Add Jinja2 as preparation for pipes.

- no longer using file suffix to choose the format
- added 'mformat' argument to 'write_xxx()' functions
- file suffix is now set by 'Message.to_file()' per default

.gitignore

@@ -106,6 +106,7 @@ celerybeat.pid
 .venv
 env/
 venv/
+.old/
 ENV/
 env.bak/
 venv.bak/

README.md

@@ -68,20 +68,21 @@ cmm question [-t OTAGS]... [-k ATAGS]... [-x XTAGS]... [-o OUTTAGS]... [-A AI_ID
 
 #### Hist
 
-The `hist` command is used to print the chat history.
+The `hist` command is used to print and manage the chat history.
 
 ```bash
-cmm hist [-t OTAGS]... [-k ATAGS]... [-x XTAGS]... [-w] [-W] [-S] [-A SUBSTRING] [-Q SUBSTRING]
+cmm hist [--print | --convert FORMAT] [-t OTAGS]... [-k ATAGS]... [-x XTAGS]... [-w] [-W] [-S] [-A SUBSTRING] [-Q SUBSTRING]
 ```
 
+* `-p, --print`: Print the DB chat history
+* `-c, --convert FORMAT`: Convert all messages to the given format
 * `-t, --or-tags OTAGS`: List of tags (one must match)
 * `-k, --and-tags ATAGS`: List of tags (all must match)
 * `-x, --exclude-tags XTAGS`: List of tags to exclude
-* `-w, --with-tags`: Print chat history with tags
+* `-w, --with-metadata`: Print chat history with metadata (tags, filenames, AI, etc.)
-* `-W, --with-files`: Print chat history with filenames
 * `-S, --source-code-only`: Only print embedded source code
-* `-A, --answer SUBSTRING`: Search for answer substring
+* `-A, --answer SUBSTRING`: Filter for answer substring
-* `-Q, --question SUBSTRING`: Search for question substring
+* `-Q, --question SUBSTRING`: Filter for question substring
 
 #### Tags
 

chatmastermind/chat.py

@@ -255,14 +255,17 @@ class Chat:
         return sum(m.tokens() for m in self.messages)
 
     def print(self, source_code_only: bool = False,
-              with_tags: bool = False, with_files: bool = False,
+              with_metadata: bool = False,
-              paged: bool = True) -> None:
+              paged: bool = True,
+              tight: bool = False) -> None:
         output: list[str] = []
         for message in self.messages:
             if source_code_only:
                 output.append(message.to_str(source_code_only=True))
                 continue
-            output.append(message.to_str(with_tags, with_files))
+            output.append(message.to_str(with_metadata))
+            if not tight:
+                output.append('\n' + ('-' * terminal_width()) + '\n')
         if paged:
             print_paged('\n'.join(output))
         else:

chatmastermind/commands/hist.py

@@ -1,13 +1,51 @@
+import sys
 import argparse
 from pathlib import Path
 from ..configuration import Config
 from ..chat import ChatDB
-from ..message import MessageFilter
+from ..message import MessageFilter, Message
 
 
-def hist_cmd(args: argparse.Namespace, config: Config) -> None:
+msg_suffix = Message.file_suffix_write  # currently '.msg'
+
+
+def convert_messages(args: argparse.Namespace, config: Config) -> None:
     """
-    Handler for the 'hist' command.
+    Convert messages to a new format. Also used to change old suffixes
+    ('.txt', '.yaml') to the latest default message file suffix ('.msg').
+    """
+    chat = ChatDB.from_dir(Path(config.cache),
+                           Path(config.db))
+    # read all known message files
+    msgs = chat.msg_gather(loc='disk', glob='*.*')
+    # make a set of all message IDs
+    msg_ids = set([m.msg_id() for m in msgs])
+    # set requested format and write all messages
+    chat.set_msg_format(args.convert)
+    # delete the current suffix
+    # -> a new one will automatically be created
+    for m in msgs:
+        if m.file_path:
+            m.file_path = m.file_path.with_suffix('')
+    chat.msg_write(msgs)
+    # read all messages with the current default suffix
+    msgs = chat.msg_gather(loc='disk', glob=f'*{msg_suffix}')
+    # make sure we converted all of the original messages
+    for mid in msg_ids:
+        if not any(mid == m.msg_id() for m in msgs):
+            print(f"Message '{mid}' has not been found after conversion. Aborting.")
+            sys.exit(1)
+    # delete messages with old suffixes
+    msgs = chat.msg_gather(loc='disk', glob='*.*')
+    for m in msgs:
+        if m.file_path and m.file_path.suffix != msg_suffix:
+            m.rm_file()
+    print(f"Successfully converted {len(msg_ids)} messages.")
+
+
+def print_chat(args: argparse.Namespace, config: Config) -> None:
+    """
+    Print the DB chat history.
     """
 
     mfilter = MessageFilter(tags_or=args.or_tags,
@@ -19,5 +57,16 @@ def hist_cmd(args: argparse.Namespace, config: Config) -> None:
                            Path(config.db),
                            mfilter=mfilter)
     chat.print(args.source_code_only,
-               args.with_tags,
+               args.with_metadata,
-               args.with_files)
+               paged=not args.no_paging,
+               tight=args.tight)
+
+
+def hist_cmd(args: argparse.Namespace, config: Config) -> None:
+    """
+    Handler for the 'hist' command.
+    """
+    if args.print:
+        print_chat(args, config)
+    elif args.convert:
+        convert_messages(args, config)

chatmastermind/commands/question.py

@@ -10,6 +10,10 @@ from ..ai_factory import create_ai
 from ..ai import AI, AIResponse
 
 
+class QuestionCmdError(Exception):
+    pass
+
+
 def add_file_as_text(question_parts: list[str], file: str) -> None:
     """
     Add the given file as plain text to the question part list.
@@ -80,7 +84,12 @@ def create_message(chat: ChatDB, args: argparse.Namespace) -> Message:
     to the cache directory.
     """
     question_parts = []
-    question_list = args.ask if args.ask is not None else []
+    if args.create is not None:
+        question_list = args.create
+    elif args.ask is not None:
+        question_list = args.ask
+    else:
+        raise QuestionCmdError("No question found")
     text_files = args.source_text if args.source_text is not None else []
     code_files = args.source_code if args.source_code is not None else []
 

chatmastermind/main.py

@@ -73,17 +73,20 @@ def create_parser() -> argparse.ArgumentParser:
 
     # 'hist' command parser
     hist_cmd_parser = cmdparser.add_parser('hist', parents=[tag_parser],
-                                           help="Print chat history.",
+                                           help="Print and manage chat history.",
                                            aliases=['h'])
     hist_cmd_parser.set_defaults(func=hist_cmd)
-    hist_cmd_parser.add_argument('-w', '--with-tags', help="Print chat history with tags.",
+    hist_group = hist_cmd_parser.add_mutually_exclusive_group(required=True)
-                                 action='store_true')
+    hist_group.add_argument('-p', '--print', help='Print the DB chat history', action='store_true')
-    hist_cmd_parser.add_argument('-W', '--with-files', help="Print chat history with filenames.",
+    hist_group.add_argument('-c', '--convert', help='Convert all message files to the given format [txt|yaml]', metavar='FORMAT')
+    hist_cmd_parser.add_argument('-w', '--with-metadata', help="Print chat history with metadata (tags, filename, AI, etc.).",
                                  action='store_true')
     hist_cmd_parser.add_argument('-S', '--source-code-only', help='Only print embedded source code',
                                  action='store_true')
-    hist_cmd_parser.add_argument('-A', '--answer', help='Search for answer substring', metavar='SUBSTRING')
+    hist_cmd_parser.add_argument('-A', '--answer', help='Print only answers with given substring', metavar='SUBSTRING')
-    hist_cmd_parser.add_argument('-Q', '--question', help='Search for question substring', metavar='SUBSTRING')
+    hist_cmd_parser.add_argument('-Q', '--question', help='Print only questions with given substring', metavar='SUBSTRING')
+    hist_cmd_parser.add_argument('-d', '--tight', help='Print without message separators', action='store_true')
+    hist_cmd_parser.add_argument('-P', '--no-paging', help='Print without paging', action='store_true')
 
     # 'tags' command parser
     tags_cmd_parser = cmdparser.add_parser('tags',

chatmastermind/message.py

@@ -422,7 +422,7 @@ class Message():
             except Exception:
                 raise MessageError(f"'{file_path}' does not contain a valid message")
 
-    def to_str(self, with_tags: bool = False, with_file: bool = False, source_code_only: bool = False) -> str:
+    def to_str(self, with_metadata: bool = False, source_code_only: bool = False) -> str:
         """
         Return the current Message as a string.
         """
@@ -432,10 +432,11 @@ class Message():
             if self.answer:
                 output.extend(self.answer.source_code(include_delims=True))
             return '\n'.join(output) if len(output) > 0 else ''
-        if with_tags:
+        if with_metadata:
             output.append(self.tags_str())
-        if with_file:
             output.append('FILE: ' + str(self.file_path))
+            output.append('AI: ' + str(self.ai))
+            output.append('MODEL: ' + str(self.model))
         output.append(Question.txt_header)
         output.append(self.question)
         if self.answer:

requirements.txt

@@ -2,3 +2,4 @@ openai
 PyYAML
 argcomplete
 pytest
+Jinja2

setup.py

@@ -2,6 +2,8 @@ from setuptools import setup, find_packages
 
 with open("README.md", "r", encoding="utf-8") as fh:
     long_description = fh.read()
+with open("requirements.txt", "r") as fh:
+    install_requirements = [line.strip() for line in fh]
 
 setup(
     name="ChatMastermind",
@@ -28,12 +30,7 @@ setup(
         "Topic :: Utilities",
         "Topic :: Text Processing",
     ],
-    install_requires=[
+    install_requires=install_requirements,
-        "openai",
-        "PyYAML",
-        "argcomplete",
-        "pytest",
-    ],
     python_requires=">=3.9",
     test_suite="tests",
     entry_points={

tests/test_chat.py

@@ -41,10 +41,14 @@ class TestChat(TestChatBase):
         self.message1 = Message(Question('Question 1'),
                                 Answer('Answer 1'),
                                 {Tag('atag1'), Tag('btag2')},
+                                ai='FakeAI',
+                                model='FakeModel',
                                 file_path=pathlib.Path(f'0001{msg_suffix}'))
         self.message2 = Message(Question('Question 2'),
                                 Answer('Answer 2'),
                                 {Tag('btag2')},
+                                ai='FakeAI',
+                                model='FakeModel',
                                 file_path=pathlib.Path(f'0002{msg_suffix}'))
         self.maxDiff = None
 
@@ -143,7 +147,7 @@ class TestChat(TestChatBase):
     @patch('sys.stdout', new_callable=StringIO)
     def test_print(self, mock_stdout: StringIO) -> None:
         self.chat.msg_add([self.message1, self.message2])
-        self.chat.print(paged=False)
+        self.chat.print(paged=False, tight=True)
         expected_output = f"""{Question.txt_header}
 Question 1
 {Answer.txt_header}
@@ -156,17 +160,21 @@ Answer 2
         self.assertEqual(mock_stdout.getvalue(), expected_output)
 
     @patch('sys.stdout', new_callable=StringIO)
-    def test_print_with_tags_and_file(self, mock_stdout: StringIO) -> None:
+    def test_print_with_metadata(self, mock_stdout: StringIO) -> None:
         self.chat.msg_add([self.message1, self.message2])
-        self.chat.print(paged=False, with_tags=True, with_files=True)
+        self.chat.print(paged=False, with_metadata=True, tight=True)
         expected_output = f"""{TagLine.prefix} atag1 btag2
 FILE: 0001{msg_suffix}
+AI: FakeAI
+MODEL: FakeModel
 {Question.txt_header}
 Question 1
 {Answer.txt_header}
 Answer 1
 {TagLine.prefix} btag2
 FILE: 0002{msg_suffix}
+AI: FakeAI
+MODEL: FakeModel
 {Question.txt_header}
 Question 2
 {Answer.txt_header}

tests/test_hist_cmd.py

@@ -0,0 +1,62 @@
+import unittest
+import argparse
+import tempfile
+import yaml
+from pathlib import Path
+from chatmastermind.message import Message, Question
+from chatmastermind.chat import ChatDB, ChatError
+from chatmastermind.configuration import Config
+from chatmastermind.commands.hist import convert_messages
+
+
+msg_suffix = Message.file_suffix_write
+
+
+class TestConvertMessages(unittest.TestCase):
+    def setUp(self) -> None:
+        self.db_dir = tempfile.TemporaryDirectory()
+        self.cache_dir = tempfile.TemporaryDirectory()
+        self.db_path = Path(self.db_dir.name)
+        self.cache_path = Path(self.cache_dir.name)
+        self.args = argparse.Namespace()
+        self.config = Config()
+        self.config.cache = self.cache_dir.name
+        self.config.db = self.db_dir.name
+        # Prepare some messages
+        self.chat = ChatDB.from_dir(Path(self.cache_path),
+                                    Path(self.db_path))
+        self.messages = [Message(Question(f'Question {i}')) for i in range(0, 6)]
+        self.chat.db_write(self.messages[0:2])
+        self.chat.cache_write(self.messages[2:])
+        # Change some of the suffixes
+        assert self.messages[0].file_path
+        assert self.messages[1].file_path
+        self.messages[0].file_path.rename(self.messages[0].file_path.with_suffix('.txt'))
+        self.messages[1].file_path.rename(self.messages[1].file_path.with_suffix('.yaml'))
+
+    def tearDown(self) -> None:
+        self.db_dir.cleanup()
+        self.cache_dir.cleanup()
+
+    def test_convert_messages(self) -> None:
+        self.args.convert = 'yaml'
+        convert_messages(self.args, self.config)
+        msgs = self.chat.msg_gather(loc='disk', glob='*.*')
+        # Check if the number of messages is the same as before
+        self.assertEqual(len(msgs), len(self.messages))
+        # Check if all messages have the requested suffix
+        for msg in msgs:
+            assert msg.file_path
+            self.assertEqual(msg.file_path.suffix, msg_suffix)
+        # Check if the message IDs are correctly maintained
+        for m_new, m_old in zip(msgs, self.messages):
+            self.assertEqual(m_new.msg_id(), m_old.msg_id())
+        # check if all messages have the new format
+        for m in msgs:
+            with open(str(m.file_path), "r") as fd:
+                yaml.load(fd, Loader=yaml.FullLoader)
+
+    def test_convert_messages_wrong_format(self) -> None:
+        self.args.convert = 'foo'
+        with self.assertRaises(ChatError):
+            convert_messages(self.args, self.config)

tests/test_message.py

@@ -856,6 +856,8 @@ class MessageToStrTestCase(unittest.TestCase):
     def setUp(self) -> None:
         self.message = Message(Question('This is a question.'),
                                Answer('This is an answer.'),
+                               ai=('FakeAI'),
+                               model=('FakeModel'),
                                tags={Tag('atag1'), Tag('btag2')},
                                file_path=pathlib.Path('/tmp/foo/bla'))
 
@@ -869,11 +871,13 @@ This is an answer."""
     def test_to_str_with_tags_and_file(self) -> None:
         expected_output = f"""{TagLine.prefix} atag1 btag2
 FILE: /tmp/foo/bla
+AI: FakeAI
+MODEL: FakeModel
 {Question.txt_header}
 This is a question.
 {Answer.txt_header}
 This is an answer."""
-        self.assertEqual(self.message.to_str(with_tags=True, with_file=True), expected_output)
+        self.assertEqual(self.message.to_str(with_metadata=True), expected_output)
 
 
 class MessageRmFileTestCase(unittest.TestCase):

tests/test_question_cmd.py

@@ -41,6 +41,8 @@ class TestMessageCreate(TestWithFakeAI):
         self.args.AI = None
         self.args.model = None
         self.args.output_tags = None
+        self.args.ask = None
+        self.args.create = None
         # File 1 : no source code block, only text
         self.source_file1 = tempfile.NamedTemporaryFile(delete=False)
         self.source_file1_content = """This is just text.
@@ -204,6 +206,21 @@ It is embedded code
 """))
 
 
+class TestCreateOption(TestMessageCreate):
+
+    def test_message_file_created(self) -> None:
+        self.args.create = ["How does question --create work?"]
+        self.args.ask = None
+        cache_dir_files = self.message_list(self.cache_dir)
+        self.assertEqual(len(cache_dir_files), 0)
+        create_message(self.chat, self.args)
+        cache_dir_files = self.message_list(self.cache_dir)
+        self.assertEqual(len(cache_dir_files), 1)
+        message = Message.from_file(cache_dir_files[0])
+        self.assertIsInstance(message, Message)
+        self.assertEqual(message.question, Question("How does question --create work?"))  # type: ignore [union-attr]
+
+
 class TestQuestionCmd(TestWithFakeAI):
 
     def setUp(self) -> None: