Add check for conditional variables in attachment content (EG416)

src/dayamlchecker/messages.py

@@ -70,6 +70,7 @@ class MessageId(StrEnum):
     MISSING_QUESTION_ID = "missing_question_id"
     MULTIPLE_MANDATORY_BLOCKS = "multiple_mandatory_blocks"
     MISSING_METADATA_FIELDS = "missing_metadata_fields"
+    ATTACHMENT_CONDITIONAL_VARIABLE = "attachment_conditional_variable"
 
     ACCESSIBILITY_COMBOBOX_NOT_ACCESSIBLE = "accessibility_combobox_not_accessible"
     ACCESSIBILITY_NO_LABEL_MULTI_FIELD = "accessibility_no_label_multi_field"
@@ -544,6 +545,13 @@ class MessageDefinition:
             "{fields}"
         ),
     ),
+    MessageId.ATTACHMENT_CONDITIONAL_VARIABLE: MessageDefinition(
+        code="EG416",
+        severity=Severity.ERROR,
+        finding_class=FindingClass.GENERAL,
+        summary="Attachment content references a conditionally asked variable",
+        template='attachment content references "{field_var}", but that field is only asked when certain conditions are met (show if/hide if). If those conditions are not met, the interview may get stuck',
+    ),
     MessageId.ACCESSIBILITY_COMBOBOX_NOT_ACCESSIBLE: MessageDefinition(
         code="EA501",
         severity=Severity.ERROR,
@@ -1194,24 +1202,18 @@ def make_finding(
         context=context,
     )
 
+
 def escape_data(value: str) -> str:
-    return (
-        value
-        .replace("%", "%25")
-        .replace("\r", "%0D")
-        .replace("\n", "%0A")
-    )
+    return value.replace("%", "%25").replace("\r", "%0D").replace("\n", "%0A")
+
 
 def escape_property(value: str) -> str:
-    return (
-        escape_data(value)
-        .replace(":", "%3A")
-        .replace(",", "%2C")
-    )
+    return escape_data(value).replace(":", "%3A").replace(",", "%2C")
+
 
 def print_github_annotation(d: Finding) -> None:
     kind = "notice" if d.severity == Severity.INFO else d.severity.value
-        
+
     props = []
 
     if getattr(d, "file_name", None):

src/dayamlchecker/yaml_structure.py

@@ -3,6 +3,7 @@
 import argparse
 from dataclasses import dataclass, field, replace
 from pathlib import Path
+from pyexpat import features
 import re
 import sys
 
@@ -1479,6 +1480,90 @@ def _has_matching_guard(active_guards: list[str], expected_guards: list[str]) ->
     return False
 
 
+def _extract_mako_guards_by_line(content: str) -> dict[int, list[str]]:
+    """Extract Mako % if conditions active at each line of a content string"""
+    guards_by_line: dict[int, list[str]] = {}
+    guard_stack: list[str] = []
+
+    for i, line in enumerate(content.splitlines(), start=1):
+        stripped = line.strip()
+        if stripped.startswith("% if ") or stripped.startswith("%if "):
+            cond = re.sub(r"^%\s*if\s+", "", stripped).rstrip(":")
+            guard_stack.append(cond)
+        elif stripped.startswith("% elif ") or stripped.startswith("%elif "):
+            if guard_stack:
+                guard_stack.pop()
+            cond = re.sub(r"^%\s*elif\s+", "", stripped).rstrip(":")
+            guard_stack.append(cond)
+        elif stripped in ("% endif", "%endif", "% endfor", "%endfor"):
+            if guard_stack:
+                guard_stack.pop()
+
+        if guard_stack:
+            guards_by_line[i] = list(guard_stack)
+
+    return guards_by_line
+
+
+def _find_unmatched_attachment_content_references(
+    doc: dict[str, Any], conditional_fields: list[dict[str, Any]]
+) -> list[tuple[str, int]]:
+    """Check attachment/attachments content blocks for unconditional references to variables that are only conditionally asked"""
+    if not conditional_fields:
+        return []
+
+    content_blocks: list[str] = []
+
+    attachment = _get_case_insensitive(doc, "attachment")
+    attachments = _get_case_insensitive(doc, "attachments")
+
+    if isinstance(attachment, dict):
+        skip = attachment.get("skip undefined")
+        if not (skip is True or (isinstance(skip, str) and skip.lower() == "true")):
+            content = attachment.get("content")
+            if isinstance(content, str):
+                content_blocks.append(content)
+    elif isinstance(attachment, list):
+        for item in attachment:
+            if isinstance(item, dict):
+                skip = item.get("skip undefined")
+                if not (
+                    skip is True or (isinstance(skip, str) and skip.lower() == "true")
+                ):
+                    content = item.get("content")
+                    if isinstance(content, str):
+                        content_blocks.append(content)
+
+    if isinstance(attachments, list):
+        for item in attachments:
+            if isinstance(item, dict):
+                skip = item.get("skip undefined")
+                if not (
+                    skip is True or (isinstance(skip, str) and skip.lower() == "true")
+                ):
+                    content = item.get("content")
+                    if isinstance(content, str):
+                        content_blocks.append(content)
+
+    if not content_blocks:
+        return []
+
+    unmatched: list[tuple[str, int]] = []
+    for content in content_blocks:
+        mako_guards_by_line = _extract_mako_guards_by_line(content)
+        for conditional in conditional_fields:
+            field_var = conditional["field_var"]
+            expected_guards = conditional["guards"]
+            ref_lines = _find_variable_reference_lines(content, field_var)
+            for ref_line in ref_lines:
+                active_guards = mako_guards_by_line.get(ref_line, [])
+                if _has_matching_guard(active_guards, expected_guards):
+                    continue
+                unmatched.append((field_var, ref_line))
+
+    return unmatched
+
+
 def _find_unmatched_interview_order_references(
     doc: dict[str, Any], conditional_fields: list[dict[str, Any]]
 ) -> list[tuple[str, int]]:
@@ -1583,6 +1668,7 @@ def find_errors_from_string(
     yaml_parser = _make_yaml_parser()
     prior_conditional_fields: list[dict[str, Any]] = []
     seen_ids: dict[str, int] = {}
+    skip_undefined = False
     parsed_docs: list[ParsedInterviewDocument] = []
     has_yaml_parse_errors = False
     line_number = 1
@@ -1737,6 +1823,28 @@ def find_errors_from_string(
                 )
             )
 
+        features = _get_case_insensitive(doc, "features")
+        if isinstance(features, dict):
+            skip_val = features.get("skip undefined")
+            if skip_val is True or (
+                isinstance(skip_val, str) and skip_val.lower() == "true"
+            ):
+                skip_undefined = True
+
+        if not skip_undefined:
+            unmatched_content_refs = _find_unmatched_attachment_content_references(
+                doc, prior_conditional_fields
+            )
+            for field_var, ref_line in unmatched_content_refs:
+                all_errors.append(
+                    make_finding(
+                        MessageId.ATTACHMENT_CONDITIONAL_VARIABLE,
+                        file_name=input_file,
+                        line_number=doc["__line__"] + line_number + ref_line,
+                        field_var=field_var,
+                    )
+                )
+
         nesting_depth = _max_screen_visibility_nesting_depth(doc)
         if nesting_depth > 2:
             all_errors.append(
@@ -2198,7 +2306,7 @@ def main(argv: Optional[list[str]] = None) -> int:
 
     had_error = False
     warning_count = sum(1 for f in all_findings if f.severity == "warning")
-    
+
     if args.format == "github":
         for finding in all_findings:
             if finding.severity == "error":
@@ -2207,19 +2315,23 @@ def main(argv: Optional[list[str]] = None) -> int:
     else:
         error_count = sum(1 for f in all_findings if f.severity == "error")
         info_count = sum(1 for f in all_findings if f.severity == "info")
-        
+
         if len(all_findings) == 0:
             print("No issues found.")
         else:
             if info_count > 0:
-                print(f"Found {len(all_findings)} issues ({error_count} errors, {warning_count} warnings, {info_count} infos):")
+                print(
+                    f"Found {len(all_findings)} issues ({error_count} errors, {warning_count} warnings, {info_count} infos):"
+                )
             elif warning_count > 0:
-                print(f"Found {len(all_findings)} issues ({error_count} errors, {warning_count} warnings):")
+                print(
+                    f"Found {len(all_findings)} issues ({error_count} errors, {warning_count} warnings):"
+                )
             else:
                 print(f"Found {len(all_findings)} errors:")
             for err in all_findings:
                 print(f"{err}")
-                
+
         if error_count > 0:
             had_error = True