Coverage for sources/mimeogram/create.py: 72%
67 statements
« prev ^ index » next coverage.py v7.6.12, created at 2025-03-03 00:13 +0000
1# vim: set filetype=python fileencoding=utf-8:
2# -*- coding: utf-8 -*-
4#============================================================================#
5# #
6# Licensed under the Apache License, Version 2.0 (the "License"); #
7# you may not use this file except in compliance with the License. #
8# You may obtain a copy of the License at #
9# #
10# http://www.apache.org/licenses/LICENSE-2.0 #
11# #
12# Unless required by applicable law or agreed to in writing, software #
13# distributed under the License is distributed on an "AS IS" BASIS, #
14# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. #
15# See the License for the specific language governing permissions and #
16# limitations under the License. #
17# #
18#============================================================================#
21''' Creation of mimeograms. '''
22# TODO? Use BSD sysexits.
25from __future__ import annotations
27from . import __
28from . import interfaces as _interfaces
29from . import tokenizers as _tokenizers
32_scribe = __.produce_scribe( __name__ )
35class Command(
36 _interfaces.CliCommand,
37 decorators = ( __.standard_dataclass, __.standard_tyro_class ),
38):
39 ''' Creates mimeogram from filesystem locations or URLs. '''
41 sources: __.typx.Annotated[
42 __.tyro.conf.Positional[ list[ str ] ],
43 __.tyro.conf.arg(
44 help = "Filesystem locations or URLs.",
45 prefix_name = False ),
46 ]
47 clip: __.typx.Annotated[
48 __.typx.Optional[ bool ],
49 __.tyro.conf.arg(
50 aliases = ( '--clipboard', '--to-clipboard' ),
51 help = "Copy mimeogram to clipboard." ),
52 ] = None
53 count_tokens: __.typx.Annotated[
54 __.typx.Optional[ bool ],
55 __.tyro.conf.arg(
56 help = "Count total tokens in mimeogram." ),
57 ] = None
58 edit: __.typx.Annotated[
59 bool,
60 __.tyro.conf.arg(
61 aliases = ( '-e', '--edit-message' ),
62 help = "Spawn editor to capture an introductory message." ),
63 ] = False
64 prepend_prompt: __.typx.Annotated[
65 bool,
66 __.tyro.conf.arg(
67 help = "Prepend mimeogram format instructions." ),
68 ] = False
69 recurse: __.typx.Annotated[
70 __.typx.Optional[ bool ],
71 __.tyro.conf.arg(
72 aliases = ( '-r', '--recurse-directories', '--recursive' ),
73 help = "Recurse into directories." ),
74 ] = None
75 strict: __.typx.Annotated[
76 __.typx.Optional[ bool ],
77 __.tyro.conf.arg(
78 aliases = ( '--fail-on-invalid', ),
79 help = "Fail on invalid contents? True, fail. False, skip." ),
80 ] = None
81 tokenizer: __.typx.Annotated[
82 __.typx.Optional[ _tokenizers.Tokenizers ],
83 __.tyro.conf.arg(
84 help = "Which tokenizer to use for counting?" ),
85 ] = None
86 tokenizer_variant: __.typx.Annotated[
87 __.typx.Optional[ str ],
88 __.tyro.conf.arg(
89 help = (
90 "Which tokenizer variant to use for counting?\n"
91 "For 'tiktoken': 'cl100k_base', 'o200k_base', etc....\n"
92 "Not all tokenizers have variants.\n"
93 "If not specified, then the default variant is used." ) ),
94 ] = None
96 async def __call__( self, auxdata: __.Globals ) -> None:
97 ''' Executes command to create mimeogram. '''
98 await create( auxdata, self )
100 def provide_configuration_edits( self ) -> __.DictionaryEdits:
101 ''' Provides edits against configuration from options. '''
102 edits: list[ __.DictionaryEdit ] = [ ]
103 if None is not self.clip:
104 edits.append( __.SimpleDictionaryEdit( # pyright: ignore
105 address = ( 'create', 'to-clipboard' ), value = self.clip ) )
106 if None is not self.count_tokens: 106 ↛ 107line 106 didn't jump to line 107 because the condition on line 106 was never true
107 edits.append( __.SimpleDictionaryEdit( # pyright: ignore
108 address = ( 'create', 'count-tokens' ),
109 value = self.count_tokens ) )
110 if None is not self.recurse:
111 edits.append( __.SimpleDictionaryEdit( # pyright: ignore
112 address = ( 'acquire-parts', 'recurse-directories' ),
113 value = self.recurse ) )
114 if None is not self.strict: 114 ↛ 115line 114 didn't jump to line 115 because the condition on line 114 was never true
115 edits.append( __.SimpleDictionaryEdit( # pyright: ignore
116 address = ( 'acquire-parts', 'fail-on-invalid' ),
117 value = self.strict ) )
118 if None is not self.tokenizer: 118 ↛ 119line 118 didn't jump to line 119 because the condition on line 118 was never true
119 edits.append( __.SimpleDictionaryEdit( # pyright: ignore
120 address = ( 'tokenizers', 'default' ),
121 value = self.tokenizer ) )
122 return tuple( edits )
125async def _acquire_prompt( auxdata: __.Globals ) -> str:
126 from .prompt import acquire_prompt
127 return await acquire_prompt( auxdata )
130async def _copy_to_clipboard( mimeogram: str ) -> None:
131 from pyperclip import copy
132 copy( mimeogram )
133 _scribe.info( "Copied mimeogram to clipboard." )
136async def _edit_message( ) -> str:
137 from .edit import edit_content
138 return edit_content( )
141async def create( # pylint: disable=too-complex,too-many-locals
142 auxdata: __.Globals,
143 command: Command, *,
144 editor: __.cabc.Callable[
145 [ ], __.cabc.Coroutine[ None, None, str ] ] = _edit_message,
146 clipcopier: __.cabc.Callable[
147 [ str ], __.cabc.Coroutine[ None, None, None ] ] = _copy_to_clipboard,
148 prompter: __.cabc.Callable[
149 [ __.Globals ],
150 __.cabc.Coroutine[ None, None, str ] ] = _acquire_prompt,
151) -> __.typx.Never:
152 ''' Creates mimeogram. '''
153 from .acquirers import acquire
154 from .formatters import format_mimeogram
155 with __.report_exceptions(
156 _scribe, "Could not acquire mimeogram parts."
157 ): parts = await acquire( auxdata, command.sources )
158 if command.edit:
159 with __.report_exceptions(
160 _scribe, "Could not acquire user message."
161 ): message = await editor( )
162 else: message = None
163 mimeogram = format_mimeogram( parts, message = message )
164 # TODO? Pass prompt to 'format_mimeogram'.
165 if command.prepend_prompt:
166 prompt = await prompter( auxdata )
167 mimeogram = f"{prompt}\n\n{mimeogram}"
168 options = auxdata.configuration.get( 'create', { } )
169 if options.get( 'count-tokens', False ): 169 ↛ 170line 169 didn't jump to line 170 because the condition on line 169 was never true
170 with __.report_exceptions(
171 _scribe, "Could not count mimeogram tokens."
172 ):
173 tokenizer = await _tokenizer_from_command( auxdata, command )
174 tokens_count = await tokenizer.count( mimeogram )
175 _scribe.info( f"Total mimeogram size is {tokens_count} tokens." )
176 if options.get( 'to-clipboard', False ):
177 with __.report_exceptions(
178 _scribe, "Could not copy mimeogram to clipboard."
179 ): await clipcopier( mimeogram )
180 else: print( mimeogram ) # TODO? Use output stream from configuration.
181 raise SystemExit( 0 )
184async def _tokenizer_from_command(
185 auxdata: __.Globals, command: Command
186) -> _tokenizers.Tokenizer:
187 options = auxdata.configuration.get( 'tokenizers', { } )
188 name = (
189 command.tokenizer.value if command.tokenizer
190 else options.get( 'default', 'tiktoken' ) )
191 variant = command.tokenizer_variant
192 args = dict( variant = variant ) if variant else { }
193 return await _tokenizers.Tokenizers.produce( name, **args )