xml-event-parser 0.1.1 → 0.2.1
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 +128 -158
- package/dist/index.cjs +1 -1
- package/dist/index.d.ts +10 -0
- package/dist/index.mjs +1 -1
- package/dist/stream.cjs +1 -0
- package/dist/stream.d.ts +61 -0
- package/dist/stream.mjs +1 -0
- package/package.json +16 -3
package/README.md
CHANGED
|
@@ -1,80 +1,122 @@
|
|
|
1
1
|
xml-event-parser
|
|
2
2
|
================
|
|
3
3
|
|
|
4
|
-
|
|
5
|
-
|
|
6
|
-
|
|
7
|
-
- **Zero dependencies**, no DOM, no `node:*` imports, no `eval`.
|
|
8
|
-
- **One pass, two views**: every element is available as a tree node and as an
|
|
9
|
-
`open`/`close` event keyed by its dotted path, so a consumer can convert while
|
|
10
|
-
parsing without a second traversal.
|
|
11
|
-
- **Streaming**: feed `write(chunk)` from a file or socket and call `end()`.
|
|
12
|
-
Chunks may split anywhere - mid-tag, mid-attribute, inside CDATA or a comment -
|
|
13
|
-
and `retain` decides whether the parser keeps the tree, one root, or nothing,
|
|
14
|
-
so a 60 MiB document can be parsed with ~0 MiB of retained heap.
|
|
15
|
-
- **Small**: 6.1 KiB minified, 2.1 KiB gzipped (ESM), streaming included.
|
|
16
|
-
- **Fastest on every shape we benchmarked**: ~1.9x `txml`'s throughput on
|
|
17
|
-
OSM-style documents at ~30% less retained heap, and ahead of `txml` on
|
|
18
|
-
text/entity-heavy input too; ~1.6x `htmlparser2`'s on element callbacks. See
|
|
19
|
-
[Benchmarks](#benchmarks).
|
|
4
|
+
**Stream XML without loading it.** A tiny parser that gives you a document tree
|
|
5
|
+
and path-based events in one pass.
|
|
20
6
|
|
|
21
|
-
|
|
22
|
-
|
|
7
|
+
- **Bounded memory**: keep only the element you are reading. A 240 MB report
|
|
8
|
+
parses in 81 MiB of RSS.
|
|
9
|
+
- **Fast**: ~3.7x `sax` and ~4x `fast-xml-parser` on the same report.
|
|
10
|
+
- **Portable**: 6.4 KiB minified / 2.2 KiB gzipped, zero dependencies, no
|
|
11
|
+
`node:*` imports in the core entry, no `eval`.
|
|
12
|
+
- **One pass, two views**: a tree node and an `open`/`close` event per element.
|
|
23
13
|
|
|
24
|
-
|
|
14
|
+
## When to use this
|
|
25
15
|
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
|
|
16
|
+
| your job | use |
|
|
17
|
+
| --- | --- |
|
|
18
|
+
| Read a file bigger than memory: CI/test reports (JUnit, SARIF, coverage), sitemaps, GPX/KML tracks, OSM extracts | **this**, `xml-event-parser/stream` |
|
|
19
|
+
| Pick elements while streaming without writing a SAX state machine | **this**, path events |
|
|
20
|
+
| Parse occasionally in a browser or Worker where a 20 KiB parser is too much | **this**, core entry |
|
|
21
|
+
| Turn a small config XML into a plain object | `fast-xml-parser` or `xml2js` |
|
|
22
|
+
| DOM semantics: namespaces, XPath, mutation | `@xmldom/xmldom` |
|
|
23
|
+
| Hand-tuned callbacks with no tree at all | `sax` |
|
|
30
24
|
|
|
31
|
-
|
|
32
|
-
```
|
|
25
|
+
## 240 MB in one second
|
|
33
26
|
|
|
34
|
-
|
|
27
|
+
Generated JUnit reports, each parser in its own process capped at 512 MB of heap:
|
|
35
28
|
|
|
36
|
-
|
|
37
|
-
|
|
38
|
-
|
|
39
|
-
|
|
40
|
-
|
|
41
|
-
|
|
42
|
-
|
|
29
|
+
| parser | 60 MB report | 240 MB report |
|
|
30
|
+
| --- | --- | --- |
|
|
31
|
+
| **xml-event-parser** | **0.27 s, 65 MiB** | **1.03 s, 81 MiB** |
|
|
32
|
+
| sax | 0.96 s, 67 MiB | 3.80 s, 79 MiB |
|
|
33
|
+
| fast-xml-parser | 1.52 s, 637 MiB | **out of memory** |
|
|
34
|
+
|
|
35
|
+
The file grows 4x; the heap does not.
|
|
36
|
+
|
|
37
|
+
## Install
|
|
38
|
+
|
|
39
|
+
```sh
|
|
40
|
+
npm install xml-event-parser
|
|
43
41
|
```
|
|
44
42
|
|
|
45
|
-
|
|
46
|
-
object per element keeps a parsed document small (see the heap column in
|
|
47
|
-
[`bench/RESULTS.md`](bench/RESULTS.md)). Names that collide with `tagName`,
|
|
48
|
-
`children` or `text` are shadowed by the element's own values; if that matters
|
|
49
|
-
for your vocabulary, read attributes off the source XML instead.
|
|
43
|
+
## Quick start
|
|
50
44
|
|
|
51
|
-
|
|
52
|
-
attached, so `node.children` is complete inside a close listener.
|
|
45
|
+
### Tree
|
|
53
46
|
|
|
54
|
-
|
|
47
|
+
```js
|
|
48
|
+
import { XmlParser } from 'xml-event-parser';
|
|
49
|
+
|
|
50
|
+
const roots = new XmlParser().parse(xml);
|
|
51
|
+
console.log(roots[0].tagName, roots[0].children);
|
|
52
|
+
```
|
|
53
|
+
|
|
54
|
+
### Stream (Node)
|
|
55
55
|
|
|
56
56
|
```js
|
|
57
57
|
import { createReadStream } from 'node:fs';
|
|
58
|
-
import {
|
|
58
|
+
import { Writable } from 'node:stream';
|
|
59
|
+
import { pipeline } from 'node:stream/promises';
|
|
60
|
+
import { createStream } from 'xml-event-parser/stream';
|
|
61
|
+
|
|
62
|
+
await pipeline(
|
|
63
|
+
createReadStream('junit.xml'),
|
|
64
|
+
createStream({ event: '</testsuites.testsuite.testcase>' }),
|
|
65
|
+
new Writable({
|
|
66
|
+
objectMode: true,
|
|
67
|
+
write(testcase, _encoding, done) {
|
|
68
|
+
// testcase is complete: attributes plus <failure>/<system-out> children
|
|
69
|
+
if (testcase.children?.some((child) => child.tagName === 'failure')) {
|
|
70
|
+
console.log(testcase.name);
|
|
71
|
+
}
|
|
72
|
+
done();
|
|
73
|
+
},
|
|
74
|
+
}),
|
|
75
|
+
);
|
|
76
|
+
```
|
|
59
77
|
|
|
60
|
-
|
|
61
|
-
|
|
62
|
-
|
|
78
|
+
### Anywhere (no Node built-ins)
|
|
79
|
+
|
|
80
|
+
```js
|
|
81
|
+
const parser = new XmlParser({
|
|
82
|
+
progressive: true,
|
|
83
|
+
retain: 'none',
|
|
84
|
+
keep: ['testsuites.testsuite.testcase'],
|
|
63
85
|
});
|
|
64
86
|
|
|
65
|
-
|
|
87
|
+
parser.on('</testsuites.testsuite.testcase>', (testcase) => { /* ... */ });
|
|
88
|
+
|
|
89
|
+
for await (const chunk of stream) {
|
|
66
90
|
parser.write(chunk);
|
|
67
91
|
}
|
|
68
92
|
parser.end();
|
|
69
93
|
```
|
|
70
94
|
|
|
71
|
-
`
|
|
72
|
-
|
|
73
|
-
|
|
74
|
-
|
|
75
|
-
|
|
95
|
+
Runnable examples: [`examples/junit.mjs`](examples/junit.mjs),
|
|
96
|
+
[`examples/sitemap.mjs`](examples/sitemap.mjs),
|
|
97
|
+
[`examples/gpx.mjs`](examples/gpx.mjs).
|
|
98
|
+
|
|
99
|
+
## API
|
|
100
|
+
|
|
101
|
+
### Node shape
|
|
102
|
+
|
|
103
|
+
```ts
|
|
104
|
+
interface XmlNode {
|
|
105
|
+
tagName: string; // element name
|
|
106
|
+
children?: XmlNode[]; // present only when the element has child elements
|
|
107
|
+
text?: string; // present only for text-only elements
|
|
108
|
+
[attribute: string]: any; // attributes, flat on the node
|
|
109
|
+
}
|
|
110
|
+
```
|
|
111
|
+
|
|
112
|
+
An element is emitted on close after its children are attached, so
|
|
113
|
+
`node.children` is complete inside a close listener. Attributes whose names
|
|
114
|
+
collide with `tagName`, `children` or `text` are shadowed by those fields.
|
|
76
115
|
|
|
77
|
-
|
|
116
|
+
### Streaming and retention
|
|
117
|
+
|
|
118
|
+
`write(chunk)` accepts a document in pieces; chunks may split anywhere, and
|
|
119
|
+
`end()` closes whatever is still open and returns the roots.
|
|
78
120
|
|
|
79
121
|
| `retain` | keeps | `end()` returns | memory |
|
|
80
122
|
| --- | --- | --- | --- |
|
|
@@ -82,142 +124,70 @@ split point).
|
|
|
82
124
|
| `'root'` | the root being built | `[]`; a finished root is emitted and dropped | largest root |
|
|
83
125
|
| `'none'` | nothing | `[]`; every element is emitted and dropped | nesting depth |
|
|
84
126
|
|
|
85
|
-
|
|
86
|
-
|
|
87
|
-
|
|
88
|
-
|
|
127
|
+
With `retain: 'none'`, `keep` restores the subtree of the elements you want, so
|
|
128
|
+
a streamed element arrives complete rather than as a bare tag:
|
|
129
|
+
|
|
130
|
+
```js
|
|
131
|
+
new XmlParser({
|
|
132
|
+
progressive: true,
|
|
133
|
+
retain: 'none',
|
|
134
|
+
keep: ['testsuites.testsuite.testcase'],
|
|
135
|
+
});
|
|
136
|
+
```
|
|
137
|
+
|
|
138
|
+
`createStream` applies `keep` for you from its `event` path.
|
|
89
139
|
|
|
90
|
-
|
|
140
|
+
### Events
|
|
91
141
|
|
|
92
142
|
| form | fires |
|
|
93
143
|
| --- | --- |
|
|
94
144
|
| `'<osm.way>'` | when a `<way>` inside `<osm>` opens |
|
|
95
145
|
| `'</osm.way>'` | when it closes, with `children` populated |
|
|
96
146
|
|
|
97
|
-
The path is the dotted chain of ancestors.
|
|
98
|
-
that have a listener, so subscribing to one path costs nothing for the elements
|
|
99
|
-
you ignore. Filter inside the listener:
|
|
147
|
+
The path is the dotted chain of ancestors. Filter inside the listener:
|
|
100
148
|
|
|
101
149
|
```js
|
|
102
150
|
parser.on('</osm.way>', (way) => {
|
|
103
|
-
if (!way.tags) return;
|
|
151
|
+
if (!way.tags) return;
|
|
104
152
|
});
|
|
105
153
|
```
|
|
106
154
|
|
|
107
|
-
|
|
155
|
+
### Options
|
|
108
156
|
|
|
109
157
|
| option | default | meaning |
|
|
110
158
|
| --- | --- | --- |
|
|
111
|
-
| `progressive` | `false` | emit events
|
|
112
|
-
| `queryParent` | `false` | track parents
|
|
113
|
-
| `retain` | `'all'` |
|
|
159
|
+
| `progressive` | `false` | emit events; without it `parse()` only builds the tree |
|
|
160
|
+
| `queryParent` | `false` | track parents for `parser.getParent(node)` |
|
|
161
|
+
| `retain` | `'all'` | `'all'`, `'root'` or `'none'` (see [Streaming and retention](#streaming-and-retention)) |
|
|
162
|
+
| `keep` | `[]` | under `retain: 'none'`, paths whose elements keep their subtree |
|
|
114
163
|
|
|
115
|
-
`parse(xml, parent?, dir?)`
|
|
116
|
-
|
|
164
|
+
`parse(xml, parent?, dir?)` can attach the parsed roots to an existing node and
|
|
165
|
+
prefix event paths.
|
|
117
166
|
|
|
118
|
-
|
|
119
|
-
|
|
120
|
-
|
|
121
|
-
|
|
122
|
-
`text` is `'text'`).
|
|
123
|
-
|
|
124
|
-
CDATA sections are content, not markup: their text is kept verbatim and is not
|
|
125
|
-
entity-decoded, while text outside them is decoded (the five predefined
|
|
126
|
-
entities and numeric character references).
|
|
167
|
+
Malformed input never throws: comments, processing instructions and doctypes
|
|
168
|
+
are skipped, and elements still open at end of input are closed, so a truncated
|
|
169
|
+
download yields the document parsed so far. CDATA content is kept verbatim and
|
|
170
|
+
not entity-decoded; text outside CDATA is.
|
|
127
171
|
|
|
128
172
|
## Benchmarks
|
|
129
173
|
|
|
130
|
-
|
|
131
|
-
nesting)
|
|
132
|
-
[`bench/RESULTS.md`](bench/RESULTS.md); rerun with `npm run bench:data && npm run bench`.
|
|
133
|
-
|
|
134
|
-
Realistic Overpass-style document, 6.0 MiB, Apple M4 Pro / Node 26:
|
|
174
|
+
Speed on six generated shapes (Overpass-style documents, children-heavy, text,
|
|
175
|
+
nesting), against `txml`, `fast-xml-parser`, `sax` and `htmlparser2`:
|
|
135
176
|
|
|
136
177
|
| workload | xml-event-parser | txml | fast-xml-parser | sax | htmlparser2 |
|
|
137
178
|
| --- | --- | --- | --- | --- | --- |
|
|
138
179
|
| tree (parse + walk) | **26.0 ms, 230 MB/s** | 50.6 ms, 118 MB/s | 167.8 ms, 36 MB/s | — | — |
|
|
139
|
-
| tree, retained heap | **30.0 MiB** | 42.5 MiB | 35.2 MiB | — | — |
|
|
140
180
|
| element callbacks | **27.8 ms, 215 MB/s** | — | — | 100.0 ms, 60 MB/s | 44.9 ms, 133 MB/s |
|
|
141
181
|
| streamed, 64 KiB chunks | **27.3 ms, 219 MB/s** | 77.4 ms, 77 MB/s | — | 103.3 ms, 58 MB/s | 50.8 ms, 118 MB/s |
|
|
142
182
|
|
|
143
|
-
|
|
144
|
-
|
|
145
|
-
- The callback and streamed rows for `xml-event-parser` use `retain: 'none'`, so
|
|
146
|
-
they do not build a document either; the tree row is the only one that holds
|
|
147
|
-
memory. `txml`'s streaming API is a Node stream (`transformStream`), measured
|
|
148
|
-
separately and asynchronously.
|
|
149
|
-
- Text-heavy documents with many entities and CDATA used to be the weak shape;
|
|
150
|
-
replacing the regex decoder with a manual scan took it from 325 to 442 MB/s,
|
|
151
|
-
ahead of `txml` (382 MB/s).
|
|
152
|
-
|
|
153
|
-
## Status
|
|
154
|
-
|
|
155
|
-
Extracted from [osm2geojson-lite](https://github.com/tibetty/osm2geojson-lite),
|
|
156
|
-
where this parser served until 2.2.0 and was then replaced by a purpose-built
|
|
157
|
-
OSM scanner. The code and its original unit tests are unchanged apart from the
|
|
158
|
-
public naming (`children`/`tagName`/`text` instead of `$innerNodes`/`$tag`/
|
|
159
|
-
`$innerText`) and the removal of the `new Function`-based conditioned-path
|
|
160
|
-
events.
|
|
161
|
-
|
|
162
|
-
Not published yet. The release workflow publishes only when the repository
|
|
163
|
-
variable `NPM_TRUSTED_PUBLISHER` is `true`; before that, a release is published
|
|
164
|
-
locally because npm cannot configure a trusted publisher for a package that does
|
|
165
|
-
not exist yet.
|
|
166
|
-
|
|
167
|
-
## Development
|
|
168
|
-
|
|
169
|
-
```sh
|
|
170
|
-
npm ci
|
|
171
|
-
npm test # unit tests
|
|
172
|
-
npm run typecheck
|
|
173
|
-
npm run build # dist/index.mjs, dist/index.cjs, dist/index.d.ts
|
|
174
|
-
npm run bench:data # generate ~27 MiB benchmark corpus (gitignored)
|
|
175
|
-
npm run bench # compare against txml / sax / htmlparser2 / fast-xml-parser
|
|
176
|
-
```
|
|
177
|
-
|
|
178
|
-
## Releasing
|
|
179
|
-
|
|
180
|
-
`release.yml` publishes with npm trusted publishing (OIDC) and is gated on the
|
|
181
|
-
repository variable `NPM_TRUSTED_PUBLISHER`, which stays unset until a trusted
|
|
182
|
-
publisher exists.
|
|
183
|
-
|
|
184
|
-
One-time bootstrap, because npm cannot attach a trusted publisher to a package
|
|
185
|
-
that does not exist yet:
|
|
186
|
-
|
|
187
|
-
1. Publish the first version locally:
|
|
188
|
-
|
|
189
|
-
```sh
|
|
190
|
-
npm login
|
|
191
|
-
npm publish --access public
|
|
192
|
-
```
|
|
193
|
-
|
|
194
|
-
2. npmjs.com -> Packages -> `xml-event-parser` -> Settings -> **Trusted
|
|
195
|
-
publishing** -> GitHub Actions:
|
|
196
|
-
|
|
197
|
-
| field | value |
|
|
198
|
-
| --- | --- |
|
|
199
|
-
| Organization or user | `tibetty` |
|
|
200
|
-
| Repository | `xml-event-parser` |
|
|
201
|
-
| Workflow filename | `release.yml` (filename only, case-sensitive) |
|
|
202
|
-
| Environment name | leave empty |
|
|
203
|
-
| Allowed actions | tick **npm publish** |
|
|
204
|
-
|
|
205
|
-
Configurations created after September 2026 default to `npm stage publish`
|
|
206
|
-
only, so `npm publish` has to be ticked explicitly. The first publish through
|
|
207
|
-
that workflow must complete within **two days**, or the configuration expires.
|
|
208
|
-
|
|
209
|
-
3. GitHub -> Settings -> Secrets and variables -> Actions -> Variables: add
|
|
210
|
-
`NPM_TRUSTED_PUBLISHER` = `true`.
|
|
183
|
+
Full tables, methodology and caveats: [`bench/RESULTS.md`](bench/RESULTS.md).
|
|
211
184
|
|
|
212
|
-
|
|
213
|
-
package Settings -> Publishing access -> *Require two-factor authentication
|
|
214
|
-
and disallow tokens*.
|
|
185
|
+
## Contributing
|
|
215
186
|
|
|
216
|
-
|
|
217
|
-
|
|
218
|
-
provenance automatically only for a public repository; from a private one the
|
|
219
|
-
package is published without an attestation.
|
|
187
|
+
See [CONTRIBUTING.md](CONTRIBUTING.md) for development commands, tests,
|
|
188
|
+
benchmarks and the release process.
|
|
220
189
|
|
|
221
190
|
## License
|
|
222
191
|
|
|
223
|
-
MIT. Includes work contributed to
|
|
192
|
+
MIT. Includes work contributed to
|
|
193
|
+
[osm2geojson-lite](https://github.com/tibetty/osm2geojson-lite) by Harel Mazor.
|
package/dist/index.cjs
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
Object.defineProperty(exports,Symbol.toStringTag,{value:`Module`});function e(e){switch(e){case`amp`:return`&`;case`lt`:return`<`;case`gt`:return`>`;case`quot`:return`"`;case`apos`:return`'`}if(e.charCodeAt(0)===35){let t=e.charCodeAt(1)===120||e.charCodeAt(1)===88?parseInt(e.slice(2),16):parseInt(e.slice(1),10);if(Number.isFinite(t)&&t>=0&&t<=1114111)return String.fromCodePoint(t)}return null}function t(t){let n=t.indexOf(`&`);if(n<0)return t;let r=t.slice(0,n);for(;n<t.length;){let i=t.indexOf(`;`,n+1);if(i<0){r+=t.slice(n);break}let a=e(t.slice(n+1,i));if(r+=a===null?t.slice(n,i+1):a,n=t.indexOf(`&`,i+1),n<0){r+=t.slice(i+1);break}r+=t.slice(i+1,n)}return r}function n(e){if(e.indexOf(`<![CDATA[`)<0)return e.indexOf(`&`)<0?e:t(e);let n=``,r=0;for(;r<e.length;){let i=e.indexOf(`<![CDATA[`,r);if(i<0){n+=t(e.slice(r));break}i>r&&(n+=t(e.slice(r,i)));let a=e.indexOf(`]]>`,i+9);if(a<0){n+=e.slice(i+9);break}n+=e.slice(i+9,a),r=a+3}return n}function r(e){return e===32||e===9||e===10||e===13}var
|
|
1
|
+
Object.defineProperty(exports,Symbol.toStringTag,{value:`Module`});function e(e){switch(e){case`amp`:return`&`;case`lt`:return`<`;case`gt`:return`>`;case`quot`:return`"`;case`apos`:return`'`}if(e.charCodeAt(0)===35){let t=e.charCodeAt(1)===120||e.charCodeAt(1)===88?parseInt(e.slice(2),16):parseInt(e.slice(1),10);if(Number.isFinite(t)&&t>=0&&t<=1114111)return String.fromCodePoint(t)}return null}function t(t){let n=t.indexOf(`&`);if(n<0)return t;let r=t.slice(0,n);for(;n<t.length;){let i=t.indexOf(`;`,n+1);if(i<0){r+=t.slice(n);break}let a=e(t.slice(n+1,i));if(r+=a===null?t.slice(n,i+1):a,n=t.indexOf(`&`,i+1),n<0){r+=t.slice(i+1);break}r+=t.slice(i+1,n)}return r}function n(e){if(e.indexOf(`<![CDATA[`)<0)return e.indexOf(`&`)<0?e:t(e);let n=``,r=0;for(;r<e.length;){let i=e.indexOf(`<![CDATA[`,r);if(i<0){n+=t(e.slice(r));break}i>r&&(n+=t(e.slice(r,i)));let a=e.indexOf(`]]>`,i+9);if(a<0){n+=e.slice(i+9);break}n+=e.slice(i+9,a),r=a+3}return n}function r(e){let t=0,n=e.length;return e.charCodeAt(0)===60&&(t=e.charCodeAt(1)===47?2:1),e.charCodeAt(n-1)===62&&n--,e.slice(t,n)}function i(e){return e===32||e===9||e===10||e===13}var a=class{constructor(e){if(this.queryParent=!1,this.progressive=!1,this.retain=`all`,this.keepPaths=/* @__PURE__ */ new Set,this.parentMap=/* @__PURE__ */ new WeakMap,this.evtListeners={},this.openListenerCount=0,this.watchedPaths=/* @__PURE__ */ new Set,this.watchedChildren=/* @__PURE__ */ new Map,this.watchedTopLevel=/* @__PURE__ */ new Set,this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=``,this.active=!1,e&&(this.queryParent=e.queryParent===!0,this.progressive=e.progressive===!0,e.retain&&(this.retain=e.retain),e.keep))for(let t of e.keep)this.keepPaths.add(r(t))}write(e){return this.active||this.begin(void 0,void 0),e.length===0||(this.buffer.length===0?this.buffer=e:this.buffer+=e,this.consume(!1)),this}end(){if(!this.active)return[];this.consume(!0),this.finish();let e=this.roots;return this.reset(),e}parse(e,t,n){return this.begin(t,n),this.write(e),this.end()}begin(e,t){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=t||``,this.top=e?{node:e,fullTag:this.basePath,textStart:0}:void 0,this.rootOwner=this.top,this.parentMap=/* @__PURE__ */ new WeakMap,this.active=!0}reset(){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.top=void 0,this.rootOwner=void 0,this.basePath=``,this.active=!1}attach(e,t,n){if(this.retain===`none`&&!n)return;if(t===void 0){this.retain===`all`&&this.roots.push(e);return}let r=t.children;r?r.push(e):t.children=[e]}consume(e){let t=this.buffer,r=t.length,i=this.progressive,a=i&&this.openListenerCount>0,o=this.cursor,s=-1;for(;o<r;){let c=t.indexOf(`<`,o);if(c<0||(s=c,o=c+1,o>=r))break;let l=t.charCodeAt(o);if(l===47){let a=t.indexOf(`>`,o);if(a<0){if(!e)break;o=r;break}let l=this.stack.pop();if(o=a+1,s=-1,!l)continue;let u=l.node,d=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,f=d?d.node:void 0;!l.hasChild&&c>l.textStart&&(u.text=n(t.slice(l.textStart,c))),i&&this.watchedPaths.has(l.fullTag)&&this.emit(`</`+l.fullTag+`>`,u,f),this.attach(u,f,d?d.keep===!0:!1),d&&(d.hasChild=!0),this.top=d;continue}if(l===63){let n=t.indexOf(`>`,o);if(n<0){if(!e)break;o=r;break}o=n+1,s=-1;continue}if(l===33){let n;if(t.charCodeAt(o+1)===45&&t.charCodeAt(o+2)===45?(n=t.indexOf(`-->`,o+3),n=n<0?-1:n+3):t.startsWith(`![CDATA[`,o)?(n=t.indexOf(`]]>`,o+9),n=n<0?-1:n+3):(n=t.indexOf(`>`,o),n=n<0?-1:n+1),n<0){if(!e)break;o=r;break}o=n,s=-1;continue}let u=this.parseOpenTag(t,o,r,e);if(u===null)break;o=u.end,s=-1;let d=u.node,f=u.tag,p=this.top?this.top.node:void 0;if(this.queryParent&&p&&this.parentMap.set(d,p),u.selfClose){if(i){let e=this.top?this.top.fullTag:this.basePath,t=e?this.watchedChildren.get(e):this.watchedTopLevel;if(t&&t.has(f)){let t=e?e+`.`+f:f;a&&this.emit(`<`+t+`>`,d,p),this.emit(`</`+t+`>`,d,p)}}this.attach(d,p,this.top?this.top.keep===!0:!1),this.top&&(this.top.hasChild=!0);continue}let m=this.top?this.top.fullTag?this.top.fullTag+`.`+f:f:this.basePath?this.basePath+`.`+f:f;a&&this.watchedPaths.has(m)&&this.emit(`<`+m+`>`,d,p);let h=this.top!==void 0&&this.top.keep===!0||this.keepPaths.has(m),g={node:d,fullTag:m,textStart:o,keep:h};this.stack.push(g),this.top=g}this.cursor=e?r:s>=0?s:o,e||this.compact()}parseOpenTag(e,n,r,a){let o=n;for(;o<r;){let t=e.charCodeAt(o);if(i(t)||t===47||t===62)break;o++}if(o>=r&&!a)return null;let s=e.slice(n,o),c={tagName:s},l=!1;for(;o<r;){let n=e.charCodeAt(o);for(;o<r&&i(n);)n=e.charCodeAt(++o);if(o>=r)break;if(n===62)return o++,{node:c,tag:s,selfClose:l,end:o};if(n===47){l=!0,o++;continue}let u=o;for(;o<r&&(n=e.charCodeAt(o),!(n===61||i(n)||n===62||n===47));)o++;if(o>=r&&!a)return null;let d=e.slice(u,o);for(;o<r&&i(e.charCodeAt(o));)o++;if(o>=r&&!a)return null;if(e.charCodeAt(o)===61){for(o++;o<r&&i(e.charCodeAt(o));)o++;if(o>=r&&!a)return null;let n=e.charCodeAt(o);if(n===34||n===39){o++;let i=e.indexOf(n===34?`"`:`'`,o);if(i<0){if(!a)return null;i=r}let s=e.slice(o,i);c[d]=s.indexOf(`&`)<0?s:t(s),o=i+1}}else d&&(c[d]=``)}return a?{node:c,tag:s,selfClose:l,end:o}:null}finish(){let e=this.buffer,t=e.length;for(;this.stack.length>0;){let r=this.stack.pop(),i=r.node;!r.hasChild&&t>r.textStart&&(i.text=n(e.slice(r.textStart,t)),t=r.textStart);let a=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,o=a?a.node:void 0;this.attach(i,o,a?a.keep===!0:!1)}}compact(){let e=this.cursor;for(let t=this.stack.length-1;t>=0;t--){let n=this.stack[t];if(!n.hasChild){e=n.textStart;break}}if(!(e<=0)){this.buffer=this.buffer.slice(e),this.cursor-=e;for(let t=0;t<this.stack.length;t++){let n=this.stack[t];n.textStart=n.textStart>e?n.textStart-e:0}this.rootOwner&&(this.rootOwner.textStart=0)}}getParent(e){return this.queryParent?this.parentMap.get(e):null}addListener(e,t){this.$addListener(e,t)}removeListener(e,t){this.$removeListener(e,t)}on(e,t){this.addListener(e,t)}off(e,t){this.removeListener(e,t)}$addListener(e,t){let n=this.evtListeners[e];if(n?n.push(t):this.evtListeners[e]=[t],e.charCodeAt(1)!==47&&this.openListenerCount++,e.charCodeAt(0)===60){let t=e.slice(e.charCodeAt(1)===47?2:1,e.length-1);this.watchedPaths.add(t);let n=t.lastIndexOf(`.`);if(n<0)this.watchedTopLevel.add(t);else{let e=t.slice(0,n),r=this.watchedChildren.get(e);r||(r=/* @__PURE__ */ new Set,this.watchedChildren.set(e,r)),r.add(t.slice(n+1))}}}$removeListener(e,t){let n=this.evtListeners[e],r=-1;n&&(r=n.indexOf(t)),r>=0&&(n.splice(r,1),e.charCodeAt(1)!==47&&this.openListenerCount--)}emit(e,t,n){let r=this.evtListeners[e];if(r)for(let e=0;e<r.length;e++)r[e](t,n)}};exports.XmlParser=a,exports.decodeEntities=t;
|
package/dist/index.d.ts
CHANGED
|
@@ -38,11 +38,21 @@ interface XmlParserOptions {
|
|
|
38
38
|
* nobody listens for is parsed and discarded.
|
|
39
39
|
*/
|
|
40
40
|
retain?: Retain;
|
|
41
|
+
/**
|
|
42
|
+
* Under `retain: 'none'`, elements whose dotted path is listed here keep
|
|
43
|
+
* their subtree (children and text), so a streamed element arrives
|
|
44
|
+
* complete instead of as a bare tag. Paths use the event form without
|
|
45
|
+
* brackets: `'testsuites.testsuite.testcase'`.
|
|
46
|
+
*
|
|
47
|
+
* Memory stays bounded by the largest matched subtree, not the document.
|
|
48
|
+
*/
|
|
49
|
+
keep?: string[];
|
|
41
50
|
}
|
|
42
51
|
export declare class XmlParser {
|
|
43
52
|
private queryParent;
|
|
44
53
|
private progressive;
|
|
45
54
|
private retain;
|
|
55
|
+
private keepPaths;
|
|
46
56
|
private parentMap;
|
|
47
57
|
private evtListeners;
|
|
48
58
|
/**
|
package/dist/index.mjs
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
function e(e){switch(e){case`amp`:return`&`;case`lt`:return`<`;case`gt`:return`>`;case`quot`:return`"`;case`apos`:return`'`}if(e.charCodeAt(0)===35){let t=e.charCodeAt(1)===120||e.charCodeAt(1)===88?parseInt(e.slice(2),16):parseInt(e.slice(1),10);if(Number.isFinite(t)&&t>=0&&t<=1114111)return String.fromCodePoint(t)}return null}function t(t){let n=t.indexOf(`&`);if(n<0)return t;let r=t.slice(0,n);for(;n<t.length;){let i=t.indexOf(`;`,n+1);if(i<0){r+=t.slice(n);break}let a=e(t.slice(n+1,i));if(r+=a===null?t.slice(n,i+1):a,n=t.indexOf(`&`,i+1),n<0){r+=t.slice(i+1);break}r+=t.slice(i+1,n)}return r}function n(e){if(e.indexOf(`<![CDATA[`)<0)return e.indexOf(`&`)<0?e:t(e);let n=``,r=0;for(;r<e.length;){let i=e.indexOf(`<![CDATA[`,r);if(i<0){n+=t(e.slice(r));break}i>r&&(n+=t(e.slice(r,i)));let a=e.indexOf(`]]>`,i+9);if(a<0){n+=e.slice(i+9);break}n+=e.slice(i+9,a),r=a+3}return n}function r(e){return e===32||e===9||e===10||e===13}var
|
|
1
|
+
function e(e){switch(e){case`amp`:return`&`;case`lt`:return`<`;case`gt`:return`>`;case`quot`:return`"`;case`apos`:return`'`}if(e.charCodeAt(0)===35){let t=e.charCodeAt(1)===120||e.charCodeAt(1)===88?parseInt(e.slice(2),16):parseInt(e.slice(1),10);if(Number.isFinite(t)&&t>=0&&t<=1114111)return String.fromCodePoint(t)}return null}function t(t){let n=t.indexOf(`&`);if(n<0)return t;let r=t.slice(0,n);for(;n<t.length;){let i=t.indexOf(`;`,n+1);if(i<0){r+=t.slice(n);break}let a=e(t.slice(n+1,i));if(r+=a===null?t.slice(n,i+1):a,n=t.indexOf(`&`,i+1),n<0){r+=t.slice(i+1);break}r+=t.slice(i+1,n)}return r}function n(e){if(e.indexOf(`<![CDATA[`)<0)return e.indexOf(`&`)<0?e:t(e);let n=``,r=0;for(;r<e.length;){let i=e.indexOf(`<![CDATA[`,r);if(i<0){n+=t(e.slice(r));break}i>r&&(n+=t(e.slice(r,i)));let a=e.indexOf(`]]>`,i+9);if(a<0){n+=e.slice(i+9);break}n+=e.slice(i+9,a),r=a+3}return n}function r(e){let t=0,n=e.length;return e.charCodeAt(0)===60&&(t=e.charCodeAt(1)===47?2:1),e.charCodeAt(n-1)===62&&n--,e.slice(t,n)}function i(e){return e===32||e===9||e===10||e===13}var a=class{constructor(e){if(this.queryParent=!1,this.progressive=!1,this.retain=`all`,this.keepPaths=/* @__PURE__ */ new Set,this.parentMap=/* @__PURE__ */ new WeakMap,this.evtListeners={},this.openListenerCount=0,this.watchedPaths=/* @__PURE__ */ new Set,this.watchedChildren=/* @__PURE__ */ new Map,this.watchedTopLevel=/* @__PURE__ */ new Set,this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=``,this.active=!1,e&&(this.queryParent=e.queryParent===!0,this.progressive=e.progressive===!0,e.retain&&(this.retain=e.retain),e.keep))for(let t of e.keep)this.keepPaths.add(r(t))}write(e){return this.active||this.begin(void 0,void 0),e.length===0||(this.buffer.length===0?this.buffer=e:this.buffer+=e,this.consume(!1)),this}end(){if(!this.active)return[];this.consume(!0),this.finish();let e=this.roots;return this.reset(),e}parse(e,t,n){return this.begin(t,n),this.write(e),this.end()}begin(e,t){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=t||``,this.top=e?{node:e,fullTag:this.basePath,textStart:0}:void 0,this.rootOwner=this.top,this.parentMap=/* @__PURE__ */ new WeakMap,this.active=!0}reset(){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.top=void 0,this.rootOwner=void 0,this.basePath=``,this.active=!1}attach(e,t,n){if(this.retain===`none`&&!n)return;if(t===void 0){this.retain===`all`&&this.roots.push(e);return}let r=t.children;r?r.push(e):t.children=[e]}consume(e){let t=this.buffer,r=t.length,i=this.progressive,a=i&&this.openListenerCount>0,o=this.cursor,s=-1;for(;o<r;){let c=t.indexOf(`<`,o);if(c<0||(s=c,o=c+1,o>=r))break;let l=t.charCodeAt(o);if(l===47){let a=t.indexOf(`>`,o);if(a<0){if(!e)break;o=r;break}let l=this.stack.pop();if(o=a+1,s=-1,!l)continue;let u=l.node,d=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,f=d?d.node:void 0;!l.hasChild&&c>l.textStart&&(u.text=n(t.slice(l.textStart,c))),i&&this.watchedPaths.has(l.fullTag)&&this.emit(`</`+l.fullTag+`>`,u,f),this.attach(u,f,d?d.keep===!0:!1),d&&(d.hasChild=!0),this.top=d;continue}if(l===63){let n=t.indexOf(`>`,o);if(n<0){if(!e)break;o=r;break}o=n+1,s=-1;continue}if(l===33){let n;if(t.charCodeAt(o+1)===45&&t.charCodeAt(o+2)===45?(n=t.indexOf(`-->`,o+3),n=n<0?-1:n+3):t.startsWith(`![CDATA[`,o)?(n=t.indexOf(`]]>`,o+9),n=n<0?-1:n+3):(n=t.indexOf(`>`,o),n=n<0?-1:n+1),n<0){if(!e)break;o=r;break}o=n,s=-1;continue}let u=this.parseOpenTag(t,o,r,e);if(u===null)break;o=u.end,s=-1;let d=u.node,f=u.tag,p=this.top?this.top.node:void 0;if(this.queryParent&&p&&this.parentMap.set(d,p),u.selfClose){if(i){let e=this.top?this.top.fullTag:this.basePath,t=e?this.watchedChildren.get(e):this.watchedTopLevel;if(t&&t.has(f)){let t=e?e+`.`+f:f;a&&this.emit(`<`+t+`>`,d,p),this.emit(`</`+t+`>`,d,p)}}this.attach(d,p,this.top?this.top.keep===!0:!1),this.top&&(this.top.hasChild=!0);continue}let m=this.top?this.top.fullTag?this.top.fullTag+`.`+f:f:this.basePath?this.basePath+`.`+f:f;a&&this.watchedPaths.has(m)&&this.emit(`<`+m+`>`,d,p);let h=this.top!==void 0&&this.top.keep===!0||this.keepPaths.has(m),g={node:d,fullTag:m,textStart:o,keep:h};this.stack.push(g),this.top=g}this.cursor=e?r:s>=0?s:o,e||this.compact()}parseOpenTag(e,n,r,a){let o=n;for(;o<r;){let t=e.charCodeAt(o);if(i(t)||t===47||t===62)break;o++}if(o>=r&&!a)return null;let s=e.slice(n,o),c={tagName:s},l=!1;for(;o<r;){let n=e.charCodeAt(o);for(;o<r&&i(n);)n=e.charCodeAt(++o);if(o>=r)break;if(n===62)return o++,{node:c,tag:s,selfClose:l,end:o};if(n===47){l=!0,o++;continue}let u=o;for(;o<r&&(n=e.charCodeAt(o),!(n===61||i(n)||n===62||n===47));)o++;if(o>=r&&!a)return null;let d=e.slice(u,o);for(;o<r&&i(e.charCodeAt(o));)o++;if(o>=r&&!a)return null;if(e.charCodeAt(o)===61){for(o++;o<r&&i(e.charCodeAt(o));)o++;if(o>=r&&!a)return null;let n=e.charCodeAt(o);if(n===34||n===39){o++;let i=e.indexOf(n===34?`"`:`'`,o);if(i<0){if(!a)return null;i=r}let s=e.slice(o,i);c[d]=s.indexOf(`&`)<0?s:t(s),o=i+1}}else d&&(c[d]=``)}return a?{node:c,tag:s,selfClose:l,end:o}:null}finish(){let e=this.buffer,t=e.length;for(;this.stack.length>0;){let r=this.stack.pop(),i=r.node;!r.hasChild&&t>r.textStart&&(i.text=n(e.slice(r.textStart,t)),t=r.textStart);let a=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,o=a?a.node:void 0;this.attach(i,o,a?a.keep===!0:!1)}}compact(){let e=this.cursor;for(let t=this.stack.length-1;t>=0;t--){let n=this.stack[t];if(!n.hasChild){e=n.textStart;break}}if(!(e<=0)){this.buffer=this.buffer.slice(e),this.cursor-=e;for(let t=0;t<this.stack.length;t++){let n=this.stack[t];n.textStart=n.textStart>e?n.textStart-e:0}this.rootOwner&&(this.rootOwner.textStart=0)}}getParent(e){return this.queryParent?this.parentMap.get(e):null}addListener(e,t){this.$addListener(e,t)}removeListener(e,t){this.$removeListener(e,t)}on(e,t){this.addListener(e,t)}off(e,t){this.removeListener(e,t)}$addListener(e,t){let n=this.evtListeners[e];if(n?n.push(t):this.evtListeners[e]=[t],e.charCodeAt(1)!==47&&this.openListenerCount++,e.charCodeAt(0)===60){let t=e.slice(e.charCodeAt(1)===47?2:1,e.length-1);this.watchedPaths.add(t);let n=t.lastIndexOf(`.`);if(n<0)this.watchedTopLevel.add(t);else{let e=t.slice(0,n),r=this.watchedChildren.get(e);r||(r=/* @__PURE__ */ new Set,this.watchedChildren.set(e,r)),r.add(t.slice(n+1))}}}$removeListener(e,t){let n=this.evtListeners[e],r=-1;n&&(r=n.indexOf(t)),r>=0&&(n.splice(r,1),e.charCodeAt(1)!==47&&this.openListenerCount--)}emit(e,t,n){let r=this.evtListeners[e];if(r)for(let e=0;e<r.length;e++)r[e](t,n)}};export{a as XmlParser,t as decodeEntities};
|
package/dist/stream.cjs
ADDED
|
@@ -0,0 +1 @@
|
|
|
1
|
+
Object.defineProperty(exports,Symbol.toStringTag,{value:`Module`});let e=require("node:stream"),t=require("node:string_decoder");function n(e){switch(e){case`amp`:return`&`;case`lt`:return`<`;case`gt`:return`>`;case`quot`:return`"`;case`apos`:return`'`}if(e.charCodeAt(0)===35){let t=e.charCodeAt(1)===120||e.charCodeAt(1)===88?parseInt(e.slice(2),16):parseInt(e.slice(1),10);if(Number.isFinite(t)&&t>=0&&t<=1114111)return String.fromCodePoint(t)}return null}function r(e){let t=e.indexOf(`&`);if(t<0)return e;let r=e.slice(0,t);for(;t<e.length;){let i=e.indexOf(`;`,t+1);if(i<0){r+=e.slice(t);break}let a=n(e.slice(t+1,i));if(r+=a===null?e.slice(t,i+1):a,t=e.indexOf(`&`,i+1),t<0){r+=e.slice(i+1);break}r+=e.slice(i+1,t)}return r}function i(e){if(e.indexOf(`<![CDATA[`)<0)return e.indexOf(`&`)<0?e:r(e);let t=``,n=0;for(;n<e.length;){let i=e.indexOf(`<![CDATA[`,n);if(i<0){t+=r(e.slice(n));break}i>n&&(t+=r(e.slice(n,i)));let a=e.indexOf(`]]>`,i+9);if(a<0){t+=e.slice(i+9);break}t+=e.slice(i+9,a),n=a+3}return t}function a(e){let t=0,n=e.length;return e.charCodeAt(0)===60&&(t=e.charCodeAt(1)===47?2:1),e.charCodeAt(n-1)===62&&n--,e.slice(t,n)}function o(e){return e===32||e===9||e===10||e===13}var s=class{constructor(e){if(this.queryParent=!1,this.progressive=!1,this.retain=`all`,this.keepPaths=/* @__PURE__ */ new Set,this.parentMap=/* @__PURE__ */ new WeakMap,this.evtListeners={},this.openListenerCount=0,this.watchedPaths=/* @__PURE__ */ new Set,this.watchedChildren=/* @__PURE__ */ new Map,this.watchedTopLevel=/* @__PURE__ */ new Set,this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=``,this.active=!1,e&&(this.queryParent=e.queryParent===!0,this.progressive=e.progressive===!0,e.retain&&(this.retain=e.retain),e.keep))for(let t of e.keep)this.keepPaths.add(a(t))}write(e){return this.active||this.begin(void 0,void 0),e.length===0||(this.buffer.length===0?this.buffer=e:this.buffer+=e,this.consume(!1)),this}end(){if(!this.active)return[];this.consume(!0),this.finish();let e=this.roots;return this.reset(),e}parse(e,t,n){return this.begin(t,n),this.write(e),this.end()}begin(e,t){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=t||``,this.top=e?{node:e,fullTag:this.basePath,textStart:0}:void 0,this.rootOwner=this.top,this.parentMap=/* @__PURE__ */ new WeakMap,this.active=!0}reset(){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.top=void 0,this.rootOwner=void 0,this.basePath=``,this.active=!1}attach(e,t,n){if(this.retain===`none`&&!n)return;if(t===void 0){this.retain===`all`&&this.roots.push(e);return}let r=t.children;r?r.push(e):t.children=[e]}consume(e){let t=this.buffer,n=t.length,r=this.progressive,a=r&&this.openListenerCount>0,o=this.cursor,s=-1;for(;o<n;){let c=t.indexOf(`<`,o);if(c<0||(s=c,o=c+1,o>=n))break;let l=t.charCodeAt(o);if(l===47){let a=t.indexOf(`>`,o);if(a<0){if(!e)break;o=n;break}let l=this.stack.pop();if(o=a+1,s=-1,!l)continue;let u=l.node,d=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,f=d?d.node:void 0;!l.hasChild&&c>l.textStart&&(u.text=i(t.slice(l.textStart,c))),r&&this.watchedPaths.has(l.fullTag)&&this.emit(`</`+l.fullTag+`>`,u,f),this.attach(u,f,d?d.keep===!0:!1),d&&(d.hasChild=!0),this.top=d;continue}if(l===63){let r=t.indexOf(`>`,o);if(r<0){if(!e)break;o=n;break}o=r+1,s=-1;continue}if(l===33){let r;if(t.charCodeAt(o+1)===45&&t.charCodeAt(o+2)===45?(r=t.indexOf(`-->`,o+3),r=r<0?-1:r+3):t.startsWith(`![CDATA[`,o)?(r=t.indexOf(`]]>`,o+9),r=r<0?-1:r+3):(r=t.indexOf(`>`,o),r=r<0?-1:r+1),r<0){if(!e)break;o=n;break}o=r,s=-1;continue}let u=this.parseOpenTag(t,o,n,e);if(u===null)break;o=u.end,s=-1;let d=u.node,f=u.tag,p=this.top?this.top.node:void 0;if(this.queryParent&&p&&this.parentMap.set(d,p),u.selfClose){if(r){let e=this.top?this.top.fullTag:this.basePath,t=e?this.watchedChildren.get(e):this.watchedTopLevel;if(t&&t.has(f)){let t=e?e+`.`+f:f;a&&this.emit(`<`+t+`>`,d,p),this.emit(`</`+t+`>`,d,p)}}this.attach(d,p,this.top?this.top.keep===!0:!1),this.top&&(this.top.hasChild=!0);continue}let m=this.top?this.top.fullTag?this.top.fullTag+`.`+f:f:this.basePath?this.basePath+`.`+f:f;a&&this.watchedPaths.has(m)&&this.emit(`<`+m+`>`,d,p);let h=this.top!==void 0&&this.top.keep===!0||this.keepPaths.has(m),g={node:d,fullTag:m,textStart:o,keep:h};this.stack.push(g),this.top=g}this.cursor=e?n:s>=0?s:o,e||this.compact()}parseOpenTag(e,t,n,i){let a=t;for(;a<n;){let t=e.charCodeAt(a);if(o(t)||t===47||t===62)break;a++}if(a>=n&&!i)return null;let s=e.slice(t,a),c={tagName:s},l=!1;for(;a<n;){let t=e.charCodeAt(a);for(;a<n&&o(t);)t=e.charCodeAt(++a);if(a>=n)break;if(t===62)return a++,{node:c,tag:s,selfClose:l,end:a};if(t===47){l=!0,a++;continue}let u=a;for(;a<n&&(t=e.charCodeAt(a),!(t===61||o(t)||t===62||t===47));)a++;if(a>=n&&!i)return null;let d=e.slice(u,a);for(;a<n&&o(e.charCodeAt(a));)a++;if(a>=n&&!i)return null;if(e.charCodeAt(a)===61){for(a++;a<n&&o(e.charCodeAt(a));)a++;if(a>=n&&!i)return null;let t=e.charCodeAt(a);if(t===34||t===39){a++;let o=e.indexOf(t===34?`"`:`'`,a);if(o<0){if(!i)return null;o=n}let s=e.slice(a,o);c[d]=s.indexOf(`&`)<0?s:r(s),a=o+1}}else d&&(c[d]=``)}return i?{node:c,tag:s,selfClose:l,end:a}:null}finish(){let e=this.buffer,t=e.length;for(;this.stack.length>0;){let n=this.stack.pop(),r=n.node;!n.hasChild&&t>n.textStart&&(r.text=i(e.slice(n.textStart,t)),t=n.textStart);let a=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,o=a?a.node:void 0;this.attach(r,o,a?a.keep===!0:!1)}}compact(){let e=this.cursor;for(let t=this.stack.length-1;t>=0;t--){let n=this.stack[t];if(!n.hasChild){e=n.textStart;break}}if(!(e<=0)){this.buffer=this.buffer.slice(e),this.cursor-=e;for(let t=0;t<this.stack.length;t++){let n=this.stack[t];n.textStart=n.textStart>e?n.textStart-e:0}this.rootOwner&&(this.rootOwner.textStart=0)}}getParent(e){return this.queryParent?this.parentMap.get(e):null}addListener(e,t){this.$addListener(e,t)}removeListener(e,t){this.$removeListener(e,t)}on(e,t){this.addListener(e,t)}off(e,t){this.removeListener(e,t)}$addListener(e,t){let n=this.evtListeners[e];if(n?n.push(t):this.evtListeners[e]=[t],e.charCodeAt(1)!==47&&this.openListenerCount++,e.charCodeAt(0)===60){let t=e.slice(e.charCodeAt(1)===47?2:1,e.length-1);this.watchedPaths.add(t);let n=t.lastIndexOf(`.`);if(n<0)this.watchedTopLevel.add(t);else{let e=t.slice(0,n),r=this.watchedChildren.get(e);r||(r=/* @__PURE__ */ new Set,this.watchedChildren.set(e,r)),r.add(t.slice(n+1))}}}$removeListener(e,t){let n=this.evtListeners[e],r=-1;n&&(r=n.indexOf(t)),r>=0&&(n.splice(r,1),e.charCodeAt(1)!==47&&this.openListenerCount--)}emit(e,t,n){let r=this.evtListeners[e];if(r)for(let e=0;e<r.length;e++)r[e](t,n)}};function c(e){if(typeof e!=`string`||e.length===0)throw TypeError(`createStream() requires { event }, e.g. { event: "</root.item>" }`);return e}function l(n){let r=c(n&&n.event),i=new s({...n,progressive:!0,retain:`none`,keep:[r]}),a=new t.StringDecoder(`utf8`),o=new e.Transform({readableObjectMode:!0,transform(e,t,n){try{i.write(typeof e==`string`?e:a.write(e)),n()}catch(e){n(e)}},flush(e){try{let t=a.end();t.length>0&&i.write(t),i.end(),e()}catch(t){e(t)}}});return i.on(r,e=>{o.push(e)}),o}exports.createStream=l;
|
package/dist/stream.d.ts
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
1
|
+
import { Transform } from "node:stream";
|
|
2
|
+
type Retain = 'all' | 'root' | 'none';
|
|
3
|
+
interface XmlParserOptions {
|
|
4
|
+
/** Emit `<path>` / `</path>` events as elements open and close. Default `false`. */
|
|
5
|
+
progressive?: boolean;
|
|
6
|
+
/** Track parents so `getParent(node)` works. Default `false`. */
|
|
7
|
+
queryParent?: boolean;
|
|
8
|
+
/**
|
|
9
|
+
* What the parser keeps after an element closes:
|
|
10
|
+
* - `'all'` (default) the whole document tree, returned by `end()`
|
|
11
|
+
* - `'root'` the root being built; a finished top-level element is emitted
|
|
12
|
+
* and dropped, so memory is bounded by the largest root
|
|
13
|
+
* - `'none'` nothing; every element is emitted and dropped, so memory is
|
|
14
|
+
* bounded by the nesting depth. Combine with listeners: an element
|
|
15
|
+
* nobody listens for is parsed and discarded.
|
|
16
|
+
*/
|
|
17
|
+
retain?: Retain;
|
|
18
|
+
/**
|
|
19
|
+
* Under `retain: 'none'`, elements whose dotted path is listed here keep
|
|
20
|
+
* their subtree (children and text), so a streamed element arrives
|
|
21
|
+
* complete instead of as a bare tag. Paths use the event form without
|
|
22
|
+
* brackets: `'testsuites.testsuite.testcase'`.
|
|
23
|
+
*
|
|
24
|
+
* Memory stays bounded by the largest matched subtree, not the document.
|
|
25
|
+
*/
|
|
26
|
+
keep?: string[];
|
|
27
|
+
}
|
|
28
|
+
export interface XmlStreamOptions extends XmlParserOptions {
|
|
29
|
+
/**
|
|
30
|
+
* Close-event path to emit, in event form: `'</testsuites.testsuite.testcase>'`.
|
|
31
|
+
*
|
|
32
|
+
* Required, and it is what keeps memory bounded: elements matching the path
|
|
33
|
+
* are handed out whole (with their children) and everything else is
|
|
34
|
+
* dropped as it closes. Memory is the largest matched element, not the file.
|
|
35
|
+
*/
|
|
36
|
+
event: string;
|
|
37
|
+
}
|
|
38
|
+
/**
|
|
39
|
+
* Parse XML from a readable stream, emitting one object per matching element.
|
|
40
|
+
*
|
|
41
|
+
* ```js
|
|
42
|
+
* import { createStream } from 'xml-event-parser/stream';
|
|
43
|
+
*
|
|
44
|
+
* pipeline(
|
|
45
|
+
* createReadStream('junit.xml'),
|
|
46
|
+
* createStream({ event: '</testsuites.testsuite.testcase>' }),
|
|
47
|
+
* new Writable({ objectMode: true, write(testcase, _enc, done) { ...; done(); } }),
|
|
48
|
+
* );
|
|
49
|
+
* ```
|
|
50
|
+
*
|
|
51
|
+
* Chunks may split anywhere; multibyte characters split across chunk boundaries
|
|
52
|
+
* are reassembled with `StringDecoder`. Backpressure is applied per chunk: the
|
|
53
|
+
* stream stops pulling input while the readable side is full, so a slow consumer
|
|
54
|
+
* does not accumulate the document.
|
|
55
|
+
*
|
|
56
|
+
* This module imports `node:stream`, which is why it is a separate entry point
|
|
57
|
+
* (`xml-event-parser/stream`): the core parser stays usable in browsers and edge
|
|
58
|
+
* runtimes that expose no Node built-ins.
|
|
59
|
+
*/
|
|
60
|
+
export declare function createStream(options: XmlStreamOptions): Transform;
|
|
61
|
+
export {};
|
package/dist/stream.mjs
ADDED
|
@@ -0,0 +1 @@
|
|
|
1
|
+
import{Transform as e}from"node:stream";import{StringDecoder as t}from"node:string_decoder";function n(e){switch(e){case`amp`:return`&`;case`lt`:return`<`;case`gt`:return`>`;case`quot`:return`"`;case`apos`:return`'`}if(e.charCodeAt(0)===35){let t=e.charCodeAt(1)===120||e.charCodeAt(1)===88?parseInt(e.slice(2),16):parseInt(e.slice(1),10);if(Number.isFinite(t)&&t>=0&&t<=1114111)return String.fromCodePoint(t)}return null}function r(e){let t=e.indexOf(`&`);if(t<0)return e;let r=e.slice(0,t);for(;t<e.length;){let i=e.indexOf(`;`,t+1);if(i<0){r+=e.slice(t);break}let a=n(e.slice(t+1,i));if(r+=a===null?e.slice(t,i+1):a,t=e.indexOf(`&`,i+1),t<0){r+=e.slice(i+1);break}r+=e.slice(i+1,t)}return r}function i(e){if(e.indexOf(`<![CDATA[`)<0)return e.indexOf(`&`)<0?e:r(e);let t=``,n=0;for(;n<e.length;){let i=e.indexOf(`<![CDATA[`,n);if(i<0){t+=r(e.slice(n));break}i>n&&(t+=r(e.slice(n,i)));let a=e.indexOf(`]]>`,i+9);if(a<0){t+=e.slice(i+9);break}t+=e.slice(i+9,a),n=a+3}return t}function a(e){let t=0,n=e.length;return e.charCodeAt(0)===60&&(t=e.charCodeAt(1)===47?2:1),e.charCodeAt(n-1)===62&&n--,e.slice(t,n)}function o(e){return e===32||e===9||e===10||e===13}var s=class{constructor(e){if(this.queryParent=!1,this.progressive=!1,this.retain=`all`,this.keepPaths=/* @__PURE__ */ new Set,this.parentMap=/* @__PURE__ */ new WeakMap,this.evtListeners={},this.openListenerCount=0,this.watchedPaths=/* @__PURE__ */ new Set,this.watchedChildren=/* @__PURE__ */ new Map,this.watchedTopLevel=/* @__PURE__ */ new Set,this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=``,this.active=!1,e&&(this.queryParent=e.queryParent===!0,this.progressive=e.progressive===!0,e.retain&&(this.retain=e.retain),e.keep))for(let t of e.keep)this.keepPaths.add(a(t))}write(e){return this.active||this.begin(void 0,void 0),e.length===0||(this.buffer.length===0?this.buffer=e:this.buffer+=e,this.consume(!1)),this}end(){if(!this.active)return[];this.consume(!0),this.finish();let e=this.roots;return this.reset(),e}parse(e,t,n){return this.begin(t,n),this.write(e),this.end()}begin(e,t){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.basePath=t||``,this.top=e?{node:e,fullTag:this.basePath,textStart:0}:void 0,this.rootOwner=this.top,this.parentMap=/* @__PURE__ */ new WeakMap,this.active=!0}reset(){this.buffer=``,this.cursor=0,this.stack=[],this.roots=[],this.top=void 0,this.rootOwner=void 0,this.basePath=``,this.active=!1}attach(e,t,n){if(this.retain===`none`&&!n)return;if(t===void 0){this.retain===`all`&&this.roots.push(e);return}let r=t.children;r?r.push(e):t.children=[e]}consume(e){let t=this.buffer,n=t.length,r=this.progressive,a=r&&this.openListenerCount>0,o=this.cursor,s=-1;for(;o<n;){let c=t.indexOf(`<`,o);if(c<0||(s=c,o=c+1,o>=n))break;let l=t.charCodeAt(o);if(l===47){let a=t.indexOf(`>`,o);if(a<0){if(!e)break;o=n;break}let l=this.stack.pop();if(o=a+1,s=-1,!l)continue;let u=l.node,d=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,f=d?d.node:void 0;!l.hasChild&&c>l.textStart&&(u.text=i(t.slice(l.textStart,c))),r&&this.watchedPaths.has(l.fullTag)&&this.emit(`</`+l.fullTag+`>`,u,f),this.attach(u,f,d?d.keep===!0:!1),d&&(d.hasChild=!0),this.top=d;continue}if(l===63){let r=t.indexOf(`>`,o);if(r<0){if(!e)break;o=n;break}o=r+1,s=-1;continue}if(l===33){let r;if(t.charCodeAt(o+1)===45&&t.charCodeAt(o+2)===45?(r=t.indexOf(`-->`,o+3),r=r<0?-1:r+3):t.startsWith(`![CDATA[`,o)?(r=t.indexOf(`]]>`,o+9),r=r<0?-1:r+3):(r=t.indexOf(`>`,o),r=r<0?-1:r+1),r<0){if(!e)break;o=n;break}o=r,s=-1;continue}let u=this.parseOpenTag(t,o,n,e);if(u===null)break;o=u.end,s=-1;let d=u.node,f=u.tag,p=this.top?this.top.node:void 0;if(this.queryParent&&p&&this.parentMap.set(d,p),u.selfClose){if(r){let e=this.top?this.top.fullTag:this.basePath,t=e?this.watchedChildren.get(e):this.watchedTopLevel;if(t&&t.has(f)){let t=e?e+`.`+f:f;a&&this.emit(`<`+t+`>`,d,p),this.emit(`</`+t+`>`,d,p)}}this.attach(d,p,this.top?this.top.keep===!0:!1),this.top&&(this.top.hasChild=!0);continue}let m=this.top?this.top.fullTag?this.top.fullTag+`.`+f:f:this.basePath?this.basePath+`.`+f:f;a&&this.watchedPaths.has(m)&&this.emit(`<`+m+`>`,d,p);let h=this.top!==void 0&&this.top.keep===!0||this.keepPaths.has(m),g={node:d,fullTag:m,textStart:o,keep:h};this.stack.push(g),this.top=g}this.cursor=e?n:s>=0?s:o,e||this.compact()}parseOpenTag(e,t,n,i){let a=t;for(;a<n;){let t=e.charCodeAt(a);if(o(t)||t===47||t===62)break;a++}if(a>=n&&!i)return null;let s=e.slice(t,a),c={tagName:s},l=!1;for(;a<n;){let t=e.charCodeAt(a);for(;a<n&&o(t);)t=e.charCodeAt(++a);if(a>=n)break;if(t===62)return a++,{node:c,tag:s,selfClose:l,end:a};if(t===47){l=!0,a++;continue}let u=a;for(;a<n&&(t=e.charCodeAt(a),!(t===61||o(t)||t===62||t===47));)a++;if(a>=n&&!i)return null;let d=e.slice(u,a);for(;a<n&&o(e.charCodeAt(a));)a++;if(a>=n&&!i)return null;if(e.charCodeAt(a)===61){for(a++;a<n&&o(e.charCodeAt(a));)a++;if(a>=n&&!i)return null;let t=e.charCodeAt(a);if(t===34||t===39){a++;let o=e.indexOf(t===34?`"`:`'`,a);if(o<0){if(!i)return null;o=n}let s=e.slice(a,o);c[d]=s.indexOf(`&`)<0?s:r(s),a=o+1}}else d&&(c[d]=``)}return i?{node:c,tag:s,selfClose:l,end:a}:null}finish(){let e=this.buffer,t=e.length;for(;this.stack.length>0;){let n=this.stack.pop(),r=n.node;!n.hasChild&&t>n.textStart&&(r.text=i(e.slice(n.textStart,t)),t=n.textStart);let a=this.stack.length?this.stack[this.stack.length-1]:this.rootOwner,o=a?a.node:void 0;this.attach(r,o,a?a.keep===!0:!1)}}compact(){let e=this.cursor;for(let t=this.stack.length-1;t>=0;t--){let n=this.stack[t];if(!n.hasChild){e=n.textStart;break}}if(!(e<=0)){this.buffer=this.buffer.slice(e),this.cursor-=e;for(let t=0;t<this.stack.length;t++){let n=this.stack[t];n.textStart=n.textStart>e?n.textStart-e:0}this.rootOwner&&(this.rootOwner.textStart=0)}}getParent(e){return this.queryParent?this.parentMap.get(e):null}addListener(e,t){this.$addListener(e,t)}removeListener(e,t){this.$removeListener(e,t)}on(e,t){this.addListener(e,t)}off(e,t){this.removeListener(e,t)}$addListener(e,t){let n=this.evtListeners[e];if(n?n.push(t):this.evtListeners[e]=[t],e.charCodeAt(1)!==47&&this.openListenerCount++,e.charCodeAt(0)===60){let t=e.slice(e.charCodeAt(1)===47?2:1,e.length-1);this.watchedPaths.add(t);let n=t.lastIndexOf(`.`);if(n<0)this.watchedTopLevel.add(t);else{let e=t.slice(0,n),r=this.watchedChildren.get(e);r||(r=/* @__PURE__ */ new Set,this.watchedChildren.set(e,r)),r.add(t.slice(n+1))}}}$removeListener(e,t){let n=this.evtListeners[e],r=-1;n&&(r=n.indexOf(t)),r>=0&&(n.splice(r,1),e.charCodeAt(1)!==47&&this.openListenerCount--)}emit(e,t,n){let r=this.evtListeners[e];if(r)for(let e=0;e<r.length;e++)r[e](t,n)}};function c(e){if(typeof e!=`string`||e.length===0)throw TypeError(`createStream() requires { event }, e.g. { event: "</root.item>" }`);return e}function l(n){let r=c(n&&n.event),i=new s({...n,progressive:!0,retain:`none`,keep:[r]}),a=new t(`utf8`),o=new e({readableObjectMode:!0,transform(e,t,n){try{i.write(typeof e==`string`?e:a.write(e)),n()}catch(e){n(e)}},flush(e){try{let t=a.end();t.length>0&&i.write(t),i.end(),e()}catch(t){e(t)}}});return i.on(r,e=>{o.push(e)}),o}export{l as createStream};
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "xml-event-parser",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.2.1",
|
|
4
4
|
"description": "Tiny dependency-free XML parser with a tree and path-based events in one pass",
|
|
5
5
|
"repository": {
|
|
6
6
|
"type": "git",
|
|
@@ -13,9 +13,15 @@
|
|
|
13
13
|
"keywords": [
|
|
14
14
|
"xml",
|
|
15
15
|
"parser",
|
|
16
|
-
"events",
|
|
17
|
-
"tree",
|
|
18
16
|
"streaming",
|
|
17
|
+
"sax",
|
|
18
|
+
"event",
|
|
19
|
+
"junit",
|
|
20
|
+
"test-report",
|
|
21
|
+
"sitemap",
|
|
22
|
+
"gpx",
|
|
23
|
+
"edge",
|
|
24
|
+
"cloudflare-workers",
|
|
19
25
|
"zero-dependency"
|
|
20
26
|
],
|
|
21
27
|
"main": "dist/index.cjs",
|
|
@@ -25,6 +31,7 @@
|
|
|
25
31
|
"typecheck": "tsc --noEmit",
|
|
26
32
|
"bench:data": "node bench/generate.mjs",
|
|
27
33
|
"bench": "node --expose-gc bench/run.mjs",
|
|
34
|
+
"bench:memory": "npm run build && node bench/memory.mjs",
|
|
28
35
|
"prepublishOnly": "npm run build && npm test"
|
|
29
36
|
},
|
|
30
37
|
"license": "MIT",
|
|
@@ -38,6 +45,7 @@
|
|
|
38
45
|
"access": "public"
|
|
39
46
|
},
|
|
40
47
|
"devDependencies": {
|
|
48
|
+
"@types/node": "^26.6.6",
|
|
41
49
|
"fast-xml-parser": "^5.12.1",
|
|
42
50
|
"htmlparser2": "^12.0.0",
|
|
43
51
|
"rolldown": "^1.3.2",
|
|
@@ -54,6 +62,11 @@
|
|
|
54
62
|
"types": "./dist/index.d.ts",
|
|
55
63
|
"import": "./dist/index.mjs",
|
|
56
64
|
"require": "./dist/index.cjs"
|
|
65
|
+
},
|
|
66
|
+
"./stream": {
|
|
67
|
+
"types": "./dist/stream.d.ts",
|
|
68
|
+
"import": "./dist/stream.mjs",
|
|
69
|
+
"require": "./dist/stream.cjs"
|
|
57
70
|
}
|
|
58
71
|
},
|
|
59
72
|
"sideEffects": false,
|