Skip to content

Commit e1d7a61

Browse files
committed
refactor(extraction): separate path and buffer inputs
Require document decoders to consume supplied bytes without pathname fallbacks. Keep explicit path-only adapters for existing callers, use the buffer-only entry point for verified reads, and separate XLSX decoding from workbook formatting. Cover disk-free buffer extraction and preserve existing document output and safety behavior.
1 parent f156d02 commit e1d7a61

6 files changed

Lines changed: 213 additions & 78 deletions

File tree

‎src/core/tools/ReadFileTool.ts‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -20,7 +20,7 @@ import { formatResponse } from "../prompts/responses"
2020
import { RecordSource } from "../context-tracking/FileContextTrackerTypes"
2121
import { isPathOutsideWorkspace } from "../../utils/pathUtils"
2222
import { getReadablePath } from "../../utils/path"
23-
import { extractTextFromFile, addLineNumbers, getSupportedBinaryFormats } from "../../integrations/misc/extract-text"
23+
import { extractTextFromBuffer, addLineNumbers, getSupportedBinaryFormats } from "../../integrations/misc/extract-text"
2424
import { readWithIndentation, readWithSlice } from "../../integrations/misc/indentation-reader"
2525
import { DEFAULT_LINE_LIMIT } from "../prompts/tools/native-tools/read_file"
2626
import type { ToolUse, PushToolResult } from "../../shared/tools"
@@ -430,7 +430,7 @@ export class ReadFileTool extends BaseTool<"read_file"> {
430430
try {
431431
const source = await file.readFile()
432432
throwIfReadFileCancelled(task)
433-
const content = await extractTextFromFile(fullPath, source)
433+
const content = await extractTextFromBuffer(source, fullPath)
434434
throwIfReadFileCancelled(task)
435435
const numberedContent = addLineNumbers(content)
436436
const lineCount = content.split("\n").length

‎src/core/tools/__tests__/readFileTool.spec.ts‎

Lines changed: 14 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -27,7 +27,11 @@ import {
2727
isSupportedImageFormat,
2828
ImageMemoryTracker,
2929
} from "../helpers/imageHelpers"
30-
import { extractTextFromFile, addLineNumbers, getSupportedBinaryFormats } from "../../../integrations/misc/extract-text"
30+
import {
31+
extractTextFromBuffer,
32+
addLineNumbers,
33+
getSupportedBinaryFormats,
34+
} from "../../../integrations/misc/extract-text"
3135
import { readWithIndentation, readWithSlice } from "../../../integrations/misc/indentation-reader"
3236

3337
// ─── Mocks ────────────────────────────────────────────────────────────────────
@@ -62,7 +66,7 @@ vi.mock("fs/promises", () => ({
6266
vi.mock("isbinaryfile")
6367

6468
vi.mock("../../../integrations/misc/extract-text", () => ({
65-
extractTextFromFile: vi.fn(),
69+
extractTextFromBuffer: vi.fn(),
6670
addLineNumbers: vi.fn().mockImplementation((text: string, startLine = 1) => {
6771
if (!text) return ""
6872
const lines = text.split("\n")
@@ -136,7 +140,7 @@ const mockedFsReadFile = vi.mocked(fsPromises.readFile)
136140
const mockedFsStat = vi.mocked(fsPromises.stat)
137141

138142
const mockedIsBinaryFile = vi.mocked(isBinaryFile)
139-
const mockedExtractTextFromFile = vi.mocked(extractTextFromFile)
143+
const mockedExtractTextFromBuffer = vi.mocked(extractTextFromBuffer)
140144
const mockedReadWithSlice = vi.mocked(readWithSlice)
141145
const mockedReadWithIndentation = vi.mocked(readWithIndentation)
142146
const mockedIsSupportedImageFormat = vi.mocked(isSupportedImageFormat)
@@ -437,23 +441,23 @@ describe("ReadFileTool", () => {
437441
const mockTask = createMockTask()
438442
const callbacks = createMockCallbacks()
439443

440-
mockedExtractTextFromFile.mockResolvedValue("PDF content here")
444+
mockedExtractTextFromBuffer.mockResolvedValue("PDF content here")
441445

442446
await readFileTool.execute({ path: "document.pdf" }, mockTask as any, callbacks)
443447

444-
expect(mockedExtractTextFromFile).toHaveBeenCalled()
448+
expect(mockedExtractTextFromBuffer).toHaveBeenCalled()
445449
expect(callbacks.pushToolResult).toHaveBeenCalledWith(expect.stringContaining("PDF content here"))
446450
})
447451

448452
it("should extract text from DOCX files", async () => {
449453
const mockTask = createMockTask()
450454
const callbacks = createMockCallbacks()
451455

452-
mockedExtractTextFromFile.mockResolvedValue("DOCX content here")
456+
mockedExtractTextFromBuffer.mockResolvedValue("DOCX content here")
453457

454458
await readFileTool.execute({ path: "document.docx" }, mockTask as any, callbacks)
455459

456-
expect(mockedExtractTextFromFile).toHaveBeenCalled()
460+
expect(mockedExtractTextFromBuffer).toHaveBeenCalled()
457461
expect(callbacks.pushToolResult).toHaveBeenCalledWith(expect.stringContaining("DOCX content here"))
458462
})
459463

@@ -473,7 +477,7 @@ describe("ReadFileTool", () => {
473477
const mockTask = createMockTask()
474478
const callbacks = createMockCallbacks()
475479

476-
mockedExtractTextFromFile.mockRejectedValue(new Error("Extraction failed"))
480+
mockedExtractTextFromBuffer.mockRejectedValue(new Error("Extraction failed"))
477481

478482
await readFileTool.execute({ path: "corrupt.pdf" }, mockTask as any, callbacks)
479483

@@ -1024,7 +1028,7 @@ describe("ReadFileTool", () => {
10241028
)
10251029
expect(mockedFsReadFile).not.toHaveBeenCalled()
10261030
expect(mockedFsStat).not.toHaveBeenCalled()
1027-
expect(mockedExtractTextFromFile).not.toHaveBeenCalled()
1031+
expect(mockedExtractTextFromBuffer).not.toHaveBeenCalled()
10281032
},
10291033
)
10301034

@@ -1049,7 +1053,7 @@ describe("ReadFileTool", () => {
10491053
)
10501054
expect(mockedFsReadFile).not.toHaveBeenCalled()
10511055
expect(mockedFsStat).not.toHaveBeenCalled()
1052-
expect(mockedExtractTextFromFile).not.toHaveBeenCalled()
1056+
expect(mockedExtractTextFromBuffer).not.toHaveBeenCalled()
10531057
})
10541058

10551059
it.each(["src/app.ts", "src/файл.ts", "src/e\u0301-📚.ts"])(
Lines changed: 33 additions & 32 deletions
Original file line numberDiff line numberDiff line change
@@ -1,9 +1,10 @@
11
import ExcelJS from "exceljs"
2-
import { extractTextFromXLSX } from "../extract-text-from-xlsx"
2+
import { formatWorkbook } from "../extract-text-from-xlsx"
3+
import { extractTextFromFile } from "../extract-text"
34

4-
describe("extractTextFromXLSX", () => {
5+
describe("formatWorkbook", () => {
56
describe("basic functionality", () => {
6-
it("should extract text with proper formatting", async () => {
7+
it("should extract text with proper formatting", () => {
78
const workbook = new ExcelJS.Workbook()
89
const worksheet = workbook.addWorksheet("Sheet1")
910

@@ -12,22 +13,22 @@ describe("extractTextFromXLSX", () => {
1213
worksheet.getCell("A2").value = "Test"
1314
worksheet.getCell("B2").value = 123
1415

15-
const result = await extractTextFromXLSX(workbook)
16+
const result = formatWorkbook(workbook)
1617

1718
expect(result).toContain("--- Sheet: Sheet1 ---")
1819
expect(result).toContain("Hello\tWorld")
1920
expect(result).toContain("Test\t123")
2021
})
2122

22-
it("should skip rows with no content", async () => {
23+
it("should skip rows with no content", () => {
2324
const workbook = new ExcelJS.Workbook()
2425
const worksheet = workbook.addWorksheet("Sheet1")
2526

2627
worksheet.getCell("A1").value = "Row 1"
2728
// Row 2 is completely empty
2829
worksheet.getCell("A3").value = "Row 3"
2930

30-
const result = await extractTextFromXLSX(workbook)
31+
const result = formatWorkbook(workbook)
3132

3233
expect(result).toContain("Row 1")
3334
expect(result).toContain("Row 3")
@@ -37,7 +38,7 @@ describe("extractTextFromXLSX", () => {
3738
})
3839

3940
describe("sheet handling", () => {
40-
it("should process multiple sheets", async () => {
41+
it("should process multiple sheets", () => {
4142
const workbook = new ExcelJS.Workbook()
4243

4344
const sheet1 = workbook.addWorksheet("First Sheet")
@@ -46,15 +47,15 @@ describe("extractTextFromXLSX", () => {
4647
const sheet2 = workbook.addWorksheet("Second Sheet")
4748
sheet2.getCell("A1").value = "Sheet 2 Data"
4849

49-
const result = await extractTextFromXLSX(workbook)
50+
const result = formatWorkbook(workbook)
5051

5152
expect(result).toContain("--- Sheet: First Sheet ---")
5253
expect(result).toContain("Sheet 1 Data")
5354
expect(result).toContain("--- Sheet: Second Sheet ---")
5455
expect(result).toContain("Sheet 2 Data")
5556
})
5657

57-
it("should skip hidden sheets", async () => {
58+
it("should skip hidden sheets", () => {
5859
const workbook = new ExcelJS.Workbook()
5960

6061
const visibleSheet = workbook.addWorksheet("Visible Sheet")
@@ -64,15 +65,15 @@ describe("extractTextFromXLSX", () => {
6465
hiddenSheet.getCell("A1").value = "Hidden Data"
6566
hiddenSheet.state = "hidden"
6667

67-
const result = await extractTextFromXLSX(workbook)
68+
const result = formatWorkbook(workbook)
6869

6970
expect(result).toContain("--- Sheet: Visible Sheet ---")
7071
expect(result).toContain("Visible Data")
7172
expect(result).not.toContain("--- Sheet: Hidden Sheet ---")
7273
expect(result).not.toContain("Hidden Data")
7374
})
7475

75-
it("should skip very hidden sheets", async () => {
76+
it("should skip very hidden sheets", () => {
7677
const workbook = new ExcelJS.Workbook()
7778

7879
const visibleSheet = workbook.addWorksheet("Visible Sheet")
@@ -82,7 +83,7 @@ describe("extractTextFromXLSX", () => {
8283
veryHiddenSheet.getCell("A1").value = "Very Hidden Data"
8384
veryHiddenSheet.state = "veryHidden"
8485

85-
const result = await extractTextFromXLSX(workbook)
86+
const result = formatWorkbook(workbook)
8687

8788
expect(result).toContain("--- Sheet: Visible Sheet ---")
8889
expect(result).toContain("Visible Data")
@@ -92,7 +93,7 @@ describe("extractTextFromXLSX", () => {
9293
})
9394

9495
describe("formatCellValue logic", () => {
95-
it("should handle null and undefined values", async () => {
96+
it("should handle null and undefined values", () => {
9697
const workbook = new ExcelJS.Workbook()
9798
const worksheet = workbook.addWorksheet("Sheet1")
9899

@@ -101,7 +102,7 @@ describe("extractTextFromXLSX", () => {
101102
worksheet.getCell("A3").value = undefined
102103
worksheet.getCell("A4").value = "After"
103104

104-
const result = await extractTextFromXLSX(workbook)
105+
const result = formatWorkbook(workbook)
105106

106107
expect(result).toContain("Before")
107108
expect(result).toContain("After")
@@ -111,43 +112,43 @@ describe("extractTextFromXLSX", () => {
111112
expect(dataLines).toHaveLength(2) // Only 'Before' and 'After' should create content
112113
})
113114

114-
it("should format dates correctly", async () => {
115+
it("should format dates correctly", () => {
115116
const workbook = new ExcelJS.Workbook()
116117
const worksheet = workbook.addWorksheet("Sheet1")
117118

118119
const testDate = new Date("2023-12-25")
119120
worksheet.getCell("A1").value = testDate
120121

121-
const result = await extractTextFromXLSX(workbook)
122+
const result = formatWorkbook(workbook)
122123

123124
expect(result).toContain("2023-12-25")
124125
})
125126

126-
it("should handle error values", async () => {
127+
it("should handle error values", () => {
127128
const workbook = new ExcelJS.Workbook()
128129
const worksheet = workbook.addWorksheet("Sheet1")
129130

130131
worksheet.getCell("A1").value = { error: "#DIV/0!" }
131132

132-
const result = await extractTextFromXLSX(workbook)
133+
const result = formatWorkbook(workbook)
133134

134135
expect(result).toContain("[Error: #DIV/0!]")
135136
})
136137

137-
it("should handle rich text", async () => {
138+
it("should handle rich text", () => {
138139
const workbook = new ExcelJS.Workbook()
139140
const worksheet = workbook.addWorksheet("Sheet1")
140141

141142
worksheet.getCell("A1").value = {
142143
richText: [{ text: "Hello " }, { text: "World", font: { bold: true } }],
143144
}
144145

145-
const result = await extractTextFromXLSX(workbook)
146+
const result = formatWorkbook(workbook)
146147

147148
expect(result).toContain("Hello World")
148149
})
149150

150-
it("should handle hyperlinks", async () => {
151+
it("should handle hyperlinks", () => {
151152
const workbook = new ExcelJS.Workbook()
152153
const worksheet = workbook.addWorksheet("Sheet1")
153154

@@ -156,45 +157,45 @@ describe("extractTextFromXLSX", () => {
156157
hyperlink: "https://roocode.com/",
157158
}
158159

159-
const result = await extractTextFromXLSX(workbook)
160+
const result = formatWorkbook(workbook)
160161

161162
expect(result).toContain("Roo Code (https://roocode.com/)")
162163
})
163164

164-
it("should handle formulas with and without results", async () => {
165+
it("should handle formulas with and without results", () => {
165166
const workbook = new ExcelJS.Workbook()
166167
const worksheet = workbook.addWorksheet("Sheet1")
167168

168169
worksheet.getCell("A1").value = { formula: "A2+A3", result: 30 }
169170
worksheet.getCell("A2").value = { formula: "SUM(B1:B10)" }
170171

171-
const result = await extractTextFromXLSX(workbook)
172+
const result = formatWorkbook(workbook)
172173

173174
expect(result).toContain("30") // Formula with result
174175
expect(result).toContain("[Formula: SUM(B1:B10)]") // Formula without result
175176
})
176177
})
177178

178179
describe("edge cases", () => {
179-
it("should handle empty workbook", async () => {
180+
it("should handle empty workbook", () => {
180181
const workbook = new ExcelJS.Workbook()
181182
workbook.addWorksheet("Empty Sheet")
182183

183-
const result = await extractTextFromXLSX(workbook)
184+
const result = formatWorkbook(workbook)
184185

185186
expect(result).toContain("--- Sheet: Empty Sheet ---")
186187
expect(result.trim()).toBe("--- Sheet: Empty Sheet ---")
187188
})
188189

189-
it("should handle workbook with only empty cells", async () => {
190+
it("should handle workbook with only empty cells", () => {
190191
const workbook = new ExcelJS.Workbook()
191192
const worksheet = workbook.addWorksheet("Sheet1")
192193

193194
// Set cells but leave them empty
194195
worksheet.getCell("A1").value = ""
195196
worksheet.getCell("B1").value = ""
196197

197-
const result = await extractTextFromXLSX(workbook)
198+
const result = formatWorkbook(workbook)
198199

199200
expect(result).toContain("--- Sheet: Sheet1 ---")
200201
// Should not contain any data rows since empty strings don't count as content
@@ -203,19 +204,19 @@ describe("extractTextFromXLSX", () => {
203204
})
204205
})
205206

206-
describe("function overloads", () => {
207-
it("should work with workbook objects", async () => {
207+
describe("workbook formatting and path adapter", () => {
208+
it("should work with workbook objects", () => {
208209
const workbook = new ExcelJS.Workbook()
209210
const worksheet = workbook.addWorksheet("Test")
210211
worksheet.getCell("A1").value = "Test Data"
211212

212-
const result = await extractTextFromXLSX(workbook)
213+
const result = formatWorkbook(workbook)
213214

214215
expect(result).toContain("Test Data")
215216
})
216217

217218
it("should reject invalid file paths", async () => {
218-
await expect(extractTextFromXLSX("/non/existent/file.xlsx")).rejects.toThrow()
219+
await expect(extractTextFromFile("/non/existent/file.xlsx")).rejects.toThrow()
219220
})
220221
})
221222
})

0 commit comments

Comments
 (0)