Skip to content

gh-144764: Made dataclasses construct automatic docstrings lazily #144765

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
adamchainz wants to merge 1 commit into from
Closed

gh-144764: Made dataclasses construct automatic docstrings lazily #144765

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
gh-144764: Made dataclasses construct automatic docstrings lazily 
Move the default docstring construction to occur on-access through a descriptor.

Verified with [tprof](https://github.com/adamchainz/tprof) and this script that generates 10k dataclasses:

```py
from dataclasses import dataclass

for i in range(10_000):
    @DataClass
    class Example:
        field1: int
        field2: str
        field3: float
```

**Before:**

```
$ tprof -t dataclasses._process_class example.py
🎯 tprof results:
 function                     calls total  mean ± σ       min … max
 dataclasses._process_class() 10000    5s 485μs ± 120μs 458μs … 6ms
 ```

After:

```
$ PYTHONPATH=Lib/ uvx tprof -t dataclasses._process_class example.py
🎯 tprof results:
 function                     calls total  mean ± σ       min … max
 dataclasses._process_class() 10000    3s 275μs ± 131μs 245μs … 6ms
```

The mean time spent in `_process_class()` has dropped from 485μs to 275μs, a ~42% time saving (admittedly skewed due to the small size of the dataclass).
  • Loading branch information
@adamchainz
adamchainz committed Feb 13, 2026
commit e2293d97f36e45ec7226c462bc492cb6567752f3
33 changes: 18 additions & 15 deletions Lib/dataclasses.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1209,23 +1209,11 @@ def _process_class(cls, init, repr, eq, order, unsafe_hash, frozen,
if hash_action:
cls.__hash__ = hash_action(cls, field_list, func_builder)

# Generate the methods and add them to the class. This needs to be done
# before the __doc__ logic below, since inspect will look at the __init__
# signature.
# Generate the methods and add them to the class.
func_builder.add_fns_to_class(cls)

if not getattr(cls, '__doc__'):
# Create a class doc-string.
try:
# In some cases fetching a signature is not possible.
# But, we surely should not fail in this case.
text_sig = str(inspect.signature(
cls,
annotation_format=annotationlib.Format.FORWARDREF,
)).replace(' -> None', '')
except (TypeError, ValueError):
text_sig = ''
cls.__doc__ = (cls.__name__ + text_sig)
if not cls.__doc__:
cls.__doc__ = _DocDescriptor()

if match_args:
# I could probably compute this once.
Expand All @@ -1243,6 +1231,21 @@ def _process_class(cls, init, repr, eq, order, unsafe_hash, frozen,
return cls


class _DocDescriptor:
def __get__(self, obj, owner):
# Create a class doc-string.
try:
# In some cases fetching a signature is not possible.
# But, we surely should not fail in this case.
text_sig = str(inspect.signature(
owner,
annotation_format=annotationlib.Format.FORWARDREF,
)).replace(' -> None', '')
except (TypeError, ValueError):
text_sig = ''
owner.__doc__ = (owner.__name__ + text_sig)
return owner.__doc__

# _dataclass_getstate and _dataclass_setstate are needed for pickling frozen
# classes with slots. These could be slightly more performant if we generated
# the code instead of iterating over fields. But that can be a project for
Expand Down
Loading