194,9 @@ EXCLUDED_URL_PREFIXES = [
     "gemini://cfdocs.wetterberg.nu/",
     "gemini://godocs.io",
 
-    # skyjakes git repos
+    # git repos
     "gemini://git.skyjake.fi",
-
+    "gemini://gemini.unlimited.pizza/git",
     # chess games or stuff
     "gemini://jsreed5.org/live/",
 ]
diff --git a/infra/update_index.sh b/infra/update_index.sh
@@ -1,10 +1,9 @@
 #!/bin/bash
 BASEDIR="$1"
-XDG_RUNTIME_DIR="/run/user/1000"
-DBUS_SESSION_BUS_ADDRESS="unix:path=${XDG_RUNTIME_DIR}/bus"
 
 cd "${BASEDIR}"
 LOCKFILE="${BASEDIR}/crawl.lock"
+LOGFILE="${BASEDIR}/crawl.log"
 
 if [[ -f "${LOCKFILE}" ]];
 then
@@ -13,14 +12,11 @@ then
 fi
 
 touch "${LOCKFILE}"
-rm -f "${BASEDIR}/crawl.log"
+rm -f "${LOGFILE}"
 "${BASEDIR}/infra/filter_seed_request.sh" "${BASEDIR}"
 
 /home/gus/.local/bin/poetry run crawl
 /home/gus/.local/bin/poetry run build_index
-/usr/bin/systemctl --user stop gus
-/usr/bin/systemctl --user stop gus-tor
-/usr/bin/systemctl --user start gus
-/usr/bin/systemctl --user start gus-tor
+/usr/bin/systemctl --user restart gus >> "${LOGFILE}" 2>&1
 
 rm "${LOCKFILE}"