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.

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

Jump to bottom

Citation De-duplication from @TaylorN15 #476

Merged
merged 1 commit into from
Jan 24, 2024
Merged

Citation De-duplication from @TaylorN15 #476

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
53 changes: 19 additions & 34 deletions app/frontend/src/components/Answer/AnswerParser.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,16 +9,14 @@ type HtmlParsedAnswer = {
citations: string[];
sourceFiles: Record<string, string>;
pageNumbers: Record<string, number>;
// sourceFiles: {};
// pageNumbers: {};
followupQuestions: string[];
};

type CitationLookup = Record<string, {
citation: string;
source_path: string;
page_number: string;
}>;
}>;

export function parseAnswerToHtml(answer: string, citation_lookup: CitationLookup, onCitationClicked: (citationFilePath: string, citationSourcePath: string, pageNumber: string) => void): HtmlParsedAnswer {
const citations: string[] = [];
Expand All @@ -36,7 +34,7 @@ export function parseAnswerToHtml(answer: string, citation_lookup: CitationLooku

// trim any whitespace from the end of the answer after removing follow-up questions
parsedAnswer = parsedAnswer.trim();

// Split the answer into parts, where the odd parts are citations
const parts = parsedAnswer.split(/\[([^\]]+)\]/g);
const fragments: string[] = parts.map((part, index) => {
Expand All @@ -45,11 +43,8 @@ export function parseAnswerToHtml(answer: string, citation_lookup: CitationLooku
return part;
} else {
// Odd parts are citations as the "FileX" moniker
// if ( typeof((citation_lookup as any)[part]) === "undefined") {

//Added this for citation Bug. aparmar
const citation = citation_lookup[part];

if (!citation) {
// if the citation reference provided by the OpenAI response does not match a key in the citation_lookup object
// then return an empty string to avoid a crash or blank citation
Expand All @@ -58,21 +53,21 @@ export function parseAnswerToHtml(answer: string, citation_lookup: CitationLooku
}
else {
let citationIndex: number;
if (citations.indexOf((citation_lookup as any)[part]) !== -1) {
citationIndex = citations.indexOf((citation_lookup as any)[part]) + 1;

// splitting the full file path from citation_lookup into an array and then slicing it to get the folders, file name, and extension
// the first 4 elements of the full file path are the "https:", "", "blob storaage url", and "container name" which are not needed in the display
let citationShortName: string = (citation_lookup)[part].citation.split("/").slice(4).join("/");

// Check if the citationShortName is already in the citations array
if (citations.includes(citationShortName)) {
// If it exists, use the existing index (add 1 because array is 0-based but citation numbers are 1-based)
citationIndex = citations.indexOf(citationShortName) + 1;
} else {
// splitting the full file path from citation_lookup into an array and then slicing it to get the folders, file name, and extension
// the first 4 elements of the full file path are the "https:", "", "blob storaage url", and "container name" which are not needed in the display

//Updated below code section for citation bug. aparmar
let citationShortName: string = (citation_lookup)[part].citation.split("/").slice(4).join("/");
citations.push(citationShortName);
// switch these to the citationShortName as key to allow dynamic lookup of the source path and page number
// The "FileX" moniker will not be used beyond this point in the UX code
sourceFiles[citationShortName] = citation.source_path;
// pageNumbers[citationShortName] = citation.page_number;

// Check if the page_number property is a valid number.
if (!isNaN(Number(citation.page_number))) {
const pageNumber: number = Number(citation.page_number);
Expand All @@ -82,31 +77,21 @@ export function parseAnswerToHtml(answer: string, citation_lookup: CitationLooku
// The page_number property is not a valid number, but we still generate a citation.
pageNumbers[citationShortName] = NaN;
}
// (sourceFiles as any)[citationShortName] = ((citation_lookup as any)[part].source_path);
// (pageNumbers as any)[citationShortName] = ((citation_lookup as any)[part].page_number);
citationIndex = citations.length;
}
const path = getCitationFilePath(citation.citation);
const sourcePath = citation.source_path;
const pageNumber = citation.page_number;

// const path = getCitationFilePath((citation_lookup as any)[part].citation);
// const sourcePath = (citation_lookup as any)[part].source_path;
// const pageNumber = (citation_lookup as any)[part].page_number;
const path = getCitationFilePath(citation.citation);
const sourcePath = citation.source_path;
const pageNumber = citation.page_number;

return renderToStaticMarkup(
// splitting the full file path from citation_lookup into an array and then slicing it to get the folders, file name, and extension
// the first 4 elements of the full file path are the "https:", "", "blob storaage url", and "container name" which are not needed in the display

return renderToStaticMarkup(
<a className="supContainer" title={citation.citation.split("/").slice(4).join("/")} onClick={() => onCitationClicked(path, sourcePath, pageNumber)}>
{/* <a className="supContainer" title={(citation_lookup as any)[part].citation.split("/").slice(4).join("/")} onClick={() => onCitationClicked(path, sourcePath, pageNumber)}> */}
<sup>{citationIndex}</sup>
</a>
);
}
}
});

});

return {
answerHtml: fragments.join(""),
Expand All @@ -115,4 +100,4 @@ export function parseAnswerToHtml(answer: string, citation_lookup: CitationLooku
pageNumbers,
followupQuestions
};
}
}
Loading