From 6b1ded41a9b0c2cdd113df3b7c9dfeb6db380653 Mon Sep 17 00:00:00 2001
From: Fabian Wimmer <github@insightby.ai>
Date: Mon, 3 Jun 2024 17:22:18 +0200
Subject: [PATCH] feat: LlamaParse: add gpt4o-mode, invalidate cache, skip
diagonal text, update supported file types (#889)
Co-authored-by: Marcus Schiesser <marcus.schiesser@googlemail.com>
---
.changeset/khaki-ears-develop.md | 5 +
packages/core/src/readers/LlamaParseReader.ts | 120 ++++++++++++++----
2 files changed, 103 insertions(+), 22 deletions(-)
create mode 100644 .changeset/khaki-ears-develop.md
diff --git a/.changeset/khaki-ears-develop.md b/.changeset/khaki-ears-develop.md
new file mode 100644
index 000000000..9c9c10e65
--- /dev/null
+++ b/.changeset/khaki-ears-develop.md
@@ -0,0 +1,5 @@
+---
+"llamaindex": patch
+---
+
+add gpt4o-mode, invalidate cache and skip diagonal text to LlamaParseReader
diff --git a/packages/core/src/readers/LlamaParseReader.ts b/packages/core/src/readers/LlamaParseReader.ts
index 2996a589e..5bc4934b1 100644
--- a/packages/core/src/readers/LlamaParseReader.ts
+++ b/packages/core/src/readers/LlamaParseReader.ts
@@ -5,45 +5,100 @@ import type { FileReader, Language, ResultType } from "./type.js";
const SupportedFiles: { [key: string]: string } = {
".pdf": "application/pdf",
+ // Documents and Presentations
+ ".602": "application/x-t602",
+ ".abw": "application/x-abiword",
+ ".cgm": "image/cgm",
+ ".cwk": "application/x-cwk",
".doc": "application/msword",
".docx":
"application/vnd.openxmlformats-officedocument.wordprocessingml.document",
".docm": "application/vnd.ms-word.document.macroEnabled.12",
".dot": "application/msword",
+ ".dotm": "application/vnd.ms-word.template.macroEnabled.12",
".dotx":
"application/vnd.openxmlformats-officedocument.wordprocessingml.template",
- ".dotm": "application/vnd.ms-word.template.macroEnabled.12",
- ".rtf": "application/rtf",
- ".wps": "application/vnd.ms-works",
- ".wpd": "application/wordperfect",
- ".sxw": "application/vnd.sun.xml.writer",
- ".stw": "application/vnd.sun.xml.writer.template",
- ".sxg": "application/vnd.sun.xml.writer.global",
- ".pages": "application/x-iwork-pages-sffpages",
+ ".hwp": "application/x-hwp",
+ ".key": "application/x-iwork-keynote-sffkey",
+ ".lwp": "application/vnd.lotus-wordpro",
".mw": "application/macwriteii",
".mcw": "application/macwriteii",
- ".uot": "application/x-uo",
- ".uof": "application/vnd.uoml+xml",
- ".uos": "application/vnd.sun.xml.calc",
- ".uop": "application/vnd.openofficeorg.presentation",
+ ".pages": "application/x-iwork-pages-sffpages",
+ ".pbd": "application/x-pagemaker",
".ppt": "application/vnd.ms-powerpoint",
+ ".pptm": "application/vnd.ms-powerpoint.presentation.macroEnabled.12",
".pptx":
"application/vnd.openxmlformats-officedocument.presentationml.presentation",
".pot": "application/vnd.ms-powerpoint",
- ".pptm": "application/vnd.ms-powerpoint.presentation.macroEnabled.12",
+ ".potm": "application/vnd.ms-powerpoint.template.macroEnabled.12",
".potx":
"application/vnd.openxmlformats-officedocument.presentationml.template",
- ".potm": "application/vnd.ms-powerpoint.template.macroEnabled.12",
- ".key": "application/x-iwork-keynote-sffkey",
- ".odp": "application/vnd.oasis.opendocument.presentation",
- ".odg": "application/vnd.oasis.opendocument.graphics",
- ".otp": "application/vnd.oasis.opendocument.presentation-template",
- ".fopd": "application/vnd.oasis.opendocument.presentation",
- ".sxi": "application/vnd.sun.xml.impress",
+ ".rtf": "application/rtf",
+ ".sda": "application/vnd.stardivision.draw",
+ ".sdd": "application/vnd.stardivision.impress",
+ ".sdp": "application/sdp",
+ ".sdw": "application/vnd.stardivision.writer",
+ ".sgl": "application/vnd.stardivision.writer",
".sti": "application/vnd.sun.xml.impress.template",
+ ".sxi": "application/vnd.sun.xml.impress",
+ ".sxw": "application/vnd.sun.xml.writer",
+ ".stw": "application/vnd.sun.xml.writer.template",
+ ".sxg": "application/vnd.sun.xml.writer.global",
+ ".txt": "text/plain",
+ ".uof": "application/vnd.uoml+xml",
+ ".uop": "application/vnd.openofficeorg.presentation",
+ ".uot": "application/x-uo",
+ ".vor": "application/vnd.stardivision.writer",
+ ".wpd": "application/wordperfect",
+ ".wps": "application/vnd.ms-works",
+ ".xml": "application/xml",
+ ".zabw": "application/x-abiword",
+ // Images
".epub": "application/epub+zip",
- ".html": "text/html",
+ ".jpg": "image/jpeg",
+ ".jpeg": "image/jpeg",
+ ".png": "image/png",
+ ".gif": "image/gif",
+ ".bmp": "image/bmp",
+ ".svg": "image/svg+xml",
+ ".tiff": "image/tiff",
+ ".webp": "image/webp",
+ // Web
".htm": "text/html",
+ ".html": "text/html",
+ // Spreadsheets
+ ".xlsx": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+ ".xls": "application/vnd.ms-excel",
+ ".xlsm": "application/vnd.ms-excel.sheet.macroEnabled.12",
+ ".xlsb": "application/vnd.ms-excel.sheet.binary.macroEnabled.12",
+ ".xlw": "application/vnd.ms-excel",
+ ".csv": "text/csv",
+ ".dif": "application/x-dif",
+ ".sylk": "text/vnd.sylk",
+ ".slk": "text/vnd.sylk",
+ ".prn": "application/x-prn",
+ ".numbers": "application/x-iwork-numbers-sffnumbers",
+ ".et": "application/vnd.ms-excel",
+ ".ods": "application/vnd.oasis.opendocument.spreadsheet",
+ ".fods": "application/vnd.oasis.opendocument.spreadsheet",
+ ".uos1": "application/vnd.uoml+xml",
+ ".uos2": "application/vnd.uoml+xml",
+ ".dbf": "application/vnd.dbf",
+ ".wk1": "application/vnd.lotus-1-2-3",
+ ".wk2": "application/vnd.lotus-1-2-3",
+ ".wk3": "application/vnd.lotus-1-2-3",
+ ".wk4": "application/vnd.lotus-1-2-3",
+ ".wks": "application/vnd.lotus-1-2-3",
+ ".123": "application/vnd.lotus-1-2-3",
+ ".wq1": "application/x-lotus",
+ ".wq2": "application/x-lotus",
+ ".wb1": "application/x-quattro-pro",
+ ".wb2": "application/x-quattro-pro",
+ ".wb3": "application/x-quattro-pro",
+ ".qpw": "application/x-quattro-pro",
+ ".xlr": "application/vnd.ms-works",
+ ".eth": "application/ethos",
+ ".tsv": "text/tab-separated-values",
};
/**
@@ -67,6 +122,14 @@ export class LlamaParseReader implements FileReader {
language: Language = "en";
// The parsing instruction for the parser.
parsingInstruction: string = "";
+ // If set to true, the parser will ignore diagonal text (when the text rotation in degrees modulo 90 is not 0).
+ skipDiagonalText: boolean = false;
+ // If set to true, the cache will be ignored and the document re-processes. All document are kept in cache for 48hours after the job was completed to avoid processing 2 time the same document.
+ invalidateCache: boolean = false;
+ // Whether to use gpt-4o extract text from documents.
+ gpt4oMode: boolean = false;
+ // The API key for the GPT-4o API. Lowers the cost of parsing.
+ gpt4oApiKey?: string;
constructor(params: Partial<LlamaParseReader> = {}) {
Object.assign(this, params);
@@ -77,6 +140,13 @@ export class LlamaParseReader implements FileReader {
);
}
this.apiKey = params.apiKey;
+
+ if (params.gpt4oMode) {
+ params.gpt4oApiKey =
+ params.gpt4oApiKey ?? getEnv("LLAMA_CLOUD_GPT4O_API_KEY");
+
+ this.gpt4oApiKey = params.gpt4oApiKey;
+ }
}
async loadData(file: string): Promise<Document[]> {
@@ -89,7 +159,13 @@ export class LlamaParseReader implements FileReader {
const body = new FormData();
body.set("file", new Blob([data], { type: mimeType }), file);
body.append("language", this.language);
- body.append("parsingInstruction", this.parsingInstruction);
+ body.append("parsing_instruction", this.parsingInstruction);
+ body.append("skip_diagonal_text", this.skipDiagonalText.toString());
+ body.append("invalidate_cache", this.invalidateCache.toString());
+ body.append("gpt4o_mode", this.gpt4oMode.toString());
+ if (this.gpt4oMode && this.gpt4oApiKey) {
+ body.append("gpt4o_api_key", this.gpt4oApiKey);
+ }
const headers = {
Authorization: `Bearer ${this.apiKey}`,
--
GitLab