@yeyuan98/opencode-bioresearcher-plugin 1.6.8-alpha.2 → 1.7.0

package/dist/skills/bioresearcher-tests/SKILL.md

@@ -38,7 +38,7 @@ Comprehensive test suite for bioresearcher plugin tools and skills.
 | Category | Tools | Status |
 |----------|-------|--------|
 | Parser | `parse_pubmed_articleSet`, `parse_obo_file` | Full tests |
-| Table | All 13 `table*` tools | Full tests + negative tests |
+| Table | All 13 `table*` tools | Full tests + negative tests + hardening tests + large table tests |
 | JSON | `jsonExtract`, `jsonValidate`, `jsonInfer` | Full tests |
 | Misc | `blockingTimer`, `calculator` | Full tests |
 | Skills | `demo-skill` | Full execution |
@@ -97,15 +97,16 @@ Record pass/fail for each test.
 
 ### Step 4: Run Table Tests
 
-Execute tests from `<skill_path>/test_cases/table_tests.md`:
+Execute tests from `<skill_path>/test_cases/table_tests.md` in order:
 
+**Basic functionality (13 tests):**
 1. `tableListSheets` - List sheets in sample Excel
 2. `tableGetSheetPreview` - Get first 6 rows
 3. `tableGetHeaders` - Get column headers
 4. `tableGetCell` - Get specific cell value
 5. `tableGetRange` - Get A1:C3 range
-6. `tableFilterRows` - Filter by Age > 25
-7. `tableSearch` - Search for "test"
+6. `tableFilterRows` - Filter by Age > 25 (verify new `total_matches_found` + `returned_count` fields)
+7. `tableSearch` - Search for "Active"
 8. `tableSummarize` - Get stats for numeric columns
 9. `tableGroupBy` - Group by Category
 10. `tableCreateFile` - Create new Excel file
@@ -113,6 +114,47 @@ Execute tests from `<skill_path>/test_cases/table_tests.md`:
 12. `tableUpdateCell` - Update cell A1
 13. `tablePivotSummary` - Create pivot table
 
+**Error / negative tests (5 tests):**
+14. Nonexistent File - error handling
+15. Invalid Cell Address - descriptive error message
+16. Invalid Filter Operator - error handling
+17. Nonexistent Column - graceful handling
+18. Out of Bounds Range - empty data
+
+**Phase 2 correctness tests (10 tests):**
+19. Pivot Avg Numeric-Only Denominator - Setup (create test file)
+20. Pivot Avg Numeric-Only Denominator - Verify (avg = 20.0, not 10.0)
+21. Pivot Max Handles Negative Values - Setup (create test file)
+22. Pivot Max Handles Negative Values - Verify (max = 3, not 0)
+23. CreateFile Overwrite Guard - rejects without `overwrite: true`
+24. CreateFile Overwrite Succeeds - works with `overwrite: true`
+25. FilterRows New Response Fields - `total_matches_found` + `returned_count`
+26. UpdateCell Invalid Address Path Traversal - rejects `../../etc/passwd`
+27. UpdateCell Null Value - accepts null without error
+28. CreateFile Invalid Data Schema - rejects non-array data
+
+**Hardening feature tests (8 tests):**
+29. Force Parameter - Filter Rows (regression check)
+30. Force Parameter - Summarize (regression check)
+31. File Size Limit - Reject Large File (generate 55MB via `dd`)
+32. File Size Limit - Bypass With Force (same file, `force: true`)
+33. Range Limit - Reject Oversized Range (`A1:ZZZ1000`)
+34. Range Limit - Bypass With Force
+35. Error Sanitization - No Resolved Paths (no `/home/`, `/root/`, `/tmp/`)
+36. Error Sanitization - Uses Original Path
+
+**Large table tests (8 tests):**
+37. Large Table - Generate Via Python (`openpyxl` row duplication, 150K+ rows)
+38. Large Table - Filter Rows Truncation (verify `truncated: true`)
+39. Large Table - Filter Rows With Force (no truncation)
+40. Large Table - Summarize Truncation
+41. Large Table - Search Truncation
+42. Large Table - GroupBy Truncation
+43. Large Table - PivotSummary Truncation
+44. Large Table - Cleanup (remove all generated test files)
+
+> **Important:** The large table tests (35-42) require `openpyxl` to be installed. If not available, install with `pip install openpyxl`. The large table is generated during the test and cleaned up afterward - no large files are shipped with the plugin.
+
 Record pass/fail for each test.
 
 ### Step 5: Run JSON Tests

package/dist/skills/bioresearcher-tests/test_cases/table_tests.md

@@ -1,5 +1,7 @@
 # Table Tool Tests
 
+## Basic Functionality Tests
+
 ## Test: List Sheets
 - Tool: tableListSheets
 - Input:
@@ -64,8 +66,9 @@
   ```
 - Validators:
   - has matched_rows
-  - has total_matches
-- Expected: Filtered rows where Age > 25
+  - has total_matches_found
+  - has returned_count
+- Expected: Filtered rows where Age > 25 with new response fields
 
 ## Test: Search
 - Tool: tableSearch
@@ -75,8 +78,9 @@
   ```
 - Validators:
   - has results array
-  - has total_found
-- Expected: Cells containing "Active"
+  - has total_matches_found
+  - has returned_count
+- Expected: Cells containing "Active" with new response fields
 
 ## Test: Summarize
 - Tool: tableSummarize
@@ -143,6 +147,8 @@
   - has columns array
 - Expected: Pivot table with Category rows, Status columns, Score sums
 
+## Error / Negative Tests
+
 ## Test: Nonexistent File
 - Tool: tableListSheets
 - Input:
@@ -161,17 +167,20 @@
   ```
 - Validators:
   - contains_string("Error")
-- Expected: Error response for invalid address
+  - contains_string("Invalid cell address")
+- Expected: Error response with descriptive message
 
-## Test: Invalid Filter Operator
+## Test: No Match Filter
 - Tool: tableFilterRows
 - Input:
   ```json
-  {"file_path": ".bioresearcher-tests/workspace/table_sample.xlsx", "column": "Age", "operator": "INVALID", "value": 25, "max_results": 100}
+  {"file_path": ".bioresearcher-tests/workspace/table_sample.xlsx", "column": "Age", "operator": ">", "value": 999999, "max_results": 100}
   ```
 - Validators:
-  - contains_string("Error")
-- Expected: Error response for invalid operator
+  - has total_matches_found
+  - has returned_count
+  - has matched_rows
+- Expected: Zero matches returned (valid query with no matching rows)
 
 ## Test: Nonexistent Column
 - Tool: tableSummarize
@@ -192,3 +201,332 @@
 - Validators:
   - has data array
 - Expected: Empty data array (out of bounds handled)
+
+## Phase 2 Correctness Tests
+
+## Test: Pivot Avg Numeric-Only Denominator - Setup
+- Tool: tableCreateFile
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/pivot_avg_test.xlsx", "sheet_name": "Sheet1", "data": [{"Group": "A", "Type": "X", "Val": 10}, {"Group": "A", "Type": "X", "Val": ""}, {"Group": "A", "Type": "X", "Val": 30}, {"Group": "B", "Type": "Y", "Val": 20}, {"Group": "B", "Type": "Y", "Val": 40}]}
+  ```
+- Validators:
+  - success_is_true
+- Expected: File created for pivot avg test
+
+## Test: Pivot Avg Numeric-Only Denominator - Verify
+- Tool: tablePivotSummary
+- Prerequisite: Test "Pivot Avg Numeric-Only Denominator - Setup" must pass first
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/pivot_avg_test.xlsx", "row_field": "Group", "col_field": "Type", "value_field": "Val", "agg": "avg"}
+  ```
+- Validators:
+  - has pivot object
+- Expected: Group A avg = 20.0 (10+30 divided by 2, not 3)
+
+## Test: Pivot Max Handles Negative Values - Setup
+- Tool: tableCreateFile
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/pivot_neg_test.xlsx", "sheet_name": "Sheet1", "data": [{"Group": "A", "Val": -5}, {"Group": "A", "Val": -10}, {"Group": "A", "Val": 3}]}
+  ```
+- Validators:
+  - success_is_true
+- Expected: File created for pivot neg test
+
+## Test: Pivot Max Handles Negative Values - Verify
+- Tool: tablePivotSummary
+- Prerequisite: Test "Pivot Max Handles Negative Values - Setup" must pass first
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/pivot_neg_test.xlsx", "row_field": "Group", "col_field": "Group", "value_field": "Val", "agg": "max"}
+  ```
+- Validators:
+  - has pivot object
+- Expected: Max is 3, not 0
+
+## Test: CreateFile Overwrite Guard
+- Prerequisite: Test "Create File" must pass first (output_created.xlsx exists)
+- Tool: tableCreateFile
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/output_created.xlsx", "sheet_name": "Sheet1", "data": [{"Name": "Override"}]}
+  ```
+- Validators:
+  - contains_string("already exists")
+  - contains_string("overwrite: true")
+- Expected: Error because file exists and overwrite is not set
+
+## Test: CreateFile Overwrite Succeeds
+- Prerequisite: Test "CreateFile Overwrite Guard" must run first
+- Tool: tableCreateFile
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/output_created.xlsx", "sheet_name": "Sheet1", "data": [{"Name": "Overwritten"}], "overwrite": true}
+  ```
+- Validators:
+  - success_is_true
+  - rows_created === 1
+- Expected: File overwritten successfully
+
+## Test: FilterRows New Response Fields
+- Tool: tableFilterRows
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/table_sample.xlsx", "column": "Age", "operator": ">=", "value": 0, "max_results": 5}
+  ```
+- Validators:
+  - has total_matches_found
+  - has returned_count
+  - has matched_rows
+- Expected: Response includes total_matches_found and returned_count (replaces old total_matches)
+
+## Test: UpdateCell Invalid Address Path Traversal
+- Tool: tableUpdateCell
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/output_created.xlsx", "cell_address": "../../etc/passwd", "value": "test"}
+  ```
+- Validators:
+  - contains_string("Error")
+  - contains_string("Invalid cell address")
+- Expected: Error response rejecting path traversal in cell address
+
+## Test: UpdateCell Null Value
+- Prerequisite: Test "CreateFile Overwrite Succeeds" must pass first
+- Tool: tableUpdateCell
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/output_created.xlsx", "cell_address": "B1", "value": null}
+  ```
+- Validators:
+  - success_is_true
+- Expected: Cell updated with null value without error
+
+## Test: CreateFile Invalid Data Schema
+- Tool: tableCreateFile
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/invalid_data.xlsx", "sheet_name": "Sheet1", "data": "not an array"}
+  ```
+- Validators:
+  - contains_string("Error")
+  - contains_string("must be an array")
+- Expected: Error because data is a string, not an array
+
+## Hardening Feature Tests
+
+## Test: Force Parameter - Filter Rows
+- Tool: tableFilterRows
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/table_sample.xlsx", "column": "Age", "operator": ">", "value": 25, "max_results": 100, "force": true}
+  ```
+- Validators:
+  - has matched_rows
+  - has total_matches_found
+  - has returned_count
+- Expected: Same results as Filter Rows without force (regression check)
+
+## Test: Force Parameter - Summarize
+- Tool: tableSummarize
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/table_sample.xlsx", "columns": ["Age", "Score"], "force": true}
+  ```
+- Validators:
+  - has summaries object
+- Expected: Same summaries as Summarize without force (regression check)
+
+## Test: File Size Limit - Reject Large File
+- Setup: Run the following Bash command to create a 55MB dummy file:
+  ```bash
+  dd if=/dev/urandom of=.bioresearcher-tests/workspace/large_dummy.xlsx bs=1M count=55 2>/dev/null && echo "Created large file"
+  ```
+- Tool: tableListSheets
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_dummy.xlsx"}
+  ```
+- Validators:
+  - contains_string("Error")
+  - contains_string("exceeding")
+  - contains_string("force: true")
+- Expected: Error rejecting file over 50MB limit with guidance to use force
+
+## Test: File Size Limit - Bypass With Force
+- Prerequisite: Test "File Size Limit - Reject Large File" must run first (large_dummy.xlsx exists)
+- Tool: tableListSheets
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_dummy.xlsx", "force": true}
+  ```
+- Validators:
+  - has sheets array
+- Expected: Size check bypassed with force: true; tool processed the file (parser may return partial results for corrupt data)
+- Cleanup: Run `rm .bioresearcher-tests/workspace/large_dummy.xlsx`
+
+## Test: Range Limit - Reject Oversized Range
+- Tool: tableGetRange
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/table_sample.xlsx", "range": "A1:ZZZ1000"}
+  ```
+- Validators:
+  - contains_string("Error")
+  - contains_string("exceeds")
+  - contains_string("force: true")
+- Expected: Error rejecting range exceeding 50K cell limit
+
+## Test: Range Limit - Bypass With Force
+- Tool: tableGetRange
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/table_sample.xlsx", "range": "A1:ZZZ1000", "force": true}
+  ```
+- Validators:
+  - has data array
+- Expected: Data array returned (mostly empty cells, but range limit bypassed)
+
+## Test: Error Sanitization - No Resolved Paths
+- Tool: tableGetCell
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/nonexistent_file.xlsx", "cell_address": "A1"}
+  ```
+- Validators:
+  - contains_string("Error")
+  - not contains_string("/home/")
+  - not contains_string("/root/")
+  - not contains_string("/tmp/")
+- Expected: Error message does not leak absolute filesystem paths
+
+## Test: Error Sanitization - Uses Original Path
+- Tool: tableListSheets
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/nonexistent_file.xlsx"}
+  ```
+- Validators:
+  - contains_string("Error")
+  - contains_string(".bioresearcher-tests/workspace/nonexistent_file.xlsx")
+- Expected: Error message references the original user-provided file path
+
+## Large Table Tests
+
+## Test: Large Table - Generate Via Python
+- Setup: Run the following Python command to generate a table with 150K+ rows by duplicating the sample data:
+  ```bash
+  pip install openpyxl -q 2>/dev/null; python3 -c "
+  import openpyxl
+  wb = openpyxl.load_workbook('.bioresearcher-tests/workspace/table_sample.xlsx')
+  ws = wb.active
+  rows = list(ws.iter_rows(values_only=True))
+  data_rows = rows[1:]
+  if not data_rows:
+      print('ERROR: no data rows found')
+  else:
+      target = 150000
+      for i in range(target // len(data_rows)):
+          for row in data_rows:
+              ws.append(row)
+      wb.save('.bioresearcher-tests/workspace/large_table.xlsx')
+      print(f'Created table with {ws.max_row} rows')
+  "
+  ```
+- Validators:
+  - contains_string("Created table with")
+- Expected: large_table.xlsx created with 150K+ rows
+
+## Test: Large Table - Filter Rows Truncation
+- Prerequisite: Test "Large Table - Generate Via Python" must pass first
+- Tool: tableFilterRows
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_table.xlsx", "column": "Age", "operator": ">", "value": 0, "max_results": 50}
+  ```
+- Validators:
+  - has matched_rows
+  - has total_matches_found
+  - has returned_count
+  - truncated === true
+  - contains_string("Set force: true")
+- Expected: Results truncated at row cap with guidance message
+
+## Test: Large Table - Filter Rows With Force
+- Prerequisite: Test "Large Table - Generate Via Python" must pass first
+- Tool: tableFilterRows
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_table.xlsx", "column": "Age", "operator": ">", "value": 0, "max_results": 50, "force": true}
+  ```
+- Validators:
+  - has matched_rows
+  - has total_matches_found
+  - has returned_count
+- Expected: All rows processed, no truncation field
+
+## Test: Large Table - Summarize Truncation
+- Prerequisite: Test "Large Table - Generate Via Python" must pass first
+- Tool: tableSummarize
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_table.xlsx", "columns": ["Age"]}
+  ```
+- Validators:
+  - has summaries object
+  - truncated === true
+  - has processed_rows
+  - contains_string("Set force: true")
+- Expected: Summary computed on capped rows with truncation notice
+
+## Test: Large Table - Search Truncation
+- Prerequisite: Test "Large Table - Generate Via Python" must pass first
+- Tool: tableSearch
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_table.xlsx", "search_term": "Active", "max_results": 10}
+  ```
+- Validators:
+  - has results array
+  - has total_matches_found
+  - has returned_count
+  - truncated === true
+  - has rows_scanned
+- Expected: Search scanned up to 100K rows, counted all matches, returned limited results
+
+## Test: Large Table - GroupBy Truncation
+- Prerequisite: Test "Large Table - Generate Via Python" must pass first
+- Tool: tableGroupBy
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_table.xlsx", "group_column": "Category", "agg_column": "Score", "agg_type": "avg"}
+  ```
+- Validators:
+  - has groups object
+  - truncated === true
+  - has processed_rows
+- Expected: GroupBy truncated at row cap
+
+## Test: Large Table - PivotSummary Truncation
+- Prerequisite: Test "Large Table - Generate Via Python" must pass first
+- Tool: tablePivotSummary
+- Input:
+  ```json
+  {"file_path": ".bioresearcher-tests/workspace/large_table.xlsx", "row_field": "Category", "col_field": "Status", "value_field": "Score", "agg": "sum"}
+  ```
+- Validators:
+  - has pivot object
+  - truncated === true
+  - has processed_rows
+- Expected: Pivot truncated at row cap
+
+## Test: Large Table - Cleanup
+- Setup: Run the following Bash command to remove the large table:
+  ```bash
+  rm -f .bioresearcher-tests/workspace/large_table.xlsx .bioresearcher-tests/workspace/large_dummy.xlsx .bioresearcher-tests/workspace/pivot_avg_test.xlsx .bioresearcher-tests/workspace/pivot_neg_test.xlsx .bioresearcher-tests/workspace/invalid_data.xlsx .bioresearcher-tests/workspace/output_created.xlsx && echo "Cleanup done"
+  ```
+- Validators:
+  - contains_string("Cleanup done")
+- Expected: All generated test files removed

package/dist/skills/bioresearcher-tests/test_runner.py

@@ -376,9 +376,17 @@ def validate_result(result: Any, validators: list[str]) -> tuple[bool, list[dict
             detail = f"has matched_rows: {passed}"
 
         elif validator == "has total_matches":
-            passed = "total_matches" in result_obj
+            passed = "total_matches_found" in result_obj or "total_matches" in result_obj
             detail = f"has total_matches: {passed}"
 
+        elif validator == "has total_matches_found":
+            passed = "total_matches_found" in result_obj
+            detail = f"has total_matches_found: {passed}"
+
+        elif validator == "has returned_count":
+            passed = "returned_count" in result_obj
+            detail = f"has returned_count: {passed}"
+
         elif validator == "has results array":
             passed = "results" in result_obj and isinstance(result_obj["results"], list)
             detail = f"has results: {passed}"
@@ -405,10 +413,37 @@ def validate_result(result: Any, validators: list[str]) -> tuple[bool, list[dict
             passed = "columns" in result_obj and isinstance(result_obj["columns"], list)
             detail = f"has columns: {passed}"
 
+        elif validator == "has truncated":
+            passed = "truncated" in result_obj
+            detail = f"has truncated: {passed}"
+
+        elif validator == "has processed_rows":
+            passed = "processed_rows" in result_obj
+            detail = f"has processed_rows: {passed}"
+
+        elif validator == "has rows_scanned":
+            passed = "rows_scanned" in result_obj
+            detail = f"has rows_scanned: {passed}"
+
+        elif validator == "truncated === true":
+            passed = result_obj.get("truncated") is True
+            detail = f"truncated={result_obj.get('truncated')}"
+
+        elif validator.startswith("not contains_string"):
+            match = re.search(r'"(.+?)"', validator)
+            if match:
+                substr = match.group(1)
+                passed = substr.lower() not in result_str.lower()
+                detail = f"does NOT contain '{substr}': {passed}"
+
         elif validator == "rows_appended === 1":
             passed = result_obj.get("rows_appended") == 1
             detail = f"rows_appended={result_obj.get('rows_appended')}"
 
+        elif validator == "rows_created === 1":
+            passed = result_obj.get("rows_created") == 1
+            detail = f"rows_created={result_obj.get('rows_created')}"
+
         elif validator == "rows_created === 2":
             passed = result_obj.get("rows_created") == 2
             detail = f"rows_created={result_obj.get('rows_created')}"