fix: handle NaN/undefined variable in integer comparisons

Wrap int(var_value) in __eval_basiccondition() with try/except to
prevent ValueError crash when a variable is undefined, empty, or
contains non-numeric text. In warn mode the comparison evaluates to
false; in stop mode a PPPInterrupt is raised with a descriptive message.

docs/CONFIG.md

@@ -34,7 +34,7 @@ With this prompt: `__quality__, 1girl, ${head:__eyes__, __hair__, __expression__
 ### General settings
 
 * **Debug level**: what to write to the console. Note: in *SD.Next* debug messages only show if you launch it with the `--debug` argument.
-* **What to do on invalid content warnings?**: warn on the console or stop the generation.
+* **What to do on invalid content warnings?**: warn on the console or stop the generation. This also affects integer comparisons with undefined or non-numeric variables: in *warn* mode the comparison evaluates to false, in *stop* mode the generation is stopped with an error.
 * **Model variant definitions**: definitions for model variants to be recognized based on strings found in the full filename.
 
     The format for each line is (with *kind* being one of the base model identifiers or not defined):

docs/SYNTAX.md

@@ -211,6 +211,8 @@ The `conditionN` can be:
 
 For a simple value the allowed operations are `eq`, `ne`, `gt`, `lt`, `ge`, `le`, `contains` and the value can be a quoted string or an integer. For a list of values the allowed operations are `contains`, `in` and the value of the variable is checked against all the elements of the list until one matches. The operation can be preceded by `not` for readability, instead of using it in the front.
 
+When a variable used in an integer comparison is undefined or cannot be converted to an integer (e.g. it contains non-numeric text or is an empty string), the behavior depends on the `on_warning` setting: in `warn` mode the comparison evaluates to false, and in `stop` mode an error is raised.
+
 You can also build complex conditions joining them with boolean operators and/or/not and parentheses.
 
 The variable can be one set with the `set` or `add` commands (user variables) or you can use system variables like these (names starting with an underscore are reserved for system variables):

ppp.py

@@ -1139,19 +1139,20 @@ class PromptPostProcessor:  # pylint: disable=too-few-public-methods,too-many-in
             )
             result = False
             for c in cond_value_adjusted:
-                var_value_adjusted = (
-                    var_value
-                    if isinstance(c, str)
-                    else (
-                        True
-                        if isinstance(c, bool) and var_value != "false" and var_value != "" and var_value is not False
-                        else (
-                            False
-                            if isinstance(c, bool) and (var_value != "true" or var_value is False)
-                            else int(var_value)
+                if isinstance(c, str):
+                    var_value_adjusted = var_value
+                elif isinstance(c, bool) and var_value != "false" and var_value != "" and var_value is not False:
+                    var_value_adjusted = True
+                elif isinstance(c, bool) and (var_value != "true" or var_value is False):
+                    var_value_adjusted = False
+                else:
+                    try:
+                        var_value_adjusted = int(var_value)
+                    except (ValueError, TypeError):
+                        self.warn_or_stop(
+                            f"Cannot convert variable value '{var_value}' to integer for comparison"
                         )
-                    )
-                )
+                        return False
                 result = comp_ops[cond_comp](var_value_adjusted, c)
                 if result:
                     break