glossary cmd / main: added --print option

Introduces several changes with the main objective of enabling OpenAI's streaming API in the chatmastermind application. This allows for the retrieval of AI responses gradually as a stream, which can significantly improve the user experience in interactions that involve large result sets.

* Added tiktoken import in 'openai.py' and modifications to the OpenAI class to support streaming. This includes the addition of a new class OpenAIAnswer to handle streaming API responses.
* Modified request function in the OpenAI class: the stream=True flag is added to the openai.ChatCompletion.create method to enable streaming API.
* Modified 'question.py' to print the answer parts as they are streamed.
* Replaced the Answer class's string data type with a generator which supports str and Generator[str, None, None] data types. Modifications are made to the Answer class methods to handle both data types accordingly.
* Updated the tests in 'test_ais_openai.py' and 'test_message.py' to reflect and validate these changes.

chatmastermind/commands/glossary.py

@@ -0,0 +1,95 @@
+import sys
+import argparse
+from pathlib import Path
+from pydoc import pager
+from ..configuration import Config
+from ..glossary import Glossary
+
+
+class GlossaryCmdError(Exception):
+    pass
+
+
+def print_paged(text: str) -> None:
+    pager(text)
+
+
+def get_glossary_file_path(name: str, config: Config) -> Path:
+    """
+    Get the complete filename for a glossary with the given path.
+    """
+    if not config.glossaries:
+        raise GlossaryCmdError("Can't create glossary name without a glossary directory")
+    return Path(config.glossaries, name).with_suffix(Glossary.file_suffix).absolute()
+
+
+def list_glossaries(args: argparse.Namespace, config: Config) -> None:
+    """
+    List existing glossaries in the 'glossaries' directory.
+    """
+    if not config.glossaries:
+        raise GlossaryCmdError("Glossaries directory missing in the configuration file")
+    glossaries = Path(config.glossaries).glob(f'*{Glossary.file_suffix}')
+    for glo in sorted(glossaries):
+        print(Glossary.from_file(glo).to_str())
+
+
+def print_glossary(args: argparse.Namespace, config: Config) -> None:
+    """
+    Print an existing glossary.
+    """
+    # sanity checks
+    if args.name is None:
+        raise GlossaryCmdError("Missing glossary name")
+    if config.glossaries is None and args.file is None:
+        raise GlossaryCmdError("Glossaries directory missing in the configuration file")
+    # create file path or use the given one
+    glo_file = Path(args.file) if args.file else get_glossary_file_path(args.name, config)
+    if not glo_file.exists():
+        raise GlossaryCmdError(f"Glossary '{glo_file}' does not exist")
+    # read glossary
+    glo = Glossary.from_file(glo_file)
+    print_paged(glo.to_str(with_entries=True))
+
+
+def create_glossary(args: argparse.Namespace, config: Config) -> None:
+    """
+    Create a new glossary and write it either to the glossaries directory
+    or the given file.
+    """
+    # sanity checks
+    if args.name is None:
+        raise GlossaryCmdError("Missing glossary name")
+    if args.source_lang is None:
+        raise GlossaryCmdError("Missing source language")
+    if args.target_lang is None:
+        raise GlossaryCmdError("Missing target language")
+    if config.glossaries is None and args.file is None:
+        raise GlossaryCmdError("Glossaries directory missing in the configuration file")
+    # create file path or use the given one
+    glo_file = Path(args.file) if args.file else get_glossary_file_path(args.name, config)
+    if glo_file.exists():
+        raise GlossaryCmdError(f"Glossary '{glo_file}' already exists")
+    glo = Glossary(name=args.name,
+                   source_lang=args.source_lang,
+                   target_lang=args.target_lang,
+                   desc=args.description,
+                   file_path=glo_file)
+    glo.to_file()
+    print(f"Successfully created new glossary '{glo_file}'.")
+
+
+def glossary_cmd(args: argparse.Namespace, config: Config) -> None:
+    """
+    Handler for the 'glossary' command.
+    """
+    try:
+        if args.create:
+            create_glossary(args, config)
+        elif args.list:
+            list_glossaries(args, config)
+        elif args.print:
+            print_glossary(args, config)
+    except GlossaryCmdError as err:
+        print(f"Error: {err}")
+        sys.exit(1)

chatmastermind/configuration.py

@@ -118,6 +118,7 @@ class Config:
     # a default configuration
     cache: str = '.'
     db: str = './db/'
+    glossaries: str | None = './glossaries/'
     ais: dict[str, AIConfig] = field(default_factory=create_default_ai_configs)
 
     @classmethod
@@ -135,7 +136,8 @@ class Config:
         return cls(
             cache=str(source['cache']) if 'cache' in source else '.',
             db=str(source['db']),
-            ais=ais
+            ais=ais,
+            glossaries=str(source['glossaries']) if 'glossaries' in source else None
         )
 
     @classmethod
@@ -148,6 +150,8 @@ class Config:
 
     @classmethod
     def from_file(cls: Type[ConfigInst], path: str) -> ConfigInst:
+        if not Path(path).exists():
+            raise ConfigError(f"Configuration file '{path}' not found. Use 'cmm config --create' to create one.")
         with open(path, 'r') as f:
             source = yaml.load(f, Loader=yaml.FullLoader)
         return cls.from_dict(source)

chatmastermind/glossary.py

@@ -30,9 +30,10 @@ class Glossary:
     """
     A glossary consists of the following parameters:
         - Name (freely selectable)
-        - Path (full file path)
+        - Path (full file path, suffix is automatically generated)
         - Source language
         - Target language
+        - Description (optional)
         - Entries (pairs of source lang and target lang terms)
         - ID (automatically generated / modified, required by DeepL)
     """
@@ -40,11 +41,16 @@ class Glossary:
     name: str
     source_lang: str
     target_lang: str
-    entries: dict[str, str] = field(default_factory=lambda: dict())
     file_path: Path | None = None
+    desc: str | None = None
+    entries: dict[str, str] = field(default_factory=lambda: dict())
     ID: str | None = None
     file_suffix: ClassVar[str] = '.glo'
 
+    def __post_init__(self) -> None:
+        # FIXME: check for valid languages
+        pass
+
     @classmethod
     def from_file(cls: Type[GlossaryInst], file_path: Path) -> GlossaryInst:
         """
@@ -56,15 +62,18 @@ class Glossary:
             raise GlossaryError(f"File type '{file_path.suffix}' is not supported")
         with open(file_path, "r") as fd:
             try:
-                data = yaml.load(fd, Loader=yaml.FullLoader)
-                # remove any quotes from the entries that YAML may have added while dumping
-                # (e. g. for special keywords like 'yes')
-                clean_entries = {key.strip('\"\' '): value for key, value in data['Entries'].items()}
+                # use BaseLoader so every entry is read as a string
+                # - disables automatic conversions
+                # - makes it possible to omit quoting for YAML keywords in entries (e. g. 'yes')
+                # - also correctly reads quoted entries
+                data = yaml.load(fd, Loader=yaml.BaseLoader)
+                clean_entries = data['Entries']
                 return cls(name=data['Name'],
                            source_lang=data['SourceLang'],
                            target_lang=data['TargetLang'],
-                           entries=clean_entries,
                            file_path=file_path,
+                           desc=data['Description'],
+                           entries=clean_entries,
                            ID=data['ID'] if data['ID'] != 'None' else None)
             except Exception:
                 raise GlossaryError(f"'{file_path}' does not contain a valid glossary")
@@ -86,6 +95,7 @@ class Glossary:
         with tempfile.NamedTemporaryFile(dir=self.file_path.parent, prefix=self.file_path.name, mode="w", delete=False) as temp_fd:
             temp_file_path = Path(temp_fd.name)
             data = {'Name': self.name,
+                    'Description': str(self.desc),
                     'ID': str(self.ID),
                     'SourceLang': self.source_lang,
                     'TargetLang': self.target_lang,
@@ -136,3 +146,20 @@ class Glossary:
                         self.entries[parts[0]] = parts[1]
         except Exception as e:
             raise GlossaryError(f"Error importing TSV: {e}")
+
+    def to_str(self, with_entries: bool = False) -> str:
+        """
+        Return the current glossary as a string.
+        """
+        output: list[str] = []
+        output.append(f'{self.name} (ID: {self.ID}):')
+        if self.desc and self.desc != 'None':
+            output.append('- ' + self.desc)
+        output.append(f'- Languages: {self.source_lang} -> {self.target_lang}')
+        if with_entries:
+            output.append('- Entries:')
+            for source, target in self.entries.items():
+                output.append(f'  {source} : {target}')
+        else:
+            output.append(f'- Entries: {len(self.entries)}')
+        return '\n'.join(output)

chatmastermind/main.py

@@ -3,11 +3,12 @@
 # vim: set fileencoding=utf-8 :
 
 import sys
+import os
 import argcomplete
 import argparse
 from pathlib import Path
 from typing import Any
-from .configuration import Config, default_config_file
+from .configuration import Config, default_config_file, ConfigError
 from .message import Message
 from .commands.question import question_cmd
 from .commands.tags import tags_cmd
@@ -15,6 +16,7 @@ from .commands.config import config_cmd
 from .commands.hist import hist_cmd
 from .commands.print import print_cmd
 from .commands.translation import translation_cmd
+from .commands.glossary import glossary_cmd
 from .chat import msg_location
 
 
@@ -53,7 +55,7 @@ def create_parser() -> argparse.ArgumentParser:
     ai_parser = argparse.ArgumentParser(add_help=False)
     ai_parser.add_argument('-A', '--AI', help='AI ID to use', metavar='AI_ID')
     ai_parser.add_argument('-M', '--model', help='Model to use', metavar='MODEL')
-    ai_parser.add_argument('-n', '--num-answers', help='Number of answers to request', type=int, default=1)
+    ai_parser.add_argument('-N', '--num-answers', help='Number of answers to request', type=int, default=1)
     ai_parser.add_argument('-m', '--max-tokens', help='Max. nr. of tokens', type=int)
     ai_parser.add_argument('-T', '--temperature', help='Temperature value', type=float)
 
@@ -139,23 +141,78 @@ def create_parser() -> argparse.ArgumentParser:
 
     # 'translation' command parser
     translation_cmd_parser = cmdparser.add_parser('translation', parents=[ai_parser, tag_parser],
-                                                  help="ask, create and repeat translations.",
+                                                  help="Ask, create and repeat translations.",
                                                   aliases=['t'])
     translation_cmd_parser.set_defaults(func=translation_cmd)
     translation_group = translation_cmd_parser.add_mutually_exclusive_group(required=True)
     translation_group.add_argument('-a', '--ask', nargs='+', help='Ask to translate the given text', metavar='TEXT')
     translation_group.add_argument('-c', '--create', nargs='+', help='Create a translation', metavar='TEXT')
     translation_group.add_argument('-r', '--repeat', nargs='*', help='Repeat a translation', metavar='MESSAGE')
-    translation_cmd_parser.add_argument('-S', '--source-lang', help="Source language", metavar="LANGUAGE", required=True)
-    translation_cmd_parser.add_argument('-T', '--target-lang', help="Target language", metavar="LANGUAGE", required=True)
-    translation_cmd_parser.add_argument('-G', '--glossaries', nargs='+', help="List of glossaries", metavar="GLOSSARY")
+    translation_cmd_parser.add_argument('-l', '--source-lang', help="Source language", metavar="LANGUAGE", required=True)
+    translation_cmd_parser.add_argument('-L', '--target-lang', help="Target language", metavar="LANGUAGE", required=True)
+    translation_cmd_parser.add_argument('-G', '--glossaries', nargs='+', help="List of glossary names", metavar="GLOSSARY")
     translation_cmd_parser.add_argument('-d', '--input-document', help="Document to translate", metavar="FILE")
     translation_cmd_parser.add_argument('-D', '--output-document', help="Path for the translated document", metavar="FILE")
 
+    # 'glossary' command parser
+    glossary_cmd_parser = cmdparser.add_parser('glossary', parents=[ai_parser],
+                                               help="Manage glossaries.",
+                                               aliases=['g'])
+    glossary_cmd_parser.set_defaults(func=glossary_cmd)
+    glossary_group = glossary_cmd_parser.add_mutually_exclusive_group(required=True)
+    glossary_group.add_argument('-c', '--create', help='Create a glossary', action='store_true')
+    glossary_cmd_parser.add_argument('-n', '--name', help="Glossary name (not ID)", metavar="NAME")
+    glossary_cmd_parser.add_argument('-l', '--source-lang', help="Source language", metavar="LANGUAGE")
+    glossary_cmd_parser.add_argument('-L', '--target-lang', help="Target language", metavar="LANGUAGE")
+    glossary_cmd_parser.add_argument('-f', '--file', help='File path of the goven glossary', metavar='GLOSSARY_FILE')
+    glossary_cmd_parser.add_argument('-D', '--description', help="Glossary description", metavar="DESCRIPTION")
+    glossary_group.add_argument('-i', '--list', help='List existing glossaries', action='store_true')
+    glossary_group.add_argument('-p', '--print', help='Print an existing glossary', action='store_true')
+
     argcomplete.autocomplete(parser)
     return parser
 
 
+def create_directories(config: Config) -> None:  # noqa: 11
+    """
+    Create the directories in the given configuration if they don't exist.
+    """
+    def make_dir(path: Path) -> None:
+        try:
+            os.makedirs(path.absolute())
+        except Exception as e:
+            print(f"Creating directory '{path.absolute()}' failed with: {e}")
+            sys.exit(1)
+    # Cache
+    cache_path = Path(config.cache)
+    if not cache_path.exists():
+        answer = input(f"Cache directory '{cache_path}' does not exist. Create it? [y/n]")
+        if answer.lower() in ['y', 'yes']:
+            make_dir(cache_path.absolute())
+        else:
+            print("Can't continue without a valid cache directory!")
+            sys.exit(1)
+    # DB
+    db_path = Path(config.db)
+    if not db_path.exists():
+        answer = input(f"DB directory '{db_path}' does not exist. Create it? [y/n]")
+        if answer.lower() in ['y', 'yes']:
+            make_dir(db_path.absolute())
+        else:
+            print("Can't continue without a valid DB directory!")
+            sys.exit(1)
+    # Glossaries
+    if config.glossaries:
+        glossaries_path = Path(config.glossaries)
+        if not glossaries_path.exists():
+            answer = input(f"Glossaries directory '{glossaries_path}' does not exist. Create it? [y/n]")
+            if answer.lower() in ['y', 'yes']:
+                make_dir(glossaries_path.absolute())
+            else:
+                print("Can't continue without a valid glossaries directory. Create it or remove it from the configuration.")
+                sys.exit(1)
+
+
 def main() -> int:
     parser = create_parser()
     args = parser.parse_args()
@@ -164,7 +221,12 @@ def main() -> int:
     if command.func == config_cmd:
         command.func(command)
     else:
-        config = Config.from_file(args.config)
+        try:
+            config = Config.from_file(args.config)
+        except ConfigError as err:
+            print(f"{err}")
+            return 1
+        create_directories(config)
         command.func(command, config)
 
     return 0

tests/test_configuration.py

@@ -71,11 +71,13 @@ class TestConfig(unittest.TestCase):
                     'frequency_penalty': 0.7,
                     'presence_penalty': 0.2
                 }
-            }
+            },
+            'glossaries': './glossaries/'
         }
         config = Config.from_dict(source_dict)
         self.assertEqual(config.cache, '.')
         self.assertEqual(config.db, './test_db/')
+        self.assertEqual(config.glossaries, './glossaries/')
         self.assertEqual(len(config.ais), 1)
         self.assertEqual(config.ais['myopenai'].name, 'openai')
         self.assertEqual(cast(OpenAIConfig, config.ais['myopenai']).system, 'Custom system')
@@ -105,6 +107,7 @@ class TestConfig(unittest.TestCase):
                     'frequency_penalty': 0.7,
                     'presence_penalty': 0.2
                 }
+                # omit glossaries, since it's optional
             }
         }
         with open(self.test_file.name, 'w') as f:
@@ -113,6 +116,8 @@ class TestConfig(unittest.TestCase):
         self.assertIsInstance(config, Config)
         self.assertEqual(config.cache, './test_cache/')
         self.assertEqual(config.db, './test_db/')
+        # missing 'glossaries' should result in 'None'
+        self.assertEqual(config.glossaries, None)
         self.assertEqual(len(config.ais), 1)
         self.assertIsInstance(config.ais['default'], AIConfig)
         self.assertEqual(cast(OpenAIConfig, config.ais['default']).system, 'Custom system')

tests/test_glossary.py

@@ -9,97 +9,136 @@ glossary_suffix: str = Glossary.file_suffix
 
 class TestGlossary(unittest.TestCase):
 
-    def test_from_file_valid_yaml(self) -> None:
-        # Prepare a temporary YAML file with valid content
+    def test_from_file_yaml_unquoted(self) -> None:
+        """
+        Test glossary creatiom from YAML with unquoted entries.
+        """
         with tempfile.NamedTemporaryFile('w', delete=False, suffix=glossary_suffix) as yaml_file:
             yaml_file.write("Name: Sample\n"
+                            "Description: A brief description\n"
                             "ID: '123'\n"
                             "SourceLang: en\n"
                             "TargetLang: es\n"
                             "Entries:\n"
                             "  hello: hola\n"
                             "  goodbye: adiós\n"
-                            "  'yes': sí\n")  # 'yes' is a YAML keyword and therefore quoted
+                            # 'yes' is a YAML keyword and would normally be quoted
+                            "  yes: sí\n"
+                            "  I'm going home: me voy a casa\n")
             yaml_file_path = Path(yaml_file.name)
-
+        # create and check valid glossary
         glossary = Glossary.from_file(yaml_file_path)
         self.assertEqual(glossary.name, "Sample")
+        self.assertEqual(glossary.desc, "A brief description")
+        self.assertEqual(glossary.ID, "123")
         self.assertEqual(glossary.source_lang, "en")
         self.assertEqual(glossary.target_lang, "es")
-        self.assertEqual(glossary.entries, {"hello": "hola", "goodbye": "adiós", "yes": "sí"})
+        self.assertEqual(glossary.entries, {"hello": "hola",
+                                            "goodbye": "adiós",
+                                            "yes": "sí",
+                                            "I'm going home": "me voy a casa"})
+        yaml_file_path.unlink()  # Remove the temporary file
+
+    def test_from_file_yaml_quoted(self) -> None:
+        """
+        Test glossary creatiom from YAML with quoted entries.
+        """
+        with tempfile.NamedTemporaryFile('w', delete=False, suffix=glossary_suffix) as yaml_file:
+            yaml_file.write("Name: Sample\n"
+                            "Description: A brief description\n"
+                            "ID: '123'\n"
+                            "SourceLang: en\n"
+                            "TargetLang: es\n"
+                            "Entries:\n"
+                            "  'hello': 'hola'\n"
+                            "  'goodbye': 'adiós'\n"
+                            "  'yes': 'sí'\n"
+                            "  \"I'm going home\": 'me voy a casa'\n")
+            yaml_file_path = Path(yaml_file.name)
+        # create and check valid glossary
+        glossary = Glossary.from_file(yaml_file_path)
+        self.assertEqual(glossary.name, "Sample")
+        self.assertEqual(glossary.desc, "A brief description")
+        self.assertEqual(glossary.ID, "123")
+        self.assertEqual(glossary.source_lang, "en")
+        self.assertEqual(glossary.target_lang, "es")
+        self.assertEqual(glossary.entries, {"hello": "hola",
+                                            "goodbye": "adiós",
+                                            "yes": "sí",
+                                            "I'm going home": "me voy a casa"})
         yaml_file_path.unlink()  # Remove the temporary file
 
     def test_to_file_writes_yaml(self) -> None:
         # Create glossary instance
-        glossary = Glossary(name="Test", source_lang="en", target_lang="fr", entries={"yes": "oui"})
+        glossary = Glossary(name="Test",
+                            desc="Test description",
+                            ID="666",
+                            source_lang="en",
+                            target_lang="fr",
+                            entries={"yes": "oui"})
 
-        with tempfile.NamedTemporaryFile('w', delete=False, suffix=glossary_suffix) as tmp_file:
+        with tempfile.NamedTemporaryFile('w', suffix=glossary_suffix) as tmp_file:
             file_path = Path(tmp_file.name)
             glossary.to_file(file_path)
-
-        with open(file_path, 'r') as file:
-            content = file.read()
-
-        self.assertIn("Name: Test", content)
-        self.assertIn("SourceLang: en", content)
-        self.assertIn("TargetLang: fr", content)
-        self.assertIn("Entries", content)
-        # 'yes' is a YAML keyword and therefore quoted
-        self.assertIn("'yes': oui", content)
-        file_path.unlink()  # Remove the temporary file
+            # read and check valid YAML
+            with open(file_path, 'r') as file:
+                content = file.read()
+            self.assertIn("Name: Test", content)
+            self.assertIn("Description: Test description", content)
+            self.assertIn("ID: '666'", content)
+            self.assertIn("SourceLang: en", content)
+            self.assertIn("TargetLang: fr", content)
+            self.assertIn("Entries", content)
+            # 'yes' is a YAML keyword and therefore quoted
+            self.assertIn("'yes': oui", content)
 
     def test_write_read_glossary(self) -> None:
         # Create glossary instance
         # -> use 'yes' in order to test if the YAML quoting is correctly removed when reading the file
         glossary_write = Glossary(name="Test", source_lang="en", target_lang="fr", entries={"yes": "oui"})
 
-        with tempfile.NamedTemporaryFile('w', delete=False, suffix=glossary_suffix) as tmp_file:
+        with tempfile.NamedTemporaryFile('w', suffix=glossary_suffix) as tmp_file:
             file_path = Path(tmp_file.name)
             glossary_write.to_file(file_path)
-
-        # create new instance from glossary file
-        glossary_read = Glossary.from_file(file_path)
-        self.assertEqual(glossary_write.name, glossary_read.name)
-        self.assertEqual(glossary_write.source_lang, glossary_read.source_lang)
-        self.assertEqual(glossary_write.target_lang, glossary_read.target_lang)
-        self.assertDictEqual(glossary_write.entries, glossary_read.entries)
-
-        file_path.unlink()  # Remove the temporary file
+            # create new instance from glossary file
+            glossary_read = Glossary.from_file(file_path)
+            self.assertEqual(glossary_write.name, glossary_read.name)
+            self.assertEqual(glossary_write.source_lang, glossary_read.source_lang)
+            self.assertEqual(glossary_write.target_lang, glossary_read.target_lang)
+            self.assertDictEqual(glossary_write.entries, glossary_read.entries)
 
     def test_import_export_csv(self) -> None:
         glossary = Glossary(name="Test", source_lang="en", target_lang="fr", entries={})
 
         # First export to CSV
-        with tempfile.NamedTemporaryFile('w', delete=False, suffix=glossary_suffix) as csvfile:
+        with tempfile.NamedTemporaryFile('w', suffix=glossary_suffix) as csvfile:
             csv_file_path = Path(csvfile.name)
-        glossary.entries = {"hello": "salut", "goodbye": "au revoir"}
-        glossary.export_csv(glossary.entries, csv_file_path)
+            glossary.entries = {"hello": "salut", "goodbye": "au revoir"}
+            glossary.export_csv(glossary.entries, csv_file_path)
 
-        # Now import CSV
-        glossary.import_csv(csv_file_path)
-        self.assertEqual(glossary.entries, {"hello": "salut", "goodbye": "au revoir"})
-        csv_file_path.unlink()  # Remove the temporary file
+            # Now import CSV
+            glossary.import_csv(csv_file_path)
+            self.assertEqual(glossary.entries, {"hello": "salut", "goodbye": "au revoir"})
 
     def test_import_export_tsv(self) -> None:
         glossary = Glossary(name="Test", source_lang="en", target_lang="fr", entries={})
 
         # First export to TSV
-        with tempfile.NamedTemporaryFile('w', delete=False, suffix=glossary_suffix) as tsvfile:
+        with tempfile.NamedTemporaryFile('w', suffix=glossary_suffix) as tsvfile:
             tsv_file_path = Path(tsvfile.name)
-        glossary.entries = {"hello": "salut", "goodbye": "au revoir"}
-        glossary.export_tsv(glossary.entries, tsv_file_path)
+            glossary.entries = {"hello": "salut", "goodbye": "au revoir"}
+            glossary.export_tsv(glossary.entries, tsv_file_path)
 
-        # Now import TSV
-        glossary.import_tsv(tsv_file_path)
-        self.assertEqual(glossary.entries, {"hello": "salut", "goodbye": "au revoir"})
-        tsv_file_path.unlink()  # Remove the temporary file
+            # Now import TSV
+            glossary.import_tsv(tsv_file_path)
+            self.assertEqual(glossary.entries, {"hello": "salut", "goodbye": "au revoir"})
 
     def test_to_file_wrong_suffix(self) -> None:
         """
         Test for exception if suffix is wrong.
         """
         glossary = Glossary(name="Test", source_lang="en", target_lang="fr", entries={"yes": "oui"})
-        with tempfile.NamedTemporaryFile('w', delete=False, suffix='.wrong') as tmp_file:
+        with tempfile.NamedTemporaryFile('w', suffix='.wrong') as tmp_file:
             file_path = Path(tmp_file.name)
             with self.assertRaises(GlossaryError) as err:
                 glossary.to_file(file_path)
@@ -110,8 +149,61 @@ class TestGlossary(unittest.TestCase):
         Test if suffix is auto-generated if omitted.
         """
         glossary = Glossary(name="Test", source_lang="en", target_lang="fr", entries={"yes": "oui"})
-        with tempfile.NamedTemporaryFile('w', delete=False, suffix='') as tmp_file:
+        with tempfile.NamedTemporaryFile('w', suffix='') as tmp_file:
             file_path = Path(tmp_file.name)
             glossary.to_file(file_path)
-        assert glossary.file_path is not None
-        self.assertEqual(glossary.file_path.suffix, glossary_suffix)
+            assert glossary.file_path is not None
+            self.assertEqual(glossary.file_path.suffix, glossary_suffix)
+            # remove glossary file (differs from 'tmp_file' because of the added suffix
+            glossary.file_path.unlink()
+
+    def test_to_str_with_id(self) -> None:
+        # Create a Glossary instance with an ID
+        glossary_with_id = Glossary(name="TestGlossary", source_lang="en", target_lang="fr",
+                                    desc="A simple test glossary", ID="1001", entries={"one": "un"})
+        glossary_str = glossary_with_id.to_str()
+        self.assertIn("TestGlossary (ID: 1001):", glossary_str)
+        self.assertIn("- A simple test glossary", glossary_str)
+        self.assertIn("- Languages: en -> fr", glossary_str)
+        self.assertIn("- Entries: 1", glossary_str)
+
+    def test_to_str_with_id_and_entries(self) -> None:
+        # Create a Glossary instance with an ID and include entries
+        glossary_with_entries = Glossary(name="TestGlossaryWithEntries", source_lang="en", target_lang="fr",
+                                         desc="Another test glossary", ID="2002",
+                                         entries={"hello": "salut", "goodbye": "au revoir"})
+        glossary_str_with_entries = glossary_with_entries.to_str(with_entries=True)
+        self.assertIn("TestGlossaryWithEntries (ID: 2002):", glossary_str_with_entries)
+        self.assertIn("- Entries:", glossary_str_with_entries)
+        self.assertIn("  hello : salut", glossary_str_with_entries)
+        self.assertIn("  goodbye : au revoir", glossary_str_with_entries)
+
+    def test_to_str_without_id(self) -> None:
+        # Create a Glossary instance without an ID
+        glossary_without_id = Glossary(name="TestGlossaryNoID", source_lang="en", target_lang="fr",
+                                       desc="A test glossary without an ID", ID=None, entries={"yes": "oui"})
+        glossary_str_no_id = glossary_without_id.to_str()
+        self.assertIn("TestGlossaryNoID (ID: None):", glossary_str_no_id)
+        self.assertIn("- A test glossary without an ID", glossary_str_no_id)
+        self.assertIn("- Languages: en -> fr", glossary_str_no_id)
+        self.assertIn("- Entries: 1", glossary_str_no_id)
+
+    def test_to_str_without_id_and_no_entries(self) -> None:
+        # Create a Glossary instance without an ID and no entries
+        glossary_no_id_no_entries = Glossary(name="EmptyGlossary", source_lang="en", target_lang="fr",
+                                             desc="An empty test glossary", ID=None, entries={})
+        glossary_str_no_id_no_entries = glossary_no_id_no_entries.to_str()
+        self.assertIn("EmptyGlossary (ID: None):", glossary_str_no_id_no_entries)
+        self.assertIn("- An empty test glossary", glossary_str_no_id_no_entries)
+        self.assertIn("- Languages: en -> fr", glossary_str_no_id_no_entries)
+        self.assertIn("- Entries: 0", glossary_str_no_id_no_entries)
+
+    def test_to_str_no_description(self) -> None:
+        # Create a Glossary instance with an ID
+        glossary_with_id = Glossary(name="TestGlossary", source_lang="en", target_lang="fr",
+                                    ID="1001", entries={"one": "un"})
+        glossary_str = glossary_with_id.to_str()
+        expected_str = """TestGlossary (ID: 1001):
+- Languages: en -> fr
+- Entries: 1"""
+        self.assertEqual(expected_str, glossary_str)

tests/test_glossary_cmd.py

@@ -0,0 +1,149 @@
+import unittest
+import argparse
+import tempfile
+import io
+from contextlib import redirect_stdout
+from chatmastermind.configuration import Config
+from chatmastermind.commands.glossary import (
+    Glossary,
+    GlossaryCmdError,
+    glossary_cmd,
+    get_glossary_file_path,
+    create_glossary,
+    print_glossary,
+    list_glossaries
+)
+
+
+class TestGlossaryCmdNoGlossaries(unittest.TestCase):
+
+    def setUp(self) -> None:
+        # create DB and cache
+        self.db_dir = tempfile.TemporaryDirectory()
+        self.cache_dir = tempfile.TemporaryDirectory()
+        self.glossaries_dir = tempfile.TemporaryDirectory()
+        # create configuration
+        self.config = Config()
+        self.config.cache = self.cache_dir.name
+        self.config.db = self.db_dir.name
+        self.config.glossaries = self.glossaries_dir.name
+        # create a mock argparse.Namespace
+        self.args = argparse.Namespace(
+            create=True,
+            list=False,
+            print=False,
+            name='new_glossary',
+            file=None,
+            source_lang='en',
+            target_lang='de',
+            description=False,
+        )
+
+    def test_glossary_create_no_glossaries_err(self) -> None:
+        self.config.glossaries = None
+        with self.assertRaises(GlossaryCmdError) as err:
+            create_glossary(self.args, self.config)
+            self.assertIn(str(err.exception).lower(), "glossaries directory missing")
+
+    def test_glossary_create_no_name_err(self) -> None:
+        self.args.name = None
+        with self.assertRaises(GlossaryCmdError) as err:
+            create_glossary(self.args, self.config)
+            self.assertIn(str(err.exception).lower(), "missing glossary name")
+
+    def test_glossary_create_no_source_lang_err(self) -> None:
+        self.args.source_lang = None
+        with self.assertRaises(GlossaryCmdError) as err:
+            create_glossary(self.args, self.config)
+            self.assertIn(str(err.exception).lower(), "missing source language")
+
+    def test_glossary_create_no_target_lang_err(self) -> None:
+        self.args.target_lang = None
+        with self.assertRaises(GlossaryCmdError) as err:
+            create_glossary(self.args, self.config)
+            self.assertIn(str(err.exception).lower(), "missing target language")
+
+    def test_glossary_print_no_name_err(self) -> None:
+        self.args.name = None
+        with self.assertRaises(GlossaryCmdError) as err:
+            print_glossary(self.args, self.config)
+            self.assertIn(str(err.exception).lower(), "missing glossary name")
+
+    def test_glossary_list_no_glossaries_err(self) -> None:
+        self.config.glossaries = None
+        with self.assertRaises(GlossaryCmdError) as err:
+            list_glossaries(self.args, self.config)
+            self.assertIn(str(err.exception).lower(), "glossaries directory missing")
+
+    def test_glossary_create(self) -> None:
+        self.args.create = True
+        self.args.list = False
+        self.args.print = False
+        glossary_cmd(self.args, self.config)
+        expected_path = get_glossary_file_path(self.args.name, self.config)
+        glo = Glossary.from_file(expected_path)
+        self.assertEqual(glo.name, self.args.name)
+        expected_path.unlink()
+
+    def test_glossary_create_twice_err(self) -> None:
+        self.args.create = True
+        self.args.list = False
+        self.args.print = False
+        glossary_cmd(self.args, self.config)
+        expected_path = get_glossary_file_path(self.args.name, self.config)
+        glo = Glossary.from_file(expected_path)
+        self.assertEqual(glo.name, self.args.name)
+        # create glossary with the same name again
+        with self.assertRaises(GlossaryCmdError) as err:
+            create_glossary(self.args, self.config)
+            self.assertIn(str(err.exception).lower(), "already exists")
+        expected_path.unlink()
+
+
+class TestGlossaryCmdWithGlossaries(unittest.TestCase):
+
+    def setUp(self) -> None:
+        # create DB and cache
+        self.db_dir = tempfile.TemporaryDirectory()
+        self.cache_dir = tempfile.TemporaryDirectory()
+        self.glossaries_dir = tempfile.TemporaryDirectory()
+        # create configuration
+        self.config = Config()
+        self.config.cache = self.cache_dir.name
+        self.config.db = self.db_dir.name
+        self.config.glossaries = self.glossaries_dir.name
+        # create a mock argparse.Namespace
+        self.args = argparse.Namespace(
+            create=True,
+            list=False,
+            print=False,
+            name='Glossary1',
+            file=None,
+            source_lang='en',
+            target_lang='de',
+            description=False,
+        )
+        # create Glossary1
+        glossary_cmd(self.args, self.config)
+        self.Glossary1_path = get_glossary_file_path('Glossary1', self.config)
+        # create Glossary2
+        self.args.name = 'Glossary2'
+        glossary_cmd(self.args, self.config)
+        self.Glossary2_path = get_glossary_file_path('Glossary2', self.config)
+
+    def test_glossaries_exist(self) -> None:
+        """
+        Test if the default glossaries created in setUp exist.
+        """
+        glo = Glossary.from_file(self.Glossary1_path)
+        self.assertEqual(glo.name, 'Glossary1')
+        glo = Glossary.from_file(self.Glossary2_path)
+        self.assertEqual(glo.name, 'Glossary2')
+
+    def test_glossaries_list(self) -> None:
+        self.args.create = False
+        self.args.list = True
+        with redirect_stdout(io.StringIO()) as list_output:
+            glossary_cmd(self.args, self.config)
+            self.assertIn('Glossary1', list_output.getvalue())
+            self.assertIn('Glossary2', list_output.getvalue())