import type { Tag } from '@/types/tag.types';
import { logger } from '@logger';
import { generateText } from 'ai';
import { readFileSync } from 'fs';
import { dirname, join } from 'path';
import { fileURLToPath } from 'url';
import { AIConfig, AIOptions } from '../aiSdk';
const __dirname = dirname(fileURLToPath(import.meta.url));
// Get the content of a file at the specified path
const getFileContent = (filePath: string) => {
return readFileSync(join(__dirname, filePath), { encoding: 'utf-8' });
};
export type AuditOptions = {
fileContent: string;
tags: Tag[];
aiConfig: AIConfig;
applicationContext?: string;
};
export type AuditFileResultData = {
fileContent: string;
tokenUsed: number;
};
// The prompt template to send to AI models
const CHAT_GPT_PROMPT = getFileContent('./PROMPT.md');
export const aiDefaultOptions: AIOptions = {
// Keep default options
};
/**
* Audits a content declaration file by constructing a prompt for AI models.
* The prompt includes details about the project's locales, file paths of content declarations,
* and requests for identifying issues or inconsistencies.
*/
export const auditDictionaryMetadata = async ({
tags,
fileContent,
applicationContext,
aiConfig,
}: AuditOptions): Promise<AuditFileResultData | undefined> => {
// Prepare the prompt for AI by replacing placeholders with actual values.
const prompt = CHAT_GPT_PROMPT.replace(
'{{tags}}',
`${JSON.stringify(
tags
.map(({ key, description }) => `- ${key}: ${description}`)
.join('\n\n'),
null,
2
)}`
)
.replace('{{contentDeclaration}}', fileContent)
.replace('{{applicationContext}}', applicationContext ?? '');
if (!aiConfig) {
logger.error('Failed to configure AI model');
return undefined;
}
// Use the AI SDK to generate the completion
const { text: newContent, usage } = await generateText({
...aiConfig,
messages: [{ role: 'system', content: prompt }],
});
logger.info(`${usage?.totalTokens ?? 0} tokens used in the request`);
return {
fileContent: newContent,
tokenUsed: usage?.totalTokens ?? 0,
};
};
