apunkt/PageIndex
1
0
Fork 0
mirror of https://github.com/VectifyAI/PageIndex.git synced 2026-06-12 19:55:17 +02:00
Code Issues Projects Releases Packages Wiki Activity

refactor(filesystem): remove reference id concept

Browse source
This commit is contained in:
BukeLy 2026-05-26 15:53:25 +08:00
parent a80b84dae4
commit b2089712b3
3 changed files with 23 additions and 32 deletions
Download patch file Download diff file Expand all files Collapse all files

5
pageindex/filesystem/commands.py
View file

@ -989,7 +989,6 @@ class PIFSCommandExecutor:
continue
hits.append(
{
"reference_id": result.reference_id,
"file_ref": result.file_ref,
"external_id": result.external_id,
"title": result.title,
@ -1058,7 +1057,6 @@ class PIFSCommandExecutor:
line_number, text = self._first_matching_source_line(path, query)
hits.append(
{
"reference_id": file_row["external_id"] or file_row["file_ref"],
"file_ref": file_row["file_ref"],
"external_id": file_row["external_id"],
"title": file_row["title"],
@ -1073,14 +1071,13 @@ class PIFSCommandExecutor:
return hits
def _grep_file_matches(self, target: str, query: str, *, limit: int) -> list[dict[str, Any]]:
file_ref = self.filesystem._resolve_reference(target)
file_ref = self.filesystem._resolve_target(target)
entry = self.filesystem.store.get_file(file_ref)
matches = []
for line_number, line in enumerate(self.filesystem.store.read_text(file_ref).splitlines(), 1):
if self._line_matches(line, query):
matches.append(
{
"reference_id": entry.external_id or file_ref,
"file_ref": file_ref,
"external_id": entry.external_id,
"title": entry.title,

48
pageindex/filesystem/core.py
View file

@ -394,7 +394,6 @@ class PageIndexFileSystem:
folder_path = self._preferred_folder_path(folder_paths, scope_path, row["folder_path"])
results.append(
SearchResult(
reference_id=row["external_id"] or row["file_ref"],
file_ref=row["file_ref"],
external_id=row["external_id"],
title=row["title"],
@ -509,11 +508,11 @@ class PageIndexFileSystem:
def find(
self,
reference_id: str,
target: str,
patterns: Union[str, list[str]],
limit: int = 20,
) -> list[OpenResult]:
file_ref = self._resolve_reference(reference_id)
file_ref = self._resolve_target(target)
patterns = [patterns] if isinstance(patterns, str) else list(patterns)
lowered_patterns = [pattern.lower() for pattern in patterns if pattern]
if not lowered_patterns:
@ -526,13 +525,13 @@ class PageIndexFileSystem:
if any(pattern in haystack for pattern in lowered_patterns):
start = max(1, i - 1)
end = min(len(lines), i + 1)
matches.append(self._open_lines(reference_id, file_ref, start, end))
matches.append(self._open_lines(file_ref, start, end))
if len(matches) >= limit:
break
return matches
def open(self, reference_id: str, location: str = "all") -> OpenResult:
file_ref = self._resolve_reference(reference_id)
def open(self, target: str, location: str = "all") -> OpenResult:
file_ref = self._resolve_target(target)
entry = self.store.get_file(file_ref)
if self._file_format(entry) in {"pdf", "markdown", "pageindex"}:
raise ValueError(
@ -540,21 +539,21 @@ class PageIndexFileSystem:
"use pageindex_structure(), pageindex_pages(), or pageindex_node()."
)
if str(location).strip().lower() in {"all", "full", "*"}:
return self._open_all(reference_id, file_ref)
return self._open_all(file_ref)
start, end = self._parse_line_range(location)
return self._open_lines(reference_id, file_ref, start, end)
return self._open_lines(file_ref, start, end)
def cat_text_artifact(self, reference_id: str, location: str = "all") -> OpenResult:
file_ref = self._resolve_reference(reference_id)
def cat_text_artifact(self, target: str, location: str = "all") -> OpenResult:
file_ref = self._resolve_target(target)
entry = self.store.get_file(file_ref)
self._require_text_artifact_file(entry, "cat --all")
if str(location).strip().lower() in {"all", "full", "*"}:
return self._open_all(reference_id, file_ref)
return self._open_all(file_ref)
start, end = self._parse_line_range(location)
return self._open_lines(reference_id, file_ref, start, end)
return self._open_lines(file_ref, start, end)
def pageindex_structure(self, reference_id: str) -> dict[str, Any]:
file_ref = self._resolve_reference(reference_id)
def pageindex_structure(self, target: str) -> dict[str, Any]:
file_ref = self._resolve_target(target)
entry = self.store.get_file(file_ref)
self._require_pageindex_document_file(entry, "cat --structure")
client, doc_id = self._pageindex_client_doc_for_entry(entry)
@ -585,8 +584,8 @@ class PageIndexFileSystem:
"structure": strip_pageindex_text_fields(structure),
}
def pageindex_node(self, reference_id: str, node_id: str) -> dict[str, Any]:
file_ref = self._resolve_reference(reference_id)
def pageindex_node(self, target: str, node_id: str) -> dict[str, Any]:
file_ref = self._resolve_target(target)
entry = self.store.get_file(file_ref)
self._require_pageindex_document_file(entry, "cat --node")
client, doc_id = self._pageindex_client_doc_for_entry(entry)
@ -637,8 +636,8 @@ class PageIndexFileSystem:
"text": text,
}
def pageindex_pages(self, reference_id: str, pages: str) -> dict[str, Any]:
file_ref = self._resolve_reference(reference_id)
def pageindex_pages(self, target: str, pages: str) -> dict[str, Any]:
file_ref = self._resolve_target(target)
entry = self.store.get_file(file_ref)
self._require_pageindex_document_file(entry, "cat --page")
client, doc_id = self._pageindex_client_doc_for_entry(entry)
@ -682,7 +681,7 @@ class PageIndexFileSystem:
}
def _stat(self, target: str) -> dict[str, Any]:
file_ref = self._resolve_reference(target)
file_ref = self._resolve_target(target)
return self.store.file_info(file_ref)
def _require_text_artifact_file(self, entry: Any, command: str) -> None:
@ -1196,14 +1195,13 @@ class PageIndexFileSystem:
self._merge_metadata_values(record["metadata"], record["derived_metadata"])
)
def _open_lines(self, reference_id: str, file_ref: str, start: int, end: int) -> OpenResult:
def _open_lines(self, file_ref: str, start: int, end: int) -> OpenResult:
entry = self.store.get_file(file_ref)
lines = self.store.read_text(file_ref).splitlines()
start = max(1, start)
end = min(max(start, end), len(lines))
text = "\n".join(lines[start - 1:end])
return OpenResult(
reference_id=reference_id,
file_ref=file_ref,
start_line=start,
end_line=end,
@ -1213,12 +1211,11 @@ class PageIndexFileSystem:
source_path=entry.source_path,
)
def _open_all(self, reference_id: str, file_ref: str) -> OpenResult:
def _open_all(self, file_ref: str) -> OpenResult:
entry = self.store.get_file(file_ref)
text = self.store.read_text(file_ref)
line_count = len(text.splitlines())
return OpenResult(
reference_id=reference_id,
file_ref=file_ref,
start_line=1,
end_line=line_count,
@ -1252,8 +1249,8 @@ class PageIndexFileSystem:
result["pages"] = pages
return result
def _resolve_reference(self, reference_id: str) -> str:
return self.store.resolve_file_ref(reference_id)
def _resolve_target(self, target: str) -> str:
return self.store.resolve_file_ref(target)
def _should_use_semantic_retrieval(
self,
@ -1319,7 +1316,6 @@ class PageIndexFileSystem:
folder_path = self._preferred_folder_path(folder_paths, scope_path, entry.folder_path)
results.append(
SearchResult(
reference_id=entry.external_id or file_ref,
file_ref=file_ref,
external_id=entry.external_id,
title=entry.title,

2
pageindex/filesystem/types.py
View file

@ -6,7 +6,6 @@ from typing import Any, Optional
@dataclass(frozen=True)
class SearchResult:
reference_id: str
file_ref: str
external_id: Optional[str]
title: str
@ -29,7 +28,6 @@ class SearchResult:
@dataclass(frozen=True)
class OpenResult:
reference_id: str
file_ref: str
start_line: int
end_line: int