fix: address xhigh code-review findings on 2d46d68..8f536cb
Verified 12 findings from an xhigh-effort review of the prior review-fix
batch; all confirmed real. Most trace back to one root cause: the
build_index() text-stripping fix (8f536cb) correctly stopped Markdown from
leaking full text by default, but broke every path that assumed text could
be re-read later.
Correctness:
- LocalBackend._fill_node_text (get_document(include_text=True)) only handled
PDF's start_index/end_index convention; Markdown nodes use line_num and got
silently empty text. Now handles both.
- get_page_content's Markdown fallback (triggered when a StorageEngine
legitimately returns None from get_pages()) read from the now-text-stripped
structure. It now re-derives from the source file, mirroring the PDF
fallback, so it no longer depends on structure text at all.
- add_document's PDF-only text-stripping branch (with the stale "markdown
needs text in structure for fallback retrieval" comment) is now dead/wrong
since build_index() already applies if_add_node_text uniformly — removed.
- _validate_llm_provider's keyless-provider allowlist was missing several
local LiteLLM providers (xinference, llamafile, triton, oobabooga,
openai_like, docker_model_runner, custom, custom_openai, petals) that need
no API key just like ollama/lm_studio; expanded.
- The three agent-tool closures (get_document, get_document_structure,
get_page_content) had three different not-found patterns; two bypassed the
backend's DocumentNotFoundError entirely. Extracted LocalBackend.
_require_document as the single existence check every method/tool now uses.
- examples/agentic_vectorless_rag_demo.py's hand-rolled Agent() didn't apply
the litellm/ prefix normalization the SDK does internally, so its own
documented "any LiteLLM provider" claim broke for non-openai models.
- cloud delete_collection's cache eviction removed the "folders unavailable"
None sentinel too, forcing a wasted re-fetch; now only pops on a real id.
Cleanup / altitude:
- build_index() skips the remove_structure_text walk entirely when text was
never added (content_based + if_add_node_summary=False + if_add_node_text=
False) instead of a guaranteed no-op tree walk.
- page_index()'s locals()-capture-as-kwargs (fragile by construction) replaced
with an explicit dict of the named parameters.
- run_pageindex.py's _cli_bool and the page_index_md.py legacy shim's
_coerce_bool were duplicate, diverging implementations; both now bind
directly to the canonical pageindex.index.page_index_md._coerce_bool.
- retrieve.py's _get_md_page_content delegated its own traversal instead of
calling the canonical get_md_page_content; now a one-line delegation.
- FileTypeError's docstring now calls out the except-ordering gotcha from
also subclassing ValueError.
17 new regression tests (tests/test_review_fixes_2.py) plus 2 updated in
tests/test_legacy_shims.py for the simplified md_to_tree shim. Full suite:
210 passed, 2 skipped.
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-08 21:56:41 +08:00
|
|
|
"""Regression tests for the second review pass (xhigh code-review of
|
|
|
|
|
2d46d68..8f536cb): the Markdown text-stripping fix's fallout, plus the other
|
|
|
|
|
directly-fixable findings from that pass."""
|
|
|
|
|
import asyncio
|
fix: CMYK image drop, empty-doc crash, page_index shadowing, sqlite hardening, flaky tests
Addresses items 9-13 and a/b/c/f from the max-effort review of PR #272.
- pdf.py: image colorspace check was `pix.n > 4`, which treats CMYK-without-
alpha (n==4, same as RGBA) as not needing RGB conversion; pix.save() as .png
then raises "unsupported colorspace", silently dropped by the surrounding
except. Fixed to `pix.n - pix.alpha >= 4` (correctly converts CMYK, leaves
RGBA untouched).
- pipeline.py: detect_strategy([]) (an empty/whitespace-only source file)
returned "content_based", routing into the PDF-oriented TOC-detection
pipeline -- wasting a real LLM call before raising IndexingError. Empty
node lists now route to level_based, whose build_tree_from_levels([])
returns an empty structure instantly with zero LLM calls.
- page_index.py (shim): pageindex/__init__.py binds the canonical `page_index`
function as the package attribute, but this file is ALSO a real submodule
of the same name -- importing it anywhere (import machinery, unconditional)
overwrites that attribute with the module object, breaking
`from pageindex import page_index; page_index(x)` for the rest of the
process. Made the shim module itself callable (delegates to the real
function via a ModuleType subclass), so whichever object ends up in that
slot is callable regardless of import order.
- storage/sqlite.py: create_collection let a raw sqlite3.IntegrityError escape
on a duplicate name (new CollectionAlreadyExistsError); the collections
table's CHECK constraint only validated the name's first character (GLOB
'*' is a wildcard, not a regex quantifier over the preceding class) --
fixed to validate the whole string, and SQLiteStorage now also validates in
Python (it's a public StorageEngine usable directly, bypassing
LocalBackend's own check).
- tests/test_review_fixes_2.py: two tests used a ContentNode with no `level`
set, so build_index took the content_based path and made real (retried,
slow, and -- with a valid key -- billable) LLM calls instead of testing the
text-stripping logic they claimed to. Mocked out _content_based_pipeline.
- retrieve.py: _parse_pages/_get_pdf_page_content were independent copies of
the canonical parse_pages/get_pdf_page_content that had already drifted
(missing the p>=1 filter and 1000-page DoS cap) -- delegate to canonical
now, so the legacy pageindex.get_page_content path can't silently regress
again.
- parser/markdown.py: a leading UTF-8 BOM broke first-header detection
(not whitespace, .strip() doesn't remove it) -- decode utf-8-sig. Only
backtick fences were recognized as code blocks, so a '#'-prefixed line
inside a ~~~-fenced block (valid CommonMark) was misparsed as a heading --
recognize both fence styles.
- run_pageindex.py: --if-thinning wasn't migrated to the bare-flag +
legacy-yes/no convention the other four --if-add-* flags got; bare usage
raised an argparse error and it never went through the shared coercion.
- types.py: DocumentDetail's `structure` field was inside the class's
total=False body, so TypedDict rules made it optional even though every
backend always populates it. Split into a required base class.
Adds regression tests for all of the above. Full suite: 244 passed, 2 skipped
(one pre-existing, unrelated flaky cloud-streaming test).
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-09 11:58:59 +08:00
|
|
|
from unittest.mock import AsyncMock
|
fix: address xhigh code-review findings on 2d46d68..8f536cb
Verified 12 findings from an xhigh-effort review of the prior review-fix
batch; all confirmed real. Most trace back to one root cause: the
build_index() text-stripping fix (8f536cb) correctly stopped Markdown from
leaking full text by default, but broke every path that assumed text could
be re-read later.
Correctness:
- LocalBackend._fill_node_text (get_document(include_text=True)) only handled
PDF's start_index/end_index convention; Markdown nodes use line_num and got
silently empty text. Now handles both.
- get_page_content's Markdown fallback (triggered when a StorageEngine
legitimately returns None from get_pages()) read from the now-text-stripped
structure. It now re-derives from the source file, mirroring the PDF
fallback, so it no longer depends on structure text at all.
- add_document's PDF-only text-stripping branch (with the stale "markdown
needs text in structure for fallback retrieval" comment) is now dead/wrong
since build_index() already applies if_add_node_text uniformly — removed.
- _validate_llm_provider's keyless-provider allowlist was missing several
local LiteLLM providers (xinference, llamafile, triton, oobabooga,
openai_like, docker_model_runner, custom, custom_openai, petals) that need
no API key just like ollama/lm_studio; expanded.
- The three agent-tool closures (get_document, get_document_structure,
get_page_content) had three different not-found patterns; two bypassed the
backend's DocumentNotFoundError entirely. Extracted LocalBackend.
_require_document as the single existence check every method/tool now uses.
- examples/agentic_vectorless_rag_demo.py's hand-rolled Agent() didn't apply
the litellm/ prefix normalization the SDK does internally, so its own
documented "any LiteLLM provider" claim broke for non-openai models.
- cloud delete_collection's cache eviction removed the "folders unavailable"
None sentinel too, forcing a wasted re-fetch; now only pops on a real id.
Cleanup / altitude:
- build_index() skips the remove_structure_text walk entirely when text was
never added (content_based + if_add_node_summary=False + if_add_node_text=
False) instead of a guaranteed no-op tree walk.
- page_index()'s locals()-capture-as-kwargs (fragile by construction) replaced
with an explicit dict of the named parameters.
- run_pageindex.py's _cli_bool and the page_index_md.py legacy shim's
_coerce_bool were duplicate, diverging implementations; both now bind
directly to the canonical pageindex.index.page_index_md._coerce_bool.
- retrieve.py's _get_md_page_content delegated its own traversal instead of
calling the canonical get_md_page_content; now a one-line delegation.
- FileTypeError's docstring now calls out the except-ordering gotcha from
also subclassing ValueError.
17 new regression tests (tests/test_review_fixes_2.py) plus 2 updated in
tests/test_legacy_shims.py for the simplified md_to_tree shim. Full suite:
210 passed, 2 skipped.
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-08 21:56:41 +08:00
|
|
|
|
|
|
|
|
import pytest
|
|
|
|
|
|
|
|
|
|
from pageindex.config import IndexConfig
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def _md_backend(tmp_path):
|
|
|
|
|
from pageindex.backend.local import LocalBackend
|
|
|
|
|
from pageindex.storage.sqlite import SQLiteStorage
|
|
|
|
|
|
|
|
|
|
backend = LocalBackend(
|
|
|
|
|
storage=SQLiteStorage(str(tmp_path / "t.db")),
|
|
|
|
|
files_dir=str(tmp_path / "f"), model="gpt-4o",
|
|
|
|
|
index_config=IndexConfig(if_add_node_summary=False, if_add_doc_description=False),
|
|
|
|
|
)
|
|
|
|
|
backend.get_or_create_collection("c")
|
|
|
|
|
return backend
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def _write_md(tmp_path, name="doc.md"):
|
|
|
|
|
path = tmp_path / name
|
|
|
|
|
path.write_text("# Title\nfirst section body\n\n## Sub\nsecond section body\n")
|
|
|
|
|
return str(path)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #1: get_document(include_text=True) must fill text for Markdown nodes ────
|
|
|
|
|
def test_get_document_include_text_fills_markdown_nodes(tmp_path):
|
|
|
|
|
backend = _md_backend(tmp_path)
|
|
|
|
|
doc_id = backend.add_document("c", _write_md(tmp_path))
|
|
|
|
|
|
|
|
|
|
def _texts(nodes):
|
|
|
|
|
for n in nodes:
|
|
|
|
|
yield n.get("text")
|
|
|
|
|
if n.get("nodes"):
|
|
|
|
|
yield from _texts(n["nodes"])
|
|
|
|
|
|
|
|
|
|
without = backend.get_document("c", doc_id, include_text=False)
|
|
|
|
|
assert not any(_texts(without["structure"]))
|
|
|
|
|
|
|
|
|
|
with_text = backend.get_document("c", doc_id, include_text=True)
|
|
|
|
|
texts = list(_texts(with_text["structure"]))
|
|
|
|
|
assert texts, "expected at least one node"
|
|
|
|
|
assert any(t for t in texts), "Markdown nodes must get real text, not all empty"
|
|
|
|
|
assert any("first section body" in t or "second section body" in t for t in texts if t)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #2: get_page_content's Markdown fallback re-derives from the source file ──
|
|
|
|
|
def test_get_page_content_markdown_fallback_reads_from_file(tmp_path):
|
|
|
|
|
backend = _md_backend(tmp_path)
|
|
|
|
|
md_path = _write_md(tmp_path)
|
|
|
|
|
doc_id = backend.add_document("c", md_path)
|
|
|
|
|
|
|
|
|
|
# Simulate a StorageEngine that doesn't cache pages (protocol explicitly
|
|
|
|
|
# allows get_pages() to return None) by clearing the cached pages column.
|
|
|
|
|
conn = backend._storage._get_conn()
|
|
|
|
|
conn.execute("UPDATE documents SET pages = NULL WHERE doc_id = ?", (doc_id,))
|
|
|
|
|
|
|
|
|
|
result = backend.get_page_content("c", doc_id, "1")
|
|
|
|
|
assert result and result[0]["content"], "fallback must return real text, not empty"
|
|
|
|
|
assert "first section body" in result[0]["content"]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #3: keyless provider allowlist covers other local LiteLLM providers ──────
|
|
|
|
|
@pytest.mark.parametrize("model", [
|
|
|
|
|
"ollama/llama3", "lm_studio/x", "xinference/llama2", "llamafile/x",
|
|
|
|
|
"triton/x", "oobabooga/x", "openai_like/x", "docker_model_runner/x",
|
|
|
|
|
])
|
|
|
|
|
def test_validate_llm_provider_accepts_more_keyless_providers(model):
|
|
|
|
|
from pageindex.client import LocalClient
|
|
|
|
|
LocalClient._validate_llm_provider(model) # must not raise
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #4: agent-tool closures consistently raise/error on a missing doc ────────
|
|
|
|
|
def test_agent_tools_consistently_report_missing_doc(tmp_path):
|
|
|
|
|
import json
|
|
|
|
|
import asyncio as _asyncio
|
|
|
|
|
from agents.tool_context import ToolContext
|
|
|
|
|
|
|
|
|
|
backend = _md_backend(tmp_path)
|
|
|
|
|
# Open-mode tools (doc_ids=None) so we probe not-found handling directly,
|
|
|
|
|
# not the separate out-of-scope rejection path.
|
|
|
|
|
tools = backend.get_agent_tools("c", doc_ids=None)
|
|
|
|
|
by_name = {t.name: t for t in tools.function_tools}
|
|
|
|
|
|
|
|
|
|
for name in ("get_document", "get_document_structure", "get_page_content"):
|
|
|
|
|
tool = by_name[name]
|
|
|
|
|
kwargs = {"doc_id": "ghost"}
|
|
|
|
|
if name == "get_page_content":
|
|
|
|
|
kwargs["pages"] = "1"
|
|
|
|
|
raw_args = json.dumps(kwargs)
|
|
|
|
|
ctx = ToolContext(context=None, tool_name=name, tool_call_id="1", tool_arguments=raw_args)
|
|
|
|
|
out = _asyncio.run(tool.on_invoke_tool(ctx, raw_args))
|
|
|
|
|
parsed = json.loads(out)
|
|
|
|
|
assert "error" in parsed and "ghost" in parsed["error"], f"{name} did not report not-found consistently: {parsed}"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #6: cloud delete_collection preserves the "folders unavailable" sentinel ──
|
|
|
|
|
def test_cloud_delete_collection_preserves_unavailable_sentinel(monkeypatch):
|
|
|
|
|
from pageindex.backend.cloud import CloudBackend
|
|
|
|
|
|
|
|
|
|
backend = CloudBackend(api_key="pi-test")
|
|
|
|
|
backend._folder_id_cache["papers"] = None # folders-unavailable sentinel
|
|
|
|
|
called = []
|
|
|
|
|
monkeypatch.setattr(backend, "_request", lambda *a, **k: called.append(a) or {})
|
|
|
|
|
backend.delete_collection("papers")
|
|
|
|
|
assert not called, "no DELETE should fire when folder_id is the unavailable sentinel"
|
|
|
|
|
assert "papers" in backend._folder_id_cache and backend._folder_id_cache["papers"] is None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_cloud_delete_collection_still_clears_real_folder_id(monkeypatch):
|
|
|
|
|
from pageindex.backend.cloud import CloudBackend
|
|
|
|
|
|
|
|
|
|
backend = CloudBackend(api_key="pi-test")
|
|
|
|
|
backend._folder_id_cache["papers"] = "folder-123"
|
|
|
|
|
monkeypatch.setattr(backend, "_request", lambda *a, **k: {})
|
|
|
|
|
backend.delete_collection("papers")
|
|
|
|
|
assert "papers" not in backend._folder_id_cache
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #7: remove_structure_text is skipped when text was never added ───────────
|
fix: CMYK image drop, empty-doc crash, page_index shadowing, sqlite hardening, flaky tests
Addresses items 9-13 and a/b/c/f from the max-effort review of PR #272.
- pdf.py: image colorspace check was `pix.n > 4`, which treats CMYK-without-
alpha (n==4, same as RGBA) as not needing RGB conversion; pix.save() as .png
then raises "unsupported colorspace", silently dropped by the surrounding
except. Fixed to `pix.n - pix.alpha >= 4` (correctly converts CMYK, leaves
RGBA untouched).
- pipeline.py: detect_strategy([]) (an empty/whitespace-only source file)
returned "content_based", routing into the PDF-oriented TOC-detection
pipeline -- wasting a real LLM call before raising IndexingError. Empty
node lists now route to level_based, whose build_tree_from_levels([])
returns an empty structure instantly with zero LLM calls.
- page_index.py (shim): pageindex/__init__.py binds the canonical `page_index`
function as the package attribute, but this file is ALSO a real submodule
of the same name -- importing it anywhere (import machinery, unconditional)
overwrites that attribute with the module object, breaking
`from pageindex import page_index; page_index(x)` for the rest of the
process. Made the shim module itself callable (delegates to the real
function via a ModuleType subclass), so whichever object ends up in that
slot is callable regardless of import order.
- storage/sqlite.py: create_collection let a raw sqlite3.IntegrityError escape
on a duplicate name (new CollectionAlreadyExistsError); the collections
table's CHECK constraint only validated the name's first character (GLOB
'*' is a wildcard, not a regex quantifier over the preceding class) --
fixed to validate the whole string, and SQLiteStorage now also validates in
Python (it's a public StorageEngine usable directly, bypassing
LocalBackend's own check).
- tests/test_review_fixes_2.py: two tests used a ContentNode with no `level`
set, so build_index took the content_based path and made real (retried,
slow, and -- with a valid key -- billable) LLM calls instead of testing the
text-stripping logic they claimed to. Mocked out _content_based_pipeline.
- retrieve.py: _parse_pages/_get_pdf_page_content were independent copies of
the canonical parse_pages/get_pdf_page_content that had already drifted
(missing the p>=1 filter and 1000-page DoS cap) -- delegate to canonical
now, so the legacy pageindex.get_page_content path can't silently regress
again.
- parser/markdown.py: a leading UTF-8 BOM broke first-header detection
(not whitespace, .strip() doesn't remove it) -- decode utf-8-sig. Only
backtick fences were recognized as code blocks, so a '#'-prefixed line
inside a ~~~-fenced block (valid CommonMark) was misparsed as a heading --
recognize both fence styles.
- run_pageindex.py: --if-thinning wasn't migrated to the bare-flag +
legacy-yes/no convention the other four --if-add-* flags got; bare usage
raised an argparse error and it never went through the shared coercion.
- types.py: DocumentDetail's `structure` field was inside the class's
total=False body, so TypedDict rules made it optional even though every
backend always populates it. Split into a required base class.
Adds regression tests for all of the above. Full suite: 244 passed, 2 skipped
(one pre-existing, unrelated flaky cloud-streaming test).
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-09 11:58:59 +08:00
|
|
|
def _mock_content_based_pipeline(monkeypatch, structure):
|
|
|
|
|
"""content_based's real path (_content_based_pipeline) drives real LLM
|
|
|
|
|
calls (TOC detection etc.) regardless of if_add_node_summary — a prior
|
|
|
|
|
version of these two tests didn't mock this out, fell through to it, and
|
|
|
|
|
made real network calls (with a dummy key: 10 retries before failing;
|
|
|
|
|
with a real key: real billable requests) on every run."""
|
|
|
|
|
from pageindex.index import pipeline
|
|
|
|
|
|
|
|
|
|
async def fake(page_list, opt):
|
|
|
|
|
return structure
|
|
|
|
|
|
|
|
|
|
monkeypatch.setattr(pipeline, "_content_based_pipeline", fake)
|
|
|
|
|
|
|
|
|
|
|
fix: address xhigh code-review findings on 2d46d68..8f536cb
Verified 12 findings from an xhigh-effort review of the prior review-fix
batch; all confirmed real. Most trace back to one root cause: the
build_index() text-stripping fix (8f536cb) correctly stopped Markdown from
leaking full text by default, but broke every path that assumed text could
be re-read later.
Correctness:
- LocalBackend._fill_node_text (get_document(include_text=True)) only handled
PDF's start_index/end_index convention; Markdown nodes use line_num and got
silently empty text. Now handles both.
- get_page_content's Markdown fallback (triggered when a StorageEngine
legitimately returns None from get_pages()) read from the now-text-stripped
structure. It now re-derives from the source file, mirroring the PDF
fallback, so it no longer depends on structure text at all.
- add_document's PDF-only text-stripping branch (with the stale "markdown
needs text in structure for fallback retrieval" comment) is now dead/wrong
since build_index() already applies if_add_node_text uniformly — removed.
- _validate_llm_provider's keyless-provider allowlist was missing several
local LiteLLM providers (xinference, llamafile, triton, oobabooga,
openai_like, docker_model_runner, custom, custom_openai, petals) that need
no API key just like ollama/lm_studio; expanded.
- The three agent-tool closures (get_document, get_document_structure,
get_page_content) had three different not-found patterns; two bypassed the
backend's DocumentNotFoundError entirely. Extracted LocalBackend.
_require_document as the single existence check every method/tool now uses.
- examples/agentic_vectorless_rag_demo.py's hand-rolled Agent() didn't apply
the litellm/ prefix normalization the SDK does internally, so its own
documented "any LiteLLM provider" claim broke for non-openai models.
- cloud delete_collection's cache eviction removed the "folders unavailable"
None sentinel too, forcing a wasted re-fetch; now only pops on a real id.
Cleanup / altitude:
- build_index() skips the remove_structure_text walk entirely when text was
never added (content_based + if_add_node_summary=False + if_add_node_text=
False) instead of a guaranteed no-op tree walk.
- page_index()'s locals()-capture-as-kwargs (fragile by construction) replaced
with an explicit dict of the named parameters.
- run_pageindex.py's _cli_bool and the page_index_md.py legacy shim's
_coerce_bool were duplicate, diverging implementations; both now bind
directly to the canonical pageindex.index.page_index_md._coerce_bool.
- retrieve.py's _get_md_page_content delegated its own traversal instead of
calling the canonical get_md_page_content; now a one-line delegation.
- FileTypeError's docstring now calls out the except-ordering gotcha from
also subclassing ValueError.
17 new regression tests (tests/test_review_fixes_2.py) plus 2 updated in
tests/test_legacy_shims.py for the simplified md_to_tree shim. Full suite:
210 passed, 2 skipped.
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-08 21:56:41 +08:00
|
|
|
def test_build_index_skips_text_strip_when_no_text_was_added(monkeypatch):
|
|
|
|
|
from pageindex.index import pipeline
|
|
|
|
|
from pageindex.parser.protocol import ContentNode, ParsedDocument
|
|
|
|
|
|
|
|
|
|
calls = []
|
|
|
|
|
# build_index() imports remove_structure_text locally (`from .utils import
|
|
|
|
|
# ...` inside the function body), so patch it on the utils module itself.
|
|
|
|
|
import pageindex.index.utils as utils_mod
|
|
|
|
|
monkeypatch.setattr(utils_mod, "remove_structure_text", lambda s: calls.append(s) or s)
|
fix: CMYK image drop, empty-doc crash, page_index shadowing, sqlite hardening, flaky tests
Addresses items 9-13 and a/b/c/f from the max-effort review of PR #272.
- pdf.py: image colorspace check was `pix.n > 4`, which treats CMYK-without-
alpha (n==4, same as RGBA) as not needing RGB conversion; pix.save() as .png
then raises "unsupported colorspace", silently dropped by the surrounding
except. Fixed to `pix.n - pix.alpha >= 4` (correctly converts CMYK, leaves
RGBA untouched).
- pipeline.py: detect_strategy([]) (an empty/whitespace-only source file)
returned "content_based", routing into the PDF-oriented TOC-detection
pipeline -- wasting a real LLM call before raising IndexingError. Empty
node lists now route to level_based, whose build_tree_from_levels([])
returns an empty structure instantly with zero LLM calls.
- page_index.py (shim): pageindex/__init__.py binds the canonical `page_index`
function as the package attribute, but this file is ALSO a real submodule
of the same name -- importing it anywhere (import machinery, unconditional)
overwrites that attribute with the module object, breaking
`from pageindex import page_index; page_index(x)` for the rest of the
process. Made the shim module itself callable (delegates to the real
function via a ModuleType subclass), so whichever object ends up in that
slot is callable regardless of import order.
- storage/sqlite.py: create_collection let a raw sqlite3.IntegrityError escape
on a duplicate name (new CollectionAlreadyExistsError); the collections
table's CHECK constraint only validated the name's first character (GLOB
'*' is a wildcard, not a regex quantifier over the preceding class) --
fixed to validate the whole string, and SQLiteStorage now also validates in
Python (it's a public StorageEngine usable directly, bypassing
LocalBackend's own check).
- tests/test_review_fixes_2.py: two tests used a ContentNode with no `level`
set, so build_index took the content_based path and made real (retried,
slow, and -- with a valid key -- billable) LLM calls instead of testing the
text-stripping logic they claimed to. Mocked out _content_based_pipeline.
- retrieve.py: _parse_pages/_get_pdf_page_content were independent copies of
the canonical parse_pages/get_pdf_page_content that had already drifted
(missing the p>=1 filter and 1000-page DoS cap) -- delegate to canonical
now, so the legacy pageindex.get_page_content path can't silently regress
again.
- parser/markdown.py: a leading UTF-8 BOM broke first-header detection
(not whitespace, .strip() doesn't remove it) -- decode utf-8-sig. Only
backtick fences were recognized as code blocks, so a '#'-prefixed line
inside a ~~~-fenced block (valid CommonMark) was misparsed as a heading --
recognize both fence styles.
- run_pageindex.py: --if-thinning wasn't migrated to the bare-flag +
legacy-yes/no convention the other four --if-add-* flags got; bare usage
raised an argparse error and it never went through the shared coercion.
- types.py: DocumentDetail's `structure` field was inside the class's
total=False body, so TypedDict rules made it optional even though every
backend always populates it. Split into a required base class.
Adds regression tests for all of the above. Full suite: 244 passed, 2 skipped
(one pre-existing, unrelated flaky cloud-streaming test).
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-09 11:58:59 +08:00
|
|
|
_mock_content_based_pipeline(monkeypatch, [{"title": "T", "start_index": 1, "end_index": 1}])
|
fix: address xhigh code-review findings on 2d46d68..8f536cb
Verified 12 findings from an xhigh-effort review of the prior review-fix
batch; all confirmed real. Most trace back to one root cause: the
build_index() text-stripping fix (8f536cb) correctly stopped Markdown from
leaking full text by default, but broke every path that assumed text could
be re-read later.
Correctness:
- LocalBackend._fill_node_text (get_document(include_text=True)) only handled
PDF's start_index/end_index convention; Markdown nodes use line_num and got
silently empty text. Now handles both.
- get_page_content's Markdown fallback (triggered when a StorageEngine
legitimately returns None from get_pages()) read from the now-text-stripped
structure. It now re-derives from the source file, mirroring the PDF
fallback, so it no longer depends on structure text at all.
- add_document's PDF-only text-stripping branch (with the stale "markdown
needs text in structure for fallback retrieval" comment) is now dead/wrong
since build_index() already applies if_add_node_text uniformly — removed.
- _validate_llm_provider's keyless-provider allowlist was missing several
local LiteLLM providers (xinference, llamafile, triton, oobabooga,
openai_like, docker_model_runner, custom, custom_openai, petals) that need
no API key just like ollama/lm_studio; expanded.
- The three agent-tool closures (get_document, get_document_structure,
get_page_content) had three different not-found patterns; two bypassed the
backend's DocumentNotFoundError entirely. Extracted LocalBackend.
_require_document as the single existence check every method/tool now uses.
- examples/agentic_vectorless_rag_demo.py's hand-rolled Agent() didn't apply
the litellm/ prefix normalization the SDK does internally, so its own
documented "any LiteLLM provider" claim broke for non-openai models.
- cloud delete_collection's cache eviction removed the "folders unavailable"
None sentinel too, forcing a wasted re-fetch; now only pops on a real id.
Cleanup / altitude:
- build_index() skips the remove_structure_text walk entirely when text was
never added (content_based + if_add_node_summary=False + if_add_node_text=
False) instead of a guaranteed no-op tree walk.
- page_index()'s locals()-capture-as-kwargs (fragile by construction) replaced
with an explicit dict of the named parameters.
- run_pageindex.py's _cli_bool and the page_index_md.py legacy shim's
_coerce_bool were duplicate, diverging implementations; both now bind
directly to the canonical pageindex.index.page_index_md._coerce_bool.
- retrieve.py's _get_md_page_content delegated its own traversal instead of
calling the canonical get_md_page_content; now a one-line delegation.
- FileTypeError's docstring now calls out the except-ordering gotcha from
also subclassing ValueError.
17 new regression tests (tests/test_review_fixes_2.py) plus 2 updated in
tests/test_legacy_shims.py for the simplified md_to_tree shim. Full suite:
210 passed, 2 skipped.
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-08 21:56:41 +08:00
|
|
|
|
|
|
|
|
nodes = [ContentNode(content="page one text", tokens=5, index=1)]
|
|
|
|
|
parsed = ParsedDocument(doc_name="d", nodes=nodes)
|
|
|
|
|
opt = IndexConfig(if_add_node_summary=False, if_add_doc_description=False,
|
|
|
|
|
if_add_node_text=False)
|
|
|
|
|
pipeline.build_index(parsed, opt=opt)
|
|
|
|
|
assert calls == [], "remove_structure_text must not run when no text was ever added"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_build_index_still_strips_text_when_summary_added_it(monkeypatch):
|
|
|
|
|
from pageindex.index import pipeline
|
|
|
|
|
from pageindex.parser.protocol import ContentNode, ParsedDocument
|
|
|
|
|
|
|
|
|
|
calls = []
|
|
|
|
|
import pageindex.index.utils as utils_mod
|
|
|
|
|
monkeypatch.setattr(utils_mod, "remove_structure_text", lambda s: calls.append(s) or s)
|
fix: CMYK image drop, empty-doc crash, page_index shadowing, sqlite hardening, flaky tests
Addresses items 9-13 and a/b/c/f from the max-effort review of PR #272.
- pdf.py: image colorspace check was `pix.n > 4`, which treats CMYK-without-
alpha (n==4, same as RGBA) as not needing RGB conversion; pix.save() as .png
then raises "unsupported colorspace", silently dropped by the surrounding
except. Fixed to `pix.n - pix.alpha >= 4` (correctly converts CMYK, leaves
RGBA untouched).
- pipeline.py: detect_strategy([]) (an empty/whitespace-only source file)
returned "content_based", routing into the PDF-oriented TOC-detection
pipeline -- wasting a real LLM call before raising IndexingError. Empty
node lists now route to level_based, whose build_tree_from_levels([])
returns an empty structure instantly with zero LLM calls.
- page_index.py (shim): pageindex/__init__.py binds the canonical `page_index`
function as the package attribute, but this file is ALSO a real submodule
of the same name -- importing it anywhere (import machinery, unconditional)
overwrites that attribute with the module object, breaking
`from pageindex import page_index; page_index(x)` for the rest of the
process. Made the shim module itself callable (delegates to the real
function via a ModuleType subclass), so whichever object ends up in that
slot is callable regardless of import order.
- storage/sqlite.py: create_collection let a raw sqlite3.IntegrityError escape
on a duplicate name (new CollectionAlreadyExistsError); the collections
table's CHECK constraint only validated the name's first character (GLOB
'*' is a wildcard, not a regex quantifier over the preceding class) --
fixed to validate the whole string, and SQLiteStorage now also validates in
Python (it's a public StorageEngine usable directly, bypassing
LocalBackend's own check).
- tests/test_review_fixes_2.py: two tests used a ContentNode with no `level`
set, so build_index took the content_based path and made real (retried,
slow, and -- with a valid key -- billable) LLM calls instead of testing the
text-stripping logic they claimed to. Mocked out _content_based_pipeline.
- retrieve.py: _parse_pages/_get_pdf_page_content were independent copies of
the canonical parse_pages/get_pdf_page_content that had already drifted
(missing the p>=1 filter and 1000-page DoS cap) -- delegate to canonical
now, so the legacy pageindex.get_page_content path can't silently regress
again.
- parser/markdown.py: a leading UTF-8 BOM broke first-header detection
(not whitespace, .strip() doesn't remove it) -- decode utf-8-sig. Only
backtick fences were recognized as code blocks, so a '#'-prefixed line
inside a ~~~-fenced block (valid CommonMark) was misparsed as a heading --
recognize both fence styles.
- run_pageindex.py: --if-thinning wasn't migrated to the bare-flag +
legacy-yes/no convention the other four --if-add-* flags got; bare usage
raised an argparse error and it never went through the shared coercion.
- types.py: DocumentDetail's `structure` field was inside the class's
total=False body, so TypedDict rules made it optional even though every
backend always populates it. Split into a required base class.
Adds regression tests for all of the above. Full suite: 244 passed, 2 skipped
(one pre-existing, unrelated flaky cloud-streaming test).
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-09 11:58:59 +08:00
|
|
|
_mock_content_based_pipeline(monkeypatch, [{"title": "T", "start_index": 1, "end_index": 1}])
|
|
|
|
|
# Summary generation itself would otherwise make a real LLM call.
|
|
|
|
|
monkeypatch.setattr(
|
|
|
|
|
utils_mod, "generate_summaries_for_structure",
|
|
|
|
|
AsyncMock(side_effect=lambda structure, model=None: [
|
|
|
|
|
n.__setitem__("summary", "fake") for n in structure
|
|
|
|
|
]),
|
|
|
|
|
)
|
fix: address xhigh code-review findings on 2d46d68..8f536cb
Verified 12 findings from an xhigh-effort review of the prior review-fix
batch; all confirmed real. Most trace back to one root cause: the
build_index() text-stripping fix (8f536cb) correctly stopped Markdown from
leaking full text by default, but broke every path that assumed text could
be re-read later.
Correctness:
- LocalBackend._fill_node_text (get_document(include_text=True)) only handled
PDF's start_index/end_index convention; Markdown nodes use line_num and got
silently empty text. Now handles both.
- get_page_content's Markdown fallback (triggered when a StorageEngine
legitimately returns None from get_pages()) read from the now-text-stripped
structure. It now re-derives from the source file, mirroring the PDF
fallback, so it no longer depends on structure text at all.
- add_document's PDF-only text-stripping branch (with the stale "markdown
needs text in structure for fallback retrieval" comment) is now dead/wrong
since build_index() already applies if_add_node_text uniformly — removed.
- _validate_llm_provider's keyless-provider allowlist was missing several
local LiteLLM providers (xinference, llamafile, triton, oobabooga,
openai_like, docker_model_runner, custom, custom_openai, petals) that need
no API key just like ollama/lm_studio; expanded.
- The three agent-tool closures (get_document, get_document_structure,
get_page_content) had three different not-found patterns; two bypassed the
backend's DocumentNotFoundError entirely. Extracted LocalBackend.
_require_document as the single existence check every method/tool now uses.
- examples/agentic_vectorless_rag_demo.py's hand-rolled Agent() didn't apply
the litellm/ prefix normalization the SDK does internally, so its own
documented "any LiteLLM provider" claim broke for non-openai models.
- cloud delete_collection's cache eviction removed the "folders unavailable"
None sentinel too, forcing a wasted re-fetch; now only pops on a real id.
Cleanup / altitude:
- build_index() skips the remove_structure_text walk entirely when text was
never added (content_based + if_add_node_summary=False + if_add_node_text=
False) instead of a guaranteed no-op tree walk.
- page_index()'s locals()-capture-as-kwargs (fragile by construction) replaced
with an explicit dict of the named parameters.
- run_pageindex.py's _cli_bool and the page_index_md.py legacy shim's
_coerce_bool were duplicate, diverging implementations; both now bind
directly to the canonical pageindex.index.page_index_md._coerce_bool.
- retrieve.py's _get_md_page_content delegated its own traversal instead of
calling the canonical get_md_page_content; now a one-line delegation.
- FileTypeError's docstring now calls out the except-ordering gotcha from
also subclassing ValueError.
17 new regression tests (tests/test_review_fixes_2.py) plus 2 updated in
tests/test_legacy_shims.py for the simplified md_to_tree shim. Full suite:
210 passed, 2 skipped.
Claude-Session: https://claude.ai/code/session_01Kx5DgKbhK1N8autqXH8SmS
2026-07-08 21:56:41 +08:00
|
|
|
|
|
|
|
|
nodes = [ContentNode(content="page one text", tokens=5, index=1)]
|
|
|
|
|
parsed = ParsedDocument(doc_name="d", nodes=nodes)
|
|
|
|
|
opt = IndexConfig(if_add_node_summary=True, if_add_doc_description=False,
|
|
|
|
|
if_add_node_text=False)
|
|
|
|
|
pipeline.build_index(parsed, opt=opt)
|
|
|
|
|
assert len(calls) == 1, "text WAS added for summary generation, so it must still be stripped"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #9/#10: run_pageindex._cli_bool and the shim's md_to_tree are the same
|
|
|
|
|
# object as the canonical implementation (no drift possible) ──────
|
|
|
|
|
def test_cli_bool_is_the_canonical_coerce_bool():
|
|
|
|
|
import run_pageindex
|
|
|
|
|
from pageindex.index.page_index_md import _coerce_bool
|
|
|
|
|
assert run_pageindex._cli_bool is _coerce_bool
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# ── #11: retrieve._get_md_page_content delegates to the canonical function ───
|
|
|
|
|
def test_retrieve_md_page_content_delegates_to_canonical():
|
|
|
|
|
from pageindex import retrieve
|
|
|
|
|
structure = [{"line_num": 5, "text": "five", "nodes": [
|
|
|
|
|
{"line_num": 40, "text": "forty", "nodes": []},
|
|
|
|
|
]}]
|
|
|
|
|
out = retrieve._get_md_page_content({"structure": structure}, [5])
|
|
|
|
|
assert [r["page"] for r in out] == [5]
|