edsl 0.1.39.dev1__py3-none-any.whl → 0.1.39.dev2__py3-none-any.whl

edsl/language_models/PriceManager.py

@@ -0,0 +1,127 @@
+from typing import Dict, Tuple, Optional, Union
+
+
+class PriceManager:
+    _instance = None
+    _price_lookup: Dict[Tuple[str, str], Dict] = {}
+    _is_initialized = False
+
+    def __new__(cls):
+        if cls._instance is None:
+            cls._instance = super(PriceManager, cls).__new__(cls)
+        return cls._instance
+
+    def __init__(self):
+        # Only initialize once, even if __init__ is called multiple times
+        if not self._is_initialized:
+            self._is_initialized = True
+            self.refresh_prices()
+
+    def refresh_prices(self) -> None:
+        """
+        Fetch fresh prices from the Coop service and update the internal price lookup.
+
+        """
+        from edsl.coop import Coop
+
+        c = Coop()
+        try:
+            self._price_lookup = c.fetch_prices()
+        except Exception as e:
+            print(f"Error fetching prices: {str(e)}")
+
+    def get_price(self, inference_service: str, model: str) -> Optional[Dict]:
+        """
+        Get the price information for a specific service and model combination.
+
+        Args:
+            inference_service (str): The name of the inference service
+            model (str): The model identifier
+
+        Returns:
+            Optional[Dict]: Price information if found, None otherwise
+        """
+        key = (inference_service, model)
+        return self._price_lookup.get(key)
+
+    def get_all_prices(self) -> Dict[Tuple[str, str], Dict]:
+        """
+        Get the complete price lookup dictionary.
+
+        Returns:
+            Dict[Tuple[str, str], Dict]: The complete price lookup dictionary
+        """
+        return self._price_lookup.copy()
+
+    def calculate_cost(
+        self,
+        inference_service: str,
+        model: str,
+        usage: Dict[str, Union[str, int]],
+        input_token_name: str,
+        output_token_name: str,
+    ) -> Union[float, str]:
+        """
+        Calculate the total cost for a model usage based on input and output tokens.
+
+        Args:
+            inference_service (str): The inference service identifier
+            model (str): The model identifier
+            usage (Dict[str, Union[str, int]]): Dictionary containing token usage information
+            input_token_name (str): Key name for input tokens in the usage dict
+            output_token_name (str): Key name for output tokens in the usage dict
+
+        Returns:
+            Union[float, str]: Total cost if calculation successful, error message string if not
+        """
+        relevant_prices = self.get_price(inference_service, model)
+        if relevant_prices is None:
+            return f"Could not find price for model {model} in the price lookup."
+
+        # Extract token counts
+        try:
+            input_tokens = int(usage[input_token_name])
+            output_tokens = int(usage[output_token_name])
+        except Exception as e:
+            return f"Could not fetch tokens from model response: {e}"
+
+        # Extract price information
+        try:
+            inverse_output_price = relevant_prices["output"]["one_usd_buys"]
+            inverse_input_price = relevant_prices["input"]["one_usd_buys"]
+        except Exception as e:
+            if "output" not in relevant_prices:
+                return f"Could not fetch prices from {relevant_prices} - {e}; Missing 'output' key."
+            if "input" not in relevant_prices:
+                return f"Could not fetch prices from {relevant_prices} - {e}; Missing 'input' key."
+            return f"Could not fetch prices from {relevant_prices} - {e}"
+
+        # Calculate input cost
+        if inverse_input_price == "infinity":
+            input_cost = 0
+        else:
+            try:
+                input_cost = input_tokens / float(inverse_input_price)
+            except Exception as e:
+                return f"Could not compute input price - {e}."
+
+        # Calculate output cost
+        if inverse_output_price == "infinity":
+            output_cost = 0
+        else:
+            try:
+                output_cost = output_tokens / float(inverse_output_price)
+            except Exception as e:
+                return f"Could not compute output price - {e}"
+
+        return input_cost + output_cost
+
+    @property
+    def is_initialized(self) -> bool:
+        """
+        Check if the PriceManager has been initialized.
+
+        Returns:
+            bool: True if initialized, False otherwise
+        """
+        return self._is_initialized

edsl/language_models/RawResponseHandler.py

@@ -0,0 +1,106 @@
+import json
+from typing import Optional, Any, List
+from edsl.exceptions.language_models import LanguageModelBadResponseError
+
+from json_repair import repair_json
+
+
+def _extract_item_from_raw_response(data, sequence):
+    if isinstance(data, str):
+        try:
+            data = json.loads(data)
+        except json.JSONDecodeError as e:
+            return data
+    current_data = data
+    for i, key in enumerate(sequence):
+        try:
+            if isinstance(current_data, (list, tuple)):
+                if not isinstance(key, int):
+                    raise TypeError(
+                        f"Expected integer index for sequence at position {i}, got {type(key).__name__}"
+                    )
+                if key < 0 or key >= len(current_data):
+                    raise IndexError(
+                        f"Index {key} out of range for sequence of length {len(current_data)} at position {i}"
+                    )
+            elif isinstance(current_data, dict):
+                if key not in current_data:
+                    raise KeyError(
+                        f"Key '{key}' not found in dictionary at position {i}"
+                    )
+            else:
+                raise TypeError(
+                    f"Cannot index into {type(current_data).__name__} at position {i}. Full response is: {data} of type {type(data)}. Key sequence is: {sequence}"
+                )
+
+            current_data = current_data[key]
+        except Exception as e:
+            path = " -> ".join(map(str, sequence[: i + 1]))
+            if "error" in data:
+                msg = data["error"]
+            else:
+                msg = f"Error accessing path: {path}. {str(e)}. Full response is: '{data}'"
+            raise LanguageModelBadResponseError(message=msg, response_json=data)
+    if isinstance(current_data, str):
+        return current_data.strip()
+    else:
+        return current_data
+
+
+class RawResponseHandler:
+    """Class to handle raw responses from language models."""
+
+    def __init__(self, key_sequence: list, usage_sequence: Optional[list] = None):
+        self.key_sequence = key_sequence
+        self.usage_sequence = usage_sequence
+
+    def get_generated_token_string(self, raw_response):
+        return _extract_item_from_raw_response(raw_response, self.key_sequence)
+
+    def get_usage_dict(self, raw_response):
+        if self.usage_sequence is None:
+            return {}
+        return _extract_item_from_raw_response(raw_response, self.usage_sequence)
+
+    def parse_response(self, raw_response: dict[str, Any]) -> "EDSLOutput":
+        """Parses the API response and returns the response text."""
+
+        from edsl.data_transfer_models import EDSLOutput
+
+        generated_token_string = self.get_generated_token_string(raw_response)
+        last_newline = generated_token_string.rfind("\n")
+
+        if last_newline == -1:
+            # There is no comment
+            edsl_dict = {
+                "answer": self.convert_answer(generated_token_string),
+                "generated_tokens": generated_token_string,
+                "comment": None,
+            }
+        else:
+            edsl_dict = {
+                "answer": self.convert_answer(generated_token_string[:last_newline]),
+                "comment": generated_token_string[last_newline + 1 :].strip(),
+                "generated_tokens": generated_token_string,
+            }
+        return EDSLOutput(**edsl_dict)
+
+    @staticmethod
+    def convert_answer(response_part):
+        import json
+
+        response_part = response_part.strip()
+
+        if response_part == "None":
+            return None
+
+        repaired = repair_json(response_part)
+        if repaired == '""':
+            # it was a literal string
+            return response_part
+
+        try:
+            return json.loads(repaired)
+        except json.JSONDecodeError as j:
+            # last resort
+            return response_part

edsl/language_models/__init__.py

@@ -1,3 +1,2 @@
 from edsl.language_models.LanguageModel import LanguageModel
 from edsl.language_models.registry import Model
-from edsl.language_models.KeyLookup import KeyLookup

edsl/language_models/key_management/KeyLookup.py

@@ -0,0 +1,63 @@
+from collections import UserDict
+from dataclasses import asdict
+from edsl.enums import service_to_api_keyname
+
+from edsl.language_models.key_management.models import LanguageModelInput
+
+
+class KeyLookup(UserDict):
+    """A class for looking up API keys and related configuration.
+
+    >>> from edsl.language_models.key_management.models import LanguageModelInput
+    >>> lookup = KeyLookup()
+    >>> lm_input = LanguageModelInput.example()
+    >>> lookup['test'] = lm_input
+    >>> lookup.to_dict()['test']['api_token']
+    'sk-abcd123'
+    >>> restored = KeyLookup.from_dict(lookup.to_dict())
+    >>> restored['test'].api_token
+    'sk-abcd123'
+    """
+
+    def to_dict(self):
+        """
+        >>> kl = KeyLookup.example()
+        >>> kl2 = KeyLookup.from_dict(kl.to_dict())
+        >>> kl2 == kl
+        True
+        >>> kl2 is kl
+        False
+        """
+        return {k: asdict(v) for k, v in self.data.items()}
+
+    @classmethod
+    def from_dict(cls, d):
+        return cls({k: LanguageModelInput(**v) for k, v in d.items()})
+
+    @classmethod
+    def example(cls):
+        return cls(
+            {
+                "test": LanguageModelInput.example(),
+                "openai": LanguageModelInput.example(),
+            }
+        )
+
+    def to_dot_env(self):
+        """Return a string representation of the key lookup collection for a .env file."""
+        lines = []
+        for service, lm_input in self.items():
+            if service != "test":
+                lines.append(f"EDSL_SERVICE_RPM_{service.upper()}={lm_input.rpm}")
+                lines.append(f"EDSL_SERVICE_TPM_{service.upper()}={lm_input.tpm}")
+                key_name = service_to_api_keyname.get(service, service)
+                lines.append(f"{key_name.upper()}={lm_input.api_token}")
+                if lm_input.api_id is not None:
+                    lines.append(f"{service.upper()}_API_ID={lm_input.api_id}")
+        return "\n".join([f"{line}" for line in lines])
+
+
+if __name__ == "__main__":
+    import doctest
+
+    doctest.testmod(optionflags=doctest.ELLIPSIS)

edsl/language_models/key_management/KeyLookupBuilder.py

@@ -0,0 +1,273 @@
+from typing import Optional, List
+from collections import UserDict
+import os
+from functools import lru_cache
+from dataclasses import dataclass, asdict
+
+from edsl.enums import service_to_api_keyname
+from edsl.exceptions.general import MissingAPIKeyError
+
+from edsl.language_models.key_management.KeyLookup import KeyLookup
+
+from edsl.language_models.key_management.models import (
+    APIKeyEntry,
+    LimitEntry,
+    APIIDEntry,
+    LanguageModelInput,
+)
+
+service_to_api_keyname["bedrock"] = "AWS_SECRET_ACCESS_KEY"
+service_to_api_id = {"bedrock": "AWS_ACCESS_KEY_ID"}
+
+api_keyname_to_service = {}
+
+for service, key in service_to_api_keyname.items():
+    if isinstance(key, list):
+        for k in key:
+            api_keyname_to_service[k] = service
+    else:
+        api_keyname_to_service[key] = service
+
+api_id_to_service = {"AWS_ACCESS_KEY_ID": "bedrock"}
+
+
+class KeyLookupBuilder:
+    """Builds KeyLookup options.
+
+    >>> builder = KeyLookupBuilder(fetch_order=("config", "env"))
+    >>> builder.DEFAULT_RPM
+    10
+    >>> builder.DEFAULT_TPM
+    2000000
+    >>> builder.fetch_order
+    ('config', 'env')
+
+    Test invalid fetch_order:
+    >>> try:
+    ...     KeyLookupBuilder(fetch_order=["config", "env"])  # Should be tuple
+    ... except ValueError as e:
+    ...     str(e)
+    'fetch_order must be a tuple'
+
+    Test service extraction:
+    >>> builder.extract_service("EDSL_SERVICE_RPM_OPENAI")
+    ('openai', 'rpm')
+    """
+
+    DEFAULT_RPM = 10
+    DEFAULT_TPM = 2000000
+
+    def __init__(self, fetch_order: Optional[tuple[str]] = None):
+        if fetch_order is None:
+            self.fetch_order = ("config", "env")
+        else:
+            self.fetch_order = fetch_order
+
+        if not isinstance(self.fetch_order, tuple):
+            raise ValueError("fetch_order must be a tuple")
+
+        self.limit_data = {}
+        self.key_data = {}
+        self.id_data = {}
+        self.process_key_value_pairs()
+
+    @property
+    def known_services(self):
+        """Get the set of known services.
+
+        >>> builder = KeyLookupBuilder()
+        >>> isinstance(builder.known_services, set)
+        True
+        """
+        return set(self.key_data.keys()) | set(self.limit_data.keys())
+
+    @lru_cache
+    def build(self) -> "KeyLookup":
+        """Build a KeyLookup instance.
+
+        >>> builder = KeyLookupBuilder()
+        >>> lookup = builder.build()
+        >>> isinstance(lookup, KeyLookup)
+        True
+        >>> lookup['test'].api_token  # Test service should always exist
+        'test'
+        """
+        d = {}
+        for service in self.known_services:
+            try:
+                d[service] = self.get_language_model_input(service)
+            except MissingAPIKeyError:
+                pass
+
+        d.update({"test": LanguageModelInput(api_token="test", rpm=10, tpm=2000000)})
+        return KeyLookup(d)
+
+    def get_language_model_input(self, service: str) -> LanguageModelInput:
+        """Get the language model input for a given service.
+
+        >>> builder = KeyLookupBuilder()
+        >>> try:
+        ...     builder.get_language_model_input("nonexistent_service")
+        ... except MissingAPIKeyError as e:
+        ...     str(e)
+        "No key found for service 'nonexistent_service'"
+        """
+        if (key_entries := self.key_data.get(service)) is None:
+            raise MissingAPIKeyError(f"No key found for service '{service}'")
+
+        if len(key_entries) == 1:
+            api_key_entry = key_entries[0]
+
+        id_entry = self.id_data.get(service)
+        id_source = id_entry.source if id_entry is not None else None
+        api_id = id_entry.value if id_entry is not None else None
+
+        if (limit_entry := self.limit_data.get(service)) is None:
+            limit_entry = LimitEntry(
+                service=service,
+                rpm=self.DEFAULT_RPM,
+                tpm=self.DEFAULT_TPM,
+                source="default",
+            )
+
+        if limit_entry.rpm is None:
+            limit_entry.rpm = self.DEFAULT_RPM
+        if limit_entry.tpm is None:
+            limit_entry.tpm = self.DEFAULT_TPM
+
+        return LanguageModelInput(
+            api_token=api_key_entry.value,
+            rpm=int(limit_entry.rpm),
+            tpm=int(limit_entry.tpm),
+            api_id=api_id,
+            token_source=api_key_entry.source,
+            limit_source=limit_entry.source,
+            id_source=id_source,
+        )
+
+    def __repr__(self):
+        return f"DataSource(key_data={self.key_data}, limit_data={self.limit_data}, id_data={self.id_data})"
+
+    def _os_env_key_value_pairs(self):
+        return dict(list(os.environ.items()))
+
+    def _coop_key_value_pairs(self):
+        from edsl.coop import Coop
+
+        c = Coop()
+        return dict(list(c.fetch_rate_limit_config_vars().items()))
+
+    def _config_key_value_pairs(self):
+        from edsl.config import CONFIG
+
+        return dict(list(CONFIG.items()))
+
+    @staticmethod
+    def extract_service(key: str) -> str:
+        """Extract the service and limit type from the key"""
+        limit_type, service_raw = key.replace("EDSL_SERVICE_", "").split("_")
+        return service_raw.lower(), limit_type.lower()
+
+    def get_key_value_pairs(self) -> dict:
+        """Get key-value pairs from configured sources."""
+        fetching_functions = {
+            "env": self._os_env_key_value_pairs,
+            "coop": self._coop_key_value_pairs,
+            "config": self._config_key_value_pairs,
+        }
+        d = {}
+        for source in self.fetch_order:
+            f = fetching_functions[source]
+            new_data = f()
+            for k, v in new_data.items():
+                d[k] = (v, source)
+        return d
+
+    def _entry_type(self, key, value) -> str:
+        """Determine the type of entry from a key.
+
+        >>> builder = KeyLookupBuilder()
+        >>> builder._entry_type("EDSL_SERVICE_RPM_OPENAI", "60")
+        'limit'
+        >>> builder._entry_type("OPENAI_API_KEY", "sk-1234")
+        'api_key'
+        >>> builder._entry_type("AWS_ACCESS_KEY_ID", "AKIA1234")
+        'api_id'
+        >>> builder._entry_type("UNKNOWN_KEY", "value")
+        'unknown'
+        """
+        if key.startswith("EDSL_SERVICE_"):
+            return "limit"
+        elif key in api_keyname_to_service:
+            return "api_key"
+        elif key in api_id_to_service:
+            return "api_id"
+        return "unknown"
+
+    def _add_id(self, key: str, value: str, source: str) -> None:
+        """Add an API ID to the id_data dictionary.
+
+        >>> builder = KeyLookupBuilder()
+        >>> builder._add_id("AWS_ACCESS_KEY_ID", "AKIA1234", "env")
+        >>> builder.id_data["bedrock"].value
+        'AKIA1234'
+        >>> try:
+        ...     builder._add_id("AWS_ACCESS_KEY_ID", "AKIA5678", "env")
+        ... except ValueError as e:
+        ...     str(e)
+        'Duplicate ID for service bedrock'
+        """
+        service = api_id_to_service[key]
+        if service not in self.id_data:
+            self.id_data[service] = APIIDEntry(
+                service=service, name=key, value=value, source=source
+            )
+        else:
+            raise ValueError(f"Duplicate ID for service {service}")
+
+    def _add_limit(self, key: str, value: str, source: str) -> None:
+        """Add a rate limit entry to the limit_data dictionary.
+
+        >>> builder = KeyLookupBuilder()
+        >>> builder._add_limit("EDSL_SERVICE_RPM_OPENAI", "60", "config")
+        >>> builder.limit_data["openai"].rpm
+        '60'
+        >>> builder._add_limit("EDSL_SERVICE_TPM_OPENAI", "100000", "config")
+        >>> builder.limit_data["openai"].tpm
+        '100000'
+        """
+        service, limit_type = self.extract_service(key)
+        if service in self.limit_data:
+            setattr(self.limit_data[service], limit_type.lower(), value)
+        else:
+            new_limit_entry = LimitEntry(
+                service=service, rpm=None, tpm=None, source=source
+            )
+            setattr(new_limit_entry, limit_type.lower(), value)
+            self.limit_data[service] = new_limit_entry
+
+    def _add_api_key(self, key: str, value: str, source: str) -> None:
+        """Add an API key entry to the key_data dictionary.
+
+        >>> builder = KeyLookupBuilder()
+        >>> builder._add_api_key("OPENAI_API_KEY", "sk-1234", "env")
+        >>> 'sk-1234' == builder.key_data["openai"][-1].value
+        True
+        """
+        service = api_keyname_to_service[key]
+        new_entry = APIKeyEntry(service=service, name=key, value=value, source=source)
+        if service not in self.key_data:
+            self.key_data[service] = [new_entry]
+        else:
+            self.key_data[service].append(new_entry)
+
+    def process_key_value_pairs(self) -> None:
+        """Process all key-value pairs from the configured sources."""
+        for key, value_pair in self.get_key_value_pairs().items():
+            value, source = value_pair
+            if (entry_type := self._entry_type(key, value)) == "limit":
+                self._add_limit(key, value, source)
+            elif entry_type == "api_key":
+                self._add_api_key(key, value, source)
+            elif entry_type == "api_id":
+                self._add_id(key, value, source)

edsl/language_models/key_management/KeyLookupCollection.py

@@ -0,0 +1,38 @@
+from collections import UserDict
+
+from edsl.language_models.key_management.KeyLookupBuilder import KeyLookupBuilder
+
+
+class KeyLookupCollection(UserDict):
+    """A singleton class that stores key-lookup objects.
+
+    This is because once a KeyLook is created once, we do not
+    need to keep re-creating it.
+
+    >>> collection = KeyLookupCollection()
+    >>> collection2 = KeyLookupCollection()
+    >>> collection is collection2  # Test singleton pattern
+    True
+    >>> collection.add_key_lookup(("config", "env"))
+    >>> ("config", "env") in collection.data
+    True
+    """
+
+    _instance = None
+
+    def __new__(cls, *args, **kwargs):
+        if cls._instance is None:
+            cls._instance = super().__new__(cls)
+        return cls._instance
+
+    def __init__(self, *args, **kwargs):
+        if not hasattr(self, "_initialized"):
+            self.data = {}
+            self._initialized = True
+            super().__init__(*args, **kwargs)
+
+    def add_key_lookup(self, fetch_order=None):
+        if fetch_order is None:
+            fetch_order = ("config", "env")
+        if fetch_order not in self.data:
+            self.data[fetch_order] = KeyLookupBuilder(fetch_order=fetch_order).build()