@codecloud-dev/pdf-lib-plus 1.18.2 → 1.18.3
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/README.md +53 -4
- package/README.zh.md +31 -2
- package/cjs/api/PDFPage.d.ts +12 -0
- package/cjs/api/PDFPage.d.ts.map +1 -1
- package/cjs/api/PDFPage.js +18 -0
- package/cjs/api/PDFPage.js.map +1 -1
- package/cjs/api/index.d.ts +1 -0
- package/cjs/api/index.d.ts.map +1 -1
- package/cjs/api/index.js +1 -0
- package/cjs/api/index.js.map +1 -1
- package/cjs/api/rotations.d.ts +1 -1
- package/cjs/api/text_extraction.d.ts +20 -0
- package/cjs/api/text_extraction.d.ts.map +1 -0
- package/cjs/api/text_extraction.js +494 -0
- package/cjs/api/text_extraction.js.map +1 -0
- package/cjs/core/index.d.ts +1 -0
- package/cjs/core/index.d.ts.map +1 -1
- package/cjs/core/index.js +2 -0
- package/cjs/core/index.js.map +1 -1
- package/cjs/core/operators/PDFOperator.d.ts +4 -0
- package/cjs/core/operators/PDFOperator.d.ts.map +1 -1
- package/cjs/core/operators/PDFOperator.js +8 -0
- package/cjs/core/operators/PDFOperator.js.map +1 -1
- package/cjs/core/parser/PDFContentStreamParser.d.ts +29 -0
- package/cjs/core/parser/PDFContentStreamParser.d.ts.map +1 -0
- package/cjs/core/parser/PDFContentStreamParser.js +97 -0
- package/cjs/core/parser/PDFContentStreamParser.js.map +1 -0
- package/cjs/core/structures/PDFContentStream.d.ts +2 -0
- package/cjs/core/structures/PDFContentStream.d.ts.map +1 -1
- package/cjs/core/structures/PDFContentStream.js +4 -0
- package/cjs/core/structures/PDFContentStream.js.map +1 -1
- package/cjs/core/structures/PDFPageLeaf.d.ts +2 -0
- package/cjs/core/structures/PDFPageLeaf.d.ts.map +1 -1
- package/cjs/core/structures/PDFPageLeaf.js +4 -0
- package/cjs/core/structures/PDFPageLeaf.js.map +1 -1
- package/cjs/utils/arrays.d.ts +1 -1
- package/dist/pdf-lib.esm.js +595 -1
- package/dist/pdf-lib.esm.js.map +1 -1
- package/dist/pdf-lib.esm.min.js +1 -1
- package/dist/pdf-lib.esm.min.js.map +1 -1
- package/dist/pdf-lib.js +598 -0
- package/dist/pdf-lib.js.map +1 -1
- package/dist/pdf-lib.min.js +1 -1
- package/dist/pdf-lib.min.js.map +1 -1
- package/docs/MIGRATION.md +80 -0
- package/docs/TEXT_EXTRACTION.md +102 -0
- package/es/api/PDFPage.d.ts +12 -0
- package/es/api/PDFPage.d.ts.map +1 -1
- package/es/api/PDFPage.js +18 -0
- package/es/api/PDFPage.js.map +1 -1
- package/es/api/index.d.ts +1 -0
- package/es/api/index.d.ts.map +1 -1
- package/es/api/index.js +1 -0
- package/es/api/index.js.map +1 -1
- package/es/api/text_extraction.d.ts +20 -0
- package/es/api/text_extraction.d.ts.map +1 -0
- package/es/api/text_extraction.js +490 -0
- package/es/api/text_extraction.js.map +1 -0
- package/es/core/index.d.ts +1 -0
- package/es/core/index.d.ts.map +1 -1
- package/es/core/index.js +1 -0
- package/es/core/index.js.map +1 -1
- package/es/core/operators/PDFOperator.d.ts +4 -0
- package/es/core/operators/PDFOperator.d.ts.map +1 -1
- package/es/core/operators/PDFOperator.js +8 -0
- package/es/core/operators/PDFOperator.js.map +1 -1
- package/es/core/parser/PDFContentStreamParser.d.ts +29 -0
- package/es/core/parser/PDFContentStreamParser.d.ts.map +1 -0
- package/es/core/parser/PDFContentStreamParser.js +95 -0
- package/es/core/parser/PDFContentStreamParser.js.map +1 -0
- package/es/core/structures/PDFContentStream.d.ts +2 -0
- package/es/core/structures/PDFContentStream.d.ts.map +1 -1
- package/es/core/structures/PDFContentStream.js +4 -0
- package/es/core/structures/PDFContentStream.js.map +1 -1
- package/es/core/structures/PDFPageLeaf.d.ts +2 -0
- package/es/core/structures/PDFPageLeaf.d.ts.map +1 -1
- package/es/core/structures/PDFPageLeaf.js +4 -0
- package/es/core/structures/PDFPageLeaf.js.map +1 -1
- package/package.json +2 -1
- package/src/api/PDFPage.ts +20 -0
- package/src/api/index.ts +1 -0
- package/src/api/text_extraction.ts +511 -0
- package/src/core/index.ts +1 -0
- package/src/core/operators/PDFOperator.ts +10 -0
- package/src/core/parser/PDFContentStreamParser.ts +106 -0
- package/src/core/structures/PDFContentStream.ts +5 -0
- package/src/core/structures/PDFPageLeaf.ts +5 -0
- package/ts3.4/cjs/api/PDFPage.d.ts +12 -0
- package/ts3.4/cjs/api/index.d.ts +1 -0
- package/ts3.4/cjs/api/rotations.d.ts +1 -1
- package/ts3.4/cjs/api/text_extraction.d.ts +20 -0
- package/ts3.4/cjs/core/index.d.ts +1 -0
- package/ts3.4/cjs/core/operators/PDFOperator.d.ts +4 -0
- package/ts3.4/cjs/core/parser/PDFContentStreamParser.d.ts +29 -0
- package/ts3.4/cjs/core/structures/PDFContentStream.d.ts +2 -0
- package/ts3.4/cjs/core/structures/PDFPageLeaf.d.ts +2 -0
- package/ts3.4/cjs/utils/arrays.d.ts +1 -1
- package/ts3.4/es/api/PDFPage.d.ts +12 -0
- package/ts3.4/es/api/index.d.ts +1 -0
- package/ts3.4/es/api/text_extraction.d.ts +20 -0
- package/ts3.4/es/core/index.d.ts +1 -0
- package/ts3.4/es/core/operators/PDFOperator.d.ts +4 -0
- package/ts3.4/es/core/parser/PDFContentStreamParser.d.ts +29 -0
- package/ts3.4/es/core/structures/PDFContentStream.d.ts +2 -0
- package/ts3.4/es/core/structures/PDFPageLeaf.d.ts +2 -0
package/ts3.4/cjs/api/index.d.ts
CHANGED
|
@@ -17,7 +17,7 @@ export declare const degreesToRadians: (degree: number) => number;
|
|
|
17
17
|
export declare const radiansToDegrees: (radian: number) => number;
|
|
18
18
|
export declare const toRadians: (rotation: Rotation) => number;
|
|
19
19
|
export declare const toDegrees: (rotation: Rotation) => number;
|
|
20
|
-
export declare const reduceRotation: (degreeAngle?: number) => 0 |
|
|
20
|
+
export declare const reduceRotation: (degreeAngle?: number) => 0 | 180 | 90 | 270;
|
|
21
21
|
export declare const adjustDimsForRotation: (dims: {
|
|
22
22
|
width: number;
|
|
23
23
|
height: number;
|
|
@@ -0,0 +1,20 @@
|
|
|
1
|
+
import PDFPage from "./PDFPage";
|
|
2
|
+
/** 提取页面文本,按行返回。 */
|
|
3
|
+
export declare const extractTextFromPage: (page: PDFPage) => string[];
|
|
4
|
+
/** 提取页面文本,合并为单个字符串(行间以 \n 连接)。 */
|
|
5
|
+
export declare const extractText: (page: PDFPage) => string;
|
|
6
|
+
/**
|
|
7
|
+
* 删除页面中「解码文本包含给定子串」的文本显示算子(文本删除 / redaction 基础)。
|
|
8
|
+
*
|
|
9
|
+
* 行为:解析页面内容流,找到 Tj / TJ / ' / " 算子中解码后含 `substring` 的,
|
|
10
|
+
* 连同其紧邻的换行算子(T* / ' / " 自带的换行语义)一并移除,其余绘图算子
|
|
11
|
+
* 原样保留,然后用一个新内容流整体替换本页内容。
|
|
12
|
+
*
|
|
13
|
+
* 边界(诚实说明):删除的是「整段显示算子」,不是逐字删除——若一段 Tj 里
|
|
14
|
+
* 只有部分文字匹配,整段都会被移除。这是内容流模型的固有限制,符合
|
|
15
|
+
* pdf-lib 一贯的「绘制指令而非文本模型」语义。
|
|
16
|
+
*
|
|
17
|
+
* @returns 被删除的显示算子数量(0 表示没有匹配,页面未改动)。
|
|
18
|
+
*/
|
|
19
|
+
export declare const removeTextContaining: (page: PDFPage, substring: string) => number;
|
|
20
|
+
//# sourceMappingURL=text_extraction.d.ts.map
|
|
@@ -39,6 +39,7 @@ export { default as PDFFlateStream } from "./structures/PDFFlateStream";
|
|
|
39
39
|
export { default as PDFOperator } from "./operators/PDFOperator";
|
|
40
40
|
export { default as PDFOperatorNames } from "./operators/PDFOperatorNames";
|
|
41
41
|
export { default as PDFObjectParser } from "./parser/PDFObjectParser";
|
|
42
|
+
export { default as PDFContentStreamParser } from "./parser/PDFContentStreamParser";
|
|
42
43
|
export { default as PDFObjectStreamParser } from "./parser/PDFObjectStreamParser";
|
|
43
44
|
export { default as PDFParser } from "./parser/PDFParser";
|
|
44
45
|
export { default as PDFXRefStreamParser } from "./parser/PDFXRefStreamParser";
|
|
@@ -11,6 +11,10 @@ declare class PDFOperator {
|
|
|
11
11
|
private readonly name;
|
|
12
12
|
private readonly args;
|
|
13
13
|
private constructor();
|
|
14
|
+
/** 只读访问算子名(如 'Tj' / 'TJ'),供内容流解释与文本提取使用。 */
|
|
15
|
+
getName(): PDFOperatorNames;
|
|
16
|
+
/** 只读访问算子操作数(返回副本,防止外部改写内部状态)。 */
|
|
17
|
+
getArgs(): PDFOperatorArg[];
|
|
14
18
|
clone(context?: PDFContext): PDFOperator;
|
|
15
19
|
toString(): string;
|
|
16
20
|
sizeInBytes(): number;
|
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
import PDFRawStream from "../objects/PDFRawStream";
|
|
2
|
+
import PDFOperator from "../operators/PDFOperator";
|
|
3
|
+
import BaseParser from "./BaseParser";
|
|
4
|
+
import ByteStream from "./ByteStream";
|
|
5
|
+
import PDFContext from "../PDFContext";
|
|
6
|
+
/**
|
|
7
|
+
* PDF 内容流(Content Stream)解析器。
|
|
8
|
+
*
|
|
9
|
+
* 上游 pdf-lib 1.x 只能解析「本库自己在内存里 push 的 PDFOperator」,
|
|
10
|
+
* 对从外部 PDF 加载进来的内容流(PDFRawStream 里的 Flate 压缩字节)没有
|
|
11
|
+
* 公开解析入口——这正是它无法实现文本提取的根因。本解析器补上这一能力:
|
|
12
|
+
* 把解码后的内容流字节还原成 PDFOperator[],供文本提取/内容分析复用。
|
|
13
|
+
*
|
|
14
|
+
* 内容流语法 = 若干「操作数 + 关键字」,关键字即算子名(Tj/TJ/BT/...)。
|
|
15
|
+
* 操作数对象(数字/字符串/十六进制串/名字/数组/字典)复用 PDFObjectParser
|
|
16
|
+
* 的对象解析,保证与本库其它解析器行为一致。
|
|
17
|
+
*/
|
|
18
|
+
declare class PDFContentStreamParser extends BaseParser {
|
|
19
|
+
static forBytes: (bytes: Uint8Array, context: PDFContext) => PDFContentStreamParser;
|
|
20
|
+
static fromPDFRawStream: (stream: PDFRawStream, context: PDFContext) => PDFContentStreamParser;
|
|
21
|
+
private readonly objectParser;
|
|
22
|
+
constructor(byteStream: ByteStream, context: PDFContext);
|
|
23
|
+
/** 解析整个内容流为算子序列。遇到无法识别的 token 跳过而非抛错(容错:真实世界 PDF 很脏)。 */
|
|
24
|
+
parseOperators(): PDFOperator[];
|
|
25
|
+
/** 读取一个关键字 token(字母/符号组成,非空白非分隔符)。 */
|
|
26
|
+
private parseKeyword;
|
|
27
|
+
}
|
|
28
|
+
export default PDFContentStreamParser;
|
|
29
|
+
//# sourceMappingURL=PDFContentStreamParser.d.ts.map
|
|
@@ -8,6 +8,8 @@ declare class PDFContentStream extends PDFFlateStream {
|
|
|
8
8
|
private constructor();
|
|
9
9
|
push(...operators: PDFOperator[]): void;
|
|
10
10
|
clone(context?: PDFContext): PDFContentStream;
|
|
11
|
+
/** 只读访问解析后的算子序列(供文本提取/内容分析使用,返回副本防改写)。 */
|
|
12
|
+
getOperators(): PDFOperator[];
|
|
11
13
|
getContentsString(): string;
|
|
12
14
|
getUnencodedContents(): Uint8Array;
|
|
13
15
|
getUnencodedContentsSize(): number;
|
|
@@ -28,6 +28,8 @@ declare class PDFPageLeaf extends PDFDict {
|
|
|
28
28
|
getInheritableAttribute(name: PDFName): PDFObject | undefined;
|
|
29
29
|
setParent(parentRef: PDFRef): void;
|
|
30
30
|
addContentStream(contentStreamRef: PDFRef): void;
|
|
31
|
+
/** 用单个内容流整体替换本页 Contents(供文本编辑/内容流重写使用)。 */
|
|
32
|
+
setContentStream(contentStreamRef: PDFRef): void;
|
|
31
33
|
wrapContentStreams(startStream: PDFRef, endStream: PDFRef): boolean;
|
|
32
34
|
addAnnot(annotRef: PDFRef): void;
|
|
33
35
|
removeAnnot(annotRef: PDFRef): void;
|
|
@@ -11,6 +11,6 @@ export declare const reverseArray: (array: Uint8Array) => Uint8Array;
|
|
|
11
11
|
export declare const sum: (array: number[] | Uint8Array) => number;
|
|
12
12
|
export declare const range: (start: number, end: number) => number[];
|
|
13
13
|
export declare const pluckIndices: <T>(arr: T[], indices: number[]) => T[];
|
|
14
|
-
export declare const canBeConvertedToUint8Array: (input: any) => input is string |
|
|
14
|
+
export declare const canBeConvertedToUint8Array: (input: any) => input is string | Uint8Array | ArrayBuffer;
|
|
15
15
|
export declare const toUint8Array: (input: string | ArrayBuffer | Uint8Array) => Uint8Array;
|
|
16
16
|
//# sourceMappingURL=arrays.d.ts.map
|
|
@@ -534,6 +534,18 @@ export default class PDFPage {
|
|
|
534
534
|
* ```
|
|
535
535
|
* @returns The default position of the page.
|
|
536
536
|
*/
|
|
537
|
+
/**
|
|
538
|
+
* 提取本页的可搜索文本(尽力而为的文本还原)。
|
|
539
|
+
*
|
|
540
|
+
* 解释内容流里的文本显示算子(Tj/TJ/'/"),按字体编码还原为 Unicode,
|
|
541
|
+
* 并按文本定位算子插入换行。支持标准 14 字体与带 ToUnicode 的嵌入字体;
|
|
542
|
+
* 对无 ToUnicode 的 CID 字体/旋转文本等复杂情形不保证逐字一致。
|
|
543
|
+
*
|
|
544
|
+
* @returns 按绘制顺序排列的文本行数组(空页返回空数组)。
|
|
545
|
+
*/
|
|
546
|
+
extractTextLines(): string[];
|
|
547
|
+
/** 提取本页文本,行间以 `\n` 连接(extractTextLines 的字符串便捷版)。 */
|
|
548
|
+
extractText(): string;
|
|
537
549
|
getPosition(): {
|
|
538
550
|
x: number;
|
|
539
551
|
y: number;
|
package/ts3.4/es/api/index.d.ts
CHANGED
|
@@ -0,0 +1,20 @@
|
|
|
1
|
+
import PDFPage from "./PDFPage";
|
|
2
|
+
/** 提取页面文本,按行返回。 */
|
|
3
|
+
export declare const extractTextFromPage: (page: PDFPage) => string[];
|
|
4
|
+
/** 提取页面文本,合并为单个字符串(行间以 \n 连接)。 */
|
|
5
|
+
export declare const extractText: (page: PDFPage) => string;
|
|
6
|
+
/**
|
|
7
|
+
* 删除页面中「解码文本包含给定子串」的文本显示算子(文本删除 / redaction 基础)。
|
|
8
|
+
*
|
|
9
|
+
* 行为:解析页面内容流,找到 Tj / TJ / ' / " 算子中解码后含 `substring` 的,
|
|
10
|
+
* 连同其紧邻的换行算子(T* / ' / " 自带的换行语义)一并移除,其余绘图算子
|
|
11
|
+
* 原样保留,然后用一个新内容流整体替换本页内容。
|
|
12
|
+
*
|
|
13
|
+
* 边界(诚实说明):删除的是「整段显示算子」,不是逐字删除——若一段 Tj 里
|
|
14
|
+
* 只有部分文字匹配,整段都会被移除。这是内容流模型的固有限制,符合
|
|
15
|
+
* pdf-lib 一贯的「绘制指令而非文本模型」语义。
|
|
16
|
+
*
|
|
17
|
+
* @returns 被删除的显示算子数量(0 表示没有匹配,页面未改动)。
|
|
18
|
+
*/
|
|
19
|
+
export declare const removeTextContaining: (page: PDFPage, substring: string) => number;
|
|
20
|
+
//# sourceMappingURL=text_extraction.d.ts.map
|
package/ts3.4/es/core/index.d.ts
CHANGED
|
@@ -39,6 +39,7 @@ export { default as PDFFlateStream } from "./structures/PDFFlateStream";
|
|
|
39
39
|
export { default as PDFOperator } from "./operators/PDFOperator";
|
|
40
40
|
export { default as PDFOperatorNames } from "./operators/PDFOperatorNames";
|
|
41
41
|
export { default as PDFObjectParser } from "./parser/PDFObjectParser";
|
|
42
|
+
export { default as PDFContentStreamParser } from "./parser/PDFContentStreamParser";
|
|
42
43
|
export { default as PDFObjectStreamParser } from "./parser/PDFObjectStreamParser";
|
|
43
44
|
export { default as PDFParser } from "./parser/PDFParser";
|
|
44
45
|
export { default as PDFXRefStreamParser } from "./parser/PDFXRefStreamParser";
|
|
@@ -11,6 +11,10 @@ declare class PDFOperator {
|
|
|
11
11
|
private readonly name;
|
|
12
12
|
private readonly args;
|
|
13
13
|
private constructor();
|
|
14
|
+
/** 只读访问算子名(如 'Tj' / 'TJ'),供内容流解释与文本提取使用。 */
|
|
15
|
+
getName(): PDFOperatorNames;
|
|
16
|
+
/** 只读访问算子操作数(返回副本,防止外部改写内部状态)。 */
|
|
17
|
+
getArgs(): PDFOperatorArg[];
|
|
14
18
|
clone(context?: PDFContext): PDFOperator;
|
|
15
19
|
toString(): string;
|
|
16
20
|
sizeInBytes(): number;
|
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
import PDFRawStream from "../objects/PDFRawStream";
|
|
2
|
+
import PDFOperator from "../operators/PDFOperator";
|
|
3
|
+
import BaseParser from "./BaseParser";
|
|
4
|
+
import ByteStream from "./ByteStream";
|
|
5
|
+
import PDFContext from "../PDFContext";
|
|
6
|
+
/**
|
|
7
|
+
* PDF 内容流(Content Stream)解析器。
|
|
8
|
+
*
|
|
9
|
+
* 上游 pdf-lib 1.x 只能解析「本库自己在内存里 push 的 PDFOperator」,
|
|
10
|
+
* 对从外部 PDF 加载进来的内容流(PDFRawStream 里的 Flate 压缩字节)没有
|
|
11
|
+
* 公开解析入口——这正是它无法实现文本提取的根因。本解析器补上这一能力:
|
|
12
|
+
* 把解码后的内容流字节还原成 PDFOperator[],供文本提取/内容分析复用。
|
|
13
|
+
*
|
|
14
|
+
* 内容流语法 = 若干「操作数 + 关键字」,关键字即算子名(Tj/TJ/BT/...)。
|
|
15
|
+
* 操作数对象(数字/字符串/十六进制串/名字/数组/字典)复用 PDFObjectParser
|
|
16
|
+
* 的对象解析,保证与本库其它解析器行为一致。
|
|
17
|
+
*/
|
|
18
|
+
declare class PDFContentStreamParser extends BaseParser {
|
|
19
|
+
static forBytes: (bytes: Uint8Array, context: PDFContext) => PDFContentStreamParser;
|
|
20
|
+
static fromPDFRawStream: (stream: PDFRawStream, context: PDFContext) => PDFContentStreamParser;
|
|
21
|
+
private readonly objectParser;
|
|
22
|
+
constructor(byteStream: ByteStream, context: PDFContext);
|
|
23
|
+
/** 解析整个内容流为算子序列。遇到无法识别的 token 跳过而非抛错(容错:真实世界 PDF 很脏)。 */
|
|
24
|
+
parseOperators(): PDFOperator[];
|
|
25
|
+
/** 读取一个关键字 token(字母/符号组成,非空白非分隔符)。 */
|
|
26
|
+
private parseKeyword;
|
|
27
|
+
}
|
|
28
|
+
export default PDFContentStreamParser;
|
|
29
|
+
//# sourceMappingURL=PDFContentStreamParser.d.ts.map
|
|
@@ -8,6 +8,8 @@ declare class PDFContentStream extends PDFFlateStream {
|
|
|
8
8
|
private constructor();
|
|
9
9
|
push(...operators: PDFOperator[]): void;
|
|
10
10
|
clone(context?: PDFContext): PDFContentStream;
|
|
11
|
+
/** 只读访问解析后的算子序列(供文本提取/内容分析使用,返回副本防改写)。 */
|
|
12
|
+
getOperators(): PDFOperator[];
|
|
11
13
|
getContentsString(): string;
|
|
12
14
|
getUnencodedContents(): Uint8Array;
|
|
13
15
|
getUnencodedContentsSize(): number;
|
|
@@ -28,6 +28,8 @@ declare class PDFPageLeaf extends PDFDict {
|
|
|
28
28
|
getInheritableAttribute(name: PDFName): PDFObject | undefined;
|
|
29
29
|
setParent(parentRef: PDFRef): void;
|
|
30
30
|
addContentStream(contentStreamRef: PDFRef): void;
|
|
31
|
+
/** 用单个内容流整体替换本页 Contents(供文本编辑/内容流重写使用)。 */
|
|
32
|
+
setContentStream(contentStreamRef: PDFRef): void;
|
|
31
33
|
wrapContentStreams(startStream: PDFRef, endStream: PDFRef): boolean;
|
|
32
34
|
addAnnot(annotRef: PDFRef): void;
|
|
33
35
|
removeAnnot(annotRef: PDFRef): void;
|