Fix super-linter: fetch full git history to resolve master branch lookup failure

[Copilot]

super-linter was fatally failing on PRs because the shallow checkout (fetch-depth: 1) left no master/origin/master ref available — which super-linter requires to diff changed files when VALIDATE_ALL_CODEBASE is unset.

Changes

• .github/workflows/linter.yml — add fetch-depth: 0 to the actions/checkout step in check-super-linter so all branches (including master) are fetched, unblocking super-linter's diff-based file filtering.
• ci/update_distros.sh — fix shfmt -i 4 -s formatting: align case patterns in _get_box_arch with the case/esac keywords (shfmt canonical style), resolving the "Check scripts format" CI failure.

---

📱 Kick off Copilot coding agent tasks wherever you are with GitHub Mobile, available on iOS and Android.

[qodo-code-review]

ⓘ You are approaching your monthly quota for Qodo. Upgrade your plan

Review Summary by Qodo

Fix CI issues: super-linter, Vagrant Cloud API, and script formatting

🐞 Bug fix ✨ Enhancement

Walkthroughs

Description

• Fix super-linter CI by fetching full git history for master branch availability
• Update Vagrant Cloud API URLs from v1 to v2 with architecture support
• Add _get_box_arch function to detect and map system architecture
• Fix shell script formatting and improve Vagrantfile cut delimiter handling

Diagram

flowchart LR
  A["CI Workflows"] -->|fetch-depth: 0| B["super-linter"]
  B -->|full git history| C["master branch available"]
  D["Vagrant Cloud API"] -->|v1 to v2 migration| E["update_distros.sh"]
  E -->|architecture detection| F["_get_box_arch function"]
  G["Vagrantfile"] -->|improved delimiter| H["VBoxManage parsing"]

Loading

File Changes

1. .github/workflows/linter.yml 🐞 Bug fix +1/-0

Enable full git history for super-linter

• Add fetch-depth: 0 to actions/checkout step in check-super-linter job
• Enables full git history fetch to make master branch available for super-linter's diff-based
 file filtering

.github/workflows/linter.yml

---

2. ci/update_distros.sh ✨ Enhancement +13/-3

Migrate to Vagrant Cloud API v2 with architecture support

• Update Vagrant Cloud API endpoint from https://app.vagrantup.com/api/v1/box/ to
 https://vagrantcloud.com/api/v2/vagrant/
• Change python to python3 for JSON parsing
• Add new _get_box_arch function to detect system architecture (x86_64→amd64, aarch64→arm64)
• Update box download URL to include architecture path: providers/$PROVIDER/$arch/vagrant.box

ci/update_distros.sh

---

3. Vagrantfile 🐞 Bug fix +1/-1

Improve VBoxManage output parsing robustness

• Replace cut -d "Name:[ ]*" -f 2 delimiter with awk '{$1=""; sub(/^[[:space:]]+/, ""); print}'
 for more robust VBoxManage output parsing
• Improves handling of whitespace in bridged interface name extraction

Vagrantfile

---

[qodo-code-review]

Code Review by Qodo

🐞 Bugs (1) 📘 Rule violations (0) 📎 Requirement gaps (0) 📐 Spec deviations (0)

1. Retry loop bypassed 🐞 Bug ⛯ Reliability

Description

In ci/update_distros.sh, _get_box_current_version is intended to retry fetching metadata, but
under set -o errexit any JSON parse error (or missing current_version.version) will abort the
script immediately, bypassing the retry loop. This can happen when the curl response is a non-JSON
error page or an API error payload that doesn’t match the expected shape.

Code

ci/update_distros.sh[R30-33]

+            metadata="$(curl -s "https://vagrantcloud.com/api/v2/vagrant/${name%/*}/${name#*/}")"
            if [ "$metadata" ]; then
-                version="$(echo "$metadata" | python -c 'import json,sys;print(json.load(sys.stdin)["current_version"]["version"])')"
+                version="$(echo "$metadata" | python3 -c 'import json,sys;print(json.load(sys.stdin)["current_version"]["version"])')"
                break

Evidence

The script enables set -o errexit, then treats any non-empty curl response body as valid JSON and
immediately runs python3 -c 'json.load(...)...["current_version"]["version"]'. If that python
process exits non-zero (invalid JSON, KeyError), bash exits the whole script due to errexit, so
the surrounding until loop never retries and max_attempts is effectively unused for these
failures.

ci/update_distros.sh[11-40]
ci/update_distros.sh[20-44]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

### Issue description
`_get_box_current_version` uses an `until` retry loop, but with `set -o errexit` the script exits on JSON parse/KeyError before any retries can occur. This makes transient API/HTTP failures fatal.

### Issue Context
- The curl call does not fail the shell on HTTP errors (no `--fail`), so error pages/bodies can still be non-empty.
- The python3 JSON extraction is unguarded and will exit non-zero on invalid JSON or unexpected response shape.

### Fix Focus Areas
- ci/update_distros.sh[11-40]
- ci/update_distros.sh[20-44]

### Implementation notes
- Capture both HTTP status and body, e.g. `curl -fsSL -w '%{http_code}'` (or `--fail-with-body`) and only attempt JSON parsing on 2xx.
- Make the JSON extraction non-fatal inside the loop, e.g.:
 - `version=$(python3 -c '...' <<<"$metadata" 2>/dev/null || true)`
 - Only `break` when `version` is non-empty.
- If parsing fails, increment `attempt_counter` and continue retrying until `max_attempts` is reached (then exit 1 with a helpful message including HTTP code).

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

ⓘ The new review experience is currently in Beta. Learn more

[qodo-code-review]

1. Retry loop bypassed 🐞 Bug ⛯ Reliability

In ci/update_distros.sh, _get_box_current_version is intended to retry fetching metadata, but
under set -o errexit any JSON parse error (or missing current_version.version) will abort the
script immediately, bypassing the retry loop. This can happen when the curl response is a non-JSON
error page or an API error payload that doesn’t match the expected shape.

Agent Prompt

### Issue description
`_get_box_current_version` uses an `until` retry loop, but with `set -o errexit` the script exits on JSON parse/KeyError before any retries can occur. This makes transient API/HTTP failures fatal.

### Issue Context
- The curl call does not fail the shell on HTTP errors (no `--fail`), so error pages/bodies can still be non-empty.
- The python3 JSON extraction is unguarded and will exit non-zero on invalid JSON or unexpected response shape.

### Fix Focus Areas
- ci/update_distros.sh[11-40]
- ci/update_distros.sh[20-44]

### Implementation notes
- Capture both HTTP status and body, e.g. `curl -fsSL -w '%{http_code}'` (or `--fail-with-body`) and only attempt JSON parsing on 2xx.
- Make the JSON extraction non-fatal inside the loop, e.g.:
  - `version=$(python3 -c '...' <<<"$metadata" 2>/dev/null || true)`
  - Only `break` when `version` is non-empty.
- If parsing fails, increment `attempt_counter` and continue retrying until `max_attempts` is reached (then exit 1 with a helpful message including HTTP code).

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools