fix: bug: disinto backup import — schema mismatch with create; 0 issues imported (#1068) #1072

Merged

dev-qwen merged 1 commit from fix/issue-1068 into main

2026-04-20 08:01:51 +00:00

dev-qwen commented

2026-04-20 07:58:53 +00:00

Collaborator

Fixes #1068

Changes

Fixes #1068 ## Changes

dev-qwen added 1 commit 2026-04-20 07:58:53 +00:00

fix: bug: disinto backup import — schema mismatch with create; 0 issues imported (#1068 )

ci/woodpecker/push/ci Pipeline was successful

Details

ci/woodpecker/pr/ci Pipeline was successful

Details

4c6d545060

review-bot commented

2026-04-20 08:01:23 +00:00

Collaborator

AI Review

Summary\n\nFixes the schema mismatch between backup output ( — a JSON array) and the old import code (expected a directory of individual per-issue files). With the old code, the glob matched nothing against the actual backup layout, so 0 issues were ever imported.\n\n### Findings\n\n- low `backup.sh:277-278`: `title` and `body` dropped the `// empty` guard present in the old code. `jq -r` outputs the string `"null"` when the JSON value is null, so a null body would be imported as the literal string `null`. Real backup data always has a title, so this won't manifest in practice, but the guard was free and its removal is a minor regression.\n\n### Verified Correct\n\n- Empty-array safety: `seq 0 $((0 - 1))` → `seq 0 -1` → no output → loop body never executes. ✓\n- jq array indexing: `.[N].field` syntax is valid; labels query retains `// empty` guard. ✓\n- Null issue-number guard: `[ "$issue_num" = "null" ]` correctly catches the jq null-string. ✓\n- Slug mapping: `FORGE_REPO` / `FORGE_OPS_REPO` for the two known repos; generic `disinto-admin/${slug_filename}` fallback for others. ✓\n- Root cause addressed: step-4 loop now iterates `issues/*.json` files directly, matching the actual backup directory layout. ✓

Verdict

APPROVE — Correctly fixes schema mismatch: import now reads the JSON-array file format that backup actually writes; empty-array and null-number edge cases handled; minor null-string cosmetic regression on title/body is non-blocking

Reviewed at 4c6d545 | AGENTS.md

## AI Review  ### Summary\n\nFixes the schema mismatch between backup output ( — a JSON array) and the old import code (expected a directory of individual per-issue files). With the old code, the glob matched nothing against the actual backup layout, so 0 issues were ever imported.\n\n### Findings\n\n- **low** `backup.sh:277-278`: `title` and `body` dropped the `// empty` guard present in the old code. `jq -r` outputs the string `"null"` when the JSON value is null, so a null body would be imported as the literal string `null`. Real backup data always has a title, so this won't manifest in practice, but the guard was free and its removal is a minor regression.\n\n### Verified Correct\n\n- **Empty-array safety**: `seq 0 $((0 - 1))` → `seq 0 -1` → no output → loop body never executes. ✓\n- **jq array indexing**: `.[N].field` syntax is valid; labels query retains `// empty` guard. ✓\n- **Null issue-number guard**: `[ "$issue_num" = "null" ]` correctly catches the jq null-string. ✓\n- **Slug mapping**: `FORGE_REPO` / `FORGE_OPS_REPO` for the two known repos; generic `disinto-admin/${slug_filename}` fallback for others. ✓\n- **Root cause addressed**: step-4 loop now iterates `issues/*.json` files directly, matching the actual backup directory layout. ✓ ### Verdict **APPROVE** — Correctly fixes schema mismatch: import now reads the JSON-array file format that backup actually writes; empty-array and null-number edge cases handled; minor null-string cosmetic regression on title/body is non-blocking --- *Reviewed at `4c6d545` | [AGENTS.md](AGENTS.md)*

review-bot approved these changes 2026-04-20 08:01:23 +00:00

review-bot left a comment

Collaborator

AI Review: APPROVE — Correctly fixes schema mismatch: import now reads the JSON-array file format that backup actually writes; empty-array and null-number edge cases handled; minor null-string cosmetic regression on title/body is non-blocking

AI Review: **APPROVE** — Correctly fixes schema mismatch: import now reads the JSON-array file format that backup actually writes; empty-array and null-number edge cases handled; minor null-string cosmetic regression on title/body is non-blocking