Files
skillshare/internal/server/handler_toggle.go
Willie 89fc4fbdf1 feat(sourcewalk): follow first-level source links when enabled
Users link a checkout directly under the skills source (e.g.
skills/_dev-skills -> ~/code/dev-skills) and expect it to behave like any
other first-level entry. sourcewalk resolved only the root, so these links
were invisible to discovery, sync, update, and the dashboard.

follow_source_links (global and project config, default false) enables a
per-operation sourcewalk.Follow policy. Walk, WalkDir, and ReadDir report a
followed link under its own name as a directory and descend into its
target, so names, RelPath, FlatName, SourcePath, tracked-repo detection, and
metadata keys stay logical. Only entries directly under the source root are
followed. The policy is built from config and passed explicitly to every
discovery caller in the CLI, server, and reconcile, so counts agree; the
server keeps a per-instance snapshot refreshed on config load and save
instead of a process-global flag.

A link is skipped, with one warning per operation, when its target is
missing or unreadable, is the source or a parent of it, or overlaps an
active skills target. A missing or unreadable target marks the operation
incomplete: sync then skips pruning and copy-orphan deletion, and reconcile
keeps metadata, so an unmounted drive or a re-clone in progress cannot
erase target entries or install records. A link to an existing file is not
a directory link and stays an ordinary entry; treating it as incomplete
would disable pruning for anyone sharing a file such as .skillignore.

doctor keeps raw link identity: with the key off it still says the link is
not followed and names follow_source_links: true; with it on it reports
followed links and skipped links with the reason.

The raw-walk allowlist gains entries for the policy path in Walk and
WalkDir: the source root walk with a policy, and the walk of a validated
link target whose paths are remapped to the logical link path. Both live
in sourcewalk itself, the package that owns source traversal.

Refs #274
2026-10-05 14:26:09 +08:00

183 lines
5.6 KiB
Go

package server
import (
"fmt"
"net/http"
"path/filepath"
"time"
"skillshare/internal/resource"
"skillshare/internal/skillignore"
ssync "skillshare/internal/sync"
)
func (s *Server) handleDisableSkill(w http.ResponseWriter, r *http.Request) {
s.handleToggleSkill(w, r, false)
}
func (s *Server) handleEnableSkill(w http.ResponseWriter, r *http.Request) {
s.handleToggleSkill(w, r, true)
}
func (s *Server) handleToggleSkill(w http.ResponseWriter, r *http.Request, enable bool) {
start := time.Now()
name := r.PathValue("name")
kind := r.URL.Query().Get("kind")
if kind != "" && kind != "agent" && kind != "skill" {
writeError(w, http.StatusBadRequest, "invalid kind: "+kind)
return
}
// Resolve under RLock — discovery is I/O-heavy, don't hold write lock
s.mu.RLock()
source := s.cfg.EffectiveSkillsSource()
agentsSource := s.agentsSource()
s.mu.RUnlock()
relPath := ""
isDisabled := false
ignorePath := ""
if kind == "agent" {
if agentsSource == "" {
writeError(w, http.StatusNotFound, "agent not found: "+name)
return
}
var err error
relPath, isDisabled, err = s.resolveAgentRelPathWithStatus(agentsSource, name)
if err != nil {
writeError(w, http.StatusNotFound, err.Error())
return
}
ignorePath = filepath.Join(agentsSource, ".agentignore")
} else {
var err error
relPath, isDisabled, err = s.resolveSkillRelPathWithStatus(source, name)
if err != nil {
writeError(w, http.StatusNotFound, err.Error())
return
}
ignorePath = filepath.Join(source, ".skillignore")
}
// Write lock only for the file mutation
s.mu.Lock()
defer s.mu.Unlock()
action := "disable"
disabled := true
if enable {
action = "enable"
disabled = false
if !isDisabled {
writeJSON(w, map[string]any{"success": true, "name": name, "disabled": false, "message": "not disabled"})
return
}
removed, err := skillignore.RemovePattern(ignorePath, relPath)
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to update .skillignore: "+err.Error())
return
}
if !removed {
writeError(w, http.StatusConflict, "skill is disabled by a glob or directory pattern in .skillignore or .skillignore.local — edit the file manually to remove the matching rule")
return
}
} else {
added, err := skillignore.AddPattern(ignorePath, relPath)
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to update .skillignore: "+err.Error())
return
}
if !added && isDisabled {
writeJSON(w, map[string]any{"success": true, "name": name, "disabled": true, "message": "already disabled"})
return
}
}
// The file write alone does not decide the outcome: a later rule, such as a
// "!" line in the .local file, can still override it.
states, err := s.disabledByRelPath(kind, source, agentsSource)
if err != nil {
writeError(w, http.StatusInternalServerError, err.Error())
return
}
logArgs := map[string]any{
"name": name,
"kind": kind,
"scope": "ui",
}
if msg := toggleOverrideError(kind, enable, states[relPath]); msg != "" {
// The ignore file may already have changed, so the attempt is still logged.
s.writeOpsLog(action, "error", start, logArgs, msg)
writeError(w, http.StatusConflict, msg)
return
}
s.writeOpsLog(action, "ok", start, logArgs, "")
writeJSON(w, map[string]any{"success": true, "name": name, "disabled": disabled})
}
// disabledByRelPath re-discovers skills or agents and reports, by relPath,
// whether each is disabled under the merged ignore and .local rules.
func (s *Server) disabledByRelPath(kind, source, agentsSource string) (map[string]bool, error) {
states := map[string]bool{}
if kind == "agent" {
discovered, err := resource.AgentKind{}.Discover(agentsSource)
if err != nil {
return nil, fmt.Errorf("failed to discover agents: %w", err)
}
for _, d := range discovered {
states[d.RelPath] = d.Disabled
}
return states, nil
}
discovered, err := ssync.DiscoverSourceSkillsAll(source, s.skillsWalk())
if err != nil {
return nil, fmt.Errorf("failed to discover skills: %w", err)
}
for _, d := range discovered {
states[d.RelPath] = d.Disabled
}
return states, nil
}
// toggleOverrideError explains why a toggle did not take effect, or returns ""
// when the resource ended up in the requested state.
func toggleOverrideError(kind string, enable, disabled bool) string {
file := ".skillignore"
if kind == "agent" {
file = ".agentignore"
}
switch {
case !enable && !disabled:
return fmt.Sprintf("still enabled: a \"!\" rule in %[1]s.local or %[1]s re-includes it — remove that rule to disable it", file)
case enable && disabled:
return fmt.Sprintf("still disabled: another rule in %[1]s or %[1]s.local matches it — remove that rule to enable it", file)
}
return ""
}
// resolveSkillRelPath finds a skill's relPath by flatName or baseName.
// Searches all skills including disabled ones. Caller must provide source path.
func (s *Server) resolveSkillRelPath(source, name string) (string, error) {
relPath, _, err := s.resolveSkillRelPathWithStatus(source, name)
return relPath, err
}
// resolveSkillRelPathWithStatus is like resolveSkillRelPath but also returns
// whether the skill is currently disabled (matched by .skillignore).
func (s *Server) resolveSkillRelPathWithStatus(source, name string) (string, bool, error) {
discovered, err := ssync.DiscoverSourceSkillsAll(source, s.skillsWalk())
if err != nil {
return "", false, fmt.Errorf("failed to discover skills: %w", err)
}
for _, d := range discovered {
baseName := filepath.Base(d.SourcePath)
if d.FlatName == name || baseName == name || d.RelPath == name {
return d.RelPath, d.Disabled, nil
}
}
return "", false, fmt.Errorf("skill not found: %s", name)
}