mirror of
https://github.com/karma-riuk/crab-webapp.git
synced 2025-07-04 22:08:12 +02:00
slimmed down dataset.py to only what's
necessary. Updated the structure to the new one
This commit is contained in:
@ -5,13 +5,6 @@ import json, uuid
|
||||
|
||||
from utils.errors import InvalidJsonFormatError
|
||||
|
||||
# fmt: off
|
||||
@dataclass
|
||||
class FileData:
|
||||
is_code_related: bool
|
||||
coverage: Dict[str, float] # jacoco-report -> coverage
|
||||
content_before_pr: str = ""
|
||||
content_after_pr: str = ""
|
||||
|
||||
@dataclass
|
||||
class Comment:
|
||||
@ -21,16 +14,18 @@ class Comment:
|
||||
to: int
|
||||
paraphrases: List[str] = field(default_factory=list)
|
||||
|
||||
|
||||
@dataclass
|
||||
class Selection:
|
||||
comment_suggests_change: bool
|
||||
diff_after_address_change: Optional[bool]
|
||||
is_code_related: bool
|
||||
|
||||
|
||||
class ArchiveState(Enum):
|
||||
BASE = "base"
|
||||
MERGED = "merged"
|
||||
|
||||
|
||||
@dataclass
|
||||
class Metadata:
|
||||
id: str
|
||||
@ -39,17 +34,20 @@ class Metadata:
|
||||
pr_title: str
|
||||
pr_body: str
|
||||
merge_commit_sha: str # to checkout for the tests
|
||||
successful: bool = True
|
||||
is_covered: Optional[bool] = None
|
||||
is_code_related: Optional[bool] = None
|
||||
successful: Optional[bool] = None
|
||||
build_system: str = ""
|
||||
reason_for_failure: str = ""
|
||||
last_cmd_error_msg: str = ""
|
||||
selection: Optional[Selection] = None
|
||||
|
||||
def archive_name(self, state: ArchiveState, only_id:bool=False):
|
||||
def archive_name(self, state: ArchiveState, only_id: bool = False):
|
||||
if only_id:
|
||||
return f"{self.id}_{state.value}.tar.gz"
|
||||
return f"{self.repo.replace('/', '_')}_{self.pr_number}_{state.value}.tar.gz"
|
||||
|
||||
|
||||
@dataclass
|
||||
class CommentGenSubmission:
|
||||
path: str
|
||||
@ -63,7 +61,9 @@ class CommentGenSubmission:
|
||||
raise InvalidJsonFormatError("Submitted json doesn't contain an object")
|
||||
if not all(k in data and isinstance(data[k], str) for k in ["path", "body"]):
|
||||
raise InvalidJsonFormatError("Submitted json doesn't contain the required fields")
|
||||
if not all(k in data and isinstance(data[k], (int, type(None))) for k in ["line_from", "line_to"]):
|
||||
if not all(
|
||||
k in data and isinstance(data[k], (int, type(None))) for k in ["line_from", "line_to"]
|
||||
):
|
||||
raise InvalidJsonFormatError("Submitted json doesn't contain the required fields")
|
||||
|
||||
return cls(
|
||||
@ -73,52 +73,19 @@ class CommentGenSubmission:
|
||||
body=data["body"],
|
||||
)
|
||||
|
||||
|
||||
@dataclass
|
||||
class DatasetEntry:
|
||||
metadata: Metadata
|
||||
files: Dict[str, FileData] # filename -> file data, files before the PR (before the first PR commits)
|
||||
diffs_before: Dict[str, str] # filename -> diff, diffs between the opening of the PR and the comment
|
||||
comments: List[Comment]
|
||||
diffs_after: Dict[str, str] # filename -> diff, changes after the comment
|
||||
|
||||
|
||||
@dataclass
|
||||
class CommentGenEntry:
|
||||
id: str
|
||||
files: Dict[str, str] # filename -> file content
|
||||
diffs: Dict[str, str] # filename -> diff, diffs between the opening of the PR and the comment
|
||||
|
||||
@staticmethod
|
||||
def from_entry(entry: DatasetEntry) -> "CommentGenEntry":
|
||||
return CommentGenEntry(
|
||||
id=entry.metadata.id,
|
||||
files={fname: fdata.content_before_pr for fname, fdata in entry.files.items()},
|
||||
diffs=entry.diffs_before,
|
||||
)
|
||||
|
||||
|
||||
@dataclass
|
||||
class CodeRefinementEntry:
|
||||
id: str
|
||||
files: Dict[str, str] # filename -> file content
|
||||
diffs: Dict[str, str] # filename -> diff, diffs between the opening of the PR and the comment
|
||||
comments: List[Comment]
|
||||
|
||||
@staticmethod
|
||||
def from_entry(entry: DatasetEntry) -> "CodeRefinementEntry":
|
||||
return CodeRefinementEntry(
|
||||
id=entry.metadata.id,
|
||||
files={fname: fdata.content_before_pr for fname, fdata in entry.files.items()},
|
||||
diffs=entry.diffs_before,
|
||||
comments=entry.comments,
|
||||
)
|
||||
|
||||
class OutputType(Enum):
|
||||
FULL = "full"
|
||||
CODE_REFINEMENT = "code_refinement"
|
||||
COMMENT_GEN = "comment_gen"
|
||||
|
||||
# fmt: on
|
||||
|
||||
@dataclass
|
||||
class Dataset:
|
||||
entries: List[DatasetEntry] = field(default_factory=list)
|
||||
@ -126,59 +93,6 @@ class Dataset:
|
||||
def __len__(self) -> int:
|
||||
return sum(1 for entry in self.entries if entry.metadata.successful)
|
||||
|
||||
def to_json(
|
||||
self,
|
||||
filename: str,
|
||||
type_: OutputType = OutputType.FULL,
|
||||
remove_non_suggesting: bool = False,
|
||||
) -> None:
|
||||
"""Serialize the dataset to a JSON file"""
|
||||
|
||||
entries_to_dump = self.entries
|
||||
|
||||
if type_ == OutputType.COMMENT_GEN:
|
||||
entries_to_dump = [
|
||||
entry
|
||||
for entry in self.entries
|
||||
if entry.metadata.selection and entry.metadata.selection.comment_suggests_change
|
||||
]
|
||||
elif type_ == OutputType.CODE_REFINEMENT:
|
||||
entries_to_dump = [
|
||||
entry
|
||||
for entry in self.entries
|
||||
if entry.metadata.selection
|
||||
and entry.metadata.selection.diff_after_address_change
|
||||
and entry.metadata.selection.is_code_related
|
||||
]
|
||||
elif type_ == OutputType.FULL and remove_non_suggesting:
|
||||
entries_to_dump = [
|
||||
entry
|
||||
for entry in self.entries
|
||||
if entry.metadata.selection and entry.metadata.selection.comment_suggests_change
|
||||
]
|
||||
|
||||
to_dump = Dataset(entries=entries_to_dump)
|
||||
# print(f"{len(entries_to_dump)} entries...", end=" ", flush=True)
|
||||
|
||||
def transform_entry(entry: Union[DatasetEntry, Dataset, Any]) -> Union[dict, list]:
|
||||
if not isinstance(entry, (DatasetEntry, Dataset)):
|
||||
return entry.__dict__
|
||||
|
||||
if type_ == OutputType.FULL:
|
||||
return entry.__dict__
|
||||
|
||||
if isinstance(entry, Dataset):
|
||||
return entry.entries
|
||||
|
||||
if type_ == OutputType.COMMENT_GEN:
|
||||
return CommentGenEntry.from_entry(entry).__dict__
|
||||
|
||||
if type_ == OutputType.CODE_REFINEMENT:
|
||||
return CodeRefinementEntry.from_entry(entry).__dict__
|
||||
|
||||
with open(filename, "w", encoding="utf-8") as f:
|
||||
json.dump(to_dump, f, default=transform_entry, indent=4)
|
||||
|
||||
@staticmethod
|
||||
def from_json(filename: str, keep_still_in_progress: bool = False) -> "Dataset":
|
||||
with open(filename, "r", encoding="utf-8") as f:
|
||||
@ -202,16 +116,11 @@ class Dataset:
|
||||
):
|
||||
continue
|
||||
|
||||
files = {fname: FileData(**fdata) for fname, fdata in entry_data["files"].items()}
|
||||
|
||||
comments = [Comment(**comment) for comment in entry_data["comments"]]
|
||||
|
||||
entry = DatasetEntry(
|
||||
metadata=metadata,
|
||||
files=files,
|
||||
diffs_before=entry_data["diffs_before"],
|
||||
comments=comments,
|
||||
diffs_after=entry_data["diffs_after"],
|
||||
)
|
||||
entries.append(entry)
|
||||
|
||||
|
Reference in New Issue
Block a user