kash-shell 0.3.11__py3-none-any.whl → 0.3.13__py3-none-any.whl

kash/actions/core/markdownify.py

@@ -1,8 +1,9 @@
 from kash.config.logger import get_logger
 from kash.exec import kash_action
-from kash.exec.preconditions import has_html_body, is_url_item
+from kash.exec.preconditions import has_html_body, is_url_resource
 from kash.model import Format, Item
 from kash.model.params_model import common_params
+from kash.utils.text_handling.markdownify_utils import markdownify_custom
 from kash.web_content.file_cache_utils import get_url_html
 from kash.web_content.web_extract_readabilipy import extract_text_readabilipy
 
@@ -10,7 +11,7 @@ log = get_logger(__name__)
 
 
 @kash_action(
-    precondition=is_url_item | has_html_body,
+    precondition=is_url_resource | has_html_body,
     params=common_params("refetch"),
     mcp_tool=True,
 )
@@ -19,12 +20,12 @@ def markdownify(item: Item, refetch: bool = False) -> Item:
     Converts a URL or raw HTML item to Markdown, fetching with the content
     cache if needed. Also uses readability to clean up the HTML.
     """
-    from markdownify import markdownify as markdownify_convert
 
     expiration_sec = 0 if refetch else None
     url, html_content = get_url_html(item, expiration_sec=expiration_sec)
     page_data = extract_text_readabilipy(url, html_content)
-    markdown_content = markdownify_convert(page_data.clean_html)
+    assert page_data.clean_html
+    markdown_content = markdownify_custom(page_data.clean_html)
 
     output_item = item.derived_copy(format=Format.markdown, body=markdown_content)
     return output_item

kash/actions/core/readability.py

@@ -1,6 +1,6 @@
 from kash.config.logger import get_logger
 from kash.exec import kash_action
-from kash.exec.preconditions import has_html_body, is_url_item
+from kash.exec.preconditions import has_html_body, is_url_resource
 from kash.model import Format, Item
 from kash.model.params_model import common_params
 from kash.web_content.file_cache_utils import get_url_html
@@ -10,7 +10,7 @@ log = get_logger(__name__)
 
 
 @kash_action(
-    precondition=is_url_item | has_html_body,
+    precondition=is_url_resource | has_html_body,
     params=common_params("refetch"),
     mcp_tool=True,
 )
@@ -20,8 +20,8 @@ def readability(item: Item, refetch: bool = False) -> Item:
     See `markdownify` to also convert to Markdown.
     """
     expiration_sec = 0 if refetch else None
-    url, html_content = get_url_html(item, expiration_sec=expiration_sec)
-    page_data = extract_text_readabilipy(url, html_content)
+    locator, html_content = get_url_html(item, expiration_sec=expiration_sec)
+    page_data = extract_text_readabilipy(locator, html_content)
 
     output_item = item.derived_copy(format=Format.html, body=page_data.clean_html)
 

kash/actions/core/render_as_html.py

@@ -1,7 +1,7 @@
 from kash.actions.core.tabbed_webpage_config import tabbed_webpage_config
 from kash.actions.core.tabbed_webpage_generate import tabbed_webpage_generate
 from kash.exec import kash_action
-from kash.exec.preconditions import has_full_html_page_body, has_text_body, is_html
+from kash.exec.preconditions import has_full_html_page_body, has_html_body, has_simple_text_body
 from kash.exec_model.args_model import ONE_OR_MORE_ARGS
 from kash.model import ActionInput, ActionResult, Param
 from kash.model.items_model import ItemType
@@ -11,10 +11,10 @@ from kash.web_gen.simple_webpage import simple_webpage_render
 
 @kash_action(
     expected_args=ONE_OR_MORE_ARGS,
-    precondition=(is_html | has_text_body) & ~has_full_html_page_body,
-    params=(Param("add_title", "Add a title to the page body.", type=bool),),
+    precondition=(has_html_body | has_simple_text_body) & ~has_full_html_page_body,
+    params=(Param("no_title", "Don't add a title to the page body.", type=bool),),
 )
-def render_as_html(input: ActionInput, add_title: bool = False) -> ActionResult:
+def render_as_html(input: ActionInput, no_title: bool = False) -> ActionResult:
     """
     Convert text, Markdown, or HTML to pretty, formatted HTML using a clean
     and simple page template. Supports GFM-flavored Markdown tables and footnotes.
@@ -27,11 +27,13 @@ def render_as_html(input: ActionInput, add_title: bool = False) -> ActionResult:
     """
     if len(input.items) == 1:
         input_item = input.items[0]
-        html_body = simple_webpage_render(input_item, add_title_h1=add_title)
+        html_body = simple_webpage_render(input_item, add_title_h1=not no_title)
         result_item = input_item.derived_copy(
             type=ItemType.export, format=Format.html, body=html_body
         )
         return ActionResult([result_item])
     else:
         config_result = tabbed_webpage_config(input)
-        return tabbed_webpage_generate(ActionInput(items=config_result.items), add_title=add_title)
+        return tabbed_webpage_generate(
+            ActionInput(items=config_result.items), add_title=not no_title
+        )

kash/actions/core/show_webpage.py

@@ -1,7 +1,7 @@
 from kash.actions.core.render_as_html import render_as_html
 from kash.commands.base.show_command import show
 from kash.exec import kash_action
-from kash.exec.preconditions import has_full_html_page_body, has_text_body, is_html
+from kash.exec.preconditions import has_full_html_page_body, has_html_body, has_simple_text_body
 from kash.exec_model.args_model import ONE_OR_MORE_ARGS
 from kash.exec_model.commands_model import Command
 from kash.exec_model.shell_model import ShellResult
@@ -10,7 +10,7 @@ from kash.model import ActionInput, ActionResult
 
 @kash_action(
     expected_args=ONE_OR_MORE_ARGS,
-    precondition=(is_html | has_text_body) & ~has_full_html_page_body,
+    precondition=(has_html_body | has_simple_text_body) & ~has_full_html_page_body,
 )
 def show_webpage(input: ActionInput) -> ActionResult:
     """

kash/actions/core/strip_html.py

@@ -1,6 +1,6 @@
 from kash.config.logger import get_logger
 from kash.exec import kash_action
-from kash.exec.preconditions import has_html_body, has_text_body
+from kash.exec.preconditions import has_html_body, has_simple_text_body
 from kash.model import Format, Item
 from kash.utils.common.format_utils import html_to_plaintext
 from kash.utils.errors import InvalidInput
@@ -8,7 +8,7 @@ from kash.utils.errors import InvalidInput
 log = get_logger(__name__)
 
 
-@kash_action(precondition=has_html_body | has_text_body)
+@kash_action(precondition=has_html_body | has_simple_text_body)
 def strip_html(item: Item) -> Item:
     """
     Strip HTML tags from HTML or Markdown. This is a simple filter, simply searching

kash/commands/base/basic_file_commands.py

@@ -2,7 +2,7 @@ import os
 
 from frontmatter_format import fmf_read_raw, fmf_strip_frontmatter
 from prettyfmt import fmt_lines
-from strif import copyfile_atomic
+from strif import atomic_output_file, copyfile_atomic
 
 from kash.config.logger import get_logger
 from kash.config.text_styles import STYLE_EMPH
@@ -28,10 +28,10 @@ log = get_logger(__name__)
 
 
 @kash_command
-def cbcopy(path: str | None = None, raw: bool = False) -> None:
+def clipboard_copy(path: str | None = None, raw: bool = False) -> None:
     """
     Copy the contents of a file (or the first file in the selection) to the OS-native
-    clipboard.
+    clipboard. Similar to `pbcopy` on macOS.
 
     :param raw: Copy the full exact contents of the file. Otherwise frontmatter is omitted.
     """
@@ -39,6 +39,8 @@ def cbcopy(path: str | None = None, raw: bool = False) -> None:
     import pyperclip
 
     input_paths = assemble_path_args(path)
+    if not input_paths:
+        raise InvalidInput("No path provided")
     input_path = input_paths[0]
 
     format = detect_file_format(input_path)
@@ -69,6 +71,25 @@ def cbcopy(path: str | None = None, raw: bool = False) -> None:
         )
 
 
+@kash_command
+def clipboard_paste(path: str = "untitled_paste.txt") -> None:
+    """
+    Paste the contents of the OS-native clipboard into a new file.
+    """
+    # TODO: Get this to work for images!
+    # And can we convert rich text to Markdown?
+    import pyperclip
+
+    contents = pyperclip.paste()
+    if not contents.strip():
+        raise InvalidInput("Clipboard is empty")
+
+    with atomic_output_file(path, backup_suffix=".{timestamp}.bak") as f:
+        f.write_text(contents)
+
+    print_status("Pasted clipboard contents to:\n%s", fmt_lines([fmt_loc(path)]))
+
+
 @kash_command
 def edit(path: str | None = None, all: bool = False) -> None:
     """

kash/commands/base/diff_commands.py

@@ -5,16 +5,51 @@ from kash.commands.workspace.selection_commands import select
 from kash.config.logger import get_logger
 from kash.exec import import_locator_args, kash_command
 from kash.exec_model.shell_model import ShellResult
-from kash.model.items_model import Item, ItemType
+from kash.model.items_model import Item, ItemRelations, ItemType
+from kash.model.paths_model import StorePath
 from kash.shell.output.shell_output import Wrap, cprint
-from kash.text_handling.unified_diffs import unified_diff_files, unified_diff_items
-from kash.utils.errors import InvalidInput, InvalidOperation
+from kash.utils.errors import ContentError, InvalidInput, InvalidOperation
 from kash.utils.file_utils.file_formats_model import Format
+from kash.utils.text_handling.unified_diffs import unified_diff, unified_diff_files
 from kash.workspaces import current_ws
 
 log = get_logger(__name__)
 
 
+def unified_diff_items(from_item: Item, to_item: Item, strict: bool = True) -> Item:
+    """
+    Generate a unified diff between two items. If `strict` is true, will raise
+    an error if the items are of different formats.
+    """
+    if not from_item.body and not to_item.body:
+        raise ContentError(f"No body to diff for {from_item} and {to_item}")
+    if not from_item.store_path or not to_item.store_path:
+        raise ContentError("No store path on items; save before diffing")
+    diff_items = [item for item in [from_item, to_item] if item.format == Format.diff]
+    if len(diff_items) == 1:
+        raise ContentError(
+            f"Cannot compare diffs to non-diffs: {from_item.format}, {to_item.format}"
+        )
+    if len(diff_items) > 0 or from_item.format != to_item.format:
+        msg = f"Diffing items of incompatible format: {from_item.format}, {to_item.format}"
+        if strict:
+            raise ContentError(msg)
+        else:
+            log.warning("%s", msg)
+
+    from_path, to_path = StorePath(from_item.store_path), StorePath(to_item.store_path)
+
+    diff = unified_diff(from_item.body, to_item.body, str(from_path), str(to_path))
+
+    return Item(
+        type=ItemType.doc,
+        title=f"Diff of {from_path} and {to_path}",
+        format=Format.diff,
+        relations=ItemRelations(diff_of=[from_path, to_path]),
+        body=diff.patch_text,
+    )
+
+
 @kash_command
 def diff_items(*paths: str, force: bool = False) -> ShellResult:
     """

kash/commands/base/files_command.py

@@ -75,7 +75,8 @@ def _print_listing_tallies(
         cprint("(use --no_max to remove cutoff)", style=STYLE_HINT)
 
 
-DEFAULT_MAX_PG = 100
+DEFAULT_MAX_PER_GROUP = 50
+"""Default maximum number of files to display per group."""
 
 
 @kash_command
@@ -301,13 +302,13 @@ def files(
     # there are lots of groups and lots of files per group.
     # Default is max 100 per group but if we have 4 * 100 items, cut to 25.
     # If we have 2 * 100 items, cut to 50.
-    final_max_pg = DEFAULT_MAX_PG if cap_per_group else max_per_group
+    final_max_pg = DEFAULT_MAX_PER_GROUP if cap_per_group else max_per_group
     max_pg_explicit = max_per_group > 0
     if not max_pg_explicit:
         group_lens = [len(group_df) for group_df in grouped]
         for ratio in [2, 4]:
-            if sum(group_lens) > ratio * DEFAULT_MAX_PG:
-                final_max_pg = int(DEFAULT_MAX_PG / ratio)
+            if sum(group_lens) > ratio * DEFAULT_MAX_PER_GROUP:
+                final_max_pg = int(DEFAULT_MAX_PER_GROUP / ratio)
 
     total_displayed = 0
     total_displayed_size = 0

kash/commands/base/reformat_command.py

@@ -8,9 +8,9 @@ from kash.config.logger import get_logger
 from kash.exec import assemble_path_args, kash_command, resolvable_paths
 from kash.exec_model.shell_model import ShellResult
 from kash.shell.output.shell_output import print_status
-from kash.text_handling.doc_normalization import normalize_text_file
 from kash.utils.common.format_utils import fmt_loc
 from kash.utils.file_utils.filename_parsing import join_filename, split_filename
+from kash.utils.text_handling.doc_normalization import normalize_text_file
 
 log = get_logger(__name__)
 

kash/commands/base/show_command.py

@@ -54,7 +54,7 @@ def show(
                 item = ws.load(input_path)
                 if thumbnail and item.thumbnail_url:
                     try:
-                        local_path, _was_cached = cache_file(item.thumbnail_url)
+                        local_path = cache_file(item.thumbnail_url).content.path
                         terminal_show_image(local_path)
                     except Exception as e:
                         log.info("Had trouble showing thumbnail image (will skip): %s", e)

kash/commands/extras/parse_uv_lock.py

@@ -1,8 +1,10 @@
+from __future__ import annotations
+
 import subprocess
 import tomllib
 from pathlib import Path
+from typing import TYPE_CHECKING
 
-import pandas as pd
 from packaging.tags import Tag, sys_tags
 from packaging.utils import parse_wheel_filename
 from prettyfmt import fmt_size_dual
@@ -13,6 +15,9 @@ from kash.config.text_styles import COLOR_STATUS
 from kash.exec import kash_command
 from kash.shell.output.shell_output import cprint
 
+if TYPE_CHECKING:
+    from pandas import DataFrame
+
 log = get_logger(__name__)
 
 
@@ -49,13 +54,15 @@ def get_platform() -> str:
     return next(sys_tags()).platform
 
 
-def parse_uv_lock(lock_path: Path) -> pd.DataFrame:
+def parse_uv_lock(lock_path: Path) -> DataFrame:
     """
     Return one row per package from a uv.lock file, selecting the best
     matching wheel for the current interpreter or falling back to the sdist.
 
     Columns: name, version, registry, file_type, url, hash, size, filename.
     """
+    from pandas import DataFrame
+
     with open(lock_path, "rb") as f:
         data = tomllib.load(f)
 
@@ -88,7 +95,7 @@ def parse_uv_lock(lock_path: Path) -> pd.DataFrame:
             }
         )
 
-    return pd.DataFrame(rows)
+    return DataFrame(rows)
 
 
 def uv_runtime_packages(
@@ -141,6 +148,8 @@ def uv_dep_info(
     By default, filters to show only 'main' dependencies from pyproject.toml.
     Helpful for looking at sizes of dependencies.
     """
+    import pandas as pd
+
     uv_lock_path = Path(uv_lock)
     pyproject_path = Path(pyproject)
 

kash/commands/workspace/selection_commands.py

@@ -2,6 +2,7 @@ from os.path import basename
 from pathlib import Path
 
 from frontmatter_format import fmf_strip_frontmatter
+from prettyfmt import plural
 from strif import copyfile_atomic
 
 from kash.config.logger import get_logger
@@ -11,7 +12,6 @@ from kash.model.paths_model import StorePath
 from kash.shell.ui.shell_results import shell_print_selection_history
 from kash.utils.common.format_utils import fmt_loc
 from kash.utils.errors import InvalidInput
-from kash.utils.lang_utils.inflection import plural
 from kash.workspaces import Selection, current_ws
 
 log = get_logger(__name__)

kash/commands/workspace/workspace_commands.py

@@ -2,7 +2,7 @@ import os
 from pathlib import Path
 
 from frontmatter_format import to_yaml_string
-from prettyfmt import fmt_lines
+from prettyfmt import fmt_lines, plural
 from rich.text import Text
 
 from kash.commands.base.basic_file_commands import trash
@@ -26,7 +26,7 @@ from kash.exec.action_registry import get_all_actions_defaults
 from kash.exec.fetch_url_metadata import fetch_url_metadata
 from kash.exec.precondition_checks import actions_matching_paths
 from kash.exec.precondition_registry import get_all_preconditions
-from kash.exec.preconditions import is_url_item
+from kash.exec.preconditions import is_url_resource
 from kash.exec_model.shell_model import ShellResult
 from kash.local_server.local_url_formatters import local_url_formatter
 from kash.media_base import media_tools
@@ -54,11 +54,11 @@ from kash.utils.common.format_utils import fmt_loc
 from kash.utils.common.obj_replace import remove_values
 from kash.utils.common.parse_key_vals import parse_key_value
 from kash.utils.common.type_utils import not_none
-from kash.utils.common.url import Url, is_url
+from kash.utils.common.url import Url, is_url, parse_http_url
 from kash.utils.errors import InvalidInput
 from kash.utils.file_formats.chat_format import tail_chat_history
 from kash.utils.file_utils.dir_info import is_nonempty_dir
-from kash.utils.lang_utils.inflection import plural
+from kash.utils.file_utils.file_formats_model import Format
 from kash.web_content.file_cache_utils import cache_file
 from kash.workspaces import (
     current_ws,
@@ -181,15 +181,15 @@ def cache_content(*urls_or_paths: str, refetch: bool = False) -> None:
     PrintHooks.spacer()
     for url_or_path in urls_or_paths:
         locator = resolve_locator_arg(url_or_path)
-        cache_path, was_cached = cache_file(locator, expiration_sec=expiration_sec)
-        cache_str = " (already cached)" if was_cached else ""
+        cache_result = cache_file(locator, expiration_sec=expiration_sec)
+        cache_str = " (already cached)" if cache_result.was_cached else ""
         cprint(f"{fmt_loc(url_or_path)}{cache_str}:", style=STYLE_EMPH, text_wrap=Wrap.NONE)
-        cprint(f"{cache_path}", text_wrap=Wrap.INDENT_ONLY)
+        cprint(f"{cache_result.content.path}", text_wrap=Wrap.INDENT_ONLY)
         PrintHooks.spacer()
 
 
 @kash_command
-def download(*urls_or_paths: str, refetch: bool = False) -> None:
+def download(*urls_or_paths: str, refetch: bool = False) -> ShellResult:
     """
     Download a URL or resource. Uses cached content if available, unless `refetch` is true.
     Inputs can be URLs or paths to URL resources.
@@ -198,31 +198,49 @@ def download(*urls_or_paths: str, refetch: bool = False) -> None:
 
     # TODO: Add option to include frontmatter metadata for text files.
     ws = current_ws()
+    saved_paths = []
     for url_or_path in urls_or_paths:
         locator = resolve_locator_arg(url_or_path)
-        url = None
+        url: Url | None = None
         if not isinstance(locator, Path) and is_url(locator):
             url = Url(locator)
         if isinstance(locator, StorePath):
             url_item = ws.load(locator)
-            if is_url_item(url_item):
+            if is_url_resource(url_item):
                 url = url_item.url
         if not url:
             raise InvalidInput(f"Not a URL or URL resource: {fmt_loc(locator)}")
 
         if is_media_url(url):
-            store_path = ws.import_item(locator, as_type=ItemType.resource)
             log.message(
-                "URL is a media URL, so added as a resource and will cache media: %s", fmt_loc(url)
+                "URL is a media URL, so adding as a resource and will cache media: %s", fmt_loc(url)
             )
+            store_path = ws.import_item(locator, as_type=ItemType.resource)
             media_tools.cache_media(url)
         else:
             log.message("Will cache file and save to workspace: %s", fmt_loc(url))
-            cache_path, _was_cached = cache_file(url, expiration_sec=expiration_sec)
-            item = Item.from_external_path(cache_path, item_type=ItemType.resource)
+            original_filename = Path(parse_http_url(url).path).name
+            cache_result = cache_file(url, expiration_sec=expiration_sec)
+            # If available, use the mime type to help set item file extension.
+            mime_type = cache_result.content.headers and cache_result.content.headers.mime_type
+            item = Item.from_external_path(
+                cache_result.content.path,
+                ItemType.resource,
+                mime_type=mime_type,
+                original_filename=original_filename,
+            )
             store_path = ws.save(item)
+            saved_paths.append(store_path)
+
+    print_status(
+        "Downloaded %s %s:\n%s",
+        len(urls_or_paths),
+        plural("item", len(urls_or_paths)),
+        fmt_lines(saved_paths),
+    )
+    select(*saved_paths)
 
-        log.info("Saved item to workspace: %s", fmt_loc(store_path))
+    return ShellResult(show_selection=True)
 
 
 @kash_command
@@ -437,7 +455,7 @@ def import_item(
     *files_or_urls: str, type: ItemType | None = None, inplace: bool = False
 ) -> ShellResult:
     """
-    Add a file or URL resource to the workspace as an item, with associated metadata.
+    Add a file or URL resource to the workspace as an item.
 
     :param inplace: If set and the item is already in the store, reimport the item,
       adding or rewriting metadata frontmatter.
@@ -464,6 +482,34 @@ def import_item(
     return ShellResult(show_selection=True)
 
 
+@kash_command
+def import_clipboard(
+    title: str | None = "pasted_text",
+    type: ItemType = ItemType.resource,
+    format: Format = Format.plaintext,
+) -> ShellResult:
+    """
+    Import the contents of the OS-native clipboard as a new item in the workspace.
+
+    :param title: The title of the new item (default: "pasted_text").
+    :param type: The type of the new item (default: resource).
+    :param format: The format of the new item (default: plaintext).
+    """
+    import pyperclip
+
+    contents = pyperclip.paste()
+    if not contents.strip():
+        raise InvalidInput("Clipboard is empty")
+
+    ws = current_ws()
+    store_path = ws.save(Item(type=type, format=format, title=title, body=contents))
+
+    print_status("Imported clipboard contents to:\n%s", fmt_lines([fmt_loc(store_path)]))
+    select(store_path)
+
+    return ShellResult(show_selection=True)
+
+
 @kash_command
 def fetch_metadata(*files_or_urls: str, refetch: bool = False) -> ShellResult:
     """

kash/config/env_settings.py

@@ -1,10 +1,7 @@
-import os
-from enum import Enum
-from pathlib import Path
-from typing import overload
+from clideps.env_vars.env_enum import EnvEnum
 
 
-class KashEnv(str, Enum):
+class KashEnv(EnvEnum):
     """
     Environment variable settings for kash. None are required, but these may be
     used to override default values.
@@ -33,40 +30,3 @@ class KashEnv(str, Enum):
 
     KASH_USER_AGENT = "KASH_USER_AGENT"
     """The user agent to use for HTTP requests."""
-
-    @overload
-    def read_str(self) -> str | None: ...
-
-    @overload
-    def read_str(self, default: str) -> str: ...
-
-    def read_str(self, default: str | None = None) -> str | None:
-        """
-        Get the value of the environment variable from the environment (with
-        optional default).
-        """
-        return os.environ.get(self.value, default)
-
-    @overload
-    def read_path(self) -> Path | None: ...
-
-    @overload
-    def read_path(self, default: Path) -> Path: ...
-
-    def read_path(self, default: Path | None = None) -> Path | None:
-        """
-        Get the value of the environment variable as a resolved path (with
-        optional default).
-        """
-        value = os.environ.get(self.value)
-        if value:
-            return Path(value).expanduser().resolve()
-        else:
-            return default.expanduser().resolve() if default else None
-
-    def read_bool(self, default: bool = False) -> bool:
-        """
-        Get the value of the environment variable as a boolean.
-        """
-        value = str(os.environ.get(self.value, default) or "").lower()
-        return bool(value and value != "0" and value != "false" and value != "no")