relationalai 0.13.0.dev0__py3-none-any.whl → 0.13.1__py3-none-any.whl

v0/relationalai/semantics/rel/rel_utils.py

@@ -1,276 +0,0 @@
-"""
-Helpers to generate Rel, many taken from the original PyRel implementation.
-"""
-from __future__ import annotations
-import re
-from v0.relationalai.semantics.metamodel import ir, types, factory as f
-
-# Rel primitive relations used by the compiler
-# TODO: how to deal with varargs?
-RelPrimitiveOuterJoin = f.relation("rel_primitive_outer_join", [])
-
-# List of higher order relations in Rel
-HIGHER_ORDER = set([
-    "::std::common::count",
-    "::std::common::sum",
-    "::std::common::mean",
-    "::std::common::max",
-    "::std::common::min",
-    "::std::common::sort",
-    "::std::common::reverse_sort",
-    "::std::common::top",
-    "::std::common::bottom",
-])
-
-# Mappings of IR operators to the equivalent in Rel
-OPERATORS = {
-    # binary
-    # TODO: maybe add the binary operators
-
-    # aggregates
-    "count": "pyrel_count",
-    "sum": "::std::common::sum",
-    "avg": "::std::common::mean",
-    "max": "::std::common::max",
-    "min": "::std::common::min",
-
-    # std
-    "range": "::std::common::range",
-    "hash": "rel_primitive_hash_tuple_uint128",
-    "uuid_to_string": "::std::common::uuid_string",
-    "parse_uuid": "::std::common::parse_uuid",
-
-    # dates
-    "date_year": "::std::common::date_year",
-    "date_quarter": "::std::common::date_quarterofyear",
-    "date_month": "::std::common::date_month",
-    "date_week": "::std::common::date_week",
-    "date_day": "::std::common::date_day",
-    "date_dayofyear": "::std::common::date_dayofyear",
-    "date_weekday": "::std::common::date_dayofweek",
-    "date_add": "::std::common::date_add",
-    "date_subtract": "::std::common::date_subtract",
-    "dates_period_days": "pyrel_dates_period_days",
-    "datetimes_period_milliseconds": "pyrel_datetimes_period_milliseconds",
-    "datetime_add": "::std::common::datetime_add",
-    "datetime_year": "::std::common::datetime_year",
-    "datetime_quarter": "::std::common::datetime_quarterofyear",
-    "datetime_month": "::std::common::datetime_month",
-    "datetime_week": "::std::common::datetime_week",
-    "datetime_day": "::std::common::datetime_day",
-    "datetime_dayofyear": "::std::common::datetime_dayofyear",
-    "datetime_hour": "::std::common::datetime_hour",
-    "datetime_minute": "::std::common::datetime_minute",
-    "datetime_second": "::std::common::datetime_second",
-    "datetime_weekday": "::std::common::datetime_dayofweek",
-    "datetime_subtract": "::std::common::datetime_subtract",
-    "construct_datetime_ms_tz": "::std::common::^DateTime",
-    "date_format": "::std::common::format_date",
-    "datetime_format": "::std::common::format_datetime",
-    "construct_date_from_datetime": "::std::common::^Date",
-    "construct_date": "::std::common::^Date",
-    "parse_date": "rel_primitive_parse_date",
-    "parse_datetime": "rel_primitive_parse_datetime",
-
-    # periods
-    "nanosecond": "::std::common::^Nanosecond",
-    "microsecond": "::std::common::^Microsecond",
-    "millisecond": "::std::common::^Millisecond",
-    "second": "::std::common::^Second",
-    "minute": "::std::common::^Minute",
-    "hour": "::std::common::^Hour",
-    "day": "::std::common::^Day",
-    "week": "::std::common::^Week",
-    "month": "::std::common::^Month",
-    "year": "::std::common::^Year",
-
-    # math
-    "abs": "::std::common::abs",
-    "natural_log": "::std::common::natural_log",
-    "log": "::std::common::log",
-    "log10": "::std::common::log10",
-    "sqrt": "std::common::sqrt",
-    "maximum": "::std::common::maximum",
-    "minimum": "::std::common::minimum",
-    "//": "::std::common::trunc_divide",
-    "isinf": "::std::common::Infinity",
-    "isnan": "::std::common::NaN",
-    "ceil": "::std::common::ceil",
-    "floor": "::std::common::floor",
-    "pow": "rel_primitive_power",
-    "cbrt": "rel_primitive_cbrt",
-    "factorial": "rel_primitive_factorial",
-    "cos": "rel_primitive_cos",
-    "cosh": "rel_primitive_cosh",
-    "acos": "rel_primitive_acos",
-    "acosh": "rel_primitive_acosh",
-    "sin": "rel_primitive_sin",
-    "sinh": "rel_primitive_sinh",
-    "asin": "rel_primitive_asin",
-    "asinh": "rel_primitive_asinh",
-    "tan": "rel_primitive_tan",
-    "tanh": "rel_primitive_tanh",
-    "atan": "rel_primitive_atan",
-    "atanh": "rel_primitive_atanh",
-    "cot": "rel_primitive_cot",
-    "acot": "rel_primitive_acot",
-    "exp": "rel_primitive_natural_exp",
-    "erf": "rel_primitive_error_function",
-    "erfinv": "rel_primitive_error_function_inverse",
-
-    # strings
-    "string": "::std::common::string",
-    "concat": "::std::common::concat",
-    "starts_with": "::std::common::starts_with",
-    "ends_with": "::std::common::ends_with",
-    "contains": "::std::common::contains",
-    "substring": "::std::common::substring",
-    "num_chars": "::std::common::num_chars",
-    "like_match": "::std::common::like_match",
-    "lower": "::std::common::lowercase",
-    "upper": "::std::common::uppercase",
-    "strip": "rel_primitive_trim",
-    "levenshtein": "rel_primitive_levenshtein",
-    "replace": "rel_primitive_replace",
-    "split": "::std::common::string_split",
-    "split_part": "::std::common::string_split",
-
-    # regex
-    "regex_match": "rel_primitive_regex_match",
-    "regex_match_all": "::std::common::regex_match_all",
-    "regex_search": "pyrel_regex_search",
-    "capture_group_by_index": "::std::common::capture_group_by_index",
-    "capture_group_by_name": "::std::common::capture_group_by_name",
-    "escape_regex_metachars": "::std::common::escape_regex_metachars",
-
-    # ints
-    "parse_int64": "::std::common::parse_int",
-    "parse_int128": "::std::common::parse_int128",
-
-    # floats
-    "parse_float": "::std::common::parse_float",
-
-    # pragmas
-    "rule_reasoner_sem_vo": "rel_primitive_force_faq_var_order",
-    "rule_reasoner_phys_vo": "rel_primitive_force_var_order",
-
-    # primitive graph algorithms
-    "infomap": "rel_primitive_infomap",
-    "louvain": "rel_primitive_louvain",
-    "label_propagation": "rel_primitive_async_label_propagation",
-}
-
-def rel_operator(ir_op):
-    """ Maps an operator from the metamodel IR into the equivalent Rel operator. """
-    if ir_op in OPERATORS:
-        return OPERATORS[ir_op]
-    else:
-        return ir_op
-
-def rel_typename(ir_type: ir.Type):
-    """ Get the name of the type to use in Rel. """
-    if isinstance(ir_type, ir.DecimalType):
-        return f'pyrel_Decimal_{ir_type.precision}_{ir_type.scale}'
-    if isinstance(ir_type, ir.ScalarType):
-        if ir_type == types.Sha1:
-            return 'Sha1'
-        elif ir_type == types.Enum:
-            return '::std::common::UInt128'
-        elif ir_type == types.EntityTypeVar:
-            return '::std::common::UInt128'
-        elif ir_type == types.Hash:
-            return "::std::common::UInt128"
-        elif ir_type == types.UInt128:
-            return '::std::common::UInt128'
-        elif ir_type == types.RowId:
-            return '::std::common::UInt128'
-        elif ir_type == types.Bool:
-            return '::std::common::Boolean'
-        elif types.is_builtin(ir_type):
-            return f'::std::common::{ir_type.name}'
-        elif ir_type.super_types:
-            # assuming the compiler narrowed this down somehow
-            return rel_typename(ir_type.super_types.some())
-        else:
-            # user-defined types without a super type are user defined entities
-            return '::std::common::UInt128'
-    else:
-        # TODO: how should we deal with Union types here?
-        return '::std::common::Any'
-
-# SEE: REL:      https://docs.relational.ai/rel/ref/lexical-symbols#keywords
-#      CORE REL: https://docs.google.com/document/d/12LUQdRed7P5EqQI1D7AYG4Q5gno9uKqy32i3kvAWPCA
-RESERVED_WORDS = [
-    "and",
-    "as",
-    "bound",
-    "declare",
-    "def",
-    "else",
-    "end",
-    "entity",
-    "exists",
-    "false",
-    "for",
-    "forall",
-    "from",
-    "ic",
-    "if",
-    "iff",
-    "implies",
-    "in",
-    "module",
-    "namespace",
-    "not",
-    "or",
-    "requires",
-    "then",
-    "true",
-    "use",
-    "where",
-    "with",
-    "xor"
-]
-
-rel_sanitize_re = re.compile(r'[^\w:\[\]\^" ,]|^(?=\d)', re.UNICODE)
-unsafe_symbol_pattern = re.compile(r"[^a-zA-Z0-9_]", re.UNICODE)
-
-def sanitize(input_string, is_rel_name_or_symbol = False):
-    # Replace non-alphanumeric characters with underscores
-    if is_rel_name_or_symbol and "[" in input_string:
-        string_parts = input_string.split('[', 1)
-        sanitized_name_or_symbol = sanitize_identifier(string_parts[0])
-        sanitized_rest = re.sub(rel_sanitize_re, "_", string_parts[1])
-        sanitized = f"{sanitized_name_or_symbol}[{sanitized_rest}"
-    else:
-        if "::" in input_string: # TODO: This is a temp solution to avoid sanitizing the namespace
-            sanitized = re.sub(rel_sanitize_re, "_", input_string)
-        else:
-            sanitized = sanitize_identifier(input_string)
-
-    # Check if the resulting string is a keyword and append an underscore if it is
-    if sanitized in RESERVED_WORDS:
-        sanitized += "_"
-
-    return sanitized
-
-def sanitize_identifier(name: str) -> str:
-    """
-    Return a string safe to use as a top level identifier in rel, such as a variable or relation name.
-
-    Args:
-        name (str): The input identifier string.
-
-    Returns:
-        str: The sanitized identifier string.
-    """
-
-    if not name:
-        return name
-
-    safe_name = ''.join(c if c.isalnum() else '_' for c in name)
-    if safe_name[0].isdigit():
-        safe_name = '_' + safe_name
-    if  safe_name in RESERVED_WORDS:
-        safe_name = safe_name + "_" # preferring the pythonic pattern of `from_` vs `_from`
-    return safe_name

v0/relationalai/semantics/snowflake/__init__.py

@@ -1,3 +0,0 @@
-from v0.relationalai.semantics.internal.snowflake import Table, is_direct_access_enabled, IcebergConfig
-
-__all__ = ["Table", "is_direct_access_enabled", "IcebergConfig"]