upgini 1.1.203a1__tar.gz → 1.1.205__tar.gz

{upgini-1.1.203a1/src/upgini.egg-info → upgini-1.1.205}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: upgini
-Version: 1.1.203a1
+Version: 1.1.205
 Summary: Intelligent data search & enrichment for Machine Learning
 Home-page: https://upgini.com/
 Author: Upgini Developers

{upgini-1.1.203a1 → upgini-1.1.205}/setup.py

@@ -40,7 +40,7 @@ def send_log(msg: str):
 
 
 here = Path(__file__).parent.resolve()
-version = "1.1.203a1"
+version = "1.1.205"
 try:
     send_log(f"Start setup PyLib version {version}")
     setup(

{upgini-1.1.203a1 → upgini-1.1.205}/src/upgini/dataset.py

@@ -80,6 +80,7 @@ class Dataset:  # (pd.DataFrame):
         endpoint: Optional[str] = None,
         api_key: Optional[str] = None,
         logger: Optional[logging.Logger] = None,
+        client_ip: Optional[str] = None,
         warning_counter: Optional[WarningCounter] = None,
         **kwargs,
     ):
@@ -123,6 +124,7 @@ class Dataset:  # (pd.DataFrame):
         else:
             self.logger = logging.getLogger()
             self.logger.setLevel("FATAL")
+        self.client_ip = client_ip
         self.warning_counter = warning_counter or WarningCounter()
 
     def __len__(self):
@@ -961,6 +963,7 @@ class Dataset:  # (pd.DataFrame):
             task_type=self.task_type,
             endpoint=self.endpoint,
             api_key=self.api_key,
+            client_ip=self.client_ip,
         )
 
     def validation(
@@ -1030,6 +1033,7 @@ class Dataset:  # (pd.DataFrame):
             initial_search_task_id=initial_search_task_id,
             endpoint=self.endpoint,
             api_key=self.api_key,
+            client_ip=self.client_ip,
         )
 
     def prepare_uploading_file(self, base_path: str) -> str:

{upgini-1.1.203a1 → upgini-1.1.205}/src/upgini/features_enricher.py

@@ -168,16 +168,18 @@ class FeaturesEnricher(TransformerMixin):
         logs_enabled: bool = True,
         raise_validation_error: bool = True,
         exclude_columns: Optional[List[str]] = None,
+        client_ip: Optional[str] = None,
         **kwargs,
     ):
         self._api_key = api_key or os.environ.get(UPGINI_API_KEY)
         if api_key is not None and not isinstance(api_key, str):
             raise ValidationError(f"api_key should be `string`, but passed: `{api_key}`")
         self.rest_client = get_rest_client(endpoint, self._api_key)
+        self.client_ip = client_ip
 
         self.logs_enabled = logs_enabled
         if logs_enabled:
-            self.logger = LoggerFactory().get_logger(endpoint, self._api_key)
+            self.logger = LoggerFactory().get_logger(endpoint, self._api_key, client_ip)
         else:
             self.logger = logging.getLogger()
             self.logger.setLevel("FATAL")
@@ -222,6 +224,7 @@ class FeaturesEnricher(TransformerMixin):
                 search_id,
                 endpoint=self.endpoint,
                 api_key=self._api_key,
+                client_ip=client_ip
             )
 
             print(bundle.get("search_by_task_id_start"))
@@ -281,7 +284,7 @@ class FeaturesEnricher(TransformerMixin):
     def _set_api_key(self, api_key: str):
         self._api_key = api_key
         if self.logs_enabled:
-            self.logger = LoggerFactory().get_logger(self.endpoint, self._api_key)
+            self.logger = LoggerFactory().get_logger(self.endpoint, self._api_key, self.client_ip)
 
     api_key = property(_get_api_key, _set_api_key)
 
@@ -1584,6 +1587,7 @@ class FeaturesEnricher(TransformerMixin):
                 api_key=self.api_key,  # type: ignore
                 date_format=self.date_format,  # type: ignore
                 logger=self.logger,
+                client_ip=self.client_ip
             )
             dataset.meaning_types = meaning_types
             dataset.search_keys = combined_search_keys
@@ -1917,6 +1921,7 @@ class FeaturesEnricher(TransformerMixin):
             date_format=self.date_format,  # type: ignore
             random_state=self.random_state,  # type: ignore
             logger=self.logger,
+            client_ip=self.client_ip,
         )
         dataset.meaning_types = meaning_types
         dataset.search_keys = combined_search_keys

{upgini-1.1.203a1 → upgini-1.1.205}/src/upgini/http.py

@@ -1,4 +1,6 @@
 import base64
+import copy
+import datetime
 import hashlib
 import logging
 import os
@@ -12,8 +14,6 @@ from http.client import HTTPConnection
 from json import dumps
 from typing import Any, Dict, List, Optional, Tuple
 from urllib.parse import urljoin
-import datetime
-import copy
 
 import jwt
 import pandas as pd
@@ -882,17 +882,18 @@ def _get_rest_client(backend_url: str, api_token: str) -> _RestClient:
 
 
 class BackendLogHandler(logging.Handler):
-    def __init__(self, rest_client: _RestClient, *args, **kwargs) -> None:
+    def __init__(self, rest_client: _RestClient, client_ip: Optional[str] = None, *args, **kwargs) -> None:
         super().__init__(*args, **kwargs)
         self.rest_client = rest_client
         self.track_metrics = None
         self.hostname = "0.0.0.0"
+        self.client_ip = client_ip
 
     def emit(self, record: logging.LogRecord) -> None:
         def task():
             try:
                 if self.track_metrics is None or len(self.track_metrics) == 0:
-                    self.track_metrics = get_track_metrics()
+                    self.track_metrics = get_track_metrics(self.client_ip)
                     self.hostname = self.track_metrics.get("ip") or "0.0.0.0"
                 text = self.format(record)
                 tags = self.track_metrics
@@ -933,7 +934,9 @@ class LoggerFactory:
         root.setLevel(logging.INFO)
         root.handlers.clear()
 
-    def get_logger(self, backend_url: Optional[str] = None, api_token: Optional[str] = None) -> logging.Logger:
+    def get_logger(
+        self, backend_url: Optional[str] = None, api_token: Optional[str] = None, client_ip: Optional[str] = None
+    ) -> logging.Logger:
         url = _resolve_backend_url(backend_url)
         token = resolve_api_token(api_token)
         key = url + token
@@ -944,7 +947,7 @@ class LoggerFactory:
         upgini_logger = logging.getLogger(f"upgini.{hash(key)}")
         upgini_logger.handlers.clear()
         rest_client = get_rest_client(backend_url, api_token)
-        datadog_handler = BackendLogHandler(rest_client)
+        datadog_handler = BackendLogHandler(rest_client, client_ip)
         json_formatter = jsonlogger.JsonFormatter(
             "%(asctime)s %(threadName)s %(name)s %(levelname)s %(message)s",
             timestamp=True,

{upgini-1.1.203a1 → upgini-1.1.205}/src/upgini/search_task.py

@@ -42,6 +42,7 @@ class SearchTask:
         task_type: Optional[ModelTaskType] = None,
         endpoint: Optional[str] = None,
         api_key: Optional[str] = None,
+        client_ip: Optional[str] = None,
     ):
         self.search_task_id = search_task_id
         self.initial_search_task_id = initial_search_task_id
@@ -53,7 +54,7 @@ class SearchTask:
         self.summary = None
         self.endpoint = endpoint
         self.api_key = api_key
-        self.logger = LoggerFactory().get_logger(endpoint, api_key)
+        self.logger = LoggerFactory().get_logger(endpoint, api_key, client_ip)
         self.provider_metadata_v2: Optional[List[ProviderTaskMetadataV2]] = None
         self.unused_features_for_generation: Optional[List[str]] = None
 

{upgini-1.1.203a1 → upgini-1.1.205}/src/upgini/utils/display_utils.py

@@ -135,9 +135,10 @@ def make_html_report(
 ):
     relevant_datasources_df = relevant_datasources_df.copy()
     relevant_datasources_df["action"] = (
-        f"""<a href="https://upgini.com/requet-a-quote?search-id={search_id}"><button>Request a quote</button></a>"""
+        f"""<a href="https://upgini.com/requet-a-quote?search-id={search_id}">"""
+        """<button type="button">Request a quote</button></a>"""
     )
-    relevant_datasources_df.rename(columns={"action": ""})
+    relevant_datasources_df.rename(columns={"action": "&nbsp;"}, inplace=True)
     return f"""<html>
         <head>
             <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />

{upgini-1.1.203a1 → upgini-1.1.205}/src/upgini/utils/track_info.py

@@ -5,6 +5,7 @@ import socket
 import sys
 from getpass import getuser
 from hashlib import sha256
+from typing import Optional
 from uuid import getnode
 
 from requests import get, post
@@ -46,7 +47,7 @@ def _get_execution_ide() -> str:
 
 
 @lru_cache()
-def get_track_metrics() -> dict:
+def get_track_metrics(client_ip: Optional[str]) -> dict:
     # default values
     track = {"ide": _get_execution_ide()}
     ident_res = "https://api.ipify.org"
@@ -93,28 +94,34 @@ def get_track_metrics() -> dict:
         except Exception as e:
             track["err"] = str(e)
             track["visitorId"] = "None"
-        try:
-            from google.colab import output  # type: ignore
-            from IPython.display import Javascript, display
-
-            display(
-                Javascript(
-                    f"""
-                        window.clientIP = fetch("{ident_res}")
-                            .then(response => response.text())
-                            .then(data => data);
-                    """
+        if client_ip:
+            track["ip"] = client_ip
+        else:
+            try:
+                from google.colab import output  # type: ignore
+                from IPython.display import Javascript, display
+
+                display(
+                    Javascript(
+                        f"""
+                            window.clientIP = fetch("{ident_res}")
+                                .then(response => response.text())
+                                .then(data => data);
+                        """
+                    )
                 )
-            )
-            track["ip"] = output.eval_js("window.clientIP", timeout_sec=10)
-        except Exception as e:
-            track["err"] = str(e)
-            track["ip"] = "0.0.0.0"
+                track["ip"] = output.eval_js("window.clientIP", timeout_sec=10)
+            except Exception as e:
+                track["err"] = str(e)
+                track["ip"] = "0.0.0.0"
 
     elif track["ide"] == "binder":
         try:
             if "CLIENT_IP" in os.environ.keys():
-                track["ip"] = os.environ["CLIENT_IP"]
+                if client_ip:
+                    track["ip"] = client_ip
+                else:
+                    track["ip"] = os.environ["CLIENT_IP"]
                 track["visitorId"] = sha256(os.environ["CLIENT_IP"].encode()).hexdigest()
         except Exception as e:
             track["err"] = str(e)
@@ -142,7 +149,10 @@ def get_track_metrics() -> dict:
             track["visitorId"] = "None"
     else:
         try:
-            track["ip"] = get(ident_res, timeout=10).text
+            if client_ip:
+                track["ip"] = client_ip
+            else:
+                track["ip"] = get(ident_res, timeout=10).text
             track["visitorId"] = sha256(str(getnode()).encode()).hexdigest()
         except Exception as e:
             track["err"] = str(e)

{upgini-1.1.203a1 → upgini-1.1.205/src/upgini.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: upgini
-Version: 1.1.203a1
+Version: 1.1.205
 Summary: Intelligent data search & enrichment for Machine Learning
 Home-page: https://upgini.com/
 Author: Upgini Developers