csv-detective 0.7.5.dev1056__py3-none-any.whl → 0.7.5.dev1069__py3-none-any.whl

csv_detective/explore_csv.py

@@ -12,6 +12,7 @@ import logging
 from time import time
 import requests
 from io import StringIO
+import pandas as pd
 
 # flake8: noqa
 from csv_detective import detect_fields, detect_labels
@@ -108,9 +109,10 @@ def routine(
     skipna: bool = True,
     output_profile: bool = False,
     output_schema: bool = False,
+    output_df: bool = False,
     verbose: bool = False,
     sheet_name: Union[str, int] = None,
-):
+) -> Union[dict, tuple[dict, pd.DataFrame]]:
     """Returns a dict with information about the csv table and possible
     column contents.
 
@@ -123,6 +125,7 @@ def routine(
         save_results: whether or not to save the results in a json file, or the path where to dump the output
         output_profile: whether or not to add the 'profile' field to the output
         output_schema: whether or not to add the 'schema' field to the output (tableschema)
+        output_df: whether or not to return the loaded DataFrame along with the analysis report
         verbose: whether or not to print process logs in console 
         sheet_name: if reading multi-sheet file (xls-like), which sheet to consider
         skipna: whether to keep NaN (empty cells) for tests
@@ -353,6 +356,8 @@ def routine(
             f'Routine completed in {round(time() - start_routine, 3)}s',
             time() - start_routine
         )
+    if output_df:
+        return return_dict, table
     return return_dict
 
 

{csv_detective-0.7.5.dev1056.data → csv_detective-0.7.5.dev1069.data}/data/share/csv_detective/CHANGELOG.md

@@ -3,6 +3,7 @@
 ## Current (in progress)
 
 - New function that creates a csv from a list of fields and constraints, or from a TableSchema [#101](https://github.com/datagouv/csv-detective/pull/101)
+- Enable outputing loaded dataframe [#102](https://github.com/datagouv/csv-detective/pull/102)
 
 ## 0.7.4 (2024-11-15)
 

{csv_detective-0.7.5.dev1056.dist-info → csv_detective-0.7.5.dev1069.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: csv_detective
-Version: 0.7.5.dev1056
+Version: 0.7.5.dev1069
 Summary: Detect CSV column content
 Home-page: https://github.com/etalab/csv_detective
 Author: Etalab
@@ -29,4 +29,3 @@ Requires-Dist: python-magic==0.4.27
 Requires-Dist: frformat==0.4.0
 Requires-Dist: faker==33.0.0
 Requires-Dist: rstr==3.2.2
-

{csv_detective-0.7.5.dev1056.dist-info → csv_detective-0.7.5.dev1069.dist-info}/RECORD

@@ -2,7 +2,7 @@ csv_detective/__init__.py,sha256=Au4bNJ_Gi6P6o0uO4R56nYdshG7M6-7Rg_xX4whLmLI,143
 csv_detective/cli.py,sha256=Ua7SE1wMH2uFUsTmfumh4nJk7O06okpMd2gvjUDO1II,1048
 csv_detective/create_example.py,sha256=358e7Q7RWMrY_eEo3pUteJWmg2smFb5edJ_AzcQPrqA,8646
 csv_detective/detection.py,sha256=AuXlPOZfzqznZY2ybAAgaXIq6qVITYd3MXf2CoigI3I,22097
-csv_detective/explore_csv.py,sha256=X5yZS3WCUsafUMcs5tOnDTeMGzMnfr0iB9vEDx7xiqg,16977
+csv_detective/explore_csv.py,sha256=6kGl1E061_CefAdei-wgwafZT1g8oKWg0eE1D5zWTOk,17216
 csv_detective/process_text.py,sha256=rsfk66BCmdpsCOd0kDJ8tmqMsEWd-OeBkEisWc4Ej9k,1246
 csv_detective/s3_utils.py,sha256=1cIVdQUYY2ovErbMwp72Gqtqx2bkB8nfVhn-QaOFTT0,1451
 csv_detective/schema_generation.py,sha256=D1Cq4QRajsKtY8EJSwbRTIB-T_Cb2ZpcmYtCrJ6DvJQ,13135
@@ -126,18 +126,18 @@ csv_detective/detect_labels/temp/date/__init__.py,sha256=GrIbo64WVM3hi7ShBRKKyKU
 csv_detective/detect_labels/temp/datetime_iso/__init__.py,sha256=Ih9l56nBcdmGLyWDavVUWuUUuVZBz9QUDE1hHzADvVg,1157
 csv_detective/detect_labels/temp/datetime_rfc822/__init__.py,sha256=DQ_h4uDW1e6qu2rATEhgGKw6O-vVi7HbDhbEDDCT9uY,1175
 csv_detective/detect_labels/temp/year/__init__.py,sha256=zPF_mvhzhXMAlHPAskS8mhuxjLj2AlKpV4ss8Q4tDms,1150
-csv_detective-0.7.5.dev1056.data/data/share/csv_detective/CHANGELOG.md,sha256=clfZIuAoLOi34VOevzdE4_bR__6SVWiNNmWOFOqpm-k,6725
-csv_detective-0.7.5.dev1056.data/data/share/csv_detective/LICENSE.AGPL.txt,sha256=2N5ReRelkdqkR9a-KP-y-shmcD5P62XoYiG-miLTAzo,34519
-csv_detective-0.7.5.dev1056.data/data/share/csv_detective/README.md,sha256=Qr8xRXc-dxQ-tdXCpCTCKp1Uliqq84r0UOlPRNuGCpI,9506
+csv_detective-0.7.5.dev1069.data/data/share/csv_detective/CHANGELOG.md,sha256=QbZKEEWbkt7a-TMHB6CpzzliDqv3BLECa_zkJgZOFkY,6820
+csv_detective-0.7.5.dev1069.data/data/share/csv_detective/LICENSE.AGPL.txt,sha256=2N5ReRelkdqkR9a-KP-y-shmcD5P62XoYiG-miLTAzo,34519
+csv_detective-0.7.5.dev1069.data/data/share/csv_detective/README.md,sha256=Qr8xRXc-dxQ-tdXCpCTCKp1Uliqq84r0UOlPRNuGCpI,9506
 tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tests/test_example.py,sha256=0NfChooJQlFxTo2nY5FOQIcsK4zzWA_SBmt2LwVQovY,2014
 tests/test_fields.py,sha256=kXel-hiyQYrJ3OLmwUMg1K3DKbbwBLvUplxZWxpp18I,10605
-tests/test_file.py,sha256=1fEOu3bArGBaarRKAoTXAF3cSIGJfFN3UIwOW6esWRs,6399
+tests/test_file.py,sha256=oQITvAxdcrqDby2wWSh_X9TCwFqdFaP34XNy92ibXyg,6725
 tests/test_labels.py,sha256=6MOKrGznkwU5fjZ_3oiB6Scmb480Eu-9geBJs0UDLds,159
 tests/test_structure.py,sha256=SVsnluVoIIprYw_67I1_gB3cp9m1wlO8C7SpdsLW8cM,1161
-csv_detective-0.7.5.dev1056.dist-info/LICENSE.AGPL.txt,sha256=2N5ReRelkdqkR9a-KP-y-shmcD5P62XoYiG-miLTAzo,34519
-csv_detective-0.7.5.dev1056.dist-info/METADATA,sha256=oNvlYf9ZGV54EfE_2tGMTHzh1jOTPaMwGt1uYyjU0BM,1146
-csv_detective-0.7.5.dev1056.dist-info/WHEEL,sha256=R06PA3UVYHThwHvxuRWMqaGcr-PuniXahwjmQRFMEkY,91
-csv_detective-0.7.5.dev1056.dist-info/entry_points.txt,sha256=JjweTReFqKJmuvkegzlew2j3D5pZzfxvbEGOtGVGmaY,56
-csv_detective-0.7.5.dev1056.dist-info/top_level.txt,sha256=M0Nv646VHo-49zWjPkwo2C48UmtfddV8_9mEZeIxy8Q,20
-csv_detective-0.7.5.dev1056.dist-info/RECORD,,
+csv_detective-0.7.5.dev1069.dist-info/LICENSE.AGPL.txt,sha256=2N5ReRelkdqkR9a-KP-y-shmcD5P62XoYiG-miLTAzo,34519
+csv_detective-0.7.5.dev1069.dist-info/METADATA,sha256=sqa9hWFoiOj9-MpBX1uuwOl5qyPCSoca3wo0RrglmNY,1145
+csv_detective-0.7.5.dev1069.dist-info/WHEEL,sha256=PZUExdf71Ui_so67QXpySuHtCi3-J3wvF4ORK6k_S8U,91
+csv_detective-0.7.5.dev1069.dist-info/entry_points.txt,sha256=JjweTReFqKJmuvkegzlew2j3D5pZzfxvbEGOtGVGmaY,56
+csv_detective-0.7.5.dev1069.dist-info/top_level.txt,sha256=M0Nv646VHo-49zWjPkwo2C48UmtfddV8_9mEZeIxy8Q,20
+csv_detective-0.7.5.dev1069.dist-info/RECORD,,

{csv_detective-0.7.5.dev1056.dist-info → csv_detective-0.7.5.dev1069.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.5.0)
+Generator: setuptools (75.6.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

tests/test_file.py

@@ -1,6 +1,7 @@
 from csv_detective import routine
 import pytest
 import responses
+import pandas as pd
 
 
 def test_columns_output_on_file():
@@ -218,3 +219,16 @@ def test_nan_values(expected_type):
         skipna=skipna,
     )
     assert output["columns"]["partly_empty"]["python_type"] == expected_type
+
+
+def test_output_df():
+    output, df = routine(
+        csv_file_path="tests/b_test_file.csv",
+        num_rows=-1,
+        output_profile=False,
+        save_results=False,
+        output_df=True,
+    )
+    assert isinstance(output, dict)
+    assert isinstance(df, pd.DataFrame)
+    assert len(df) == 6