edsl 0.1.14__py3-none-any.whl → 0.1.40__py3-none-any.whl

edsl/utilities/naming_utilities.py

@@ -0,0 +1,263 @@
+import re
+import keyword
+
+stop_words = {
+    "into",
+    "mustn't",
+    "there",
+    "you'll",
+    "don",
+    "have",
+    "at",
+    "if",
+    "on",
+    "some",
+    "with",
+    "in",
+    "can",
+    "mightn",
+    "off",
+    "few",
+    "not",
+    "d",
+    "hadn",
+    "shan't",
+    "t",
+    "re",
+    "where",
+    "s",
+    "won't",
+    "mustn",
+    "wasn't",
+    "didn't",
+    "has",
+    "same",
+    "too",
+    "will",
+    "you've",
+    "all",
+    "haven't",
+    "isn't",
+    "over",
+    "of",
+    "about",
+    "its",
+    "being",
+    "it",
+    "her",
+    "should",
+    "himself",
+    "wasn",
+    "out",
+    "theirs",
+    "aren",
+    "that",
+    "our",
+    "shouldn't",
+    "you'd",
+    "such",
+    "above",
+    "my",
+    "the",
+    "any",
+    "been",
+    "as",
+    "very",
+    "herself",
+    "o",
+    "weren",
+    "until",
+    "their",
+    "shouldn",
+    "up",
+    "wouldn",
+    "couldn't",
+    "hasn't",
+    "no",
+    "than",
+    "hadn't",
+    "had",
+    "you",
+    "here",
+    "yourself",
+    "yourselves",
+    "during",
+    "ain",
+    "once",
+    "aren't",
+    "what",
+    "so",
+    "hers",
+    "that'll",
+    "other",
+    "ours",
+    "yours",
+    "nor",
+    "him",
+    "doesn",
+    "doesn't",
+    "he",
+    "them",
+    "for",
+    "ll",
+    "isn",
+    "this",
+    "or",
+    "who",
+    "only",
+    "itself",
+    "they",
+    "between",
+    "against",
+    "under",
+    "me",
+    "now",
+    "mightn't",
+    "those",
+    "needn't",
+    "these",
+    "when",
+    "before",
+    "his",
+    "she's",
+    "having",
+    "be",
+    "don't",
+    "haven",
+    "won",
+    "while",
+    "both",
+    "didn",
+    "by",
+    "ourselves",
+    "m",
+    "your",
+    "then",
+    "myself",
+    "we",
+    "it's",
+    "should've",
+    "through",
+    "why",
+    "from",
+    "and",
+    "hasn",
+    "more",
+    "how",
+    "ve",
+    "most",
+    "because",
+    "did",
+    "y",
+    "i",
+    "an",
+    "but",
+    "whom",
+    "below",
+    "further",
+    "am",
+    "which",
+    "just",
+    "ma",
+    "you're",
+    "couldn",
+    "do",
+    "shan",
+    "own",
+    "again",
+    "are",
+    "weren't",
+    "down",
+    "is",
+    "were",
+    "each",
+    "needn",
+    "themselves",
+    "she",
+    "after",
+    "does",
+    "wouldn't",
+    "to",
+    "a",
+    "was",
+    "doing",
+}
+
+
+def sanitize_string(input_string, max_length=35):
+    """Return a sanitized version of the input string that can be used as a variable name.
+
+    >>> candidate_names = ["How are you doing this morning, Dave? What is your favorite kind of coffee?", "class", "def", "here_is_some_text"]
+    >>> [sanitize_string(name) for name in candidate_names]
+    ['morning_dave_favorite_kind_coffee', 'class_modified', 'def_modified', 'here_is_some_text']
+    """
+
+    # Ensure nltk stopwords are downloaded
+    # try:
+    #     from nltk.corpus import stopwords
+    # except ImportError or ModuleNotFoundError:
+    #     print(
+    #         "nltk is not installed. Please install it using 'pip install nltk' to use these features."
+    #     )
+    #     raise
+
+    # try:
+    #     stop_words = set(stopwords.words("english"))
+    # except LookupError:
+    #     nltk.download("stopwords")
+    #     stop_words = set(stopwords.words("english"))
+    #     # raise LookupError("Stopwords not found. Please download them using nltk.download('stopwords')")
+
+    # # Define the list of stopwords
+
+    # Replace special characters with spaces and split into words
+    words = re.sub(r"\W+", " ", input_string).split()
+
+    # Remove stopwords
+    important_words = [word for word in words if word.lower() not in stop_words]
+
+    # Join words with underscores
+    sanitized_string = "_".join(important_words)
+
+    # Ensure the length is less than 25 characters
+    if len(sanitized_string) > max_length:
+        # split off the last word and remove it
+        words = sanitized_string[:max_length].split("_")
+        if len(words) == 1:
+            sanitized_string = words[0]
+        else:
+            sanitized_string = "_".join(words[:-1])
+
+    # Remove leading and trailing underscores
+    sanitized_string = sanitized_string.strip("_")
+
+    # Check if the string is a Python keyword
+    if keyword.iskeyword(sanitized_string):
+        sanitized_string += "_modified"
+
+    result = sanitized_string.lower()
+    return result
+
+
+# Example usage
+# input_string = "This is a sample variable-name@123 for testing"
+# sanitized_string = sanitize_string(input_string)
+# print(sanitized_string)  # Output might be: sample_variable_name_123
+
+# if __name__ == "__main__":
+#     candidate_names = [
+#         "How are you doing this morning, Dave? What is your favorite kind of coffee?",
+#         "class",
+#         "def",
+#         "here_is_some_text",
+#     ]
+#     for name in candidate_names:
+#         print(f"Original: {name}")
+#         print(f"Sanitized: {sanitize_string(name)}")
+#         print()
+
+if __name__ == "__main__":
+    # from edsl.conjure.InputData import InputDataABC
+    import doctest
+
+    doctest.testmod(optionflags=doctest.ELLIPSIS)

edsl/utilities/remove_edsl_version.py

@@ -0,0 +1,24 @@
+from functools import wraps
+
+
+def remove_edsl_version(func):
+    """
+    Decorator for the EDSL objects' `from_dict` method.
+    - Removes the EDSL version and class name from the dictionary.
+    - Ensures backwards compatibility with older versions of EDSL.
+    """
+
+    @wraps(func)
+    def wrapper(cls, data, *args, **kwargs):
+        data_copy = dict(data)
+        edsl_version = data_copy.pop("edsl_version", None)
+        edsl_classname = data_copy.pop("edsl_class_name", None)
+
+        # Version- and class-specific logic here
+        if edsl_classname == "Survey":
+            if edsl_version is None or edsl_version <= "0.1.20":
+                data_copy["question_groups"] = {}
+
+        return func(cls, data_copy, *args, **kwargs)
+
+    return wrapper

edsl/utilities/repair_functions.py

@@ -0,0 +1,28 @@
+import json
+from edsl.utilities.utilities import valid_json
+
+
+def extract_json_from_string(s):
+    """Extract a JSON string from a string."""
+    # Find the first occurrence of '{'
+    start_idx = s.find("{")
+    # Find the last occurrence of '}'
+    end_idx = s.rfind("}")
+    # If both '{' and '}' are found in the string
+    if start_idx != -1 and end_idx != -1 and start_idx < end_idx:
+        # Extract the substring from start_idx to end_idx (inclusive)
+        json_str = s[start_idx : end_idx + 1]
+        try:
+            return json.loads(json_str)
+        except json.JSONDecodeError:
+            raise ValueError("Invalid JSON string")
+    else:
+        raise ValueError("No JSON object found in string")
+
+
+if __name__ == "__main__":
+    text = (
+        'Sure - here is some JSON { "key": "value", "number": 123, "array": [1, 2, 3] }'
+    )
+    extracted_json = extract_json_from_string(text)
+    d = extracted_json

edsl/utilities/restricted_python.py

@@ -0,0 +1,70 @@
+from RestrictedPython import compile_restricted, safe_globals
+from RestrictedPython.Guards import (
+    safe_builtins,
+    full_write_guard,
+    guarded_iter_unpack_sequence,
+)
+
+
+def guarded_iter(obj, allowed_types=(list, tuple, set, dict, range)):
+    """Ensures iteration is only performed on safe, allowable types."""
+    if not isinstance(obj, allowed_types):
+        raise TypeError(f"Iteration over {type(obj).__name__} not allowed.")
+    return iter(obj)
+
+
+class FunctionAForbiddenAttributeException(Exception):
+    """Exception for errors during function execution when forbidden key is used."""
+
+
+def default_guarded_getitem(ob, key):
+    sensitive_python_keys = [
+        "__dict__",
+        "__class__",
+        "__module__",
+        "__bases__",
+        "__mro__",
+        "__subclasses__",
+        "__func__",
+        "__self__",
+        "__closure__",
+        "__code__",
+        "__globals__",
+        "__call__",
+        "__getattr__",
+        "__getattribute__",
+        "__delattr__",
+        "__setattr__",
+    ]
+    if key in sensitive_python_keys:
+        raise FunctionAForbiddenAttributeException(
+            f"Access denied for attribute: {key}"
+        )
+
+    return ob.get(key)
+
+
+def create_restricted_function(function_name, source_code, loop_activated=True):
+    """Activate the function using RestrictedPython with basic restrictions."""
+    safe_env = safe_globals.copy()
+    safe_env["__builtins__"] = {**safe_builtins}
+    safe_env["_getitem_"] = default_guarded_getitem
+
+    if loop_activated:
+        safe_env["_getiter_"] = guarded_iter
+        safe_env["_iter_unpack_sequence_"] = guarded_iter_unpack_sequence
+
+    tmp_source_code = source_code.split("def ")
+    if len(tmp_source_code) >= 2:
+        source_code = "def " + tmp_source_code[1]
+
+    byte_code = compile_restricted(source_code, "<string>", "exec")
+    loc = {}
+    try:
+        exec(byte_code, safe_env, loc)
+        func = loc[function_name]
+    except Exception as e:
+        print("Creating restricted function error:", e)
+        raise e
+
+    return func