Chore: process line stream

This commit is contained in:
SukkaW 2024-11-26 21:22:14 +08:00
parent e2920de2fa
commit f64fa201e9
12 changed files with 67 additions and 43 deletions

View File

@ -5,18 +5,14 @@ import { task } from './trace';
import { SHARED_DESCRIPTION } from './constants/description';
import { appendArrayInPlace } from './lib/append-array-in-place';
import { SOURCE_DIR } from './constants/dir';
import { processLine } from './lib/process-line';
import { DomainsetOutput } from './lib/create-file';
import { CRASHLYTICS_WHITELIST } from './constants/reject-data-source';
const getS3OSSDomainsPromise = (async (): Promise<string[]> => {
const trie = new HostnameTrie();
for await (const line of await fetchRemoteTextByLine('https://publicsuffix.org/list/public_suffix_list.dat')) {
const tmp = processLine(line);
if (tmp) {
trie.add(tmp);
}
for await (const line of await fetchRemoteTextByLine('https://publicsuffix.org/list/public_suffix_list.dat', true)) {
trie.add(line);
}
/**

View File

@ -1,5 +1,4 @@
import { fetchRemoteTextByLine } from './lib/fetch-text-by-line';
import { processLineFromReadline } from './lib/process-line';
import { task } from './trace';
import { contains as containsCidr, exclude as excludeCidr } from 'fast-cidr-tools';
@ -19,8 +18,8 @@ const PROBE_CHN_CIDR_V4 = [
export const getChnCidrPromise = createMemoizedPromise(cachedOnlyFail(
async function getChnCidr() {
const [_cidr4, cidr6] = await Promise.all([
fetchRemoteTextByLine('https://raw.githubusercontent.com/misakaio/chnroutes2/master/chnroutes.txt').then(processLineFromReadline),
fetchRemoteTextByLine('https://gaoyifan.github.io/china-operator-ip/china6.txt').then(processLineFromReadline)
fetchRemoteTextByLine('https://raw.githubusercontent.com/misakaio/chnroutes2/master/chnroutes.txt', true).then(Array.fromAsync<string>),
fetchRemoteTextByLine('https://gaoyifan.github.io/china-operator-ip/china6.txt', true).then(Array.fromAsync<string>)
]);
const cidr4 = excludeCidr(

View File

@ -15,7 +15,7 @@ const getBogusNxDomainIPsPromise: Promise<[ipv4: string[], ipv6: string[]]> = $f
const ipv4: string[] = [];
const ipv6: string[] = [];
for await (const line of createReadlineInterfaceFromResponse(resp)) {
for await (const line of createReadlineInterfaceFromResponse(resp, true)) {
if (line.startsWith('bogus-nxdomain=')) {
const ip = line.slice(15).trim();
if (isProbablyIpv4(ip)) {

View File

@ -1,7 +1,6 @@
// @ts-check
import { createReadlineInterfaceFromResponse } from './lib/fetch-text-by-line';
import { isProbablyIpv4, isProbablyIpv6 } from './lib/is-fast-ip';
import { processLine } from './lib/process-line';
import { task } from './trace';
import { SHARED_DESCRIPTION } from './constants/description';
import { createMemoizedPromise } from './lib/memo-promise';
@ -16,10 +15,7 @@ export const getTelegramCIDRPromise = createMemoizedPromise(async () => {
const ipcidr: string[] = [];
const ipcidr6: string[] = [];
for await (const line of createReadlineInterfaceFromResponse(resp)) {
const cidr = processLine(line);
if (!cidr) continue;
for await (const cidr of createReadlineInterfaceFromResponse(resp, true)) {
const [subnet] = cidr.split('/');
if (isProbablyIpv4(subnet)) {
ipcidr.push(cidr);

View File

@ -1,5 +1,4 @@
import { fetchRemoteTextByLine } from './fetch-text-by-line';
import { processLineFromReadline } from './process-line';
import createKeywordFilter from './aho-corasick';
@ -36,7 +35,7 @@ if (require.main === module) {
(async () => {
const { bench, group, run } = await import('mitata');
const data = await processLineFromReadline(await fetchRemoteTextByLine('https://easylist.to/easylist/easylist.txt'));
const data = await Array.fromAsync(await fetchRemoteTextByLine('https://easylist.to/easylist/easylist.txt', true));
console.log({ dataLen: data.length });
const keywordsSet = [
'!',

View File

@ -1,4 +1,3 @@
import { processLine, processLineFromReadline } from './process-line';
import { readFileByLine, readFileByLineLegacy } from './fetch-text-by-line';
import path from 'node:path';
import fsp from 'node:fs/promises';
@ -10,9 +9,9 @@ const file = path.join(SOURCE_DIR, 'domainset/cdn.conf');
const { bench, group, run } = await import('mitata');
group(() => {
bench('readFileByLine', () => processLineFromReadline(readFileByLine(file)));
bench('readFileByLineLegacy', () => processLineFromReadline(readFileByLineLegacy(file)));
bench('fsp.readFile', () => fsp.readFile(file, 'utf-8').then((content) => content.split('\n').filter(processLine)));
bench('readFileByLine', () => Array.fromAsync(readFileByLine(file)));
bench('readFileByLineLegacy', () => Array.fromAsync(readFileByLineLegacy(file)));
bench('fsp.readFile', () => fsp.readFile(file, 'utf-8').then((content) => content.split('\n')));
});
run();

View File

@ -6,7 +6,7 @@ import readline from 'node:readline';
import { TextLineStream } from './text-line-transform-stream';
import type { ReadableStream } from 'node:stream/web';
import { TextDecoderStream } from 'node:stream/web';
import { processLine } from './process-line';
import { processLine, ProcessLineStream } from './process-line';
import { $fetch } from './make-fetch-happen';
import type { NodeFetchResponse } from './make-fetch-happen';
import type { UndiciResponseData } from './fetch-retry';
@ -40,7 +40,7 @@ function ensureResponseBody<T extends NodeFetchResponse | UndiciResponseData | U
return resp.body;
}
export const createReadlineInterfaceFromResponse: ((resp: NodeFetchResponse | UndiciResponseData | UnidiciWebResponse) => AsyncIterable<string>) = (resp) => {
export const createReadlineInterfaceFromResponse: ((resp: NodeFetchResponse | UndiciResponseData | UnidiciWebResponse, processLine?: boolean) => ReadableStream<string>) = (resp, processLine = false) => {
const stream = ensureResponseBody(resp);
const webStream: ReadableStream<Uint8Array> = 'getReader' in stream
@ -51,13 +51,18 @@ export const createReadlineInterfaceFromResponse: ((resp: NodeFetchResponse | Un
: Readable.toWeb(new Readable().wrap(stream))
);
return webStream
const resultStream = webStream
.pipeThrough(new TextDecoderStream())
.pipeThrough(new TextLineStream());
if (processLine) {
return resultStream.pipeThrough(new ProcessLineStream());
}
return resultStream;
};
export function fetchRemoteTextByLine(url: string) {
return $fetch(url).then(createReadlineInterfaceFromResponse);
export function fetchRemoteTextByLine(url: string, processLine = false): Promise<AsyncIterable<string>> {
return $fetch(url).then(resp => createReadlineInterfaceFromResponse(resp, processLine));
}
export async function readFileIntoProcessedArray(file: string /* | FileHandle */) {

View File

@ -19,7 +19,7 @@ export function extractDomainsFromFelixDnsmasq(line: string): string | null {
export async function parseFelixDnsmasqFromResp(resp: NodeFetchResponse | UndiciResponseData | Response): Promise<string[]> {
const results: string[] = [];
for await (const line of createReadlineInterfaceFromResponse(resp)) {
for await (const line of createReadlineInterfaceFromResponse(resp, true)) {
const domain = extractDomainsFromFelixDnsmasq(line);
if (domain && isDomainLoose(domain)) {
results.push(domain);

View File

@ -1,3 +1,5 @@
import { TransformStream } from 'node:stream/web';
export function processLine(line: string): string | null {
if (!line) {
return null;
@ -11,8 +13,7 @@ export function processLine(line: string): string | null {
const line_0: string = trimmed[0];
if (
line_0 === '#'
|| line_0 === ' '
line_0 === ' '
|| line_0 === '\r'
|| line_0 === '\n'
|| line_0 === '!'
@ -21,16 +22,48 @@ export function processLine(line: string): string | null {
return null;
}
if (line_0 === '#') {
if (trimmed[1] !== '#') {
// # Comment
return null;
}
if (trimmed[2] === '#' && trimmed[3] === '#') {
// ################## EOF ##################
return null;
}
/**
* AdGuard Filter can be:
*
* ##.class
* ##tag.class
* ###id
*/
}
return trimmed;
}
export async function processLineFromReadline(rl: AsyncIterable<string>): Promise<string[]> {
const res: string[] = [];
for await (const line of rl) {
const l: string | null = processLine(line);
if (l) {
res.push(l);
}
export class ProcessLineStream extends TransformStream<string, string> {
// private __buf = '';
constructor() {
super({
transform(l, controller) {
const line = processLine(l);
if (line) {
controller.enqueue(line);
}
}
});
}
return res;
}
// export class ProcessLineNodeStream extends Transform {
// _transform(chunk: string, encoding: BufferEncoding, callback: TransformCallback) {
// // Convert chunk to string and then to uppercase
// const upperCased = chunk.toUpperCase();
// // Push transformed data to readable side
// this.push(upperCased);
// // Call callback when done
// callback();
// }
// }

View File

@ -1,10 +1,9 @@
import { fetchRemoteTextByLine } from './fetch-text-by-line';
import { processLineFromReadline } from './process-line';
import { bench, group, run } from 'mitata';
(async () => {
const data = await processLineFromReadline(await fetchRemoteTextByLine('https://osint.digitalside.it/Threat-Intel/lists/latestdomains.txt'));
const data = await Array.fromAsync(await fetchRemoteTextByLine('https://osint.digitalside.it/Threat-Intel/lists/latestdomains.txt', true));
group(() => {
bench('setAddFromArray', () => {

View File

@ -1,11 +1,10 @@
import { fetchRemoteTextByLine } from './fetch-text-by-line';
import { processLineFromReadline } from './process-line';
import { sortDomains } from './stable-sort-domain';
import { bench, group, run } from 'mitata';
(async () => {
const data = await processLineFromReadline(await fetchRemoteTextByLine('https://osint.digitalside.it/Threat-Intel/lists/latestdomains.txt'));
const data = await Array.fromAsync(await fetchRemoteTextByLine('https://osint.digitalside.it/Threat-Intel/lists/latestdomains.txt', true));
group(() => {
bench('sortDomains', () => sortDomains(data));

View File

@ -1,5 +1,4 @@
import { fetchRemoteTextByLine } from './fetch-text-by-line';
import { processLineFromReadline } from './process-line';
import { bench, group, run } from 'mitata';
@ -7,7 +6,7 @@ import * as tldts from 'tldts';
import * as tldtsExperimental from 'tldts-experimental';
(async () => {
const data = await processLineFromReadline(await fetchRemoteTextByLine('https://osint.digitalside.it/Threat-Intel/lists/latestdomains.txt'));
const data = await Array.fromAsync(await fetchRemoteTextByLine('https://osint.digitalside.it/Threat-Intel/lists/latestdomains.txt', true));
const tldtsOpt: Parameters<typeof tldts.getDomain>[1] = {
allowPrivateDomains: false,