Issue/123 improve test coverage

docs/KNOWN_ISSUES_AND_UNCOVERED_CODE.md

@@ -0,0 +1,393 @@
+# Known Issues and Uncovered Code
+
+**Status**: Documentation of bugs, peculiarities, and technical debt for future iteration in the Python client
+**Date**: January 2026
+
+This document comes from improving test coverage in the python client and exists to document issues that remain to be addressed. 
+
+## Overview
+
+This document catalogs known issues, bugs, peculiarities, and uncovered code paths in the TerminusDB Python client. Issues are organized by category and client area to facilitate future systematic fixes.
+
+---
+
+## 1. Known Bugs
+
+### 1.1 Query Builder - Args Introspection Issues
+
+**Area**: `woqlquery/woql_query.py`  
+**Severity**: Medium  
+**Lines**: 1056, 1092, 1128, 1775, 1807
+
+**Issue**: Quad methods (`quad`, `added_quad`, `removed_quad`, `delete_quad`, `add_quad`) attempt to call `.append()` or `.concat()` on `WOQLQuery` objects when handling args introspection which is a deprecated feature that is inconsistently implemented across javascript and python (disabled in the javascript client).
+
+**Details**:
+- When `sub == "args"`, methods call `triple()` which returns a `WOQLQuery` object
+- Code then tries: `return arguments.append("graph")` or `return arguments.concat(["graph"])`
+- `WOQLQuery` objects don't have `append()` or `concat()` methods
+- Should return a list like: `return triple_args + ["graph"]`
+
+**Blocked Tests**:
+- `test_woql_graph_operations.py::test_removed_quad_with_args_subject`
+- `test_woql_graph_operations.py::test_added_quad_with_args_subject`
+- `test_woql_path_operations.py::test_quad_with_special_args_subject`
+
+**Impact**: Args introspection feature doesn't work for quad operations
+
+---
+
+### 1.2 Query Builder - Missing Method
+
+**Area**: `woqlquery/woql_query.py`  
+**Severity**: Medium  
+**Line**: 3230
+
+**Issue**: `graph()` method calls non-existent `_set_context()` method.
+
+**Details**:
+- Line 3230: `return self._set_context({"graph": g})`
+- Method `_set_context()` is not defined in the class
+- Should directly set context: `self._triple_builder_context["graph"] = g`
+
+**Blocked Tests**:
+- `test_woql_graph_operations.py::test_graph_method_basic`
+- `test_woql_graph_operations.py::test_graph_with_subquery`
+- `test_woql_graph_operations.py::test_multiple_graph_operations`
+
+**Impact**: Graph context setting is completely broken
+
+---
+
+### 1.3 Query Validation - Unreachable Logic
+
+**Area**: `woqlquery/woql_query.py`  
+**Severity**: High  
+**Lines**: 784-785, 821-822
+
+**Issue**: Logically impossible validation conditions prevent proper error handling.
+
+**Details**:
+```python
+# Line 784 in select() method:
+if queries != [] and not queries:
+    raise ValueError("Select must be given a list of variable names")
+
+# Line 821-822 in distinct() method:
+if queries != [] and not queries:
+    raise ValueError("Distinct must be given a list of variable names")
+```
+
+**Analysis**:
+- After `queries = list(args)`, queries is always a list
+- If `queries == []`, then `queries != []` is False
+- If `queries != []`, then `not queries` is False
+- Condition can never be True, so ValueError is never raised
+- Should be: `if not queries:` or `if len(queries) == 0:`
+
+Same issue across both. Probably `select()` should allow to have no variables (to verify against terminusdb), and `distinct()` should have at least one variable (it can't be distinct otherwise…).
+
+**Blocked Tests**:
+- `test_woql_schema_validation.py::test_select_with_no_arguments_should_raise_error`
+
+**Impact**: Invalid queries with no arguments are not properly validated
+
+---
+
+### 1.4 From Method - Cursor Wrapping Edge Case
+
+**Area**: `woqlquery/woql_query.py`  
+**Severity**: Low  
+**Line**: 907
+
+**Issue**: The `woql_from()` method's cursor wrapping logic is not covered by tests.
+
+**Details**:
+```python
+# Line 906-907 in woql_from() method:
+if self._cursor.get("@type"):
+    self._wrap_cursor_with_and()  # Line 907 - uncovered
+```
+
+**Analysis**:
+- Line 907 wraps the existing cursor with an `And` operation when the cursor already has a `@type`
+- This is defensive programming to handle chaining `from()` after other operations
+- JavaScript client has identical logic: `if (this.cursor['@type']) this.wrapCursorWithAnd();`
+- Test attempts `query.limit(10).start(5)` but doesn't test `from()` with existing cursor
+
+**Correct Test Scenario**:
+```python
+query = WOQLQuery()
+query.triple("v:X", "v:P", "v:O")  # Set cursor @type
+result = query.woql_from("admin/mydb")  # Should trigger line 907
+```
+
+**Blocked Tests**:
+- `test_woql_advanced_features.py::test_subquery_with_limit_offset` (incorrectly named/implemented)
+
+**Impact**: Low - defensive code path, basic functionality works
+
+---
+
+## 2. Deprecated Code (Technical Debt)
+
+### 2.1 Data Value List Method
+
+**Area**: `woqlquery/woql_query.py`  
+**Severity**: Low (Dead Code)  
+**Lines**: 357-361
+
+**Issue**: `_data_value_list()` method is never called anywhere in the codebase.
+
+**Details**:
+- Method exists but has no callers
+- Similar functionality exists in `_value_list()` which is actively used
+- Originally had a bug (called `clean_data_value` instead of `_clean_data_value`)
+- Bug was fixed but method remains unused
+
+**Blocked Tests**:
+- `test_woql_path_operations.py::test_data_value_list_with_various_types`
+- `test_woql_query_builder.py::test_data_value_list_with_mixed_items`
+
+**Recommendation**: Remove in future major version after deprecation period
+
+---
+
+## 3. Uncovered Edge Cases
+
+### 3.1 Args Introspection Paths
+
+**Area**: `woqlquery/woql_query.py`  
+**Lines**: 907, 1572, 1693, 2301, 2560, 2562, 2584, 2586, 2806, 2832, 2836, 2877, 2897, 2932, 2957, 2959, 3001, 3008, 3015, 3021, 3062, 3065, 3108, 3135, 3137, 3157, 3159, 3230
+
+**Issue**: Args introspection feature (when first param == "args") is not covered by tests for many methods.
+
+**Details**:
+- Args introspection allows API discovery by passing "args" as first parameter
+- Methods return list of parameter names instead of executing
+- Many methods have this feature but it's not tested
+- Tests exist in `test_woql_remaining_edge_cases.py` but some paths remain uncovered
+
+**Methods Affected**:
+- `woql_or()` (907)
+- `start()` (1572)
+- `comment()` (1693)
+- `limit()` (2301)
+- `get()` (2560)
+- `put()` (2562)
+- `file()` (2584)
+- `remote()` (2586)
+- Math operations: `minus()`, `divide()`, `div()` (2806, 2832, 2836)
+- Comparison: `less()`, `lte()` (2877, 2897)
+- Logical: `once()`, `count()`, `cast()` (2932, 2957, 2959)
+- Type operations: `type_of()`, `order_by()`, `group_by()`, `length()` (3001, 3008, 3015, 3021)
+- String operations: `lower()`, `pad()` (3062, 3065)
+- Regex: `split()`, `regexp()`, `like()` (3108, 3135, 3137)
+- Substring operations (3157, 3159)
+- `trim()` (3230)
+
+**Impact**: Low - feature works but lacks test coverage
+
+---
+
+### 3.2 As() Method Edge Cases
+
+**Area**: `woqlquery/woql_query.py`  
+**Lines**: 1490-1495, 1501, 1508
+
+**Issue**: Complex argument handling in `woql_as()` method not fully covered.
+
+**Details**:
+- Lines 1490-1495: List of lists with optional type parameter
+- Line 1501: XSD prefix in second argument
+- Line 1508: Objects with `to_dict()` method
+
+**Impact**: Low - basic functionality tested, edge cases uncovered
+
+---
+
+### 3.3 Cursor Wrapping Edge Cases
+
+**Area**: `woqlquery/woql_query.py`  
+**Lines**: 2652, 2679, 2713
+
+**Issue**: `_wrap_cursor_with_and()` calls when cursor already has @type.
+
+**Details**:
+- `join()` line 2652
+- `sum()` line 2679
+- `slice()` line 2713
+
+**Impact**: Low - defensive programming, rarely triggered
+
+---
+
+### 3.4 Utility Method Edge Cases
+
+**Area**: `woqlquery/woql_query.py`  
+**Lines**: 3247-3254, 3280-3283, 3328-3358
+
+**Issue**: Complex utility methods with uncovered branches.
+
+**Details**:
+- Lines 3247-3254: `_find_last_subject()` with And query iteration
+- Lines 3280-3283: `_same_entry()` dictionary comparison
+- Lines 3328-3358: `_add_partial()` triple builder context logic
+
+**Impact**: Low - internal utilities, basic paths covered
+
+---
+
+### 3.5 Vocabulary and Type Handling
+
+**Area**: `woqlquery/woql_query.py`  
+**Lines**: 338, 706, 785, 3389
+
+**Issue**: Specific edge cases in type and vocabulary handling.
+
+**Details**:
+- Line 338: `_clean_arithmetic()` with dict having `to_dict` method
+- Line 706: Vocabulary extraction with prefixed terms
+- Line 785: Select validation (see bug 1.3)
+- Line 3389: Final line of file (likely closing brace or comment)
+
+**Impact**: Very Low - rare edge cases
+
+---
+
+## 4. Infrastructure and Integration Issues
+
+### 4.1 Cloud Infrastructure Tests
+
+**Area**: `integration_tests/`  
+**Severity**: N/A (External Dependency)
+
+**Issue**: TerminusX cloud infrastructure no longer operational (use DFRNT.com instead)
+
+**Skipped Tests**:
+- `test_client.py::test_diff_ops_no_auth`
+- `test_client.py::test_terminusx`
+- `test_client.py::test_terminusx_crazy_path`
+- `test_scripts.py::test_script_happy_path`
+
+**Impact**: Cannot test cloud integration features
+
+---
+
+### 4.2 JWT Authentication Tests
+
+**Area**: `integration_tests/test_client.py`  
+**Severity**: N/A (Optional Feature)
+
+**Issue**: JWT tests require environment variable `TERMINUSDB_TEST_JWT=1`.
+
+**Skipped Tests**:
+- `test_client.py::test_jwt`
+
+**Impact**: JWT authentication not tested in CI
+
+---
+
+## 5. Schema and Type System Peculiarities
+
+### 5.1 Relaxed Type Checking
+
+**Area**: `test_Schema.py`  
+**Severity**: N/A (Design Decision)
+
+**Issue**: Type constraints intentionally relaxed.
+
+**Skipped Tests**:
+- `test_Schema.py::test_abstract_class`
+- `test_Schema.py::test_type_check`
+
+**Details**: Tests skipped with reason "Relaxing type constraints" and "relaxing type checking"
+
+**Impact**: Type system is more permissive than originally designed
+
+---
+
+### 5.2 Backend-Dependent Features
+
+**Area**: `test_Schema.py`  
+**Severity**: N/A (Backend Dependency)
+
+**Issue**: Import objects feature requires backend implementation.
+
+**Skipped Tests**:
+- `test_Schema.py::test_import_objects`
+
+**Impact**: Cannot test object import without backend
+
+---
+
+### 5.3 Client Triple Retrieval
+
+**Area**: `test_Client.py`  
+**Severity**: N/A (Temporarily Unavailable)
+
+**Issue**: Triple retrieval features temporarily unavailable.
+
+**Skipped Tests**:
+- `test_Client.py::test_get_triples`
+- `test_Client.py::test_get_triples_with_enum`
+
+**Impact**: Cannot test triple retrieval functionality
+
+---
+
+## 6. Summary Statistics
+
+### Coverage Metrics
+- **Total Lines**: 1478
+- **Covered Lines**: 1388
+- **Uncovered Lines**: 90
+- **Coverage**: 94%
+
+### Test Metrics
+- **Total Tests**: 1356 passing
+- **Skipped Tests**: 20
+- **Test Files**: 50+
+
+### Issue Breakdown
+- **Critical Bugs**: 3 (validation logic, missing method, args introspection)
+- **Deprecated Code**: 1 (dead code to remove)
+- **Uncovered Edge Cases**: 60+ lines (mostly args introspection)
+- **Infrastructure Issues**: 5 (cloud/JWT tests)
+- **Design Decisions**: 4 (relaxed constraints, backend dependencies)
+
+---
+
+## 7. Recommendations for Future Iteration
+
+### High Priority
+1. **Fix validation logic** (lines 784-785, 821-822) - Critical for query correctness
+2. **Implement `_set_context()` or fix `graph()` method** (line 3230) - Broken feature
+3. **Fix quad args introspection** (lines 1056, 1092, 1128, 1775, 1807) - Consistent API
+
+### Medium Priority
+4. **Add tests for args introspection** - Improve coverage to 95%+
+5. **Investigate subquery limit/offset** (line 903) - Unknown issue
+6. **Document type system decisions** - Clarify relaxed constraints
+
+### Low Priority
+7. **Remove deprecated `_data_value_list()`** - Clean up dead code
+8. **Add edge case tests** - Cover remaining utility methods
+9. **Document cloud infrastructure status** - Update integration test docs
+
+### Future Considerations
+10. **JWT test automation** - Add to CI pipeline
+11. **Backend integration tests** - Coordinate with backend team
+12. **Triple retrieval feature** - Investigate "temporarily unavailable" status
+
+---
+
+## 8. Maintenance Notes
+
+**Last Updated**: January 2026  
+**Next Review**: After addressing high-priority bugs  
+
+**Note**: This document should be updated whenever:
+- Bugs are fixed (move to resolved section)
+- New issues are discovered
+- Coverage improves
+- Design decisions change