Add filter creation in configuration creation command.

.gitignore

@@ -157,4 +157,7 @@ cython_debug/
 #  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
-#.idea/
+.idea/
+
+# Project Specific
+scripts/

requirements.txt

@@ -1,4 +1,5 @@
 click
 genanki
 pandas
-pyyaml
+pyyaml
+bullet

setup.py

@@ -27,6 +27,7 @@ setup(
         "genanki",
         "pandas",
         "pyyaml",
+        "bullet"
     ],
     long_description_content_type='text/markdown',
 )

src/ankimaker/commands/__init__.py

@@ -1,10 +1,3 @@
-import click
-
-
-@click.group("cli")
-def cli():
-    pass
-
-
-from ..commands.from_csv import generate_anki
-from ..commands.make_config import make_csv_config
+from .base_click import cli
+from .from_csv import generate_anki
+from .make_config import make_csv_config

src/ankimaker/commands/base_click.py

@@ -0,0 +1,6 @@
+import click
+
+
+@click.group("cli")
+def cli():
+    pass

src/ankimaker/commands/from_csv.py

@@ -1,5 +1,6 @@
-import click
 import re
+import click
+
 from ankimaker.commands import cli
 from ankimaker.tasks import basic_pandas_to_anki
 

src/ankimaker/config/configuration.py

@@ -14,12 +14,15 @@ class AnkimakerConfig(yaml.YAMLObject):
     separators = ','
     filters: List[List[FilterConfig]] = list()
 
-    def __init__(self, header=None, answer_column=None, question_column=None, filters=tuple()):
-        AnkimakerConfig.answer_column = answer_column
-        AnkimakerConfig.question_column = question_column
-        AnkimakerConfig.header = header
-        AnkimakerConfig.AnkimakerConfig = AnkimakerConfig
-        AnkimakerConfig.filters = list(map(lambda x: FilterConfig, filters))
+    def __init__(
+            self, separators=',', header=None, answer_column=None, question_column=None,
+            filters=tuple(), *args, **karhs
+    ):
+        self.answer_column = answer_column
+        self.question_column = question_column
+        self.header = header
+        self.separators = separators
+        self.filters = _conditionally_create_new_filters(filters)
 
     @staticmethod
     def loader(configuration_content):
@@ -31,8 +34,19 @@ class AnkimakerConfig(yaml.YAMLObject):
         AnkimakerConfig.question_column = content.question_column
         AnkimakerConfig.answer_column = content.answer_column
         AnkimakerConfig.separators = content.separators
-        AnkimakerConfig.filters = [
-            [FilterConfig(**x) for x in or_filter]
-            for or_filter in content.filters
-        ]
+        AnkimakerConfig.filters = _conditionally_create_new_filters(content.filters)
 
+
+def _conditionally_create_new_filters(filters):
+    conf_has_filters = len(filters) > 0
+    if conf_has_filters:
+        should_cast_filter = not isinstance(filters[0][0], FilterConfig)
+        if should_cast_filter:
+            new_filters = [
+                [FilterConfig(**x) for x in or_filter]
+                for or_filter in filters
+            ]
+        else:
+            new_filters = filters
+        return new_filters
+    return list()

src/ankimaker/config/filters.py

@@ -1,7 +1,10 @@
+import yaml
+
 from typing import List, Union
 
 
-class FilterConfig:
+class FilterConfig(yaml.YAMLObject):
+    yaml_tag = '!fitlerconfig'
     column: Union[str, int]
     values: Union[List[Union[int, str]], Union[int, str]]
 
@@ -10,7 +13,7 @@ class FilterConfig:
         self.values = values
 
     def __str__(self):
-        return f'<ankimaker.config.filters.FilterConfig {self.column}: {self.values} >'
+        return f'<F({self.column}:{self.values})>'
 
     def __repr__(self):
         return self.__str__()

src/ankimaker/tasks/basic_csv_to_anki.py

@@ -3,11 +3,11 @@ import pandas as pd
 from typing import List
 from functools import reduce
 
-from ankimaker.config import Config, FilterConfig
 from ankimaker import generator, config
+from ankimaker.config import Config, FilterConfig
 
 
-def load_csv(path):
+def load_csv(path: str) -> pd.DataFrame:
     df = pd.read_csv(path, header=Config.header, sep=Config.separators)
     df_columns_are_unnamed = all(map(lambda x: str(x).isnumeric(), df.columns))
     if df_columns_are_unnamed:
@@ -17,7 +17,7 @@ def load_csv(path):
     return df
 
 
-def add_df_to_deck(df: pd.DataFrame, deck: genanki.Deck):
+def add_df_to_deck(df: pd.DataFrame, deck: genanki.Deck) -> genanki.Deck:
     model = generator.create_model()
 
     for entry in df.to_dict('records'):
@@ -29,7 +29,7 @@ def add_df_to_deck(df: pd.DataFrame, deck: genanki.Deck):
     return deck
 
 
-def handle_config(config_file_path):
+def handle_config(config_file_path: str):
     if config_file_path is None:
         Config.header = None
         Config.question_column = 0
@@ -40,9 +40,10 @@ def handle_config(config_file_path):
 
 def apply_filters(df: pd.DataFrame) -> pd.DataFrame:
     """
-
-    :param df:
-    :return:
+    Returns filtered dataframe removing any row that does not correspond to at least one
+    of the filter  groups defined in Configuration.
+    :param df: Original dataframe.
+    :return: Filtered Dataframe.
     """
     there_are_no_filter_to_apply = len(Config.filters) == 0
     if there_are_no_filter_to_apply:
@@ -53,7 +54,12 @@ def apply_filters(df: pd.DataFrame) -> pd.DataFrame:
 
 
 def load_filter_from_config(df: pd.DataFrame) -> pd.Series:
-
+    """
+    Given a dataframe, returns a series indicating which rows should be kept according to loaded
+    Config [AnkimakerConfig]. The rows presented in any filter group should be kept.
+    :param df: Original dataframe.
+    :return pd.Series: Boolean Series to filter df.
+    """
     group_filters: List[pd.Series] = list()
     for group in Config.filters:
         if len(group) > 0:
@@ -66,22 +72,23 @@ def load_filter_from_config(df: pd.DataFrame) -> pd.Series:
 
 def create_group_filter(df: pd.DataFrame, group: List[FilterConfig]) -> pd.Series:
     """
-
-    :param df:
-    :param group:
-    :return:
+    Creates a boolean series indicating which rows are in the filters configuration defined
+    group to be used to filter the dataframe.
+    :param df: Input dataframe to be filtered.
+    :param group: Filter defined Group.
+    :return: Series of boolean indicating rows that are in the group.
     """
     rule: FilterConfig
     query: List[pd.Series] = list()
     for rule in group:
-        assert_rule_is_valid(df, rule)
+        __assert_rule_is_valid(df, rule)
         is_in_rule = df[rule.column].apply(lambda x: x in rule.values)
         query.append(is_in_rule)
     is_in_group = reduce(lambda a, b: a & b, query)
     return is_in_group
 
 
-def assert_rule_is_valid(df: pd.DataFrame, rule: FilterConfig):
+def __assert_rule_is_valid(df: pd.DataFrame, rule: FilterConfig):
     assert rule.column in df.columns
 
 

src/ankimaker/tasks/config_tasks/create_config.py

@@ -1,10 +1,12 @@
 import os
+
 import yaml
 import click
 import pandas as pd
-from typing import Type
+from typing import Type, List
+from bullet import Bullet, Input, YesNo
 
-from ankimaker.config import Config
+from ankimaker.config import Config, FilterConfig
 
 
 __CONFIRMATION_QUESTION = """
@@ -25,25 +27,93 @@ __COMMAND_SAMPLE = """ankimaker csv \
 """
 
 
+__ADD_FILTER_QUESTION = """Do you want do add a filter to the configuration?"""
+
+
 def create_config(input_file, output_path):
-    new_config = Config()
 
-    new_config.separators = handle_read_option(
-        input_file, read_option='sep', sep=new_config.separators
+    separators = handle_read_option(
+        input_file, read_option='sep', sep=','
     )
-    new_config.header = handle_read_option(
-        input_file, read_option='header', header=new_config.header,
-        sep=new_config.separators, option_type=int
+    header = handle_read_option(
+        input_file, read_option='header', header=None,
+        sep=separators, option_type=int
     )
 
-    new_config.question_column = get_column('question')
-    new_config.answer_column = get_column('answer')
+    question_column = get_column('question')
+    answer_column = get_column('answer')
 
+    filters = process_filters(input_file, header, separators)
+
+    new_config = Config(
+        separators=separators,
+        header=header,
+        question_column=question_column,
+        answer_column=answer_column,
+        filters=filters
+    )
     save_file(new_config, output_path)
+
     finish_message = __SUCCESS_MESSAGE.format(command=make_sample_command(input_file, output_path))
+    click.clear()
     click.echo(finish_message)
 
 
+def process_filters(input_file, header, separators):
+    df = pd.read_csv(input_file, header=header, sep=separators)
+    filters = add_filters_to_config(df)
+    return filters
+
+
+def __inline_yes_or_no_question(question):
+    answer = YesNo(prompt=question, default='n').launch()
+    return answer
+
+
+def add_filters_to_config(df: pd.DataFrame) -> List[List[FilterConfig]]:
+    config = Config()
+    should_add_filter = __inline_yes_or_no_question(__ADD_FILTER_QUESTION)
+    while should_add_filter:
+        config = add_filter_to_or_create_filter_group(df, config)
+        should_add_filter = __inline_yes_or_no_question(__ADD_FILTER_QUESTION)
+    return config.filters
+
+
+def add_filter_to_or_create_filter_group(df: pd.DataFrame, config: Config) -> Config:
+    config_has_filters = len(config.filters) > 0
+    chosen_group = -1
+    if config_has_filters:
+        filter_options = [f'({"|".join(map(str, group)):.45s})' for group in config.filters]
+        filter_options = [f'Group{i+1}{s}' for i, s in enumerate(filter_options)]
+        cli = Bullet(
+            prompt="Select group: ",
+            choices=["Create new", *filter_options],
+            return_index=True,
+        )
+        chosen_group = cli.launch()[1] - 1
+    new_filter = create_filter_config(df)
+    if chosen_group < 0:
+        config.filters.append([new_filter])
+    else:
+        config.filters[chosen_group].append(new_filter)
+    return config
+
+
+def create_filter_config(df: pd.DataFrame) -> FilterConfig:
+    options = list(df.columns)
+    cli = Bullet(
+        prompt="Select a columns to filter: ",
+        choices=list(map(str, options)),
+        return_index=True
+    )
+    chosen = cli.launch()[1]
+    filter_column = options[chosen]
+    columns_values = df[filter_column].unique()
+    values = Input(f'Which values fo filter out? values[{columns_values}]: ').launch()
+    new_filter = FilterConfig(column=filter_column, values=values)
+    return new_filter
+
+
 def get_column(name: str) -> str:
     answer = click.prompt(f'Which is your {name} column?', type=str, confirmation_prompt=True)
     return answer

src/ankimaker/utils/files.py

@@ -1,3 +1,3 @@
-def get_fyle_type(filename):
+def get_fyle_type(filename: str) -> str:
     filetype = filename.split('.')[-1] if len(filename.split('.')) > 0 else None
     return filetype