From e7d1185836b79a8ccfdcb0afa844ba056e7002a6 Mon Sep 17 00:00:00 2001
From: "John R. D'Orazio" <priest@johnromanodorazio.com>
Date: Wed, 29 Apr 2026 19:48:27 +0200
Subject: [PATCH] fix(deploy): make verify step tolerate transient curl errors
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The verify step exited with code 56 (CURLE_RECV_ERROR) during the
first big-batch poll, killing the entire workflow run on a single
transient blip. Three robustness fixes:

- Append `|| echo '{}'` to each curl call so `set -e` doesn't kill
  the script when curl fails for any reason — the empty JSON makes
  jq return empty strings, which keeps the page in PENDING for the
  next attempt (correct semantics: "not yet translated").
- Add `--max-time 15 --retry 3 --retry-delay 5 --retry-all-errors`
  so curl handles short network blips at the transport layer.
- Append `2>/dev/null || echo ""` to jq calls for the same reason.

Also bump MAX_ATTEMPTS from 20 to 60 (10 min → 30 min). The current
deploy enqueues up to 50 translations at a time (10 docs × 5 langs
on workflow_dispatch / first deploy), and observed throughput is
~1 translation every 5–8 seconds — 50 jobs needs ~5 minutes minimum,
plus retry budget for OpenAI capacity issues.

Progress messages now show "X/N done" so the run status is readable
at a glance.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---
 .github/workflows/deploy-docs.yml | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/.github/workflows/deploy-docs.yml b/.github/workflows/deploy-docs.yml
index 3650e89..89362e5 100644
--- a/.github/workflows/deploy-docs.yml
+++ b/.github/workflows/deploy-docs.yml
@@ -308,20 +308,26 @@ jobs:
           # without this step a silent worker failure (OpenAI timeout, missing
           # API key, dead Redis worker) lets the deploy report success while
           # leaving translations stale.
-          MAX_ATTEMPTS=20      # 20 x 30s = 10 minutes
+          MAX_ATTEMPTS=60      # 60 x 30s = 30 minutes
           SLEEP_SECONDS=30
 
           declare -A PENDING
           for ID in $TRANSLATED_IDS; do PENDING[$ID]=1; done
+          INITIAL_COUNT=${#PENDING[@]}
 
           for attempt in $(seq 1 "$MAX_ATTEMPTS"); do
             REMAINING=()
             for ID in "${!PENDING[@]}"; do
-              RESPONSE=$(curl -s \
+              # `|| echo '{}'` keeps `set -e` from killing the whole step on
+              # transient curl failures (we observed exit 56 / CURLE_RECV_ERROR
+              # mid-poll). `--retry` handles short network blips at the curl
+              # layer; the `|| echo` is the safety net for everything else.
+              RESPONSE=$(curl -s --max-time 15 --retry 3 --retry-delay 5 --retry-all-errors \
                 -u "$WP_APP_USERNAME:$WP_APP_PASSWORD" \
-                "$WP_REST_URL/wp/v2/pages/${ID}?_fields=id,modified_gmt,status&context=edit")
-              MODIFIED=$(echo "$RESPONSE" | jq -r '.modified_gmt // empty')
-              STATUS=$(echo "$RESPONSE" | jq -r '.status // empty')
+                "$WP_REST_URL/wp/v2/pages/${ID}?_fields=id,modified_gmt,status&context=edit" \
+                || echo '{}')
+              MODIFIED=$(echo "$RESPONSE" | jq -r '.modified_gmt // empty' 2>/dev/null || echo "")
+              STATUS=$(echo "$RESPONSE" | jq -r '.status // empty' 2>/dev/null || echo "")
 
               # modified_gmt is ISO 8601 with no offset; lexicographic compare
               # against DEPLOY_START_GMT (also UTC, no offset) is correct.
@@ -343,7 +349,8 @@ jobs:
               exit 0
             fi
 
-            echo "Attempt ${attempt}/${MAX_ATTEMPTS}: ${#PENDING[@]} pending - sleeping ${SLEEP_SECONDS}s..."
+            COMPLETED=$((INITIAL_COUNT - ${#PENDING[@]}))
+            echo "Attempt ${attempt}/${MAX_ATTEMPTS}: ${COMPLETED}/${INITIAL_COUNT} done, ${#PENDING[@]} pending - sleeping ${SLEEP_SECONDS}s..."
             sleep "$SLEEP_SECONDS"
           done