Files
seshat/backend/nestjs-seshat-api/src/providers/google/google.service.ts

120 lines
4.7 KiB
TypeScript

import { Injectable } from '@nestjs/common';
import { BookSearchResultDto } from '../dto/book-search-result.dto';
import { HttpService } from '@nestjs/axios';
import { firstValueFrom, map, timeout } from 'rxjs';
import { AxiosResponse } from 'axios';
import { GoogleSearchContext } from '../contexts/google.search.context';
@Injectable()
export class GoogleService {
constructor(
private readonly http: HttpService,
) { }
async searchRaw(searchQuery: string): Promise<BookSearchResultDto[]> {
const queryParams = 'langRestrict=en&printType=books&maxResults=10&fields=items(kind,id,volumeInfo(title,description,authors,publisher,publishedDate,industryIdentifiers,language,categories,maturityRating,imageLinks,canonicalVolumeLink,seriesInfo))&q=';
return await firstValueFrom(
this.http.get('https://www.googleapis.com/books/v1/volumes?' + queryParams + searchQuery)
.pipe(
timeout({ first: 5000 }),
map(value => this.transform(value)),
)
);
}
async search(context: GoogleSearchContext): Promise<BookSearchResultDto[]> {
if (!context) {
return null;
}
const defaultQueryParams = 'langRestrict=en&printType=books&fields=items(kind,id,volumeInfo(title,description,authors,publisher,publishedDate,industryIdentifiers,language,categories,maturityRating,imageLinks,canonicalVolumeLink,seriesInfo))';
const customQueryParams = context.generateQueryParams();
return await firstValueFrom(
this.http.get('https://www.googleapis.com/books/v1/volumes?' + defaultQueryParams + '&' + customQueryParams)
.pipe(
timeout({ first: 5000 }),
map(value => this.transform(value)),
)
);
}
private transform(response: AxiosResponse): BookSearchResultDto[] {
if (!response.data?.items) {
return [];
}
return response.data.items
//.filter(item => item.volumeInfo?.canonicalVolumeLink?.startsWith('https://play.google.com/store/books/details'))
.map(item => this.extract(item));
}
private extract(item: any): BookSearchResultDto {
const result: BookSearchResultDto = {
providerBookId: item.id,
providerSeriesId: item.volumeInfo.seriesInfo?.volumeSeries[0].seriesId,
title: item.volumeInfo.title,
desc: item.volumeInfo.description,
volume: item.volumeInfo.seriesInfo?.bookDisplayNumber ? parseInt(item.volumeInfo.seriesInfo?.bookDisplayNumber, 10) : undefined,
publisher: item.volumeInfo.publisher,
authors: item.volumeInfo.authors,
categories: item.volumeInfo.categories ?? [],
mediaType: null,
maturityRating: item.volumeInfo.maturityRating,
industryIdentifiers: item.volumeInfo.industryIdentifiers ? Object.assign({}, ...item.volumeInfo.industryIdentifiers.map(i => i.type == 'OTHER' ? { [i.identifier.split(':')[0]]: i.identifier.split(':')[1] } : { [i.type]: i.identifier })) : [],
publishedAt: new Date(item.volumeInfo.publishedDate),
language: item.volumeInfo.language,
thumbnail: item.volumeInfo.imageLinks?.thumbnail,
url: item.volumeInfo.canonicalVolumeLink,
provider: 'google'
}
const regex = this.getRegexByPublisher(result.publisher);
const match = result.title.match(regex);
if (match?.groups) {
result.title = match.groups['title'].trim();
if (!result.volume || isNaN(result.volume)) {
result.volume = parseInt(match.groups['volume'], 10);
}
}
if (match?.groups && 'media_type' in match.groups) {
result.mediaType = match.groups['media_type'];
} else if (result.categories.includes('Comics & Graphic Novels')) {
result.mediaType = 'Comics & Graphic Novels';
} else if (result.categories.includes('Fiction') || result.categories.includes('Young Adult Fiction')) {
result.mediaType = 'Novel';
} else {
result.mediaType = 'Book';
}
if (result.mediaType) {
if (result.mediaType.toLowerCase() == "light novel") {
result.mediaType = 'Light Novel';
} else if (result.mediaType.toLowerCase() == 'manga') {
result.mediaType = 'Manga';
}
}
return result;
}
private getRegexByPublisher(publisher: string): RegExp {
switch (publisher) {
case 'J-Novel Club':
return /^(?<title>.+?):?\sVolume\s(?<volume>\d+)$/i;
case 'Yen On':
case 'Yen Press':
case 'Yen Press LLC':
return /^(?<title>.+?)(?:,?\sVol\.\s(?<volume>\d+))?\s\((?<media_type>[\w\s]+)\)$/;
case 'Hanashi Media':
return /^(?<title>.+?)\s\((?<media_type>[\w\s]+)\),?\sVol\.\s(?<volume>\d+)$/
case 'Regin\'s Chronicles':
return /^(?<title>.+?)\s\((?<media_type>[\w\s]+)\)(?<subtitle>\:\s.+?)?$/
default:
return /^(?<title>.+?)(?:,|:|\s\-)?\s(?:Vol(?:\.|ume)?)?\s(?<volume>\d+)$/;
}
}
}