mirror of
https://github.com/xlionjuan/searxng-mcp-go.git
synced 2026-11-17 04:17:51 +00:00
* chore(deps): update dependency go to v1.27.1 * Fixed all lints; just verify passes. Co-authored-by: xlionjuan <[email protected]> * Fixed `.golangci.yml` paths→path; `just verify` passes. Co-authored-by: xlionjuan <[email protected]> * Parallelized golden & error-path tests; verify passes. Co-authored-by: xlionjuan <[email protected]> * Verify passes; HTTPClient/Logger nil set; exhaustruct covers Config. Co-authored-by: xlionjuan <[email protected]> --------- Co-authored-by: xlion-bot-renovate[bot] <218815870+xlion-bot-renovate[bot]@users.noreply.github.com> Co-authored-by: opencode-agent[bot] <opencode-agent[bot]@users.noreply.github.com> Co-authored-by: xlionjuan <[email protected]>
79 lines
3.5 KiB
Go
79 lines
3.5 KiB
Go
// Package searxng provides a client library for interacting with SearXNG search instances.
|
|
package searxng
|
|
|
|
import "time"
|
|
|
|
// DefaultTimeout is the fallback HTTP client timeout for SearXNG requests.
|
|
const DefaultTimeout = 8 * time.Second
|
|
|
|
// DefaultMaxRetries is the fallback number of retries after the initial request.
|
|
const DefaultMaxRetries = 5
|
|
|
|
// DefaultRetryDelay is the fallback base delay for exponential backoff.
|
|
const DefaultRetryDelay = time.Second
|
|
|
|
// DefaultMaxRetryDelay is the fallback upper bound for retry backoff delays.
|
|
const DefaultMaxRetryDelay = 30 * time.Second
|
|
|
|
// DefaultResultLimit is the default value applied when callers omit a result
|
|
// limit. It is the single source of truth used by CLI flag defaulting, MCP
|
|
// handler defaulting, and the canonical ParamDef metadata in params.go.
|
|
var DefaultResultLimit = 10
|
|
|
|
// MaxContentRunes is the CLI text-mode formatting limit, in Unicode runes,
|
|
// applied to result `content` and infobox `content` fields by
|
|
// `formatResults` (root package). It is purely a rendering budget for
|
|
// terminal output: the typed `SearchResponse` returned by normalization
|
|
// and surfaced to JSON and MCP consumers is NOT truncated, so downstream
|
|
// consumers always see the full upstream text.
|
|
//
|
|
// The rune-safe truncation itself is implemented by `TruncateRunes`
|
|
// (truncate.go, with the implementation in the answer subpackage) so the
|
|
// answer deduplication prefix match and the CLI formatter share a single
|
|
// helper. See docs/adr/011-max-content-runes-cli-only.md for the scope
|
|
// decision.
|
|
const MaxContentRunes = 4000
|
|
|
|
// MaxErrorBodySize is the maximum response body size read for error handling.
|
|
const MaxErrorBodySize = 100 * 1024
|
|
|
|
// MaxResponseBodySize is the maximum successful search response body size.
|
|
const MaxResponseBodySize = 2 * 1024 * 1024
|
|
|
|
// MaxAnswers caps the number of answers retained after JSON unmarshalling.
|
|
//
|
|
// The answer/infobox deduplication pass runs in O(len(answers)*len(infoboxes))
|
|
// and, for each answer, scans every infobox twice with strings.Contains.
|
|
// Real SearXNG responses contain at most a handful of each, so a generous cap
|
|
// keeps deduplication work bounded and prevents an adversarial or
|
|
// mis-configured upstream from forcing O(n*m) substring scans inside the
|
|
// MaxResponseBodySize envelope. Truncation is performed before
|
|
// deduplication and a warning is logged when it triggers.
|
|
const MaxAnswers = 100
|
|
|
|
// MaxInfoboxes caps the number of infoboxes retained after JSON
|
|
// unmarshalling. See MaxAnswers for the rationale; the cap exists to bound
|
|
// the answer/infobox deduplication work the same way.
|
|
const MaxInfoboxes = 100
|
|
|
|
// MaxErrorDisplayBytes caps the size of error body previews retained on
|
|
// SearXNGError.ResponseBody (and surfaced in debug logs and error messages).
|
|
//
|
|
// This is a byte ceiling, not a rune/character count: callers truncate with
|
|
// truncateBytesToValidUTF8, which slices at most MaxErrorDisplayBytes bytes
|
|
// and walks back to a valid UTF-8 rune boundary so multi-byte sequences are
|
|
// never split.
|
|
const MaxErrorDisplayBytes = 200
|
|
|
|
// ResultSizeEstimate is the approximate JSON result size used for response preallocation.
|
|
const ResultSizeEstimate = 200
|
|
|
|
// DebugBodyPreviewBytes is the maximum response body preview length emitted in debug logs.
|
|
const DebugBodyPreviewBytes = 500
|
|
|
|
// schemeHTTP and schemeHTTPS are the URL scheme strings compared across the
|
|
// redirect policy, base-URL validation, and the plaintext-http warning.
|
|
const (
|
|
schemeHTTP = "http"
|
|
schemeHTTPS = "https"
|
|
)
|