csvpath 0.0.522__tar.gz → 0.0.524__tar.gz

{csvpath-0.0.522 → csvpath-0.0.524}/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.1
 Name: csvpath
-Version: 0.0.522
+Version: 0.0.524
 Summary: An edge governance framework for managing and validating CSV, Excel, and other tabular data files
 Author: David Kershaw
 Author-email: dk107dk@hotmail.com
-Requires-Python: >3.10
+Requires-Python: >3.10,<4.0
 Classifier: Development Status :: 3 - Alpha
 Classifier: Environment :: Console
 Classifier: Intended Audience :: Developers
@@ -32,6 +32,7 @@ Requires-Dist: metaphone (>=0.6,<0.7)
 Requires-Dist: openlineage-python (>=1.25.0,<2.0.0)
 Requires-Dist: opentelemetry-distro[otlp] (>=0.50b0,<0.51)
 Requires-Dist: paramiko (>=3.5.0,<4.0.0)
+Requires-Dist: pdoc (>=15.0.1,<16.0.0)
 Requires-Dist: ply (>=3.11,<4.0)
 Requires-Dist: prompt-toolkit (>=3.0.50,<4.0.0)
 Requires-Dist: pylightxl (>=1.61,<2.0)
@@ -49,9 +50,9 @@ Description-Content-Type: text/markdown
 
 # <a href='https://www.csvpath.org/'><img src='https://github.com/csvpath/csvpath/blob/main/docs/images/logo-wordmark-4.svg'/></a>
 
-The CsvPath Language defines a declarative syntax for inspecting and validating CSV and Excel files, and other tabular data.
+The CsvPath Framework helps you close the gap between Managed File Transfer and your data lake and/or applications. CsvPath Language is the core of the Framework. It defines a declarative syntax for inspecting and validating CSV and Excel files, and other tabular data.
 
-The CsvPath Framework makes it easy to setup a <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics.pdf'>Collect, Store, Validate Pattern</a> flat-file landing zone that:
+The CsvPath Framework makes it easy to setup a <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics-88gj.pdf'>Collect, Store, Validate Pattern</a> flat-file landing zone that:
 - Registers files for **clear and durable identity**
 - **Validates the data** against expectations
 - Provides **operations and validity reports**
@@ -93,6 +94,7 @@ If you need help, use the <a href='https://www.csvpath.org/getting-started/get-h
 
 - [Motivation](#motivation)
 - [Install](#install)
+- [Python Interface Docs (pdocs)](#pdocs)
 - [High-level Description](#description)
 - [Running CsvPath](#running)
    - [Validation](#validating)
@@ -126,9 +128,9 @@ This project tackles two needs:
 - A more robust validation language for delimited and tabular data
 - A systems integration framework for flat-file landing and staging
 
-CsvPath Language is first and foremost a validation language. It supports both schema definitions and rules-based validation. CsvPath Language describes data so you can easily tell if a file is valid. CsvPath can also extract and shape data and create reports. Overall the goal is to automate human judgement out of the processing loop and instead move it to the process definition stage.
+CsvPath Language is the core of the CsvPath Framework. It is a simple validation language for delimited data. It supports both schema definitions and rules-based validation. CsvPath Language describes data so you can easily tell if a file is valid. CsvPath can also extract and shape data and create reports. Overall the goal is to automate human judgement out of the processing loop and instead move it to the process definition stage.
 
-The CsvPath Framework implements CsvPath Language, but goes far beyond it to provide a full <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics.pdf'>Collect, Store, Validate Pattern</a> framework for landing flat files, registering them, validating them, shaping them to a consistent and comparable form, and staging them for a data lake. In that way, CsvPath fills the gap commonly found between an organization's MFT (managed file transfer) and a typical data lake architecture.
+The CsvPath Framework implements CsvPath Language, but goes far beyond it to provide a full <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics-88gj.pdf'>Collect, Store, Validate Pattern</a> framework for landing flat files, registering them, validating them, shaping them to a consistent and comparable form, and staging them for a data lake. In that way, CsvPath fills the gap commonly found between an organization's MFT (managed file transfer) and a typical data lake architecture.
 
 CsvPath's goal is to make simple validations almost trivial and more complex situations more manageable. It is a library and framework, not a system, so it relies on being easy to integrate with other DataOps tools.
 
@@ -150,6 +152,12 @@ CsvPath has an optional dependency on Pandas. Pandas data frames can be used as
 
 Pandas and its dependencies can make it harder to use CsvPath in certain specific MFT use cases. For e.g., using Pandas in an AWS Lambda layer may be less straightforward. If you need the capability, though, it is easy to install.
 
+
+<a name="pdocs"></a>
+# Python Interface Docs
+<a href='https://csvpath.github.io/csvpath/' target='_blank'>Python docs are here</a>.
+The CsvPath Framework's public interface is streamlined. csvpath.CsvPath and csvpath.CsvPaths are where most of the magic happens. Docs for deeper levels will be added over time.
+
 # Description
 <a name="description"></a>
 

{csvpath-0.0.522 → csvpath-0.0.524}/README.md

@@ -1,9 +1,9 @@
 
 # <a href='https://www.csvpath.org/'><img src='https://github.com/csvpath/csvpath/blob/main/docs/images/logo-wordmark-4.svg'/></a>
 
-The CsvPath Language defines a declarative syntax for inspecting and validating CSV and Excel files, and other tabular data.
+The CsvPath Framework helps you close the gap between Managed File Transfer and your data lake and/or applications. CsvPath Language is the core of the Framework. It defines a declarative syntax for inspecting and validating CSV and Excel files, and other tabular data.
 
-The CsvPath Framework makes it easy to setup a <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics.pdf'>Collect, Store, Validate Pattern</a> flat-file landing zone that:
+The CsvPath Framework makes it easy to setup a <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics-88gj.pdf'>Collect, Store, Validate Pattern</a> flat-file landing zone that:
 - Registers files for **clear and durable identity**
 - **Validates the data** against expectations
 - Provides **operations and validity reports**
@@ -45,6 +45,7 @@ If you need help, use the <a href='https://www.csvpath.org/getting-started/get-h
 
 - [Motivation](#motivation)
 - [Install](#install)
+- [Python Interface Docs (pdocs)](#pdocs)
 - [High-level Description](#description)
 - [Running CsvPath](#running)
    - [Validation](#validating)
@@ -78,9 +79,9 @@ This project tackles two needs:
 - A more robust validation language for delimited and tabular data
 - A systems integration framework for flat-file landing and staging
 
-CsvPath Language is first and foremost a validation language. It supports both schema definitions and rules-based validation. CsvPath Language describes data so you can easily tell if a file is valid. CsvPath can also extract and shape data and create reports. Overall the goal is to automate human judgement out of the processing loop and instead move it to the process definition stage.
+CsvPath Language is the core of the CsvPath Framework. It is a simple validation language for delimited data. It supports both schema definitions and rules-based validation. CsvPath Language describes data so you can easily tell if a file is valid. CsvPath can also extract and shape data and create reports. Overall the goal is to automate human judgement out of the processing loop and instead move it to the process definition stage.
 
-The CsvPath Framework implements CsvPath Language, but goes far beyond it to provide a full <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics.pdf'>Collect, Store, Validate Pattern</a> framework for landing flat files, registering them, validating them, shaping them to a consistent and comparable form, and staging them for a data lake. In that way, CsvPath fills the gap commonly found between an organization's MFT (managed file transfer) and a typical data lake architecture.
+The CsvPath Framework implements CsvPath Language, but goes far beyond it to provide a full <a href='https://www.atestaanalytics.com/s/The-Collect-Store-Validate-Pattern-Atesta-Analytics-88gj.pdf'>Collect, Store, Validate Pattern</a> framework for landing flat files, registering them, validating them, shaping them to a consistent and comparable form, and staging them for a data lake. In that way, CsvPath fills the gap commonly found between an organization's MFT (managed file transfer) and a typical data lake architecture.
 
 CsvPath's goal is to make simple validations almost trivial and more complex situations more manageable. It is a library and framework, not a system, so it relies on being easy to integrate with other DataOps tools.
 
@@ -102,6 +103,12 @@ CsvPath has an optional dependency on Pandas. Pandas data frames can be used as
 
 Pandas and its dependencies can make it harder to use CsvPath in certain specific MFT use cases. For e.g., using Pandas in an AWS Lambda layer may be less straightforward. If you need the capability, though, it is easy to install.
 
+
+<a name="pdocs"></a>
+# Python Interface Docs
+<a href='https://csvpath.github.io/csvpath/' target='_blank'>Python docs are here</a>.
+The CsvPath Framework's public interface is streamlined. csvpath.CsvPath and csvpath.CsvPaths are where most of the magic happens. Docs for deeper levels will be added over time.
+
 # Description
 <a name="description"></a>
 

csvpath-0.0.524/csvpath/__init__.py

@@ -0,0 +1,18 @@
+"""
+    The CsvPath Framework makes it easy to pre-board external data files. It sits between Managed File Transfer and the data lake or applications. CsvPath provides durable dataset identification, validation and canonicalization, and stages data for internal use as a known-good raw data source. The goal is to automate the pre-boarding process.
+
+CsvPath Language is the core validation and canonicalization engine of the Framework. The validation files can be developed without coding using the CLI. When you are ready to automate pre-boarding you will use the classes documented here, in particular csvpath.CsvPath, csvpath.CsvPaths, and the managers in csvpath.managers:
+
+- csvpath.managers.files.file_manager
+- csvpath.managers.paths.paths_manager
+- csvpath.managers.results.results_manager
+
+You access the managers from a csvpath.CsvPaths instance. You should not construct your own.
+
+There are many other classes you could potentially use in some specific and narrow cases, such as building a new integration, a new function, or a new type of printer. But for 99% of automation use cases these classes are all you need.
+"""
+
+from csvpath.csvpath import CsvPath
+from csvpath.csvpaths import CsvPaths
+
+__all__ = ["CsvPath", "CsvPaths"]

{csvpath-0.0.522 → csvpath-0.0.524}/csvpath/cli/asker.py

@@ -14,7 +14,7 @@ from prompt_toolkit.completion import WordCompleter
 
 
 class Asker:
-    def __init__(self, cli, *, name_type) -> None:
+    def __init__(self, cli, *, name_type, prompt=None) -> None:
         self._cli = cli
         names = None
         self.name_type = name_type
@@ -24,18 +24,24 @@ class Asker:
         elif name_type == "paths":
             names = self._cli.csvpaths.paths_manager.named_paths_names
             names.sort()
+        elif name_type == "none":
+            names = []
         else:
             raise ValueError("Name type must be files or paths")
         self.completer = WordCompleter(names, ignore_case=True)
         self.result = None
+        self.prompt = prompt
 
     def create_prompt_application(self, prompt_text=None):
         if prompt_text is None:
-            prompt_text = (
-                "Named-file name? "
-                if self.name_type == "files"
-                else "Named-paths name? "
-            )
+            if self.prompt is None:
+                prompt_text = (
+                    "Named-file name? "
+                    if self.name_type == "files"
+                    else "Named-paths name? "
+                )
+            else:
+                prompt_text = self.prompt
         # Create a buffer to store input
         buffer = Buffer(completer=self.completer, complete_while_typing=True)
         # Create key bindings
@@ -83,8 +89,9 @@ class Asker:
             # complete_while_typing=True
         )
 
-    def ask(self) -> str:
+    def ask(self, prompt=None) -> str:
         try:
+            self.prompt = prompt
             app = self.create_prompt_application()
             app.run()
             return self.result

{csvpath-0.0.522 → csvpath-0.0.524}/csvpath/cli/cli.py

@@ -6,13 +6,18 @@ from csvpath import CsvPaths
 from .drill_down import DrillDown
 from .selecter import Selecter
 from .debug_config import DebugConfig
+from csvpath.util.nos import Nos
+from .asker import Asker
+from .function_lister import FunctionLister
+from .const import Const
 
 
 class Cli:
     def __init__(self):
         self.csvpaths = CsvPaths()
         self.clear()
-        splash = """
+        """
+        splash = ""
           *** *            ******        **  **
         ***  **   *        **  **      **** **
        **        ** **  * ** *** ***** **  *****
@@ -24,7 +29,7 @@ CsvPath Command Line Interface
 Try tab completion and menu-by-key.
 For help see https://www.csvpath.org
 """
-        print(splash)
+        print(Const.SPLASH)
         self._return_to_cont()
         self.clear()
 
@@ -38,33 +43,33 @@ For help see https://www.csvpath.org
         time.sleep(0.5)
 
     ITALIC = "\033[3m"
-    SIDEBAR_COLOR = "\033[36m"
-    REVERT = "\033[0m"
-    STOP_HERE = f"{SIDEBAR_COLOR}{ITALIC}... done picking dir{REVERT}"
-    STOP_HERE2 = "👍 pick this dir"
-    CANCEL = f"{SIDEBAR_COLOR}{ITALIC}... cancel{REVERT}"
-    CANCEL2 = "← cancel"
-    QUIT = "← quit"
-    NAMED_FILES = "register data"
-    NAMED_PATHS = "load csvpaths"
-    ARCHIVE = "access the archive"
+    # SIDEBAR_COLOR = "\033[36m"
+    # REVERT = "\033[0m"
+    # STOP_HERE = f"{Const.SIDEBAR_COLOR}{ITALIC}... done picking dir{Const.REVERT}"
+    # STOP_HERE2 = "👍 pick this dir"
+    # CANCEL = f"{Const.SIDEBAR_COLOR}{ITALIC}... cancel{Const.REVERT}"
+    # CANCEL2 = "← cancel"
+    # QUIT = "← quit"
+    # NAMED_FILES = "register data"
+    # NAMED_PATHS = "load csvpaths"
+    # ARCHIVE = "access the archive"
 
     def _return_to_cont(self):
         print(
-            f"\n{Cli.SIDEBAR_COLOR}{Cli.ITALIC}... Hit return to continue{Cli.REVERT}\n"
+            f"\n{Const.SIDEBAR_COLOR}{Cli.ITALIC}... Hit return to continue{Const.REVERT}\n"
         )
         self._input("")
 
     def _response(self, text: str) -> None:
-        sys.stdout.write(f"\u001b[30;1m{text}{Cli.REVERT}\n")
+        sys.stdout.write(f"\u001b[30;1m{text}{Const.REVERT}\n")
 
     def action(self, text: str) -> None:
-        sys.stdout.write(f"\033[36m{text}{Cli.REVERT}\n")
+        sys.stdout.write(f"\033[36m{text}{Const.REVERT}\n")
 
     def _input(self, prompt: str) -> str:
         try:
             response = input(f"{prompt}\033[93m")
-            sys.stdout.write(Cli.REVERT)
+            sys.stdout.write(Const.REVERT)
             return response.strip()
         except KeyboardInterrupt:
             return "cancel"
@@ -78,12 +83,12 @@ For help see https://www.csvpath.org
             return
         if q is not None:
             print(q)
-        if choices[len(choices) - 1] == Cli.CANCEL:
-            choices[len(choices) - 1] = Cli.CANCEL2
-        if choices[len(choices) - 2] == Cli.STOP_HERE:
-            choices[len(choices) - 2] = Cli.STOP_HERE2
+        if choices[len(choices) - 1] == Const.CANCEL:
+            choices[len(choices) - 1] = Const.CANCEL2
+        if choices[len(choices) - 2] == Const.STOP_HERE:
+            choices[len(choices) - 2] = Const.STOP_HERE2
         cs = [(s, s) for s in choices]
-        t = Selecter().ask(title="", values=cs, cancel_value="CANCEL")
+        t = Selecter().ask(title="", values=cs, cancel_value=Const.CANCEL)
         self.clear()
         return t
 
@@ -92,47 +97,53 @@ For help see https://www.csvpath.org
             t = None
             try:
                 choices = [
-                    Cli.NAMED_FILES,
-                    Cli.NAMED_PATHS,
-                    Cli.ARCHIVE,
+                    Const.NAMED_FILES,
+                    Const.NAMED_PATHS,
+                    Const.ARCHIVE,
                     "run",
                     "config",
-                    self.QUIT,
+                    "functions",
+                    Const.QUIT,
                 ]
                 t = self.ask(choices)
             except KeyboardInterrupt:
                 self.end()
                 return
             t = self._do(t)
-            if t == Cli.QUIT:
+            if t == Const.QUIT:
                 self.end()
                 return
 
     def _do(self, t: str) -> str | None:
-        if t == Cli.QUIT:
+        if t == Const.QUIT:
             return t
         try:
             if t == "run":
                 self.run()
-            if t == Cli.NAMED_FILES:
+            if t == Const.NAMED_FILES:
                 self._files()
-            if t == Cli.NAMED_PATHS:
+            if t == Const.NAMED_PATHS:
                 self._paths()
-            if t == Cli.ARCHIVE:
+            if t == Const.ARCHIVE:
                 self._results()
             if t == "config":
                 self._config()
+            if t == "functions":
+                self._functions()
         except KeyboardInterrupt:
-            return Cli.QUIT
+            return Const.QUIT
         except Exception:
             print(traceback.format_exc())
             self._return_to_cont()
 
+    def _functions(self) -> None:
+        FunctionLister(self).list_functions()
+
     def _config(self) -> None:
         DebugConfig(self).show()
 
     def _files(self) -> None:
-        choices = ["add named-file", "list named-files", self.CANCEL2]
+        choices = ["add named-file", "list named-files", Const.CANCEL2]
         t = self.ask(choices)
         if t == "add named-file":
             DrillDown(self).name_file()
@@ -140,7 +151,7 @@ For help see https://www.csvpath.org
             self.list_named_files()
 
     def _paths(self) -> None:
-        choices = ["add named-paths", "list named-paths", self.CANCEL2]
+        choices = ["add named-paths", "list named-paths", Const.CANCEL2]
         t = self.ask(choices)
         if t == "add named-paths":
             DrillDown(self).name_paths()
@@ -148,7 +159,7 @@ For help see https://www.csvpath.org
             self.list_named_paths()
 
     def _results(self) -> None:
-        choices = ["open named-result", "list named-results", self.CANCEL2]
+        choices = ["open named-result", "list named-results", Const.CANCEL2]
         t = self.ask(choices)
         if t == "open named-result":
             self.open_named_result()
@@ -171,14 +182,17 @@ For help see https://www.csvpath.org
             names = self.csvpaths.results_manager.list_named_results()
             names = [n for n in names if n.find(".") == -1]
             print(f"{len(names)} named-results names:")
-            names.append(self.CANCEL)
+            names.append(Const.CANCEL)
             t = self.ask(names)
-            if t == self.CANCEL:
+            if t == Const.CANCEL:
                 return
             t = f"{self.csvpaths.config.archive_path}{os.sep}{t}"
             self.action(f"Opening results at {t}...")
             self.short_pause()
-            c = f"open {t}"
+            #
+            # not sure if this works for the linux desktop user
+            #
+            c = f"open {t}" if os.sep == "/" else f"explorer {t}"
             os.system(c)
         except Exception:
             print(traceback.format_exc())
@@ -208,19 +222,42 @@ For help see https://www.csvpath.org
             input("You must add a named-file. Press any key to continue.")
             return
         files.sort()
-        file = self.ask(files, q="What named-file? ")
+        file = self.ask(
+            files, q="What named-file? \n(enter $ on any line to build a reference) "
+        )
+        #
+        # if '$' user wants to use a reference, possibly for replay
+        #
+        if file.startswith("$"):
+            # find the run and instance
+            file = self.complete_file_reference()
+            if file is None:
+                input(
+                    "Could not build the reference. Check if all files are present. Press any key to continue."
+                )
+                return
+
         self.clear()
         allpaths = self.csvpaths.paths_manager.named_paths_names
         if len(allpaths) == 0:
             input("You must add a named-paths file. Press any key to continue.")
             return
         allpaths.sort()
-        paths = self.ask(allpaths, q="What named-paths? ")
+        paths = self.ask(
+            allpaths,
+            q="What named-paths? \n(enter $ on any line to build a reference) ",
+        )
+        #
+        # if '$' user wants to use a reference to rewind
+        #
+        if paths.startswith("$"):
+            paths = self.complete_paths_reference(paths)
+
         self.clear()
-        choices = ["collect", "fast-forward", Cli.CANCEL2]
+        choices = ["collect", "fast-forward", Const.CANCEL2]
         method = self.ask(choices, q="What method? ")
         self.clear()
-        if method == Cli.CANCEL2:
+        if method == Const.CANCEL2:
             return
         self.action(f"Running {paths} against {file} using {method}\n")
         self.pause()
@@ -249,6 +286,48 @@ For help see https://www.csvpath.org
                 self.clear()
         self._return_to_cont()
 
+    def complete_file_reference(self) -> str:
+        allpaths = self.csvpaths.paths_manager.named_paths_names
+        allpaths.sort()
+        file = self.ask(allpaths, q="Building a reference. Use what results? ")
+        file = file.lstrip("$")
+        results = self.csvpaths.config.get(section="results", name="archive")
+        results = f"{results}{os.sep}{file}"
+        if not Nos(results).dir_exists():
+            return None
+        runs = Nos(results).listdir()
+        runs.sort()
+        run = self.ask(runs, q="Which run? ")
+        run = run.lstrip("$")
+        results = f"{results}{os.sep}{run}"
+        instances = Nos(results).listdir()
+        instances = [i for i in instances if i.find(".json") == -1]
+        instance = self.ask(instances, q="Which csvpath? ")
+        instance = instance.lstrip("$")
+        return f"${file}.results.{run}.{instance}"
+
+    def complete_paths_reference(self, paths) -> str:
+        allpaths = self.csvpaths.paths_manager.named_paths_names
+        allpaths.sort()
+        path = self.ask(allpaths, q="Building a reference. Use what paths? ")
+        path = path.lstrip("$")
+        instances = self.csvpaths.paths_manager.get_identified_paths_in(path)
+        ids = [i[0] for i in instances]
+        run = self.ask(ids, q="Which csvpath? ")
+        run = run.lstrip("$")
+        # ft = Asker(self, name_type="none").ask("from:, to:, or neither? ")
+        ft = self.ask(
+            ["from", "to", "neither"], q=f"Limit csvpaths in group relative to {run}? "
+        )
+        ft = ft.lstrip("$")
+        ft = ft.replace(":", "")
+        ft = ft.strip()
+        if ft in ["from", "to"]:
+            ft = f":{ft}"
+        else:
+            ft = ""
+        return f"${path}.csvpaths.{run}{ft}"
+
 
 def run():
     cli = Cli()

csvpath-0.0.524/csvpath/cli/const.py

@@ -0,0 +1,26 @@
+class Const:
+
+    SPLASH = """
+          *** *            ******        **  **
+        ***  **   *        **  **      **** **
+       **        ** **  * ** *** ***** **  *****
+       **    * **** **** ***** *** ** **  ** **
+       **   **** ****** **     ** ** *** ** ** **
+         ***   **** *  **      *** ** ****   **
+***************************
+CsvPath Command Line Interface
+Try tab completion and menu-by-key.
+For help see https://www.csvpath.org
+"""
+
+    ITALIC = "\033[3m"
+    SIDEBAR_COLOR = "\033[36m"
+    REVERT = "\033[0m"
+    STOP_HERE = f"{SIDEBAR_COLOR}{ITALIC}... done picking dir{REVERT}"
+    STOP_HERE2 = "👍 pick this dir"
+    CANCEL = f"{SIDEBAR_COLOR}{ITALIC}... cancel{REVERT}"
+    CANCEL2 = "← cancel"
+    QUIT = "← quit"
+    NAMED_FILES = "register data"
+    NAMED_PATHS = "load csvpaths"
+    ARCHIVE = "access the archive"

{csvpath-0.0.522 → csvpath-0.0.524}/csvpath/cli/drill_down.py

@@ -2,6 +2,7 @@ import os
 import traceback
 from .debug_config import DebugConfig
 from .asker import Asker
+from .const import Const
 
 
 class DrillDown:
@@ -16,14 +17,16 @@ class DrillDown:
         #
         # get name
         #
+        t = self._get_add_type()
+        if t == Const.CANCEL2:
+            return
         self._cli.clear()
-        name = Asker(self._cli, name_type="files").ask()
+        name = None
+        if t == "file":
+            name = Asker(self._cli, name_type="files").ask()
         #
         # get path
         #
-        t = self._get_add_type()
-        if t == self._cli.CANCEL2:
-            return
         p = self._get_path(t, self._cli.csvpaths.config.csv_file_extensions)
         if p is False:
             return
@@ -63,17 +66,19 @@ class DrillDown:
     # ============================
 
     def name_paths(self):
+        t = self._get_add_type()
+        if t == Const.CANCEL2:
+            return
         #
         # get name
         #
         self._cli.clear()
-        name = Asker(self._cli, name_type="paths").ask()
+        name = None
+        if t == "file":
+            name = Asker(self._cli, name_type="paths").ask()
         #
         # get path
         #
-        t = self._get_add_type()
-        if t == self._cli.CANCEL2:
-            return
         exts = self._cli.csvpaths.config.csvpath_file_extensions
         p = self._get_path(t, exts)
         if p is False:
@@ -145,7 +150,7 @@ class DrillDown:
 
     def _get_add_type(self) -> str:
         self._cli.clear()
-        choices = ["dir", "file", "json", self._cli.CANCEL2]
+        choices = ["dir", "file", "json", Const.CANCEL2]
         t = None
         t = self._cli.ask(choices)
         return t
@@ -160,9 +165,9 @@ class DrillDown:
         names.sort()
         names = self._decorate(path, names, select_dir=dir_only)
         t = self._cli.ask(names)
-        if t in [self._cli.STOP_HERE, self._cli.STOP_HERE2]:
+        if t in [Const.STOP_HERE, Const.STOP_HERE2]:
             return (path, True)
-        if t in [self._cli.CANCEL, self._cli.CANCEL2]:
+        if t in [Const.CANCEL, Const.CANCEL2]:
             return (path, False)
         if t.startswith("📂 ") or t.startswith("📄 "):
             t = t[2:]
@@ -171,7 +176,7 @@ class DrillDown:
     def _decorate(self, path, names, select_dir=False) -> list[str]:
         ns = []
         for n in names:
-            if n in [self._cli.STOP_HERE, self._cli.STOP_HERE2]:
+            if n in [Const.STOP_HERE, Const.STOP_HERE2]:
                 pass
             elif os.path.isfile(os.path.join(path, n)):
                 n = f"📄 {n}"
@@ -179,8 +184,8 @@ class DrillDown:
                 n = f"📂 {n}"
             ns.append(n)
         if select_dir is True:
-            ns.append(self._cli.STOP_HERE)
-        ns.append(self._cli.CANCEL)
+            ns.append(Const.STOP_HERE)
+        ns.append(Const.CANCEL)
         return ns
 
     def _filter_hidden(self, names) -> list[str]: