luxorasap 0.1.17__py3-none-any.whl → 0.1.18__py3-none-any.whl

luxorasap/__init__.py

@@ -13,7 +13,7 @@ from types import ModuleType
 try:
     __version__: str = metadata.version(__name__)
 except metadata.PackageNotFoundError:  # editable install
-    __version__ = "0.1.17"
+    __version__ = "0.1.18"
 
 # ─── Lazy loader ─────────────────────────────────────────────────
 def __getattr__(name: str) -> ModuleType:

luxorasap/ingest/cloud/__init__.py

@@ -7,6 +7,7 @@ from luxorasap.utils.storage import BlobParquetClient
 from luxorasap.utils.dataframe import prep_for_save
 from luxorasap.datareader import LuxorQuery
 
+
 __all__ = ["save_table", "incremental_load"]
 
 _client = BlobParquetClient()   # instância única para o módulo
@@ -21,8 +22,7 @@ def save_table(
     index_name: str = "index",
     normalize_columns: bool = True,
     directory: str = "enriched/parquet",
-    override=False,
-    large_df: bool = False
+    override=False
 ):
     """Salva DataFrame como Parquet em ADLS (sobrescrevendo)."""
     
@@ -35,7 +35,10 @@ def save_table(
             return
     
     df = prep_for_save(df, index=index, index_name=index_name, normalize=normalize_columns)
-    _client.write_df(df.astype(str), f"{directory}/{table_name}.parquet", large_df=large_df)
+    
+    #_client.write_df(df.astype(str), f"{directory}/{table_name}.parquet")
+    _client.write_df(df, f"{directory}/{table_name}.parquet")
+
 
 
 def incremental_load(
@@ -47,8 +50,7 @@ def incremental_load(
     index: bool = False,
     index_name: str = "index",
     normalize_columns: bool = True,
-    directory: str = "enriched/parquet",
-    large_df: bool = False
+    directory: str = "enriched/parquet"
 ):
     """Concatena novos dados aos existentes, cortando duplicados pela data."""
     df["Last_Updated"] = dt.datetime.now()
@@ -66,6 +68,5 @@ def incremental_load(
         index_name=index_name,
         normalize_columns=normalize_columns,
         directory=directory,
-        override=True,
-        large_df=large_df
+        override=True
     )

luxorasap/utils/dataframe/__init__.py

@@ -1,4 +1,4 @@
-from .transforms import prep_for_save, persist_column_formatting, text_to_lowercase
+from .transforms import prep_for_save, persist_column_formatting, text_to_lowercase_inplace
 from .reader import read_bytes
 
-__all__ = ["prep_for_save", "persist_column_formatting", "text_to_lowercase", "read_bytes"]
+__all__ = ["prep_for_save", "persist_column_formatting", "text_to_lowercase_inplace", "read_bytes"]

luxorasap/utils/dataframe/transforms.py

@@ -1,42 +1,48 @@
 import pandas as pd
+from pandas.api.types import is_object_dtype, is_string_dtype
 
-def text_to_lowercase(t: pd.DataFrame) -> pd.DataFrame:
+
+def text_to_lowercase_inplace(df: pd.DataFrame, cols: list[str]) -> None:
     """
-    Converte todas as colunas de texto para lowercase
-    Args:
-        t (pd.DataFrame): pandas DataFrame
-    Returns:
-        pd.DataFrame
+    Converte para lower+strip apenas as células que são str.
+    Não tenta aplicar `.str` se a coluna (ou célula) não for string.
+    Opera in-place; não devolve nada.
     """
+    for col in cols:
+        # Precisa ser coluna potencialmente textual
+        if not (is_object_dtype(df[col]) or is_string_dtype(df[col])):
+            continue
 
-    return t.map(lambda x: x.lower().strip() if isinstance(x, str) else x)
+        # Cria máscara com valores realmente str (ignora NaN, ints, decimals…)
+        mask = df[col].apply(lambda x: isinstance(x, str))
 
+        if mask.any():                              # só se houver algo a tratar
+            df.loc[mask, col] = (
+                df.loc[mask, col]
+                  .str.lower()
+                  .str.strip()
+            )
 
-def persist_column_formatting(t: pd.DataFrame, columns_to_persist_override : set = {}) -> pd.DataFrame:
-    """
-    Persiste a formatacao de algumas colunas, e transforma o resto em lowercase
-    Args:
-        t (pd.DataFrame): pandas DataFrame
-    Returns:
-        pd.DataFrame
-    """
 
-    columns_to_persist = {"Name", "Class", "Vehicles", "Segment"}
-    columns_to_persist = columns_to_persist.union(columns_to_persist_override)
-
-    if len(set(t.columns).intersection(columns_to_persist)) > 0:
-        # Vamos persistir a formatacao de algumas colunas
-        columns_order = list(t.columns)
-        columns_to_persist = list(set(t.columns).intersection(columns_to_persist))
-        persistent_data = t[columns_to_persist].copy()
-
-        columns_to_normalize = list(set(columns_order) - set(columns_to_persist))
-        t = text_to_lowercase(t[columns_to_normalize])
-        t.loc[:,columns_to_persist] = persistent_data
-        return t[columns_order]
-    
-    # Nos outros casos, transformaremos tudo em lowercase
-    return text_to_lowercase(t)
+def persist_column_formatting(df: pd.DataFrame,
+        columns_to_persist_override: set | None = None) -> pd.DataFrame:
+    if columns_to_persist_override is None:
+        columns_to_persist_override = set()
+
+    cols_keep_case = {
+        "Name", "Class", "Vehicles", "Segment"
+    }.union(columns_to_persist_override)
+
+    # Só colunas objeto/string candidatas
+    candidate_cols = [
+        c for c in df.columns
+        if c not in cols_keep_case and
+           (df[c].dtype == "object" or pd.api.types.is_string_dtype(df[c]))
+    ]
+
+    text_to_lowercase_inplace(df, candidate_cols)
+
+    return df  # mesma referência; alterações foram in-place
 
 
 def prep_for_save(

luxorasap/utils/storage/blob.py

@@ -4,7 +4,7 @@ from datetime import timezone
 import pandas as pd
 import pyarrow as pa, pyarrow.parquet as pq
 from azure.storage.blob import BlobServiceClient
-import tempfile
+import io
 
 from ..dataframe import read_bytes
 
@@ -34,29 +34,14 @@ class BlobParquetClient:
             return None, False
             
 
-    def write_df(self, df, blob_path: str, large_df: bool = False):
-        if not large_df:
-            table = pa.Table.from_pandas(df)
-            buf = io.BytesIO()
-            pq.write_table(table, buf)
-            buf.seek(0)
-            self._blob(blob_path).upload_blob(buf, overwrite=True)
+    def write_df(self, df, blob_path: str):
         
-        else:
-            with tempfile.NamedTemporaryFile(delete=False, suffix=".parquet") as f:
-                writer = None
-                chunk_size = 100_000
-                for i in range(0, len(df), chunk_size):
-                    chunk = pa.Table.from_pandas(df.iloc[i:i+chunk_size])
-                    if writer is None:
-                        writer = pq.ParquetWriter(f.name, chunk.schema)
-                    writer.write_table(chunk)
-                writer.close()
-                
-                with open(f.name, "rb") as f_read:
-                    self._blob(blob_path).upload_blob(f_read, overwrite=True)
-                
-                os.remove(f.name)
+        blob = self._blob(blob_path)
+        table = pa.Table.from_pandas(df, preserve_index=False)
+        buf = io.BytesIO()
+        pq.write_table(table, buf)
+        buf.seek(0)
+        blob.upload_blob(buf, overwrite=True)
             
         
     def get_df_update_time(self, blob_path: str) -> float:
@@ -102,12 +87,10 @@ class BlobParquetClient:
             Checa se uma tabela existe no blob storage.
         """
         return self.exists_df(table_path)
-        
-            
-            
     
     
     # ---------- interno --------------
     def _blob(self, path: str):
         path = str(PurePosixPath(path))
-        return self._svc.get_blob_client(self._container, path)
+        return self._svc.get_blob_client(self._container, path)
+    

{luxorasap-0.1.17.dist-info → luxorasap-0.1.18.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: luxorasap
-Version: 0.1.17
+Version: 0.1.18
 Summary: Toolbox da Luxor para ingestão, análise e automação de dados financeiros.
 Author-email: Luxor Group <backoffice@luxor.com.br>
 License: Proprietary – All rights reserved

{luxorasap-0.1.17.dist-info → luxorasap-0.1.18.dist-info}/RECORD

@@ -1,4 +1,4 @@
-luxorasap/__init__.py,sha256=aPCQIYh22iaw5xRNSlLfrz65PJNvNMc4NXPrDfeLITs,1356
+luxorasap/__init__.py,sha256=K9mc2r4AggUR7HX2DzwrsIQ2O68lH_nPqSOY3NZo0u4,1356
 luxorasap/btgapi/__init__.py,sha256=QUlfb5oiBY6K1Q5x4-a-x2wECe1At5wc2962I5odOJk,620
 luxorasap/btgapi/auth.py,sha256=PvyCtbEyBO2B1CIeAlNXWugKW1OgiKfPcVzS6K5FBnQ,1872
 luxorasap/btgapi/reports.py,sha256=ZVEMLoJPXc0r3XjPJPMsKQN0zZd1Npd7umNpAj1bncs,8040
@@ -6,16 +6,16 @@ luxorasap/btgapi/trades.py,sha256=956HZ9BvN9C_VQvKTyBLN0x6ZygwVqBZN11F7OnNbDI,59
 luxorasap/datareader/__init__.py,sha256=41RAvbrQ4R6oj67S32CrKqolx0CJ2W8cbOF6g5Cqm2g,120
 luxorasap/datareader/core.py,sha256=P8AjtRFRRmUrqjbjfKRb0wTLW2eHcUva8iWid4uh4PE,155123
 luxorasap/ingest/__init__.py,sha256=XhxDTN2ar-u6UCPhnxNU_to-nWiit-SpQ6cA_N9eMSs,795
-luxorasap/ingest/cloud/__init__.py,sha256=P4GSvfC4JFMbTSpevhfvbfyn-zqiBMIMWhc_U0mQCFc,2153
+luxorasap/ingest/cloud/__init__.py,sha256=CT1lTyr5_kxy9BwWWO0QjxBYzFHNJRcAK8eiGPJdmwM,2121
 luxorasap/ingest/legacy_local/dataloader.py,sha256=zKPhuiBSFwkuWN6d8g2s60KkbVk1R_1cGMCtQM9j-0c,11908
 luxorasap/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-luxorasap/utils/dataframe/__init__.py,sha256=dU_RwTTOi6F3mlhM-0MYWM_qexBN9BmmKc_yrDE1Lwc,207
+luxorasap/utils/dataframe/__init__.py,sha256=wRY4e0rNOp4JlzIxqR_tjwPW8leFMCm7SifzuxLwY0o,223
 luxorasap/utils/dataframe/reader.py,sha256=Vzjdw-AeS1lnWEHQ8RZNh0kK93NWTp0NWVi_B6mN5N0,616
-luxorasap/utils/dataframe/transforms.py,sha256=Bm_cv9L9923QIXH82Fa_M4pM94f2AJRPu62Vv_i7tto,1684
+luxorasap/utils/dataframe/transforms.py,sha256=bqGhMOB-fnJl_SrTT9JTg7eC8oDsqKlKWuO6rnRtGeA,1857
 luxorasap/utils/storage/__init__.py,sha256=U3XRq94yzRp3kgBSUcRzs2tQgJ4o8h8a1ZzwiscA5XM,67
-luxorasap/utils/storage/blob.py,sha256=MVGOXnZR62jJPNWIM8sH_-DF-WuxMFVlpmi1llfos78,3949
-luxorasap-0.1.17.dist-info/METADATA,sha256=aqn2ND5T-_1YUIx1HuRWAxStin3ikanLYS0e9dTJ2F4,3804
-luxorasap-0.1.17.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-luxorasap-0.1.17.dist-info/entry_points.txt,sha256=XFh-dOwUhlya9DmGvgookMI0ezyUJjcOvTIHDEYS44g,52
-luxorasap-0.1.17.dist-info/top_level.txt,sha256=9YOL6bUIpzY06XFBRkUW1e4rgB32Ds91fQPGwUEjxzU,10
-luxorasap-0.1.17.dist-info/RECORD,,
+luxorasap/utils/storage/blob.py,sha256=hy18amzBqJtJqKGTaO74tHuy4_7FfvyN83yIC_Dnz7g,3206
+luxorasap-0.1.18.dist-info/METADATA,sha256=c2ioyv688f7TbbRGkh3hJXFNn18JqrYKzOc8FsEKF0c,3804
+luxorasap-0.1.18.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+luxorasap-0.1.18.dist-info/entry_points.txt,sha256=XFh-dOwUhlya9DmGvgookMI0ezyUJjcOvTIHDEYS44g,52
+luxorasap-0.1.18.dist-info/top_level.txt,sha256=9YOL6bUIpzY06XFBRkUW1e4rgB32Ds91fQPGwUEjxzU,10
+luxorasap-0.1.18.dist-info/RECORD,,