csv-detective 0.9.3.dev2258__py3-none-any.whl → 0.9.3.dev2348__py3-none-any.whl

csv_detective/detection/formats.py

@@ -7,7 +7,7 @@ from csv_detective.detection.variables import (
     detect_categorical_variable,
     # detect_continuous_variable,
 )
-from csv_detective.load_tests import return_all_tests
+from csv_detective.format import Format, FormatsManager
 from csv_detective.output.utils import prepare_output_dict
 from csv_detective.parsing.columns import (
     MAX_NUMBER_CATEGORICAL_VALUES,
@@ -16,12 +16,14 @@ from csv_detective.parsing.columns import (
     test_label,
 )
 
+fmtm = FormatsManager()
+
 
 def detect_formats(
     table: pd.DataFrame,
     analysis: dict,
     file_path: str,
-    user_input_tests: str | list[str] = "ALL",
+    tags: list[str] | None = None,
     limited_output: bool = True,
     skipna: bool = True,
     verbose: bool = False,
@@ -29,15 +31,12 @@ def detect_formats(
     in_chunks = analysis.get("total_lines") is None
 
     # list testing to be performed
-    all_tests_fields = return_all_tests(
-        user_input_tests, detect_type="detect_fields"
-    )  # list all tests for the fields
-    all_tests_labels = return_all_tests(
-        user_input_tests, detect_type="detect_labels"
-    )  # list all tests for the labels
+    formats: dict[str, Format] = (
+        fmtm.get_formats_from_tags(tags) if tags is not None else fmtm.formats
+    )
 
     # if no testing then return
-    if not all_tests_fields and not all_tests_labels:
+    if len(formats) == 0:
         return analysis, None
 
     # Perform testing on fields
@@ -45,7 +44,7 @@ def detect_formats(
         # table is small enough to be tested in one go
         scores_table_fields = test_col(
             table=table,
-            all_tests=all_tests_fields,
+            formats=formats,
             limited_output=limited_output,
             skipna=skipna,
             verbose=verbose,
@@ -62,7 +61,7 @@ def detect_formats(
             table=table,
             file_path=file_path,
             analysis=analysis,
-            all_tests=all_tests_fields,
+            formats=formats,
             limited_output=limited_output,
             skipna=skipna,
             verbose=verbose,
@@ -70,9 +69,7 @@ def detect_formats(
     analysis["columns_fields"] = prepare_output_dict(scores_table_fields, limited_output)
 
     # Perform testing on labels
-    scores_table_labels = test_label(
-        analysis["header"], all_tests_labels, limited_output, verbose=verbose
-    )
+    scores_table_labels = test_label(analysis["header"], formats, limited_output, verbose=verbose)
     analysis["columns_labels"] = prepare_output_dict(scores_table_labels, limited_output)
 
     # Multiply the results of the fields by 1 + 0.5 * the results of the labels.
@@ -115,7 +112,7 @@ def detect_formats(
         "float": "float",
         "string": "string",
         "json": "json",
-        "json_geojson": "json",
+        "geojson": "json",
         "datetime_aware": "datetime",
         "datetime_naive": "datetime",
         "datetime_rfc822": "datetime",

csv_detective/explore_csv.py

@@ -15,7 +15,7 @@ logging.basicConfig(level=logging.INFO)
 def routine(
     file_path: str,
     num_rows: int = 500,
-    user_input_tests: str | list[str] = "ALL",
+    tags: list[str] | None = None,
     limited_output: bool = True,
     save_results: bool | str = True,
     encoding: str | None = None,
@@ -28,14 +28,13 @@ def routine(
     verbose: bool = False,
     sheet_name: str | int | None = None,
 ) -> dict | tuple[dict, pd.DataFrame]:
-    """Returns a dict with information about the table and possible
-    column contents, and if requested the DataFrame with columns cast according to analysis.
+    """
+    Returns a dict with information about the table and possible column contents, and if requested the DataFrame with columns cast according to analysis.
 
     Args:
         file_path: local path or URL to file
-        num_rows: number of rows to sample from the file for analysis ; -1 for analysis
-        of the whole file
-        user_input_tests: tests to run on the file
+        num_rows: number of rows to sample from the file for analysis ; -1 for analysis of the whole file
+        tags: tags to filter formats (for instance ["geo", "fr] to run only the checks related to geo and French formats)
         limited_output: whether or not to return all possible types or only the most likely one for each column
         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
@@ -74,7 +73,7 @@ def routine(
         table=table,
         analysis=analysis,
         file_path=file_path,
-        user_input_tests=user_input_tests,
+        tags=tags,
         limited_output=limited_output,
         skipna=skipna,
         verbose=verbose,
@@ -107,7 +106,7 @@ def validate_then_detect(
     file_path: str,
     previous_analysis: dict,
     num_rows: int = 500,
-    user_input_tests: str | list[str] = "ALL",
+    tags: list[str] | None = None,
     limited_output: bool = True,
     save_results: bool | str = True,
     skipna: bool = True,
@@ -117,6 +116,26 @@ def validate_then_detect(
     cast_json: bool = True,
     verbose: bool = False,
 ):
+    """
+    Performs a validation of the given file against the given analysis.
+    If the validation fails, performs a full analysis and return it.
+    Otherwise return the previous analysis (which is therefore still valid).
+    NB: if asked, the profile is recreated in both cases.
+
+    Args:
+        file_path: the path of the file to validate.
+        previous_analysis: the previous analysis to validate against (expected in the same structure as the output of the routine)
+        num_rows: number of rows to sample from the file for analysis ; -1 for analysis of the whole file
+        tags: tags to filter formats (for instance ["geo", "fr] to run only the checks related to geo and French formats)
+        limited_output: whether or not to return all possible types or only the most likely one for each column
+        save_results: whether or not to save the results in a json file, or the path where to dump the output
+        skipna: whether to ignore NaN values in the checks
+        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
+        cast_json: whether or not to cast json columns into objects (otherwise they are returned as strings)
+        verbose: whether the code displays the steps it's going through
+    """
     if verbose:
         start_routine = time()
         if is_url(file_path):
@@ -140,7 +159,7 @@ def validate_then_detect(
             table=table,
             analysis=analysis,
             file_path=file_path,
-            user_input_tests=user_input_tests,
+            tags=tags,
             limited_output=limited_output,
             skipna=skipna,
             verbose=verbose,

csv_detective/format.py

@@ -0,0 +1,67 @@
+from typing import Any, Callable
+
+from csv_detective.parsing.text import header_score
+
+
+class Format:
+    def __init__(
+        self,
+        name: str,
+        func: Callable[[Any], bool],
+        _test_values: dict[bool, list[str]],
+        labels: list[str] = [],
+        proportion: float = 1,
+        tags: list[str] = [],
+    ) -> None:
+        """
+        Instanciates a Format object.
+
+        Args:
+            name: the name of the format.
+            func: the value test for the format (returns whether a string is valid).
+            _test_values: lists of valid and invalid values, used in the tests
+            labels: the list of hint headers for the header score
+            proportion: the tolerance (between 0 and 1) to say a column is valid for a format. (1 => 100% of the column has to pass the func check for the column to be considered valid)
+            tags: to allow users to submit a file to only a subset of formats
+        """
+        self.name: str = name
+        self.func: Callable = func
+        self._test_values: dict[bool, list[str]] = _test_values
+        self.labels: list[str] = labels
+        self.proportion: float = proportion
+        self.tags: list[str] = tags
+
+    def is_valid_label(self, val: str) -> float:
+        return header_score(val, self.labels)
+
+
+class FormatsManager:
+    formats: dict[str, Format]
+
+    def __init__(self) -> None:
+        import csv_detective.formats as formats
+
+        format_labels = [f for f in dir(formats) if "_is" in dir(getattr(formats, f))]
+        self.formats = {
+            label: Format(
+                name=label,
+                func=(module := getattr(formats, label))._is,
+                _test_values=module._test_values,
+                **{
+                    attr: val
+                    for attr in ["labels", "proportion", "tags"]
+                    if (val := getattr(module, attr, None))
+                },
+            )
+            for label in format_labels
+        }
+
+    def get_formats_from_tags(self, tags: list[str]) -> dict[str, Format]:
+        return {
+            label: fmt
+            for label, fmt in self.formats.items()
+            if all(tag in fmt.tags for tag in tags)
+        }
+
+    def available_tags(self) -> set[str]:
+        return set(tag for format in self.formats.values() for tag in format.tags)

csv_detective/formats/__init__.py

@@ -0,0 +1,9 @@
+import importlib
+import os
+
+for file in os.listdir(os.path.dirname(__file__)):
+    if file.endswith(".py") and not file.startswith("_"):
+        module_name = file[:-3]
+        module = importlib.import_module(f"csv_detective.formats.{module_name}")
+        globals()[module_name] = module
+        del module

csv_detective/{detect_fields/FR/geo/adresse/__init__.py → formats/adresse.py}

@@ -1,100 +1,116 @@
-from csv_detective.parsing.text import _process_text
-
-PROPORTION = 0.55
-# ajouts d'espaces en fin de mots pour s'assurer que le str n'est pas juste une substr d'un mot plus long
-voies = {
-    "aire ",
-    "allee ",
-    "avenue ",
-    "base ",
-    "boulevard ",
-    "cami ",
-    "carrefour ",
-    "chemin ",
-    "cheminement ",
-    "chaussee ",
-    "cite ",
-    "clos ",
-    "coin ",
-    "corniche ",
-    "cote ",
-    "cour ",
-    "cours ",
-    "domaine ",
-    "descente ",
-    "ecart ",
-    "esplanade ",
-    "faubourg ",
-    "gare ",
-    "grande rue",
-    "hameau ",
-    "halle ",
-    "ilot ",
-    "impasse ",
-    "lieu dit",
-    "lotissement ",
-    "marche ",
-    "montee ",
-    "parc ",
-    "passage ",
-    "place ",
-    "plan ",
-    "plaine ",
-    "plateau ",
-    "pont ",
-    "port ",
-    "promenade ",
-    "parvis ",
-    "quartier ",
-    "quai ",
-    "residence ",
-    "ruelle ",
-    "rocade ",
-    "rond point",
-    "route ",
-    "rue ",
-    # 'sente - sentier',
-    "square ",
-    "tour ",
-    # 'terre-plein',
-    "traverse ",
-    "villa ",
-    "village ",
-    "voie ",
-    "zone artisanale",
-    "zone d’amenagement concerte",
-    "zone d’amenagement differe",
-    "zone industrielle",
-    "zone ",
-    # 'r',
-    "av ",
-    "pl ",
-    "bd ",
-    "cami ",
-    # 'che',
-    "chs ",
-    "dom ",
-    "ham ",
-    "ld ",
-    # 'pro',
-    # 'rte',
-    "vlge ",
-    "za ",
-    "zac ",
-    "zad ",
-    "zi ",
-    # 'car',
-    "fg ",
-    # 'lot',
-    "imp ",
-    # 'qu',
-    "mte",
-}
-
-
-def _is(val):
-    """Repere des adresses"""
-    if not isinstance(val, str) or len(val) > 150:
-        return False
-    val = _process_text(val)
-    return any(x in val for x in voies)
+from csv_detective.parsing.text import _process_text
+
+proportion = 0.55
+tags = ["fr", "geo"]
+labels = [
+    "adresse",
+    "localisation",
+    "adresse postale",
+    "adresse geographique",
+    "adr",
+    "adresse complete",
+    "adresse station",
+]
+
+voies = {
+    "aire ",
+    "allee ",
+    "avenue ",
+    "base ",
+    "boulevard ",
+    "cami ",
+    "carrefour ",
+    "chemin ",
+    "cheminement ",
+    "chaussee ",
+    "cite ",
+    "clos ",
+    "coin ",
+    "corniche ",
+    "cote ",
+    "cour ",
+    "cours ",
+    "domaine ",
+    "descente ",
+    "ecart ",
+    "esplanade ",
+    "faubourg ",
+    "gare ",
+    "grande rue",
+    "hameau ",
+    "halle ",
+    "ilot ",
+    "impasse ",
+    "lieu dit",
+    "lotissement ",
+    "marche ",
+    "montee ",
+    "parc ",
+    "passage ",
+    "place ",
+    "plan ",
+    "plaine ",
+    "plateau ",
+    "pont ",
+    "port ",
+    "promenade ",
+    "parvis ",
+    "quartier ",
+    "quai ",
+    "residence ",
+    "ruelle ",
+    "rocade ",
+    "rond point",
+    "route ",
+    "rue ",
+    # 'sente - sentier',
+    "square ",
+    "tour ",
+    # 'terre-plein',
+    "traverse ",
+    "villa ",
+    "village ",
+    "voie ",
+    "zone artisanale",
+    "zone d’amenagement concerte",
+    "zone d’amenagement differe",
+    "zone industrielle",
+    "zone ",
+    # 'r',
+    "av ",
+    "pl ",
+    "bd ",
+    "cami ",
+    # 'che',
+    "chs ",
+    "dom ",
+    "ham ",
+    "ld ",
+    # 'pro',
+    # 'rte',
+    "vlge ",
+    "za ",
+    "zac ",
+    "zad ",
+    "zi ",
+    # 'car',
+    "fg ",
+    # 'lot',
+    "imp ",
+    # 'qu',
+    "mte",
+}
+
+
+def _is(val):
+    """Repere des adresses"""
+    if not isinstance(val, str) or len(val) > 150:
+        return False
+    val = _process_text(val)
+    return any(x in val for x in voies)
+
+
+_test_values = {
+    True: ["rue du martyr"],
+    False: ["un batiment"],
+}

csv_detective/{detect_fields/other/booleen/__init__.py → formats/booleen.py}

@@ -1,27 +1,35 @@
-PROPORTION = 1
-bool_mapping = {
-    "1": True,
-    "0": False,
-    "vrai": True,
-    "faux": False,
-    "true": True,
-    "false": False,
-    "oui": True,
-    "non": False,
-    "yes": True,
-    "no": False,
-    "y": True,
-    "n": False,
-    "o": True,
-}
-
-liste_bool = set(bool_mapping.keys())
-
-
-def bool_casting(val: str) -> bool:
-    return bool_mapping.get(val.lower())
-
-
-def _is(val: str) -> bool:
-    """Détecte les booléens"""
-    return isinstance(val, str) and val.lower() in liste_bool
+proportion = 1
+tags = ["type"]
+labels = ["is ", "has ", "est "]
+
+bool_mapping = {
+    "1": True,
+    "0": False,
+    "vrai": True,
+    "faux": False,
+    "true": True,
+    "false": False,
+    "oui": True,
+    "non": False,
+    "yes": True,
+    "no": False,
+    "y": True,
+    "n": False,
+    "o": True,
+}
+
+liste_bool = set(bool_mapping.keys())
+
+
+def bool_casting(val: str) -> bool:
+    return bool_mapping.get(val.lower())
+
+
+def _is(val):
+    return isinstance(val, str) and val.lower() in liste_bool
+
+
+_test_values = {
+    True: ["oui", "0", "1", "yes", "false", "True"],
+    False: ["nein", "ja", "2", "-0"],
+}

csv_detective/formats/code_commune_insee.py

@@ -0,0 +1,26 @@
+from frformat import CodeCommuneInsee, Millesime
+
+proportion = 0.75
+tags = ["fr", "geo"]
+labels = [
+    "code commune insee",
+    "code insee",
+    "codes insee",
+    "code commune",
+    "code insee commune",
+    "insee",
+    "code com",
+    "com",
+]
+
+_code_commune_insee = CodeCommuneInsee(Millesime.LATEST)
+
+
+def _is(val):
+    return isinstance(val, str) and _code_commune_insee.is_valid(val)
+
+
+_test_values = {
+    True: ["91471", "01053"],
+    False: ["914712", "01000"],
+}

csv_detective/{detect_fields/FR/other/code_csp_insee/__init__.py → formats/code_csp_insee.py}

@@ -1,29 +1,36 @@
-import re
-
-from csv_detective.parsing.text import _process_text
-
-PROPORTION = 1
-
-
-def _is(val):
-    """Repère les code csp telles que définies par l'INSEE"""
-    if not isinstance(val, str):
-        return False
-    val = _process_text(val)
-    if len(val) != 4:
-        return False
-    a = bool(re.match(r"^[123456][0-9]{2}[abcdefghijkl]$", val))
-    b = val in {
-        "7100",
-        "7200",
-        "7400",
-        "7500",
-        "7700",
-        "7800",
-        "8100",
-        "8300",
-        "8400",
-        "8500",
-        "8600",
-    }
-    return a or b
+import re
+
+from csv_detective.parsing.text import _process_text
+
+proportion = 1
+tags = ["fr"]
+labels = ["code csp insee", "code csp"]
+
+
+def _is(val):
+    if not isinstance(val, str):
+        return False
+    val = _process_text(val)
+    if len(val) != 4:
+        return False
+    a = bool(re.match(r"^[123456][0-9]{2}[abcdefghijkl]$", val))
+    b = val in {
+        "7100",
+        "7200",
+        "7400",
+        "7500",
+        "7700",
+        "7800",
+        "8100",
+        "8300",
+        "8400",
+        "8500",
+        "8600",
+    }
+    return a or b
+
+
+_test_values = {
+    True: ["121f"],
+    False: ["121x"],
+}

csv_detective/{detect_fields/FR/geo/code_departement/__init__.py → formats/code_departement.py}

@@ -1,15 +1,29 @@
-from frformat import Millesime, NumeroDepartement, Options
-
-PROPORTION = 1
-
-_options = Options(
-    ignore_case=True,
-    ignore_accents=True,
-    replace_non_alphanumeric_with_space=True,
-    ignore_extra_whitespace=True,
-)
-_numero_departement = NumeroDepartement(Millesime.LATEST, _options)
-
-
-def _is(val):
-    return isinstance(val, str) and _numero_departement.is_valid(val)
+from frformat import Millesime, NumeroDepartement, Options
+
+proportion = 1
+tags = ["fr", "geo"]
+labels = [
+    "code departement",
+    "code_departement",
+    "dep",
+    "departement",
+    "dept",
+]
+
+_options = Options(
+    ignore_case=True,
+    ignore_accents=True,
+    replace_non_alphanumeric_with_space=True,
+    ignore_extra_whitespace=True,
+)
+_numero_departement = NumeroDepartement(Millesime.LATEST, _options)
+
+
+def _is(val):
+    return isinstance(val, str) and _numero_departement.is_valid(val)
+
+
+_test_values = {
+    True: ["75", "2A", "2b", "974", "01"],
+    False: ["00", "96", "101"],
+}