Skip to content
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.

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

Feature/show proper source citation #36

Merged
merged 7 commits into from
Sep 26, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
7 changes: 5 additions & 2 deletions mind_palace/app.py
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,10 @@ def load_index(model):

st.markdown("### Sources")
for i, source_node in enumerate(response.source_nodes):
with st.expander(f"[{i + 1}] {source_node.node.node_id}"):
st.write(f"relevancy score: {source_node.score}")
with st.expander(f"[{i + 1}] {source_node.node.metadata['citation']}"):
if source_node.metadata["paragraph_number"]:
st.write(
f"paragraph number: {source_node.metadata['paragraph_number']}"
)
st.write("original text:")
st.write(source_node.node.get_text().split(":", 1)[1])
37 changes: 36 additions & 1 deletion mind_palace/docs.py
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
from llama_index.schema import TextNode, NodeRelationship, RelatedNodeInfo
import grobid_tei_xml
from llama_index.schema import NodeRelationship, RelatedNodeInfo, TextNode


def load_tei_xml(file_path):
Expand All @@ -8,6 +8,22 @@ def load_tei_xml(file_path):
return grobid_tei_xml.parse_document_xml(xml_file.read())


def cite_authors(xml):
first_author = xml.header.authors[0]
first_author_display_name = f"{first_author.surname}, {first_author.given_name}"
return first_author_display_name + (
", et al" if len(xml.header.authors) > 1 else ""
)


def cite_journal(xml):
return f"{xml.header.journal_abbrev} {xml.header.date};{xml.header.volume}({xml.header.issue}). doi:{xml.header.doi}"


def cite(xml):
return f"{cite_authors(xml)}. {xml.header.title}. {cite_journal(xml)}."


def title(xml, doc_id):
return TextNode(
text=xml.header.title,
Expand Down Expand Up @@ -44,6 +60,7 @@ def body(xml, doc_id):
return [
TextNode(
text=line,
metadata={"paragraph_number": index + 1},
id_=f"{doc_id}-body-paragraph-{index}",
)
for index, line in enumerate(xml.body.split("\n"))
Expand All @@ -64,3 +81,21 @@ def set_relationships(title_node, abstract_node, body_nodes):
set_prev_relationships(body_nodes)
set_next_relationships(body_nodes)
return


def _set_citation_metadata(citation, node):
node.metadata["citation"] = citation
return


def set_citations(xml, nodes):
citation = cite(xml)
for node in nodes:
if isinstance(node, list):
for sub_node in node:
_set_citation_metadata(citation, sub_node)
continue
else:
_set_citation_metadata(citation, node)

return
4 changes: 3 additions & 1 deletion mind_palace/extract.py
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
import os
from llama_index.schema import TextNode

import docs
from llama_index.schema import TextNode


def _gen_document_dict(file_path) -> dict[str, TextNode]:
Expand All @@ -15,6 +15,8 @@ def _gen_document_dict(file_path) -> dict[str, TextNode]:
body_nodes = docs.body(xml, doi)

docs.set_relationships(title_node, abstract_node, body_nodes)
docs.set_citations(xml=xml, nodes=[title_node, abstract_node, body_nodes])

return {
"title": title_node,
"abstract": abstract_node,
Expand Down
50 changes: 48 additions & 2 deletions tests/unit/test_docs.py
Original file line number Diff line number Diff line change
@@ -1,10 +1,41 @@
from tests.context import docs
from unittest.mock import MagicMock, patch

import grobid_tei_xml.types as grobid_types
from llama_index.schema import TextNode, NodeRelationship
from llama_index.schema import NodeRelationship, TextNode

from tests.context import docs

XML_PATH = "./resources/xmls/12-pdfs-from-steve-aug-22/"


def test_cite_authors():
xml = MagicMock()
xml.header.authors = [
MagicMock(surname="Doe", given_name="John"),
MagicMock(surname="Smith", given_name="Jane"),
]
assert docs.cite_authors(xml) == "Doe, John, et al"

xml.header.authors = [MagicMock(surname="Doe", given_name="John")]
assert docs.cite_authors(xml) == "Doe, John"


def test_cite():
xml = MagicMock()

with patch("docs.cite_authors", return_value="Doe, John"):
xml.header.title = "This is a title"
xml.header.journal_abbrev = "J. Abbrev."
xml.header.date = "2023"
xml.header.volume = "1"
xml.header.issue = "9"
xml.header.doi = "10.1234/1234"
assert (
docs.cite(xml)
== "Doe, John. This is a title. J. Abbrev. 2023;1(9). doi:10.1234/1234."
)


def gen_nodes():
return [
TextNode(text="this is first"),
Expand Down Expand Up @@ -63,6 +94,21 @@ def test_node_relationships():
)


def test_set_citations():
xml = docs.load_tei_xml(
XML_PATH
+ "2010_PhysRevLett_Pulsating Tandem Microbubble for Localized and Directional Single-Cell Membrane Poration.pdf.tei.xml"
)
title_node = TextNode(text="this is title")
abstract_node = TextNode(text="this is abstract")
body_nodes = gen_nodes()
docs.set_citations(xml=xml, nodes=[title_node, abstract_node, body_nodes])
assert title_node.metadata["citation"] == docs.cite(xml)
assert abstract_node.metadata["citation"] == docs.cite(xml)
for body_node in body_nodes:
assert body_node.metadata["citation"] == docs.cite(xml)


def test_load_tei_xml():
xml = docs.load_tei_xml(
XML_PATH
Expand Down