Files
WeKnora/cli/cmd/doc/list.go
T
nullkey c90f4e895e refactor(cli): resource-command refactors — --kb resolver, MCP session_ask, session stop (BREAKING)
Unify the --kb resolver (accepts name or UUID) across kb/doc/chunk/session/
search/chat/link; `agent create` attaches knowledge bases via --attach-kb
(no longer --kb). Rename the MCP tool agent_invoke → session_ask and drop dead
MCP error codes. Add `session stop` (StopSession), completing the
start / continue-stream / stop triad.

Across these commands, bring the agent contract up to par: --dry-run previews
on mutations (incl. kb pin/unpin, session stop), agent-help blobs on the
read/health surface, accurate destructive-confirmation verbs, and search
has_more truncation signalling so an agent can tell its results were capped.
2026-06-08 20:18:43 +08:00

259 lines
9.8 KiB
Go

package doc
import (
"context"
"fmt"
"slices"
"sort"
"strings"
"text/tabwriter"
"time"
"github.com/spf13/cobra"
"github.com/Tencent/WeKnora/cli/internal/cmdutil"
"github.com/Tencent/WeKnora/cli/internal/iostreams"
"github.com/Tencent/WeKnora/cli/internal/output"
"github.com/Tencent/WeKnora/cli/internal/text"
sdk "github.com/Tencent/WeKnora/client"
)
// docListFields enumerates the fields surfaced for `--format json` discovery on
// `doc list`. Filter applies to each Knowledge object in the bare array.
var docListFields = []string{
"id", "knowledge_base_id", "tag_id", "type", "title", "description",
"source", "channel", "parse_status", "summary_status", "enable_status",
"embedding_model_id", "file_name", "file_type", "file_size", "file_hash",
"file_path", "storage_size",
"created_at", "updated_at", "processed_at", "error_message",
}
type ListOptions struct {
PageSize int // Items per server batch. With --all-pages, controls
// per-request load. Without, controls the single page size.
Status string // --status: filter by parse_status (server-side query param)
// Limit caps the returned items client-side (default 30; 0 = no cap).
// Applied after pagination / --all-pages accumulation and sort.
Limit int
// AllPages walks server pages internally, accumulating items until
// total exhausted or --limit hit.
AllPages bool
// Additional server-side filters (each maps 1:1 to a sdk.KnowledgeListFilter
// field). Empty / zero values are omitted from the request.
Keyword string
FileType string
Source string
TagID string
StartTime string // raw RFC3339; parsed into filter.StartTime
EndTime string // raw RFC3339; parsed into filter.EndTime
}
// rfc3339Example is the canonical RFC3339 hint surfaced when --start-time /
// --end-time fail to parse. Picked to match Go's reference time docs.
const rfc3339Example = "2006-01-02T15:04:05Z"
// docListStatusValues mirrors internal/types/knowledge.go ParseStatus*
// constants - these are the values the server accepts on the
// ?parse_status= query. Kept in sync manually since the SDK doesn't
// re-export the enum.
var docListStatusValues = []string{"pending", "processing", "completed", "failed"}
// ListService is the narrow SDK surface this command depends on.
// *sdk.Client satisfies it.
type ListService interface {
ListKnowledgeWithFilter(ctx context.Context, kbID string, page, pageSize int, filter sdk.KnowledgeListFilter) ([]sdk.Knowledge, int64, error)
}
// NewCmdList builds `weknora doc list`.
func NewCmdList(f *cmdutil.Factory) *cobra.Command {
opts := &ListOptions{}
cmd := &cobra.Command{
Use: "list",
Short: "List documents in a knowledge base",
Long: `Lists documents (uploaded files / web pages / inline text) in the
resolved knowledge base. KB resolution follows the standard 4-level chain:
--kb flag > WEKNORA_KB_ID env > .weknora/project.yaml > error. The --kb
flag accepts either a KB UUID (passed through) or a name (resolved via list).
Default sort is updated_at desc so the most recent uploads surface first;
backend storage order is not guaranteed and varies between deployments.`,
Example: ` weknora doc list # uses project link / env
weknora doc list --kb a32a63ff-fb36-4874-bcaa-30f48570a694 # explicit UUID
weknora doc list --kb my-kb # resolved by name
weknora doc list --all-pages --format json # walk every page`,
Args: cobra.NoArgs,
RunE: func(c *cobra.Command, _ []string) error {
fopts, err := cmdutil.CheckFormatFlag(c)
if err != nil {
return err
}
fopts.ResolveDefault(iostreams.IO.IsStdoutTTY())
kbID, err := f.ResolveKB(c)
if err != nil {
return err
}
cli, err := f.Client()
if err != nil {
return err
}
return runList(c.Context(), opts, fopts, cli, kbID)
},
}
cmdutil.AddKBFlag(cmd)
cmd.Flags().IntVar(&opts.PageSize, "page-size", 50, "Items per server batch (1..1000)")
cmd.Flags().IntVarP(&opts.Limit, "limit", "L", 30, "Maximum results to return (1..10000)")
cmd.Flags().BoolVar(&opts.AllPages, "all-pages", false, "Walk all server pages until exhausted (or --limit hit)")
cmd.Flags().StringVar(&opts.Status, "status", "", "Filter by parse status: pending | processing | completed | failed")
cmd.Flags().StringVar(&opts.Keyword, "keyword", "", "Server-side substring match against title / file_name (case-sensitive)")
cmd.Flags().StringVar(&opts.FileType, "file-type", "", `Filter by file extension (e.g. "pdf", "md")`)
cmd.Flags().StringVar(&opts.Source, "source", "", `Filter by ingestion source (e.g. "api", "web")`)
cmd.Flags().StringVar(&opts.TagID, "tag-id", "", "Filter by tag association")
cmd.Flags().StringVar(&opts.StartTime, "start-time", "", "Include docs with updated_at >= this RFC3339 timestamp (e.g. 2006-01-02T15:04:05Z)")
cmd.Flags().StringVar(&opts.EndTime, "end-time", "", "Include docs with updated_at <= this RFC3339 timestamp (e.g. 2006-01-02T15:04:05Z)")
cmdutil.AddFormatFlag(cmd, docListFields...)
cmdutil.SetAgentHelp(cmd, cmdutil.AgentHelp{
UsedFor: "List documents in the resolved knowledge base. Results come with meta.count; use --limit to cap, --all-pages to walk every server page, --status/--keyword to filter server-side.",
Examples: []string{"weknora doc list --format json", "weknora doc list --all-pages --limit 200 --format json"},
Output: "envelope.data is an array of Knowledge objects with id, title, file_name, parse_status; meta.count is the total returned",
})
return cmd
}
func runList(ctx context.Context, opts *ListOptions, fopts *cmdutil.FormatOptions, svc ListService, kbID string) error {
if opts.PageSize < 1 || opts.PageSize > 1000 {
return &cmdutil.Error{
Code: cmdutil.CodeInputInvalidArgument,
Message: fmt.Sprintf("--page-size must be in 1..1000, got %d", opts.PageSize),
}
}
if opts.Limit < 1 || opts.Limit > 10000 {
return &cmdutil.Error{
Code: cmdutil.CodeInputInvalidArgument,
Message: fmt.Sprintf("--limit must be in 1..10000, got %d", opts.Limit),
}
}
if opts.Status != "" && !validDocListStatus(opts.Status) {
return &cmdutil.Error{
Code: cmdutil.CodeInputInvalidArgument,
Message: fmt.Sprintf("--status must be one of: %s - got %q",
strings.Join(docListStatusValues, " | "), opts.Status),
}
}
filter := sdk.KnowledgeListFilter{
ParseStatus: opts.Status,
Keyword: opts.Keyword,
FileType: opts.FileType,
Source: opts.Source,
TagID: opts.TagID,
}
if opts.StartTime != "" {
t, err := time.Parse(time.RFC3339, opts.StartTime)
if err != nil {
return cmdutil.NewError(cmdutil.CodeInputInvalidArgument,
fmt.Sprintf("--start-time must be RFC3339 (e.g. %s), got %q", rfc3339Example, opts.StartTime))
}
filter.StartTime = t
}
if opts.EndTime != "" {
t, err := time.Parse(time.RFC3339, opts.EndTime)
if err != nil {
return cmdutil.NewError(cmdutil.CodeInputInvalidArgument,
fmt.Sprintf("--end-time must be RFC3339 (e.g. %s), got %q", rfc3339Example, opts.EndTime))
}
filter.EndTime = t
}
// Pagination is always 1-indexed internally. --all-pages walks; the
// non-walking path returns the first page only.
var items []sdk.Knowledge
if opts.AllPages {
accum := make([]sdk.Knowledge, 0)
for page := 1; ; page++ {
chunk, total, err := svc.ListKnowledgeWithFilter(ctx, kbID, page, opts.PageSize, filter)
if err != nil {
return cmdutil.WrapHTTP(err, "list documents")
}
accum = append(accum, chunk...)
if opts.Limit > 0 && len(accum) >= opts.Limit {
accum = accum[:opts.Limit]
break
}
if int64(len(accum)) >= total || len(chunk) == 0 {
break
}
}
items = accum
} else {
chunk, _, err := svc.ListKnowledgeWithFilter(ctx, kbID, 1, opts.PageSize, filter)
if err != nil {
return cmdutil.WrapHTTP(err, "list documents")
}
items = chunk
}
if items == nil {
items = []sdk.Knowledge{} // ensure JSON [] not null
}
// Default sort: updated_at desc. Server return order is not guaranteed,
// so client-side sort makes output deterministic regardless of backend
// storage choices. Mirrors `weknora kb list`.
sort.Slice(items, func(i, j int) bool {
return items[i].UpdatedAt.After(items[j].UpdatedAt)
})
// --limit applies after sort so users get the top-N most-recent items
// when combined with a single-page fetch where page_size > limit.
truncated := false
if opts.Limit > 0 && len(items) > opts.Limit {
items = items[:opts.Limit]
truncated = true
}
if fopts.WantsJSON() {
meta := &output.Meta{Count: len(items), HasMore: truncated}
return fopts.Emit(iostreams.IO.Out, items, meta)
}
if len(items) == 0 {
fmt.Fprintln(iostreams.IO.Out, "(no documents)")
return nil
}
tw := tabwriter.NewWriter(iostreams.IO.Out, 0, 0, 2, ' ', 0)
fmt.Fprintln(tw, "ID\tNAME\tSTATUS\tSIZE\tUPDATED")
now := time.Now()
for _, k := range items {
name := text.Truncate(40, text.KnowledgeDisplayName(k.FileName, k.Title, k.ID))
updated := text.FuzzyAgo(now, k.UpdatedAt)
fmt.Fprintf(tw, "%s\t%s\t%s\t%s\t%s\n", k.ID, name, k.ParseStatus, formatSize(k.FileSize), updated)
}
return tw.Flush()
}
// validDocListStatus reports whether s matches one of the server-accepted
// parse_status enum values surfaced via --status.
func validDocListStatus(s string) bool {
return slices.Contains(docListStatusValues, s)
}
// formatSize renders a byte count as a short human string (KB / MB).
// Kept tiny on purpose - go-humanize would pull a transitive dep just for one
// column. A "-" placeholder hides zero-size entries (URL / text).
func formatSize(bytes int64) string {
if bytes <= 0 {
return "-"
}
const (
kb = 1 << 10
mb = 1 << 20
gb = 1 << 30
)
switch {
case bytes >= gb:
return fmt.Sprintf("%.1fGB", float64(bytes)/float64(gb))
case bytes >= mb:
return fmt.Sprintf("%.1fMB", float64(bytes)/float64(mb))
case bytes >= kb:
return fmt.Sprintf("%.1fKB", float64(bytes)/float64(kb))
}
return fmt.Sprintf("%dB", bytes)
}