Merge branch 'main' into mixpanel

HISTORY.md

@@ -1,6 +1,19 @@
 
 # Release history
 
+### main branch
+
+- Infinite output for DeepSeek Coder, Mistral models in addition to Anthropic's models.
+- New `--deepseek` switch to use DeepSeek Coder.
+- New `--chat-mode <mode>` switch to launch in ask/help/code modes.
+- New `/code <message>` command request a code edit while in `ask` mode.
+- Web scraper is more robust if page never idles.
+- Improved token and cost reporting for infinite output.
+- Improvements and bug fixes for `/read` only files.
+- Bug fix to persist files added during `/ask`.
+- Bug fix for chat history size in `/tokens`.
+
+
 ### Aider v0.49.1
 
 - Bugfix to `/help`.

aider/args.py

@@ -109,6 +109,14 @@ def get_parser(default_config_files, git_root):
         const=gpt_3_model_name,
         help=f"Use {gpt_3_model_name} model for the main chat",
     )
+    deepseek_model = "deepseek/deepseek-coder"
+    group.add_argument(
+        "--deepseek",
+        action="store_const",
+        dest="model",
+        const=deepseek_model,
+        help=f"Use {deepseek_model} model for the main chat",
+    )
 
     ##########
     group = parser.add_argument_group("Model Settings")

aider/coders/base_coder.py

@@ -73,6 +73,9 @@ class Coder:
     multi_response_content = ""
     partial_response_content = ""
     commit_before_message = []
+    message_cost = 0.0
+    message_tokens_sent = 0
+    message_tokens_received = 0
 
     @classmethod
     def create(
@@ -149,7 +152,10 @@ class Coder:
         main_model = self.main_model
         weak_model = main_model.weak_model
         prefix = "Model:"
-        output = f" {main_model.name} with {self.edit_format} edit format"
+        output = f" {main_model.name} with"
+        if main_model.info.get("supports_assistant_prefill"):
+            output += " ♾️"
+        output += f" {self.edit_format} edit format"
         if weak_model is not main_model:
             prefix = "Models:"
             output += f", weak model {weak_model.name}"
@@ -993,7 +999,7 @@ class Coder:
                     return
                 except FinishReasonLength:
                     # We hit the output limit!
-                    if not self.main_model.can_prefill:
+                    if not self.main_model.info.get("supports_assistant_prefill"):
                         exhausted = True
                         break
 
@@ -1002,7 +1008,9 @@ class Coder:
                     if messages[-1]["role"] == "assistant":
                         messages[-1]["content"] = self.multi_response_content
                     else:
-                        messages.append(dict(role="assistant", content=self.multi_response_content))
+                        messages.append(
+                            dict(role="assistant", content=self.multi_response_content, prefix=True)
+                        )
                 except Exception as err:
                     self.io.tool_error(f"Unexpected error: {err}")
                     traceback.print_exc()
@@ -1017,8 +1025,7 @@ class Coder:
 
         self.io.tool_output()
 
-        if self.usage_report:
+        self.show_usage_report()
-            self.io.tool_output(self.usage_report)
 
         if exhausted:
             self.show_exhausted_error()
@@ -1241,7 +1248,6 @@ class Coder:
 
         self.io.log_llm_history("TO LLM", format_messages(messages))
 
-        interrupted = False
         try:
             hash_object, completion = send_completion(
                 model.name,
@@ -1258,9 +1264,9 @@ class Coder:
                 yield from self.show_send_output_stream(completion)
             else:
                 self.show_send_output(completion)
-        except KeyboardInterrupt:
+        except KeyboardInterrupt as kbi:
             self.keyboard_interrupt()
-            interrupted = True
+            raise kbi
         finally:
             self.io.log_llm_history(
                 "LLM RESPONSE",
@@ -1275,9 +1281,6 @@ class Coder:
                 if args:
                     self.io.ai_output(json.dumps(args, indent=4))
 
-        if interrupted:
-            raise KeyboardInterrupt
-
             self.calculate_and_show_tokens_and_cost(messages, completion)
 
     def show_send_output(self, completion):
@@ -1390,13 +1393,19 @@ class Coder:
             prompt_tokens = self.main_model.token_count(messages)
             completion_tokens = self.main_model.token_count(self.partial_response_content)
 
-        self.usage_report = f"Tokens: {prompt_tokens:,} sent, {completion_tokens:,} received."
+        self.message_tokens_sent += prompt_tokens
+        self.message_tokens_received += completion_tokens
+
+        tokens_report = (
+            f"Tokens: {self.message_tokens_sent:,} sent, {self.message_tokens_received:,} received."
+        )
 
         if self.main_model.info.get("input_cost_per_token"):
             cost += prompt_tokens * self.main_model.info.get("input_cost_per_token")
             if self.main_model.info.get("output_cost_per_token"):
                 cost += completion_tokens * self.main_model.info.get("output_cost_per_token")
             self.total_cost += cost
+            self.message_cost += cost
 
             def format_cost(value):
                 if value == 0:
@@ -1407,9 +1416,20 @@ class Coder:
                 else:
                     return f"{value:.{max(2, 2 - int(math.log10(magnitude)))}f}"
 
-            self.usage_report += (
+            cost_report = (
-                f" Cost: ${format_cost(cost)} request, ${format_cost(self.total_cost)} session."
+                f" Cost: ${format_cost(self.message_cost)} message,"
+                f" ${format_cost(self.total_cost)} session."
             )
+            self.usage_report = tokens_report + cost_report
+        else:
+            self.usage_report = tokens_report
+
+    def show_usage_report(self):
+        if self.usage_report:
+            self.io.tool_output(self.usage_report)
+            self.message_cost = 0.0
+            self.message_tokens_sent = 0
+            self.message_tokens_received = 0
 
         self.event(
             "message_send",

aider/main.py

@@ -423,11 +423,11 @@ def main(argv=None, input=None, output=None, force_git_root=None, return_coder=F
             return main(argv, input, output, right_repo_root, return_coder=return_coder)
 
     if args.just_check_update:
-        update_available = check_version(io, just_check=True)
+        update_available = check_version(io, just_check=True, verbose=args.verbose)
         return 0 if not update_available else 1
 
     if args.check_update:
-        check_version(io)
+        check_version(io, verbose=args.verbose)
 
     if args.models:
         models.print_matching_models(io, args.models)

aider/models.py

@@ -70,7 +70,6 @@ class ModelSettings:
     lazy: bool = False
     reminder_as_sys_msg: bool = False
     examples_as_sys_msg: bool = False
-    can_prefill: bool = False
     extra_headers: Optional[dict] = None
     max_tokens: Optional[int] = None
 
@@ -248,7 +247,6 @@ MODEL_SETTINGS = [
         weak_model_name="claude-3-haiku-20240307",
         use_repo_map=True,
         send_undo_reply=True,
-        can_prefill=True,
     ),
     ModelSettings(
         "openrouter/anthropic/claude-3-opus",
@@ -256,13 +254,11 @@ MODEL_SETTINGS = [
         weak_model_name="openrouter/anthropic/claude-3-haiku",
         use_repo_map=True,
         send_undo_reply=True,
-        can_prefill=True,
     ),
     ModelSettings(
         "claude-3-sonnet-20240229",
         "whole",
         weak_model_name="claude-3-haiku-20240307",
-        can_prefill=True,
     ),
     ModelSettings(
         "claude-3-5-sonnet-20240620",
@@ -270,7 +266,6 @@ MODEL_SETTINGS = [
         weak_model_name="claude-3-haiku-20240307",
         use_repo_map=True,
         examples_as_sys_msg=True,
-        can_prefill=True,
         accepts_images=True,
         max_tokens=8192,
         extra_headers={"anthropic-beta": "max-tokens-3-5-sonnet-2024-07-15"},
@@ -281,7 +276,6 @@ MODEL_SETTINGS = [
         weak_model_name="claude-3-haiku-20240307",
         use_repo_map=True,
         examples_as_sys_msg=True,
-        can_prefill=True,
         max_tokens=8192,
         extra_headers={
             "anthropic-beta": "max-tokens-3-5-sonnet-2024-07-15",
@@ -295,7 +289,6 @@ MODEL_SETTINGS = [
         weak_model_name="openrouter/anthropic/claude-3-haiku-20240307",
         use_repo_map=True,
         examples_as_sys_msg=True,
-        can_prefill=True,
         accepts_images=True,
         max_tokens=8192,
         extra_headers={
@@ -312,7 +305,6 @@ MODEL_SETTINGS = [
         weak_model_name="vertex_ai/claude-3-haiku@20240307",
         use_repo_map=True,
         examples_as_sys_msg=True,
-        can_prefill=True,
         accepts_images=True,
     ),
     ModelSettings(
@@ -321,13 +313,11 @@ MODEL_SETTINGS = [
         weak_model_name="vertex_ai/claude-3-haiku@20240307",
         use_repo_map=True,
         send_undo_reply=True,
-        can_prefill=True,
     ),
     ModelSettings(
         "vertex_ai/claude-3-sonnet@20240229",
         "whole",
         weak_model_name="vertex_ai/claude-3-haiku@20240307",
-        can_prefill=True,
     ),
     # Cohere
     ModelSettings(
@@ -486,14 +476,10 @@ class Model:
         if "gpt-3.5" in model or "gpt-4" in model:
             self.reminder_as_sys_msg = True
 
-        if "anthropic" in model:
-            self.can_prefill = True
-
         if "3.5-sonnet" in model or "3-5-sonnet" in model:
             self.edit_format = "diff"
             self.use_repo_map = True
             self.examples_as_sys_msg = True
-            self.can_prefill = True
 
         # use the defaults
         if self.edit_format == "diff":

aider/sendchat.py

@@ -57,6 +57,7 @@ def send_completion(
         temperature=temperature,
         stream=stream,
     )
+
     if functions is not None:
         kwargs["functions"] = functions
     if extra_headers is not None:

aider/versioncheck.py

@@ -10,12 +10,15 @@ from aider import utils
 from aider.dump import dump  # noqa: F401
 
 
-def check_version(io, just_check=False):
+def check_version(io, just_check=False, verbose=False):
     fname = Path.home() / ".aider" / "caches" / "versioncheck"
     if not just_check and fname.exists():
         day = 60 * 60 * 24
         since = time.time() - fname.stat().st_mtime
         if since < day:
+            if verbose:
+                hours = since / 60 / 60
+                io.tool_output(f"Too soon to check version: {hours:.1f} hours")
             return
 
     # To keep startup fast, avoid importing this unless needed
@@ -27,7 +30,7 @@ def check_version(io, just_check=False):
         latest_version = data["info"]["version"]
         current_version = aider.__version__
 
-        if just_check:
+        if just_check or verbose:
             io.tool_output(f"Current version: {current_version}")
             io.tool_output(f"Latest version: {latest_version}")
 
@@ -41,11 +44,13 @@ def check_version(io, just_check=False):
         fname.parent.mkdir(parents=True, exist_ok=True)
         fname.touch()
 
-    if just_check:
+    if just_check or verbose:
         if is_update_available:
             io.tool_output("Update available")
         else:
             io.tool_output("No update available")
+
+    if just_check:
         return is_update_available
 
     if not is_update_available:

aider/website/HISTORY.md

@@ -16,6 +16,19 @@ cog.out(text)
 
 # Release history
 
+### main branch
+
+- Infinite output for DeepSeek Coder, Mistral models in addition to Anthropic's models.
+- New `--deepseek` switch to use DeepSeek Coder.
+- New `--chat-mode <mode>` switch to launch in ask/help/code modes.
+- New `/code <message>` command request a code edit while in `ask` mode.
+- Web scraper is more robust if page never idles.
+- Improved token and cost reporting for infinite output.
+- Improvements and bug fixes for `/read` only files.
+- Bug fix to persist files added during `/ask`.
+- Bug fix for chat history size in `/tokens`.
+
+
 ### Aider v0.49.1
 
 - Bugfix to `/help`.

aider/website/assets/sample.aider.conf.yml

@@ -47,6 +47,9 @@
 ## Use gpt-3.5-turbo model for the main chat
 #35turbo: false
 
+## Use deepseek/deepseek-coder model for the main chat
+#deepseek: false
+
 #################
 # Model Settings:
 

aider/website/assets/sample.env

@@ -51,6 +51,9 @@
 ## Use gpt-3.5-turbo model for the main chat
 #AIDER_35TURBO=
 
+## Use deepseek/deepseek-coder model for the main chat
+#AIDER_DEEPSEEK=
+
 #################
 # Model Settings:
 

aider/website/docs/config/aider_conf.md

@@ -86,6 +86,9 @@ cog.outl("```")
 ## Use gpt-3.5-turbo model for the main chat
 #35turbo: false
 
+## Use deepseek/deepseek-coder model for the main chat
+#deepseek: false
+
 #################
 # Model Settings:
 

aider/website/docs/config/dotenv.md

@@ -93,6 +93,9 @@ cog.outl("```")
 ## Use gpt-3.5-turbo model for the main chat
 #AIDER_35TURBO=
 
+## Use deepseek/deepseek-coder model for the main chat
+#AIDER_DEEPSEEK=
+
 #################
 # Model Settings:
 

aider/website/docs/config/options.md

@@ -27,7 +27,7 @@ cog.out(get_md_help())
 ```
 usage: aider [-h] [--openai-api-key] [--anthropic-api-key] [--model]
              [--opus] [--sonnet] [--4] [--4o] [--mini] [--4-turbo]
-             [--35turbo] [--models] [--openai-api-base]
+             [--35turbo] [--deepseek] [--models] [--openai-api-base]
              [--openai-api-type] [--openai-api-version]
              [--openai-api-deployment-id] [--openai-organization-id]
              [--model-settings-file] [--model-metadata-file]
@@ -118,6 +118,10 @@ Aliases:
   - `--3`
   - `-3`
 
+### `--deepseek`
+Use deepseek/deepseek-coder model for the main chat  
+Environment variable: `AIDER_DEEPSEEK`  
+
 ## Model Settings:
 
 ### `--models MODEL`

aider/website/docs/llms/deepseek.md

@@ -6,7 +6,7 @@ nav_order: 500
 # DeepSeek
 
 Aider can connect to the DeepSeek.com API.
-The DeepSeek Coder V2 model gets the top score on aider's code editing benchmark.
+The DeepSeek Coder V2 model has a top score on aider's code editing benchmark.
 
 ```
 python -m pip install aider-chat
@@ -15,10 +15,6 @@ export DEEPSEEK_API_KEY=<key> # Mac/Linux
 setx   DEEPSEEK_API_KEY <key> # Windows, restart shell after setx
 
 # Use DeepSeek Coder V2
-aider --model deepseek/deepseek-coder
+aider --deepseek
 ```
 
-See the [model warnings](warnings.html)
-section for information on warnings which will occur
-when working with models that aider is not familiar with.
-

requirements.txt

@@ -6,7 +6,7 @@
 #
 aiohappyeyeballs==2.3.5
     # via aiohttp
-aiohttp==3.10.2
+aiohttp==3.10.3
     # via litellm
 aiosignal==1.3.1
     # via aiohttp
@@ -92,7 +92,7 @@ jsonschema==4.23.0
     #   litellm
 jsonschema-specifications==2023.12.1
     # via jsonschema
-litellm==1.43.4
+litellm==1.43.9
     # via -r requirements/requirements.in
 markdown-it-py==3.0.0
     # via rich
@@ -112,7 +112,7 @@ numpy==1.26.4
     # via
     #   -r requirements/requirements.in
     #   scipy
-openai==1.40.2
+openai==1.40.6
     # via litellm
 packaging==24.1
     # via
@@ -125,7 +125,9 @@ pathspec==0.12.1
 pillow==10.4.0
     # via -r requirements/requirements.in
 prompt-toolkit==3.0.47
-    # via -r requirements/requirements.in
+    # via
+    #   -r requirements/requirements.in
+    #   pypager
 pycodestyle==2.12.1
     # via flake8
 pycparser==2.22
@@ -139,7 +141,11 @@ pydantic-core==2.20.1
 pyflakes==3.2.0
     # via flake8
 pygments==2.18.0
-    # via rich
+    # via
+    #   pypager
+    #   rich
+pypager==3.0.1
+    # via -r requirements/requirements.in
 pypandoc==1.13
     # via -r requirements/requirements.in
 pyperclip==1.9.0
@@ -176,7 +182,7 @@ sniffio==1.3.1
     #   anyio
     #   httpx
     #   openai
-sounddevice==0.4.7
+sounddevice==0.5.0
     # via -r requirements/requirements.in
 soundfile==0.12.1
     # via -r requirements/requirements.in
@@ -210,5 +216,5 @@ wcwidth==0.2.13
     # via prompt-toolkit
 yarl==1.9.4
     # via aiohttp
-zipp==3.19.2
+zipp==3.20.0
     # via importlib-metadata

requirements/requirements-browser.txt

@@ -4,7 +4,7 @@
 #
 #    pip-compile --output-file=requirements/requirements-browser.txt requirements/requirements-browser.in
 #
-altair==5.3.0
+altair==5.4.0
     # via streamlit
 attrs==24.2.0
     # via
@@ -64,10 +64,11 @@ mdurl==0.1.2
     # via
     #   -c requirements/../requirements.txt
     #   markdown-it-py
+narwhals==1.3.0
+    # via altair
 numpy==1.26.4
     # via
     #   -c requirements/../requirements.txt
-    #   altair
     #   pandas
     #   pyarrow
     #   pydeck
@@ -78,9 +79,7 @@ packaging==24.1
     #   altair
     #   streamlit
 pandas==2.2.2
-    # via
+    # via streamlit
-    #   altair
-    #   streamlit
 pillow==10.4.0
     # via
     #   -c requirements/../requirements.txt
@@ -129,13 +128,12 @@ tenacity==8.5.0
     # via streamlit
 toml==0.10.2
     # via streamlit
-toolz==0.12.1
-    # via altair
 tornado==6.4.1
     # via streamlit
 typing-extensions==4.12.2
     # via
     #   -c requirements/../requirements.txt
+    #   altair
     #   streamlit
 tzdata==2024.1
     # via pandas
@@ -143,5 +141,5 @@ urllib3==2.2.2
     # via
     #   -c requirements/../requirements.txt
     #   requests
-watchdog==4.0.1
+watchdog==4.0.2
     # via -r requirements/requirements-browser.in

requirements/requirements-dev.txt

@@ -75,7 +75,7 @@ markupsafe==2.1.5
     # via
     #   -c requirements/../requirements.txt
     #   jinja2
-matplotlib==3.9.1.post1
+matplotlib==3.9.2
     # via -r requirements/requirements-dev.in
 mdurl==0.1.2
     # via

requirements/requirements-help.txt

@@ -8,7 +8,7 @@ aiohappyeyeballs==2.3.5
     # via
     #   -c requirements/../requirements.txt
     #   aiohttp
-aiohttp==3.10.2
+aiohttp==3.10.3
     # via
     #   -c requirements/../requirements.txt
     #   huggingface-hub
@@ -112,11 +112,11 @@ joblib==1.4.2
     # via
     #   nltk
     #   scikit-learn
-llama-index-core==0.10.63
+llama-index-core==0.10.65
     # via
     #   -r requirements/requirements-help.in
     #   llama-index-embeddings-huggingface
-llama-index-embeddings-huggingface==0.2.2
+llama-index-embeddings-huggingface==0.2.3
     # via -r requirements/requirements-help.in
 markupsafe==2.1.5
     # via
@@ -142,7 +142,7 @@ networkx==3.2.1
     #   -c requirements/../requirements.txt
     #   llama-index-core
     #   torch
-nltk==3.8.1
+nltk==3.8.2
     # via llama-index-core
 numpy==1.26.4
     # via
@@ -153,7 +153,7 @@ numpy==1.26.4
     #   scipy
     #   sentence-transformers
     #   transformers
-openai==1.40.2
+openai==1.40.6
     # via
     #   -c requirements/../requirements.txt
     #   llama-index-core
@@ -224,7 +224,7 @@ sqlalchemy[asyncio]==2.0.32
     # via
     #   llama-index-core
     #   sqlalchemy
-sympy==1.13.1
+sympy==1.13.2
     # via torch
 tenacity==8.5.0
     # via llama-index-core

requirements/requirements-playwright.txt

@@ -6,7 +6,7 @@
 #
 greenlet==3.0.3
     # via playwright
-playwright==1.45.1
+playwright==1.46.0
     # via -r requirements/requirements-playwright.in
 pyee==11.1.0
     # via playwright

tests/basic/test_main.py

@@ -224,6 +224,15 @@ class TestMain(TestCase):
 
                     main(["--yes", fname, "--encoding", "iso-8859-15"])
 
+    def test_main_exit_calls_version_check(self):
+        with GitTemporaryDirectory():
+            with patch("aider.main.check_version") as mock_check_version, patch(
+                "aider.main.InputOutput"
+            ) as mock_input_output:
+                main(["--exit"], input=DummyInput(), output=DummyOutput())
+                mock_check_version.assert_called_once()
+                mock_input_output.assert_called_once()
+
     @patch("aider.main.InputOutput")
     @patch("aider.coders.base_coder.Coder.run")
     def test_main_message_adds_to_input_history(self, mock_run, MockInputOutput):