toolslm 0.3.8__py3-none-any.whl → 0.3.10__py3-none-any.whl

toolslm/__init__.py

@@ -1 +1 @@
-__version__ = "0.3.8"
+__version__ = "0.3.10"

toolslm/funccall.py

@@ -5,7 +5,7 @@ __all__ = ['empty', 'custom_types', 'get_schema', 'python', 'mk_ns', 'call_func'
            'mk_tool']
 
 # %% ../01_funccall.ipynb
-import inspect, json
+import inspect, json, ast
 from collections import abc
 from fastcore.utils import *
 from fastcore.docments import docments
@@ -37,12 +37,20 @@ def _types(t:type)->tuple[str,Optional[str]]:
     else: return tmap.get(t.__name__, "object"), None
 
 # %% ../01_funccall.ipynb
-def _param(name, info):
-    "json schema parameter given `name` and `info` from docments full dict."
+def _param(
+    name, # param name
+    info, # dict from docments
+    evalable=False): # stringify defaults that can't be literal_eval'd?
+    "json schema parameter given `name` and `info` from docments full dict"
     paramt,itemt = _types(info.anno)
     pschema = dict(type=paramt, description=info.docment or "")
     if itemt: pschema["items"] = {"type": itemt}
-    if info.default is not empty: pschema["default"] = info.default
+    if info.default is not empty:
+        if evalable:
+            try: ast.literal_eval(repr(info.default))
+            except: pschema["default"] = str(info.default)
+            else: pschema["default"] = info.default
+        else: pschema["default"] = info.default
     return pschema
 
 # %% ../01_funccall.ipynb
@@ -90,9 +98,9 @@ def _handle_container(origin, args, defs):
     return None
 
 # %% ../01_funccall.ipynb
-def _process_property(name, obj, props, req, defs):
+def _process_property(name, obj, props, req, defs, evalable=False):
     "Process a single property of the schema"
-    p = _param(name, obj)
+    p = _param(name, obj, evalable=evalable)
     props[name] = p
     if obj.default is empty: req[name] = True
 
@@ -103,14 +111,14 @@ def _process_property(name, obj, props, req, defs):
         p.update(_handle_type(obj.anno, defs))
 
 # %% ../01_funccall.ipynb
-def _get_nested_schema(obj):
+def _get_nested_schema(obj, evalable=False):
     "Generate nested JSON schema for a class or function"
     d = docments(obj, full=True)
     props, req, defs = {}, {}, {}
 
     for n, o in d.items():
         if n != 'return' and n != 'self':
-            _process_property(n, o, props, req, defs)
+            _process_property(n, o, props, req, defs, evalable=evalable)
 
     tkw = {}
     if isinstance(obj, type): tkw['title']=obj.__name__
@@ -120,10 +128,14 @@ def _get_nested_schema(obj):
     return schema
 
 # %% ../01_funccall.ipynb
-def get_schema(f:Union[callable,dict], pname='input_schema')->dict:
+def get_schema(
+    f:Union[callable,dict], # Function to get schema for
+    pname='input_schema',   # Key name for parameters
+    evalable=False   # stringify defaults that can't be literal_eval'd?
+)->dict:
     "Generate JSON schema for a class, function, or method"
     if isinstance(f, dict): return f
-    schema = _get_nested_schema(f)
+    schema = _get_nested_schema(f, evalable=evalable)
     desc = f.__doc__
     assert desc, "Docstring missing!"
     d = docments(f, full=True)

toolslm/xml.py

@@ -102,48 +102,56 @@ def mk_doc(index:int,  # The document index
 def docs_xml(docs:list[str],  # The content of each document
              srcs:Optional[list]=None,  # URLs, filenames, etc; each one defaults to `md5(content)` if not provided
              prefix:bool=True, # Include Anthropic's suggested prose intro?
-             details:Optional[list]=None # Optional list of dicts with additional attrs for each doc
+             details:Optional[list]=None, # Optional list of dicts with additional attrs for each doc
+             title:str=None # Optional title attr for Documents element
             )->str:
     "Create an XML string containing `docs` in Anthropic's recommended format"
     pre = 'Here are some documents for you to reference for your task:\n\n' if prefix else ''
     if srcs is None: srcs = [None]*len(docs)
     if details is None: details = [{}]*len(docs)
     docs = (mk_doc(i+1, d, s, **kw) for i,(d,s,kw) in enumerate(zip(docs,srcs,details)))
-    return pre + to_xml(Documents(docs), do_escape=False)
+    kw = dict(title=title) if title else {}
+    return pre + to_xml(Documents(*docs, **kw), do_escape=False)
 
 # %% ../00_xml.ipynb
-def read_file(fname, out=True):
+def read_file(fname, out=True, max_size=None):
     "Read file content, converting notebooks to XML if needed"
     fname = Path(fname)
-    if fname.suffix == '.ipynb': return nb2xml(fname, out=out)
-    return fname.read_text()
+    if fname.suffix == '.ipynb': res = nb2xml(fname, out=out)
+    else: res = fname.read_text()
+    if max_size and len(res)>max_size: return f"[Skipped: {fname.name} exceeds {max_size} bytes]"
+    return res
 
 # %% ../00_xml.ipynb
 def files2ctx(
     fnames:list[Union[str,Path]], # List of file names to add to context
     prefix:bool=True, # Include Anthropic's suggested prose intro?
     out:bool=True, # Include notebook cell outputs?
-    srcs:Optional[list]=None # Use the labels instead of `fnames`
+    srcs:Optional[list]=None, # Use the labels instead of `fnames`
+    title:str=None, # Optional title attr for Documents element
+    max_size:int=None # Skip files larger than this (bytes)
 )->str: # XML for LM context
     "Convert files to XML context, handling notebooks"
     fnames = [Path(o) for o in fnames]
-    contents = [read_file(o, out=out) for o in fnames]
-    return docs_xml(contents, srcs or fnames, prefix=prefix)
+    contents = [read_file(o, out=out, max_size=max_size) for o in fnames]
+    return docs_xml(contents, srcs or fnames, prefix=prefix, title=title)
 
 # %% ../00_xml.ipynb
 @delegates(globtastic)
 def folder2ctx(
     folder:Union[str,Path],
-    prefix:bool=True,
-    out:bool=True,
-    include_base:bool=True,
+    prefix:bool=True, # Include Anthropic's suggested prose intro?
+    out:bool=True, # Include notebook cell outputs?
+    include_base:bool=True, # Include full path in src?
+    title:str=None, # Optional title attr for Documents element
+    max_size:int=100_000, # Skip files larger than this (bytes)
     **kwargs
 )->str:
     "Convert folder contents to XML context, handling notebooks"
     folder = Path(folder)
     fnames = globtastic(folder, **kwargs)
     srcs = fnames if include_base else [Path(f).relative_to(folder) for f in fnames]
-    return files2ctx(fnames, prefix=prefix, out=out, srcs=srcs)
+    return files2ctx(fnames, prefix=prefix, out=out, srcs=srcs, title=title, max_size=max_size)
 
 # %% ../00_xml.ipynb
 @delegates(folder2ctx)
@@ -158,11 +166,15 @@ def repo2ctx(
     api = GhApi()
     if ref is None: ref = api.repos.get(owner, repo).default_branch
     data = api.repos.download_tarball_archive(owner, repo, ref)
+    parts = ' | '.join(f"{k}: {', '.join(v) if isinstance(v, (list,tuple)) else v}"
+        for k,v in kwargs.items() if v)
+    title = f"GitHub repository contents from {owner}/{repo} at ref '{ref}'"
+    if parts: title += f" (filters applied: {parts})"
     tf = tarfile.open(fileobj=io.BytesIO(data))
     with tempfile.TemporaryDirectory() as tmp:
         tf.extractall(tmp, filter='data')
         subdir = Path(tmp) / tf.getmembers()[0].name.split('/')[0]
-        return folder2ctx(subdir, include_base=False, **kwargs)
+        return folder2ctx(subdir, include_base=False, title=title, **kwargs)
 
 # %% ../00_xml.ipynb
 @call_parse

{toolslm-0.3.8.dist-info → toolslm-0.3.10.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: toolslm
-Version: 0.3.8
+Version: 0.3.10
 Summary: Tools to make language models a bit easier to use
 Home-page: https://github.com/AnswerDotAI/toolslm
 Author: Jeremy Howard

toolslm-0.3.10.dist-info/RECORD

@@ -0,0 +1,13 @@
+toolslm/__init__.py,sha256=h9TycTJK2pK49s87IMbNRq4lTqRt3xctcJl2jxCe3sU,23
+toolslm/_modidx.py,sha256=kpgsDpj-Tvn90wezrHaMttyzhNcyNVgw_dQgK10qotI,5308
+toolslm/download.py,sha256=g3BxUSxylC_575M7RFSJ1GI3Co3EwPDdEeWzxaf2Czk,4451
+toolslm/funccall.py,sha256=KO08GspCaVROb5Z29cxHRz3MaZdTUFj4j08sbbEyUSs,10968
+toolslm/md_hier.py,sha256=r_NPezhgfxjRmSYFlu_ND42hXt1qSbaPWHTcjbviOn4,11010
+toolslm/shell.py,sha256=dGInuRKvexu21VmtZkw_0S3BGiTsbAongUG-yG4YHpc,1566
+toolslm/xml.py,sha256=tAHoqXrTRiX8i3pR-9KpHoBb8QXJ_TKEVyTEOPviudE,8095
+toolslm-0.3.10.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+toolslm-0.3.10.dist-info/METADATA,sha256=pqfkL4TIn_jnIpuL06aWoCO5WW8k1yorccDYdnHPpfg,2404
+toolslm-0.3.10.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+toolslm-0.3.10.dist-info/entry_points.txt,sha256=xFz0Eymlo5X7BGpaO6DI9gMxvN5A7faebzrlr8ctp5I,95
+toolslm-0.3.10.dist-info/top_level.txt,sha256=4hRTrFWayz_Kz5221XjvlpCwVFrW3WPi1P0fllkTq9s,8
+toolslm-0.3.10.dist-info/RECORD,,

toolslm-0.3.8.dist-info/RECORD

@@ -1,13 +0,0 @@
-toolslm/__init__.py,sha256=7dTW0A5-FkrEuNOotvR8oW59M2lvIwYouVqfJzvXpKk,22
-toolslm/_modidx.py,sha256=kpgsDpj-Tvn90wezrHaMttyzhNcyNVgw_dQgK10qotI,5308
-toolslm/download.py,sha256=g3BxUSxylC_575M7RFSJ1GI3Co3EwPDdEeWzxaf2Czk,4451
-toolslm/funccall.py,sha256=0OBrx6KzI0KK13L-5Hn69yah9oZhgTsKchmMenCoT0A,10421
-toolslm/md_hier.py,sha256=r_NPezhgfxjRmSYFlu_ND42hXt1qSbaPWHTcjbviOn4,11010
-toolslm/shell.py,sha256=dGInuRKvexu21VmtZkw_0S3BGiTsbAongUG-yG4YHpc,1566
-toolslm/xml.py,sha256=TO3i6QD1g_ya8B7Wxwib2ZWv7pwVpfyaAalw1qrKb74,7148
-toolslm-0.3.8.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-toolslm-0.3.8.dist-info/METADATA,sha256=j4UPSM5q7rP16BlkIQzb8_4czu7-EnLDSqRGeIgPdaU,2403
-toolslm-0.3.8.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-toolslm-0.3.8.dist-info/entry_points.txt,sha256=xFz0Eymlo5X7BGpaO6DI9gMxvN5A7faebzrlr8ctp5I,95
-toolslm-0.3.8.dist-info/top_level.txt,sha256=4hRTrFWayz_Kz5221XjvlpCwVFrW3WPi1P0fllkTq9s,8
-toolslm-0.3.8.dist-info/RECORD,,