From 61ccea029cdf787aa7a0775c94be5ed35d06863f Mon Sep 17 00:00:00 2001
From: Timothee Groleau <timothee.groleau@gmail.com>
Date: Thu, 14 Mar 2024 18:23:38 +0800
Subject: [PATCH] feat: extract tests and notes in one pass

---
 scripts/release_prep.sh | 36 +++++++++++++-----------------------
 1 file changed, 13 insertions(+), 23 deletions(-)

diff --git a/scripts/release_prep.sh b/scripts/release_prep.sh
index ea1db2a42..ca34d8cad 100644
--- a/scripts/release_prep.sh
+++ b/scripts/release_prep.sh
@@ -85,41 +85,31 @@ fi
 # Login to github to be able to query PR info
 gh auth login
 
-# Extract test procedures from each feature PR
-echo "" >> ${pr_body_file_grouped}
-echo "## Tests" >> ${pr_body_file_grouped}
-echo "" >> ${pr_body_file_grouped}
+repo_url=$(gh repo view --json url --jq '.url')
+
+# Extract tests and deploy notes from each feature PR
+tests_section="## Tests\n\n"
+notes_section="## Deploy Notes\n\n"
+
 grep -v -E -- '- [a-z]+\(deps(-dev)?\)' ${pr_body_file} | grep -v -E -- '- (release|backport) v' | grep -v -E -- '- \d+\.\d+\.\d+ ' | while read line_item; do
   pr_id=$(echo ${line_item} | grep -o -E '\[`#\d+`\]' | grep -o -E '\d+')
   pr_content=$(gh pr view ${pr_id})
+
   tests=$(echo "$pr_content" | awk "/^## Test/{flag=1;next}/^## /{flag=0}flag" | sed -E "s/\[[Xx]\]/[ ]/" | sed -E "s/^(###+) /\1## /")
   if [[ ${tests} =~ [^[:space:]] ]]; then
-    echo ${line_item} | sed "s/^- /### /" >> ${pr_body_file_grouped}
-    echo "${tests}" >> ${pr_body_file_grouped}
-    echo "" >> ${pr_body_file_grouped}
+    tests_section+="$(echo ${line_item} | sed "s/^- /### /")\n"
+    tests_section+="${tests}\n\n"
   fi
-done
 
-# Extract deploy notes from each feature PR
-# (contains duplication from previous blocks for tests, but makes it easy to understand the similarity and what's going on)
-# TODO: refactor to remove duplication
-echo "" >> ${pr_body_file_grouped}
-echo "## Deploy Notes" >> ${pr_body_file_grouped}
-echo "" >> ${pr_body_file_grouped}
-grep -v -E -- '- [a-z]+\(deps(-dev)?\)' ${pr_body_file} | grep -v -E -- '- (release|backport) v' | grep -v -E -- '- \d+\.\d+\.\d+ ' | while read line_item; do
-  pr_id=$(echo ${line_item} | grep -o -E '\[`#\d+`\]' | grep -o -E '\d+')
-  pr_content=$(gh pr view ${pr_id})
   notes=$(echo "$pr_content" | awk "/^## Deploy Notes/{flag=1;next}/^## /{flag=0}flag" | sed -E "s/\[[Xx]\]/[ ]/" | sed -E "s/^(###+) /\1## /")
   if [[ ${notes} =~ [^[:space:]] ]]; then
-    echo ${line_item} | sed "s/^- /### /" >> ${pr_body_file_grouped}
-    echo "${notes}" >> ${pr_body_file_grouped}
-    echo "" >> ${pr_body_file_grouped}
+    notes_section+="$(echo ${line_item} | sed "s/^- /### /")\n"
+    notes_section+="${notes}\n\n"
   fi
-done
-
 
-repo_url=$(gh repo view --json url --jq '.url')
 
+echo "${tests_section}\n" >> ${pr_body_file_grouped}
+echo "${notes_section}\n" >> ${pr_body_file_grouped}
 echo "" >> ${pr_body_file_grouped}
 echo "" >> ${pr_body_file_grouped}
 echo "**Full Changelog**: ${repo_url}/compare/${current_version}..${release_version}" >> ${pr_body_file_grouped}