This commit is contained in:
lasta
2026-05-22 00:22:03 +09:00
parent 8d77bc659f
commit d841fb823a
49 changed files with 2732 additions and 3763 deletions

View File

@@ -44,6 +44,11 @@ const startCrawlSchema = z.object({
max_depth: z.number().int().min(0).max(10),
max_pages: z.number().int().min(1).max(500),
same_domain_only: z.boolean(),
extraction_mode: z.enum(["rule_only", "llm_only", "hybrid", "compare"]),
extractor_provider: z.enum(["lm_studio", "openai", "ollama"]),
extractor_model: z.string().optional(),
extractor_base_url: z.string().optional(),
fallback_to_rules: z.boolean(),
});
type StartCrawlFormValues = z.infer<typeof startCrawlSchema>;
@@ -96,14 +101,22 @@ export default function CrawlPage() {
max_depth: 2,
max_pages: 30,
same_domain_only: true,
extraction_mode: "hybrid",
extractor_provider: "lm_studio",
extractor_model: "",
extractor_base_url: "http://localhost:1234/v1",
fallback_to_rules: true,
},
});
const onStart = async (values: StartCrawlFormValues) => {
try {
const usesLlm = values.extraction_mode !== "rule_only";
const created = await startCrawl.mutateAsync({
project_name: projectName,
...values,
extractor_model: usesLlm ? values.extractor_model || null : null,
extractor_base_url: usesLlm ? values.extractor_base_url || null : null,
});
setActiveJobId(created.job_id);
toast.success(
@@ -136,6 +149,8 @@ export default function CrawlPage() {
const sources = project?.sources ?? [];
const sourceName = watch("source_name");
const extractionMode = watch("extraction_mode");
const usesLlm = extractionMode !== "rule_only";
const selectedSource = sources.find((s) => s.name === sourceName);
const progress = job?.progress;
const visited = progress?.visited_count ?? 0;
@@ -340,6 +355,60 @@ export default function CrawlPage() {
</span>
</label>
<div className="space-y-3 rounded-md border bg-background-subtle p-3">
<div className="grid grid-cols-2 gap-3">
<div className="space-y-1.5">
<Label htmlFor="extraction_mode">Extraction mode</Label>
<Select id="extraction_mode" {...register("extraction_mode")}>
<option value="hybrid">Hybrid</option>
<option value="rule_only">Rule only</option>
<option value="llm_only">LLM only</option>
<option value="compare">Compare</option>
</Select>
</div>
<div className="space-y-1.5">
<Label htmlFor="extractor_provider">Provider</Label>
<Select
id="extractor_provider"
disabled={!usesLlm}
{...register("extractor_provider")}
>
<option value="lm_studio">LM Studio</option>
<option value="openai">OpenAI</option>
<option value="ollama">Ollama</option>
</Select>
</div>
</div>
{usesLlm && (
<>
<div className="space-y-1.5">
<Label htmlFor="extractor_model">Model</Label>
<Input
id="extractor_model"
placeholder="deepseek-r1-distill-qwen-7b"
{...register("extractor_model")}
/>
</div>
<div className="space-y-1.5">
<Label htmlFor="extractor_base_url">Base URL</Label>
<Input
id="extractor_base_url"
placeholder="http://localhost:1234/v1"
{...register("extractor_base_url")}
/>
</div>
<label className="flex items-center gap-2 text-sm">
<input
type="checkbox"
className="h-4 w-4 rounded border-input"
{...register("fallback_to_rules")}
/>
<span>Fallback to rules</span>
</label>
</>
)}
</div>
<Button
type="submit"
className="w-full"