@credal/actions 0.2.116 → 0.2.118
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/actions/autogen/templates.js +28 -0
- package/dist/actions/autogen/types.d.ts +9 -0
- package/dist/actions/autogen/types.js +23 -1
- package/dist/actions/providers/confluence/updatePage.js +5 -2
- package/dist/actions/providers/firecrawl/scrapeUrl.js +63 -2
- package/dist/actions/providers/generic/fillTemplateAction.d.ts +7 -0
- package/dist/actions/providers/generic/fillTemplateAction.js +18 -0
- package/dist/actions/providers/generic/genericApiCall.d.ts +3 -0
- package/dist/actions/providers/{jira/createTicket.js → generic/genericApiCall.js} +22 -18
- package/dist/actions/providers/google-oauth/getDriveContentById.d.ts +3 -0
- package/dist/actions/providers/google-oauth/getDriveContentById.js +161 -0
- package/dist/actions/providers/google-oauth/searchAndGetDriveContentByKeywords.d.ts +3 -0
- package/dist/actions/providers/google-oauth/searchAndGetDriveContentByKeywords.js +47 -0
- package/dist/actions/providers/google-oauth/searchDriveAndGetContentByKeywords.d.ts +3 -0
- package/dist/actions/providers/google-oauth/searchDriveAndGetContentByKeywords.js +110 -0
- package/dist/actions/providers/google-oauth/searchDriveAndGetContentByQuery.d.ts +3 -0
- package/dist/actions/providers/google-oauth/searchDriveAndGetContentByQuery.js +78 -0
- package/dist/actions/providers/google-oauth/utils/extractContentFromDriveFileId.d.ts +15 -0
- package/dist/actions/providers/google-oauth/utils/extractContentFromDriveFileId.js +129 -0
- package/dist/actions/providers/googlemaps/nearbysearch.js +31 -31
- package/dist/actions/providers/slack/listConversations.d.ts +1 -1
- package/dist/actions/providers/snowflake/runSnowflakeQueryWriteResultsToS3.d.ts +3 -0
- package/dist/actions/providers/snowflake/runSnowflakeQueryWriteResultsToS3.js +154 -0
- package/dist/actions/providers/x/scrapeTweetDataWithNitter.d.ts +3 -0
- package/dist/actions/providers/x/scrapeTweetDataWithNitter.js +45 -0
- package/package.json +1 -1
- package/dist/actions/autogen/definitions.d.ts +0 -5
- package/dist/actions/autogen/definitions.js +0 -132
- package/dist/actions/definitions.js +0 -35
- package/dist/actions/invokeMapper.d.ts +0 -9
- package/dist/actions/invokeMapper.js +0 -33
- package/dist/actions/providers/google-oauth/getSheetValue.d.ts +0 -3
- package/dist/actions/providers/google-oauth/getSheetValue.js +0 -50
- package/dist/actions/providers/google-oauth/getSheetValues.d.ts +0 -3
- package/dist/actions/providers/google-oauth/getSheetValues.js +0 -50
- package/dist/actions/providers/google-oauth/listGmailThreads.d.ts +0 -3
- package/dist/actions/providers/google-oauth/listGmailThreads.js +0 -98
- package/dist/actions/providers/google-oauth/searchGmailMessages.d.ts +0 -3
- package/dist/actions/providers/google-oauth/searchGmailMessages.js +0 -91
- package/dist/actions/providers/jamf/types.d.ts +0 -8
- package/dist/actions/providers/jamf/types.js +0 -7
- package/dist/actions/providers/jira/createTicket.d.ts +0 -3
- package/dist/actions/providers/salesforce/getSalesforceRecordByQuery.d.ts +0 -3
- package/dist/actions/providers/salesforce/getSalesforceRecordByQuery.js +0 -43
- package/dist/actions/providers/slack/list_conversations.d.ts +0 -3
- package/dist/actions/providers/slack/list_conversations.js +0 -60
- package/dist/actions/providers/slack/summarizeChannel.d.ts +0 -3
- package/dist/actions/providers/slack/summarizeChannel.js +0 -51
- package/dist/actions/schema.js +0 -6
- package/dist/actions/types.js +0 -2
- package/dist/main.js +0 -11
|
@@ -3203,6 +3203,34 @@ export const firecrawlScrapeUrlDefinition = {
|
|
|
3203
3203
|
type: "string",
|
|
3204
3204
|
description: "The URL to scrape",
|
|
3205
3205
|
},
|
|
3206
|
+
waitMs: {
|
|
3207
|
+
type: "number",
|
|
3208
|
+
description: "Optional wait time in milliseconds before scraping the page",
|
|
3209
|
+
minimum: 0,
|
|
3210
|
+
},
|
|
3211
|
+
onlyMainContent: {
|
|
3212
|
+
type: "boolean",
|
|
3213
|
+
description: "Extract only the main content of the page, excluding headers, footers, and navigation",
|
|
3214
|
+
},
|
|
3215
|
+
formats: {
|
|
3216
|
+
type: "array",
|
|
3217
|
+
description: "Array of formats to return",
|
|
3218
|
+
items: {
|
|
3219
|
+
type: "string",
|
|
3220
|
+
enum: [
|
|
3221
|
+
"content",
|
|
3222
|
+
"json",
|
|
3223
|
+
"html",
|
|
3224
|
+
"screenshot",
|
|
3225
|
+
"markdown",
|
|
3226
|
+
"rawHtml",
|
|
3227
|
+
"links",
|
|
3228
|
+
"screenshot@fullPage",
|
|
3229
|
+
"extract",
|
|
3230
|
+
"changeTracking",
|
|
3231
|
+
],
|
|
3232
|
+
},
|
|
3233
|
+
},
|
|
3206
3234
|
},
|
|
3207
3235
|
},
|
|
3208
3236
|
output: {
|
|
@@ -3141,10 +3141,19 @@ export type firecrawlDeepResearchOutputType = z.infer<typeof firecrawlDeepResear
|
|
|
3141
3141
|
export type firecrawlDeepResearchFunction = ActionFunction<firecrawlDeepResearchParamsType, AuthParamsType, firecrawlDeepResearchOutputType>;
|
|
3142
3142
|
export declare const firecrawlScrapeUrlParamsSchema: z.ZodObject<{
|
|
3143
3143
|
url: z.ZodString;
|
|
3144
|
+
waitMs: z.ZodOptional<z.ZodNumber>;
|
|
3145
|
+
onlyMainContent: z.ZodOptional<z.ZodBoolean>;
|
|
3146
|
+
formats: z.ZodOptional<z.ZodArray<z.ZodEnum<["content", "json", "html", "screenshot", "markdown", "rawHtml", "links", "screenshot@fullPage", "extract", "changeTracking"]>, "many">>;
|
|
3144
3147
|
}, "strip", z.ZodTypeAny, {
|
|
3145
3148
|
url: string;
|
|
3149
|
+
waitMs?: number | undefined;
|
|
3150
|
+
onlyMainContent?: boolean | undefined;
|
|
3151
|
+
formats?: ("content" | "json" | "html" | "screenshot" | "markdown" | "rawHtml" | "links" | "screenshot@fullPage" | "extract" | "changeTracking")[] | undefined;
|
|
3146
3152
|
}, {
|
|
3147
3153
|
url: string;
|
|
3154
|
+
waitMs?: number | undefined;
|
|
3155
|
+
onlyMainContent?: boolean | undefined;
|
|
3156
|
+
formats?: ("content" | "json" | "html" | "screenshot" | "markdown" | "rawHtml" | "links" | "screenshot@fullPage" | "extract" | "changeTracking")[] | undefined;
|
|
3148
3157
|
}>;
|
|
3149
3158
|
export type firecrawlScrapeUrlParamsType = z.infer<typeof firecrawlScrapeUrlParamsSchema>;
|
|
3150
3159
|
export declare const firecrawlScrapeUrlOutputSchema: z.ZodObject<{
|
|
@@ -976,7 +976,29 @@ export const firecrawlDeepResearchOutputSchema = z.object({
|
|
|
976
976
|
}))
|
|
977
977
|
.describe("The sources of the research"),
|
|
978
978
|
});
|
|
979
|
-
export const firecrawlScrapeUrlParamsSchema = z.object({
|
|
979
|
+
export const firecrawlScrapeUrlParamsSchema = z.object({
|
|
980
|
+
url: z.string().describe("The URL to scrape"),
|
|
981
|
+
waitMs: z.number().gte(0).describe("Optional wait time in milliseconds before scraping the page").optional(),
|
|
982
|
+
onlyMainContent: z
|
|
983
|
+
.boolean()
|
|
984
|
+
.describe("Extract only the main content of the page, excluding headers, footers, and navigation")
|
|
985
|
+
.optional(),
|
|
986
|
+
formats: z
|
|
987
|
+
.array(z.enum([
|
|
988
|
+
"content",
|
|
989
|
+
"json",
|
|
990
|
+
"html",
|
|
991
|
+
"screenshot",
|
|
992
|
+
"markdown",
|
|
993
|
+
"rawHtml",
|
|
994
|
+
"links",
|
|
995
|
+
"screenshot@fullPage",
|
|
996
|
+
"extract",
|
|
997
|
+
"changeTracking",
|
|
998
|
+
]))
|
|
999
|
+
.describe("Array of formats to return")
|
|
1000
|
+
.optional(),
|
|
1001
|
+
});
|
|
980
1002
|
export const firecrawlScrapeUrlOutputSchema = z.object({ content: z.string().describe("The content of the URL") });
|
|
981
1003
|
export const firecrawlSearchAndScrapeParamsSchema = z.object({
|
|
982
1004
|
query: z.string().describe("The query to search for"),
|
|
@@ -20,8 +20,11 @@ function getConfluenceRequestConfig(baseUrl, username, apiToken) {
|
|
|
20
20
|
};
|
|
21
21
|
}
|
|
22
22
|
const confluenceUpdatePage = (_a) => __awaiter(void 0, [_a], void 0, function* ({ params, authParams, }) {
|
|
23
|
-
const { pageId,
|
|
24
|
-
const { baseUrl, authToken } = authParams;
|
|
23
|
+
const { pageId, content, title } = params;
|
|
24
|
+
const { baseUrl, authToken, username } = authParams;
|
|
25
|
+
if (!baseUrl || !authToken || !username) {
|
|
26
|
+
throw new Error("Missing required authentication information");
|
|
27
|
+
}
|
|
25
28
|
const config = getConfluenceRequestConfig(baseUrl, username, authToken);
|
|
26
29
|
// Get current version number
|
|
27
30
|
const response = yield axiosClient_1.axiosClient.get(`/api/v2/pages/${pageId}`, config);
|
|
@@ -13,9 +13,70 @@ const scrapeUrl = (_a) => __awaiter(void 0, [_a], void 0, function* ({ params, a
|
|
|
13
13
|
const firecrawl = new FirecrawlApp({
|
|
14
14
|
apiKey: authParams.apiKey,
|
|
15
15
|
});
|
|
16
|
-
const result = yield firecrawl.scrapeUrl(params.url
|
|
16
|
+
const result = yield firecrawl.scrapeUrl(params.url, Object.assign(Object.assign(Object.assign({}, (params.waitMs !== undefined && {
|
|
17
|
+
actions: [{ type: "wait", milliseconds: params.waitMs }],
|
|
18
|
+
})), (params.onlyMainContent !== undefined && {
|
|
19
|
+
onlyMainContent: params.onlyMainContent,
|
|
20
|
+
})), (params.formats !== undefined &&
|
|
21
|
+
params.formats.length > 0 && {
|
|
22
|
+
formats: params.formats,
|
|
23
|
+
})));
|
|
24
|
+
console.log("Result is: ", result);
|
|
25
|
+
if (!result.success) {
|
|
26
|
+
return firecrawlScrapeUrlOutputSchema.parse({
|
|
27
|
+
content: "",
|
|
28
|
+
});
|
|
29
|
+
}
|
|
30
|
+
// Extract content based on requested formats
|
|
31
|
+
let content = "";
|
|
32
|
+
if (params.formats && params.formats.length > 0) {
|
|
33
|
+
const contentParts = [];
|
|
34
|
+
for (const format of params.formats) {
|
|
35
|
+
let formatContent = undefined;
|
|
36
|
+
// Handle different format mappings
|
|
37
|
+
switch (format) {
|
|
38
|
+
case "rawHtml":
|
|
39
|
+
formatContent = result.rawHtml;
|
|
40
|
+
break;
|
|
41
|
+
case "markdown":
|
|
42
|
+
formatContent = result.markdown;
|
|
43
|
+
break;
|
|
44
|
+
case "html":
|
|
45
|
+
formatContent = result.html;
|
|
46
|
+
break;
|
|
47
|
+
case "links":
|
|
48
|
+
formatContent = Array.isArray(result.links)
|
|
49
|
+
? result.links.map(link => (typeof link === "string" ? link : JSON.stringify(link))).join("\n")
|
|
50
|
+
: JSON.stringify(result.links);
|
|
51
|
+
break;
|
|
52
|
+
case "json":
|
|
53
|
+
formatContent = result.json ? JSON.stringify(result.json, null, 2) : undefined;
|
|
54
|
+
break;
|
|
55
|
+
case "extract":
|
|
56
|
+
formatContent = result.extract ? JSON.stringify(result.extract, null, 2) : undefined;
|
|
57
|
+
break;
|
|
58
|
+
case "screenshot":
|
|
59
|
+
formatContent = result.screenshot;
|
|
60
|
+
break;
|
|
61
|
+
case "changeTracking":
|
|
62
|
+
formatContent = result.changeTracking ? JSON.stringify(result.changeTracking, null, 2) : undefined;
|
|
63
|
+
break;
|
|
64
|
+
default:
|
|
65
|
+
formatContent = result[format];
|
|
66
|
+
}
|
|
67
|
+
if (formatContent) {
|
|
68
|
+
const formatHeader = `=== ${format.toUpperCase()} ===`;
|
|
69
|
+
contentParts.push(`${formatHeader}\n${formatContent}`);
|
|
70
|
+
}
|
|
71
|
+
}
|
|
72
|
+
content = contentParts.join("\n\n");
|
|
73
|
+
}
|
|
74
|
+
else {
|
|
75
|
+
// Default to markdown if no formats specified
|
|
76
|
+
content = result.markdown || "";
|
|
77
|
+
}
|
|
17
78
|
return firecrawlScrapeUrlOutputSchema.parse({
|
|
18
|
-
content
|
|
79
|
+
content,
|
|
19
80
|
});
|
|
20
81
|
});
|
|
21
82
|
export default scrapeUrl;
|
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
"use strict";
|
|
2
|
+
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
|
|
3
|
+
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
|
|
4
|
+
return new (P || (P = Promise))(function (resolve, reject) {
|
|
5
|
+
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
|
|
6
|
+
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
|
|
7
|
+
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
|
|
8
|
+
step((generator = generator.apply(thisArg, _arguments || [])).next());
|
|
9
|
+
});
|
|
10
|
+
};
|
|
11
|
+
Object.defineProperty(exports, "__esModule", { value: true });
|
|
12
|
+
const fillTemplateAction = (_a) => __awaiter(void 0, [_a], void 0, function* ({ template }) {
|
|
13
|
+
// Simply return the template without any modification
|
|
14
|
+
return {
|
|
15
|
+
result: template,
|
|
16
|
+
};
|
|
17
|
+
});
|
|
18
|
+
exports.default = fillTemplateAction;
|
|
@@ -13,22 +13,26 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
|
|
|
13
13
|
};
|
|
14
14
|
Object.defineProperty(exports, "__esModule", { value: true });
|
|
15
15
|
const axios_1 = __importDefault(require("axios"));
|
|
16
|
-
const
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
|
|
31
|
-
|
|
32
|
-
|
|
16
|
+
const genericApiCall = (_a) => __awaiter(void 0, [_a], void 0, function* ({ params, }) {
|
|
17
|
+
try {
|
|
18
|
+
const { endpoint, method, headers, body } = params;
|
|
19
|
+
const response = yield (0, axios_1.default)({
|
|
20
|
+
url: endpoint,
|
|
21
|
+
method,
|
|
22
|
+
headers,
|
|
23
|
+
data: method !== "GET" ? body : undefined,
|
|
24
|
+
});
|
|
25
|
+
return {
|
|
26
|
+
statusCode: response.status,
|
|
27
|
+
headers: response.headers,
|
|
28
|
+
data: response.data,
|
|
29
|
+
};
|
|
30
|
+
}
|
|
31
|
+
catch (error) {
|
|
32
|
+
if (axios_1.default.isAxiosError(error)) {
|
|
33
|
+
throw Error("Axios Error: " + (error.message || "Failed to make API call"));
|
|
34
|
+
}
|
|
35
|
+
throw Error("Error: " + (error || "Failed to make API call"));
|
|
36
|
+
}
|
|
33
37
|
});
|
|
34
|
-
exports.default =
|
|
38
|
+
exports.default = genericApiCall;
|
|
@@ -0,0 +1,161 @@
|
|
|
1
|
+
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
|
|
2
|
+
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
|
|
3
|
+
return new (P || (P = Promise))(function (resolve, reject) {
|
|
4
|
+
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
|
|
5
|
+
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
|
|
6
|
+
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
|
|
7
|
+
step((generator = generator.apply(thisArg, _arguments || [])).next());
|
|
8
|
+
});
|
|
9
|
+
};
|
|
10
|
+
import pdf from "pdf-parse/lib/pdf-parse.js";
|
|
11
|
+
import { axiosClient } from "../../util/axiosClient.js";
|
|
12
|
+
import mammoth from "mammoth";
|
|
13
|
+
import { MISSING_AUTH_TOKEN } from "../../util/missingAuthConstants.js";
|
|
14
|
+
const getDriveFileContentByID = (_a) => __awaiter(void 0, [_a], void 0, function* ({ params, authParams, }) {
|
|
15
|
+
if (!authParams.authToken) {
|
|
16
|
+
return { success: false, error: MISSING_AUTH_TOKEN };
|
|
17
|
+
}
|
|
18
|
+
const { fileId, limit } = params;
|
|
19
|
+
try {
|
|
20
|
+
// First, get file metadata to determine the file type
|
|
21
|
+
const metadataUrl = `https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}?fields=name,mimeType,size`;
|
|
22
|
+
const metadataRes = yield axiosClient.get(metadataUrl, {
|
|
23
|
+
headers: {
|
|
24
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
25
|
+
},
|
|
26
|
+
});
|
|
27
|
+
const { name: fileName, mimeType, size } = metadataRes.data;
|
|
28
|
+
// Check if file is too large (50MB limit for safety)
|
|
29
|
+
if (size && parseInt(size) > 50 * 1024 * 1024) {
|
|
30
|
+
return {
|
|
31
|
+
success: false,
|
|
32
|
+
error: "File too large (>50MB)",
|
|
33
|
+
};
|
|
34
|
+
}
|
|
35
|
+
let content = "";
|
|
36
|
+
// Handle different file types - read content directly
|
|
37
|
+
if (mimeType === "application/vnd.google-apps.document") {
|
|
38
|
+
// Google Docs - download as plain text
|
|
39
|
+
const downloadUrl = `https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}?alt=media&format=txt`;
|
|
40
|
+
const downloadRes = yield axiosClient.get(downloadUrl, {
|
|
41
|
+
headers: {
|
|
42
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
43
|
+
},
|
|
44
|
+
responseType: 'text',
|
|
45
|
+
});
|
|
46
|
+
content = downloadRes.data;
|
|
47
|
+
}
|
|
48
|
+
else if (mimeType === "application/vnd.google-apps.spreadsheet") {
|
|
49
|
+
// Google Sheets - download as CSV
|
|
50
|
+
const downloadUrl = `https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}?alt=media&format=csv`;
|
|
51
|
+
const downloadRes = yield axiosClient.get(downloadUrl, {
|
|
52
|
+
headers: {
|
|
53
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
54
|
+
},
|
|
55
|
+
responseType: 'text',
|
|
56
|
+
});
|
|
57
|
+
content = downloadRes.data;
|
|
58
|
+
}
|
|
59
|
+
else if (mimeType === "application/vnd.google-apps.presentation") {
|
|
60
|
+
// Google Slides - download as plain text
|
|
61
|
+
const downloadUrl = `https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}?alt=media&format=txt`;
|
|
62
|
+
const downloadRes = yield axiosClient.get(downloadUrl, {
|
|
63
|
+
headers: {
|
|
64
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
65
|
+
},
|
|
66
|
+
responseType: 'text',
|
|
67
|
+
});
|
|
68
|
+
content = downloadRes.data;
|
|
69
|
+
}
|
|
70
|
+
else if (mimeType === "application/pdf") {
|
|
71
|
+
// PDF files - use pdf-parse
|
|
72
|
+
const downloadUrl = `https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}?alt=media`;
|
|
73
|
+
const downloadRes = yield axiosClient.get(downloadUrl, {
|
|
74
|
+
headers: {
|
|
75
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
76
|
+
},
|
|
77
|
+
responseType: 'arraybuffer',
|
|
78
|
+
});
|
|
79
|
+
try {
|
|
80
|
+
const pdfData = yield pdf(downloadRes.data);
|
|
81
|
+
content = pdfData.text;
|
|
82
|
+
}
|
|
83
|
+
catch (pdfError) {
|
|
84
|
+
return {
|
|
85
|
+
success: false,
|
|
86
|
+
error: `Failed to parse PDF: ${pdfError instanceof Error ? pdfError.message : 'Unknown PDF error'}`,
|
|
87
|
+
};
|
|
88
|
+
}
|
|
89
|
+
}
|
|
90
|
+
else if (mimeType === "application/vnd.openxmlformats-officedocument.wordprocessingml.document" ||
|
|
91
|
+
mimeType === "application/msword") {
|
|
92
|
+
// Word documents (.docx or .doc) - download and extract text using mammoth
|
|
93
|
+
const downloadUrl = `https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}?alt=media`;
|
|
94
|
+
const downloadRes = yield axiosClient.get(downloadUrl, {
|
|
95
|
+
headers: {
|
|
96
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
97
|
+
},
|
|
98
|
+
responseType: 'arraybuffer',
|
|
99
|
+
});
|
|
100
|
+
try {
|
|
101
|
+
// mammoth works with .docx files. It will ignore formatting and return raw text
|
|
102
|
+
const result = yield mammoth.extractRawText({ buffer: Buffer.from(downloadRes.data) });
|
|
103
|
+
content = result.value; // raw text
|
|
104
|
+
}
|
|
105
|
+
catch (wordError) {
|
|
106
|
+
return {
|
|
107
|
+
success: false,
|
|
108
|
+
error: `Failed to parse Word document: ${wordError instanceof Error ? wordError.message : 'Unknown Word error'}`,
|
|
109
|
+
};
|
|
110
|
+
}
|
|
111
|
+
}
|
|
112
|
+
else if (mimeType === "text/plain" ||
|
|
113
|
+
mimeType === "text/html" ||
|
|
114
|
+
mimeType === "application/rtf" ||
|
|
115
|
+
(mimeType === null || mimeType === void 0 ? void 0 : mimeType.startsWith("text/"))) {
|
|
116
|
+
// Text-based files
|
|
117
|
+
const downloadUrl = `https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}?alt=media`;
|
|
118
|
+
const downloadRes = yield axiosClient.get(downloadUrl, {
|
|
119
|
+
headers: {
|
|
120
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
121
|
+
},
|
|
122
|
+
responseType: 'text',
|
|
123
|
+
});
|
|
124
|
+
content = downloadRes.data;
|
|
125
|
+
}
|
|
126
|
+
else if (mimeType === null || mimeType === void 0 ? void 0 : mimeType.startsWith("image/")) {
|
|
127
|
+
// Skip images
|
|
128
|
+
return {
|
|
129
|
+
success: false,
|
|
130
|
+
error: "Image files are not supported for text extraction",
|
|
131
|
+
};
|
|
132
|
+
}
|
|
133
|
+
else {
|
|
134
|
+
// Unsupported file type
|
|
135
|
+
return {
|
|
136
|
+
success: false,
|
|
137
|
+
error: `Unsupported file type: ${mimeType}`,
|
|
138
|
+
};
|
|
139
|
+
}
|
|
140
|
+
content = content.trim();
|
|
141
|
+
const originalLength = content.length;
|
|
142
|
+
// Naive way to truncate content
|
|
143
|
+
if (limit && content.length > limit) {
|
|
144
|
+
content = content.substring(0, limit);
|
|
145
|
+
}
|
|
146
|
+
return {
|
|
147
|
+
success: true,
|
|
148
|
+
content,
|
|
149
|
+
fileName,
|
|
150
|
+
fileLength: originalLength,
|
|
151
|
+
};
|
|
152
|
+
}
|
|
153
|
+
catch (error) {
|
|
154
|
+
console.error("Error getting Google Drive file content", error);
|
|
155
|
+
return {
|
|
156
|
+
success: false,
|
|
157
|
+
error: error instanceof Error ? error.message : "Unknown error",
|
|
158
|
+
};
|
|
159
|
+
}
|
|
160
|
+
});
|
|
161
|
+
export default getDriveFileContentByID;
|
|
@@ -0,0 +1,47 @@
|
|
|
1
|
+
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
|
|
2
|
+
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
|
|
3
|
+
return new (P || (P = Promise))(function (resolve, reject) {
|
|
4
|
+
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
|
|
5
|
+
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
|
|
6
|
+
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
|
|
7
|
+
step((generator = generator.apply(thisArg, _arguments || [])).next());
|
|
8
|
+
});
|
|
9
|
+
};
|
|
10
|
+
import { axiosClient } from "../../util/axiosClient.js";
|
|
11
|
+
import { MISSING_AUTH_TOKEN } from "../../util/missingAuthConstants.js";
|
|
12
|
+
const searchDriveByKeywords = (_a) => __awaiter(void 0, [_a], void 0, function* ({ params, authParams, }) {
|
|
13
|
+
var _b;
|
|
14
|
+
if (!authParams.authToken) {
|
|
15
|
+
return { success: false, error: MISSING_AUTH_TOKEN, files: [] };
|
|
16
|
+
}
|
|
17
|
+
const { keywords, limit } = params;
|
|
18
|
+
// Build the query: fullText contains 'keyword1' or fullText contains 'keyword2' ...
|
|
19
|
+
const query = keywords.map(kw => `fullText contains '${kw.replace(/'/g, "\\'")}'`).join(" or ");
|
|
20
|
+
const url = `https://www.googleapis.com/drive/v3/files?q=${encodeURIComponent(query)}&fields=files(id,name,mimeType,webViewLink)&supportsAllDrives=true&includeItemsFromAllDrives=true`;
|
|
21
|
+
// 1. Get the file metadata from google drive search
|
|
22
|
+
let files = [];
|
|
23
|
+
try {
|
|
24
|
+
const res = yield axiosClient.get(url, {
|
|
25
|
+
headers: {
|
|
26
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
27
|
+
},
|
|
28
|
+
});
|
|
29
|
+
files =
|
|
30
|
+
((_b = res.data.files) === null || _b === void 0 ? void 0 : _b.map((file) => ({
|
|
31
|
+
id: file.id || "",
|
|
32
|
+
name: file.name || "",
|
|
33
|
+
mimeType: file.mimeType || "",
|
|
34
|
+
url: file.webViewLink || "",
|
|
35
|
+
}))) || [];
|
|
36
|
+
}
|
|
37
|
+
catch (error) {
|
|
38
|
+
console.error("Error searching Google Drive", error);
|
|
39
|
+
return {
|
|
40
|
+
success: false,
|
|
41
|
+
error: error instanceof Error ? error.message : "Unknown error",
|
|
42
|
+
files: [],
|
|
43
|
+
};
|
|
44
|
+
}
|
|
45
|
+
files = limit ? files.splice(0, limit) : files;
|
|
46
|
+
});
|
|
47
|
+
export default searchDriveByKeywords;
|
|
@@ -0,0 +1,110 @@
|
|
|
1
|
+
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
|
|
2
|
+
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
|
|
3
|
+
return new (P || (P = Promise))(function (resolve, reject) {
|
|
4
|
+
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
|
|
5
|
+
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
|
|
6
|
+
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
|
|
7
|
+
step((generator = generator.apply(thisArg, _arguments || [])).next());
|
|
8
|
+
});
|
|
9
|
+
};
|
|
10
|
+
import { axiosClient } from "../../util/axiosClient.js";
|
|
11
|
+
import { MISSING_AUTH_TOKEN } from "../../util/missingAuthConstants.js";
|
|
12
|
+
import extractContentFromDriveFileId from "./utils/extractContentFromDriveFileId.js";
|
|
13
|
+
import { normalizeText } from "../../../utils/string.js";
|
|
14
|
+
/** Intelligently selects a section of text around the median occurrence of keywords */
|
|
15
|
+
const intelligentSelectByMedianSection = (text, keywords, limit) => {
|
|
16
|
+
if (!text || text.length <= limit)
|
|
17
|
+
return text;
|
|
18
|
+
if (!(keywords === null || keywords === void 0 ? void 0 : keywords.length))
|
|
19
|
+
return text.substring(0, limit);
|
|
20
|
+
// Find all keyword positions (case-insensitive, limited to first 1000 matches)
|
|
21
|
+
const positions = [];
|
|
22
|
+
const normalizedText = normalizeText(text);
|
|
23
|
+
for (const keyword of keywords) {
|
|
24
|
+
if (keyword.length < 3)
|
|
25
|
+
continue; // Skip very short keywords
|
|
26
|
+
let pos = -1;
|
|
27
|
+
const normalizedKeyword = normalizeText(keyword);
|
|
28
|
+
while ((pos = normalizedText.indexOf(normalizedKeyword, pos + 1)) !== -1 && positions.length < 1000) {
|
|
29
|
+
positions.push(pos);
|
|
30
|
+
}
|
|
31
|
+
}
|
|
32
|
+
if (!positions.length)
|
|
33
|
+
return text.substring(0, limit);
|
|
34
|
+
// Find median position (using sort for simplicity, still fast for 1000 elements)
|
|
35
|
+
positions.sort((a, b) => a - b);
|
|
36
|
+
const medianPos = positions[Math.floor(positions.length / 2)];
|
|
37
|
+
// Return window around median
|
|
38
|
+
const half = Math.floor(limit / 2);
|
|
39
|
+
const start = Math.max(0, medianPos - half);
|
|
40
|
+
const end = Math.min(text.length, start + limit);
|
|
41
|
+
return text.substring(start, end);
|
|
42
|
+
};
|
|
43
|
+
const searchDriveAndGetContentByKeywords = (_a) => __awaiter(void 0, [_a], void 0, function* ({ params, authParams, }) {
|
|
44
|
+
var _b;
|
|
45
|
+
if (!authParams.authToken) {
|
|
46
|
+
return { success: false, error: MISSING_AUTH_TOKEN, files: [] };
|
|
47
|
+
}
|
|
48
|
+
const { keywords, fileLimit, fileSizeLimit } = params;
|
|
49
|
+
let files = [];
|
|
50
|
+
// 1. Search for files and get their metadata
|
|
51
|
+
// Build the query: fullText contains 'keyword1' or fullText contains 'keyword2' ...
|
|
52
|
+
const query = keywords.map(kw => `fullText contains '${kw.replace(/'/g, "\\'")}'`).join(" or ");
|
|
53
|
+
const url = `https://www.googleapis.com/drive/v3/files?q=${encodeURIComponent(query)}&fields=files(id,name,mimeType,webViewLink)&supportsAllDrives=true&includeItemsFromAllDrives=true`;
|
|
54
|
+
try {
|
|
55
|
+
const res = yield axiosClient.get(url, {
|
|
56
|
+
headers: {
|
|
57
|
+
Authorization: `Bearer ${authParams.authToken}`,
|
|
58
|
+
},
|
|
59
|
+
});
|
|
60
|
+
files =
|
|
61
|
+
((_b = res.data.files) === null || _b === void 0 ? void 0 : _b.map((file) => ({
|
|
62
|
+
id: file.id,
|
|
63
|
+
name: file.name,
|
|
64
|
+
mimeType: file.mimeType,
|
|
65
|
+
url: file.webViewLink,
|
|
66
|
+
}))) || [];
|
|
67
|
+
}
|
|
68
|
+
catch (error) {
|
|
69
|
+
console.error("Error searching Google Drive", error);
|
|
70
|
+
return {
|
|
71
|
+
success: false,
|
|
72
|
+
error: error instanceof Error ? error.message : "Unknown error",
|
|
73
|
+
files: [],
|
|
74
|
+
};
|
|
75
|
+
}
|
|
76
|
+
files = fileLimit ? files.splice(0, fileLimit) : files;
|
|
77
|
+
// 2. Extract content from files and do some smart range selection
|
|
78
|
+
const processedFiles = yield Promise.all(files
|
|
79
|
+
.filter((file) => file.id && file.mimeType)
|
|
80
|
+
.map((file) => __awaiter(void 0, void 0, void 0, function* () {
|
|
81
|
+
const content = yield extractContentFromDriveFileId({
|
|
82
|
+
params: { fileId: file.id, mimeType: file.mimeType },
|
|
83
|
+
authParams,
|
|
84
|
+
});
|
|
85
|
+
if (content.success) {
|
|
86
|
+
let selectedContent = content.content;
|
|
87
|
+
if (fileSizeLimit && selectedContent && selectedContent.length > fileSizeLimit) {
|
|
88
|
+
selectedContent = intelligentSelectByMedianSection(selectedContent, keywords, fileSizeLimit);
|
|
89
|
+
}
|
|
90
|
+
return {
|
|
91
|
+
id: file.id || "",
|
|
92
|
+
name: file.name || "",
|
|
93
|
+
mimeType: file.mimeType || "",
|
|
94
|
+
url: file.url || "",
|
|
95
|
+
content: selectedContent,
|
|
96
|
+
};
|
|
97
|
+
}
|
|
98
|
+
else {
|
|
99
|
+
return {
|
|
100
|
+
id: file.id || "",
|
|
101
|
+
name: file.name || "",
|
|
102
|
+
mimeType: file.mimeType || "",
|
|
103
|
+
url: file.url || "",
|
|
104
|
+
error: content.error,
|
|
105
|
+
};
|
|
106
|
+
}
|
|
107
|
+
})));
|
|
108
|
+
return { success: true, files: processedFiles };
|
|
109
|
+
});
|
|
110
|
+
export default searchDriveAndGetContentByKeywords;
|