From 8772e25d919eb6234e4f0abcc6afce71e6955d74 Mon Sep 17 00:00:00 2001
From: Patrick Lewis <4015312+locus313@users.noreply.github.com>
Date: Wed, 5 Aug 2026 12:42:31 -0700
Subject: [PATCH] refactor: extract csv_escape helper to lib/github-common.sh

Deduplicates the 'sed s/"/""/g' CSV double-quote escaping pattern that
was hand-rolled 8 times across github-dockerfile-discovery.sh (6 call
sites), github-get-repo-list.sh, and github-archive-old-repos.sh into a
single shared csv_escape helper, documented in the shared library docs.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 .github/copilot-instructions.md                    |  1 +
 AGENTS.md                                          |  2 +-
 README.md                                          |  1 +
 .../github-dockerfile-discovery.sh                 | 12 ++++++------
 lib/github-common.sh                               | 10 ++++++++++
 .../github-archive-old-repos.sh                    |  2 +-
 .../github-get-repo-list/github-get-repo-list.sh   |  2 +-
 tests/test_common.bats                             | 14 ++++++++++++++
 8 files changed, 35 insertions(+), 9 deletions(-)

diff --git a/.github/copilot-instructions.md b/.github/copilot-instructions.md
index 498742d..39099f7 100644
--- a/.github/copilot-instructions.md
+++ b/.github/copilot-instructions.md
@@ -33,6 +33,7 @@ Provides reusable helpers sourced by scripts:
 | `validate_github_token [bearer]` | Verify GITHUB_TOKEN via /user endpoint |
 | `validate_slug <value> <label>` | Reject values with non-alphanumeric/hyphen/underscore chars |
 | `is_bsd_date` | Returns 0 if `date` is BSD-style (macOS), 1 for GNU (Linux); branch between `date -v` and `date -d` syntax |
+| `csv_escape <value>` | Doubles embedded double-quotes for a quoted CSV field |
 | `gh_api <path> [--api-version V] [curl args...]` | Bearer-auth REST helper with 5-retry rate-limit handling; optional `--api-version` overrides the default `2022-11-28` header; returns literal `__404__` or `__422__` for those HTTP statuses — callers must check for these sentinels |
 | `gh_api_paginate <path> [filter] [version]` | Paginated REST helper, follows Link headers, streams items; returns silently with empty output on 404/422 |
 | `get_enterprise_orgs` | Three-tier enterprise org resolver (REST → GraphQL → /user/orgs) |
diff --git a/AGENTS.md b/AGENTS.md
index 853f283..bc100f5 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -93,7 +93,7 @@ bats tests/test_common.bats
 
 | File | What it covers |
 |------|----------------|
-| `tests/test_common.bats` | `lib/github-common.sh` — pure-logic functions (`validate_slug`, `require_env_var`, `require_command`, `err`, `configure_gh_auth`, `validate_token`, `get_repo_page_count`, `is_bsd_date`) and API helpers (`gh_api` sentinels, `gh_api_paginate`) |
+| `tests/test_common.bats` | `lib/github-common.sh` — pure-logic functions (`validate_slug`, `require_env_var`, `require_command`, `err`, `configure_gh_auth`, `validate_token`, `get_repo_page_count`, `is_bsd_date`, `csv_escape`) and API helpers (`gh_api` sentinels, `gh_api_paginate`) |
 | `tests/test_script_validation.bats` | Every script — missing required env vars exit 1, invalid CLI args exit 1, `--help` exits 0, script-specific enum/allowlist validation |
 | `tests/mock_curl.sh` | Universal drop-in curl mock (used by both test files); response data via env vars `MOCK_CURL_CODE`, `MOCK_CURL_BODY`, `MOCK_CURL_LINK` |
 
diff --git a/README.md b/README.md
index 78054a5..ae23eea 100644
--- a/README.md
+++ b/README.md
@@ -938,6 +938,7 @@ All scripts can leverage a shared utility library for common operations like val
 - `validate_github_token [bearer]` — Convenience wrapper for `GITHUB_TOKEN` validation
 - `validate_slug <value> <label>` — Exit if value contains characters other than alphanumeric, hyphen, or underscore
 - `is_bsd_date` — Return 0 if the system `date` is BSD-style (macOS), 1 for GNU (Linux); use to branch between `date -v` and `date -d` syntax
+- `csv_escape <value>` — Double embedded double-quotes so a value is safe to wrap in a quoted CSV field
 
 **Auth helpers:**
 - `configure_gh_auth [scope_hint]` — Bridge `GITHUB_TOKEN→GH_TOKEN` for scripts that use the `gh` CLI, or verify an active `gh` auth session if no token is set
diff --git a/enterprise/github-dockerfile-discovery/github-dockerfile-discovery.sh b/enterprise/github-dockerfile-discovery/github-dockerfile-discovery.sh
index 5ec74f6..417a9fa 100755
--- a/enterprise/github-dockerfile-discovery/github-dockerfile-discovery.sh
+++ b/enterprise/github-dockerfile-discovery/github-dockerfile-discovery.sh
@@ -188,13 +188,13 @@ parse_dockerfile_content() {
 
     # Sanitize fields for CSV (escape double-quotes, wrap in quotes)
     local safe_path safe_url
-    safe_path=$(echo "${dockerfile_path}" | sed 's/"/""/g')
-    safe_url=$(echo "${html_url}" | sed 's/"/""/g')
+    safe_path=$(csv_escape "${dockerfile_path}")
+    safe_url=$(csv_escape "${html_url}")
     local safe_image safe_tag safe_digest safe_base
-    safe_image=$(echo "${image}" | sed 's/"/""/g')
-    safe_tag=$(echo "${tag}" | sed 's/"/""/g')
-    safe_digest=$(echo "${digest}" | sed 's/"/""/g')
-    safe_base=$(echo "${base_image}" | sed 's/"/""/g')
+    safe_image=$(csv_escape "${image}")
+    safe_tag=$(csv_escape "${tag}")
+    safe_digest=$(csv_escape "${digest}")
+    safe_base=$(csv_escape "${base_image}")
     local repo_name
     repo_name=$(echo "${repo_full_name}" | cut -d/ -f2)
 
diff --git a/lib/github-common.sh b/lib/github-common.sh
index 19f9859..2c0d045 100755
--- a/lib/github-common.sh
+++ b/lib/github-common.sh
@@ -19,6 +19,7 @@
 #   validate_token <VAR_NAME>                 — verify a secondary token variable
 #   validate_slug <value> [label]             — exit if value contains unsafe chars
 #   is_bsd_date                                — 0 if `date` is BSD-style (macOS), 1 for GNU (Linux)
+#   csv_escape <value>                         — doubles embedded double-quotes for a quoted CSV field
 #   gh_api <path|url> [--api-version V] [curl args…] — Bearer-auth REST helper with retry;
 #                                               returns "__404__"/"__422__" (exit 0) for those codes
 #   gh_api_paginate <path> [filter] [version] — paginated REST, follows Link headers;
@@ -163,6 +164,15 @@ is_bsd_date() {
   date -v-1d > /dev/null 2>&1
 }
 
+###
+## csv_escape <value>
+## Doubles embedded double-quotes so a value is safe to wrap in a quoted
+## CSV field (e.g. printf '"%s"' "$(csv_escape "$value")").
+###
+csv_escape() {
+  echo "$1" | sed 's/"/""/g'
+}
+
 ###
 ## validate_slug <value> <label>
 ## Exits with status 1 if the value contains characters other than
diff --git a/org-admin/github-archive-old-repos/github-archive-old-repos.sh b/org-admin/github-archive-old-repos/github-archive-old-repos.sh
index 2860895..b708842 100755
--- a/org-admin/github-archive-old-repos/github-archive-old-repos.sh
+++ b/org-admin/github-archive-old-repos/github-archive-old-repos.sh
@@ -112,7 +112,7 @@ fetch_old_repos() {
                 REPO_HTMLURL=$(echo "$REPO_PAYLOAD"  | jq -r .html_url)
                 REPO_DESCRIPTION=$(echo "$REPO_PAYLOAD" | jq -r .description)
                 REPO_FORK=$(echo "$REPO_PAYLOAD"     | jq -r .fork)
-                ESCAPED_DESC=$(echo "$REPO_DESCRIPTION" | sed 's/"/""/g')
+                ESCAPED_DESC=$(csv_escape "$REPO_DESCRIPTION")
                 echo "${REPO_NAME},${REPO_FULLNAME},${REPO_PRIVATE},${REPO_ARCHIVED},${REPO_HTMLURL},\"${ESCAPED_DESC}\",${REPO_FORK},${REPO_UPDATEDAT},${DAYS_SINCE}" >> "$REPORT_FILE"
 
                 total_old_repos=$((total_old_repos + 1))
diff --git a/org-admin/github-get-repo-list/github-get-repo-list.sh b/org-admin/github-get-repo-list/github-get-repo-list.sh
index b766ff3..41a00ff 100755
--- a/org-admin/github-get-repo-list/github-get-repo-list.sh
+++ b/org-admin/github-get-repo-list/github-get-repo-list.sh
@@ -71,7 +71,7 @@ process_repos () {
       REPO_PUSHEDAT=$(echo "${REPO_PAYLOAD}" | jq -r .pushed_at)
       REPO_CREATEDAT=$(echo "${REPO_PAYLOAD}" | jq -r .created_at)
       REPO_UPDATEDAT=$(echo "${REPO_PAYLOAD}" | jq -r .updated_at)
-      ESCAPED_DESCRIPTION=$(echo "${REPO_DESCRIPTION}" | sed 's/"/""/g')
+      ESCAPED_DESCRIPTION=$(csv_escape "${REPO_DESCRIPTION}")
 
       printf '%s,%s,%s,%s,%s,"%s",%s,%s,%s,%s\n' \
         "${i}" "${REPO_FULLNAME}" "${REPO_OWNER}" "${REPO_PRIVATE}" "${REPO_HTMLURL}" \
diff --git a/tests/test_common.bats b/tests/test_common.bats
index c00bb17..637469a 100644
--- a/tests/test_common.bats
+++ b/tests/test_common.bats
@@ -102,6 +102,20 @@ _mock_curl() {
   fi
 }
 
+# ─── csv_escape ───────────────────────────────────────────────────────────────
+
+@test "csv_escape: leaves plain values unchanged" {
+  run bash -c "GITHUB_TOKEN=x source '${LIB_PATH}' 2>/dev/null; csv_escape 'plain value'"
+  [ "$status" -eq 0 ]
+  [ "$output" = "plain value" ]
+}
+
+@test "csv_escape: doubles embedded double-quotes" {
+  run bash -c "GITHUB_TOKEN=x source '${LIB_PATH}' 2>/dev/null; csv_escape 'she said \"hi\"'"
+  [ "$status" -eq 0 ]
+  [ "$output" = 'she said ""hi""' ]
+}
+
 # ─── require_env_var ──────────────────────────────────────────────────────────
 
 @test "require_env_var: exits 1 when variable is unset" {