ShadowsOverWestgate/sow-tools
7
0
Fork 0
Code Issues 2 Pull Requests Actions Packages Projects Releases 38 Activity

Immutable Extracts

Browse Source
This commit is contained in:
archvillainette
2026-05-20 20:20:07 +02:00
parent b1b7a08848
commit c37b27b720
5 changed files with 603 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files
internal/project/project.go
+81 -7
View File
@@ -247,13 +247,30 @@ type TopDataWikiManagedRegionConfig struct {
}
type ExtractConfig struct {
IgnoreExtensions []string `json:"ignore_extensions" yaml:"ignore_extensions"`
Archives []string `json:"archives,omitempty" yaml:"archives,omitempty"`
Layout string `json:"layout" yaml:"layout"`
HAKDiscovery string `json:"hak_discovery" yaml:"hak_discovery"`
CleanupStale *bool `json:"cleanup_stale,omitempty" yaml:"cleanup_stale,omitempty"`
ConsumeArchives *bool `json:"consume_archives,omitempty" yaml:"consume_archives,omitempty"`
DeleteModuleArchiveAfterSuccess bool `json:"delete_module_archive_after_success,omitempty" yaml:"delete_module_archive_after_success,omitempty"`
IgnoreExtensions []string `json:"ignore_extensions" yaml:"ignore_extensions"`
Archives []string `json:"archives,omitempty" yaml:"archives,omitempty"`
Layout string `json:"layout" yaml:"layout"`
HAKDiscovery string `json:"hak_discovery" yaml:"hak_discovery"`
CleanupStale *bool `json:"cleanup_stale,omitempty" yaml:"cleanup_stale,omitempty"`
ConsumeArchives *bool `json:"consume_archives,omitempty" yaml:"consume_archives,omitempty"`
DeleteModuleArchiveAfterSuccess bool `json:"delete_module_archive_after_success,omitempty" yaml:"delete_module_archive_after_success,omitempty"`
Merge ExtractMergeConfig `json:"merge" yaml:"merge"`
}
type ExtractMergeConfig struct {
GFFJSON []ExtractGFFJSONMergeRule `json:"gff_json" yaml:"gff_json"`
}
type ExtractGFFJSONMergeRule struct {
Target string `json:"target" yaml:"target"`
PreserveFields []string `json:"preserve_fields" yaml:"preserve_fields"`
MergeLists []ExtractListMergeRule `json:"merge_lists" yaml:"merge_lists"`
}
type ExtractListMergeRule struct {
Field string `json:"field" yaml:"field"`
KeyField string `json:"key_field" yaml:"key_field"`
Strategy string `json:"strategy" yaml:"strategy"`
}
type AutogenConfig struct {
@@ -598,6 +615,7 @@ func (p *Project) ValidateLayout() error {
failures = append(failures, fmt.Errorf("extract.hak_discovery %q is not supported", effective.Extract.HAKDiscovery))
}
failures = append(failures, validateGlobList("extract.archives", effective.Extract.Archives)...)
failures = append(failures, validateExtractMergeConfig(effective.Extract.Merge)...)
if _, err := time.ParseDuration(effective.Autogen.Cache.MaxAge); err != nil {
failures = append(failures, fmt.Errorf("autogen.cache.max_age must be a duration: %w", err))
}
@@ -669,6 +687,49 @@ func (p *Project) ValidateLayout() error {
return nil
}
func validateExtractMergeConfig(config ExtractMergeConfig) []error {
var failures []error
seenTargets := map[string]struct{}{}
for ruleIndex, rule := range config.GFFJSON {
prefix := fmt.Sprintf("extract.merge.gff_json[%d]", ruleIndex)
target := strings.TrimSpace(rule.Target)
switch {
case target == "":
failures = append(failures, fmt.Errorf("%s.target must not be empty", prefix))
case filepath.IsAbs(filepath.FromSlash(target)):
failures = append(failures, fmt.Errorf("%s.target %q must be relative to paths.source", prefix, rule.Target))
default:
clean := filepath.Clean(filepath.FromSlash(target))
if clean == "." || clean == ".." || strings.HasPrefix(clean, ".."+string(filepath.Separator)) {
failures = append(failures, fmt.Errorf("%s.target %q must not escape paths.source", prefix, rule.Target))
}
normalized := filepath.ToSlash(clean)
if _, exists := seenTargets[normalized]; exists {
failures = append(failures, fmt.Errorf("extract.merge.gff_json target %q is configured more than once", normalized))
}
seenTargets[normalized] = struct{}{}
}
for listIndex, listRule := range rule.MergeLists {
listPrefix := fmt.Sprintf("%s.merge_lists[%d]", prefix, listIndex)
if strings.TrimSpace(listRule.Field) == "" {
failures = append(failures, fmt.Errorf("%s.field must not be empty", listPrefix))
}
if strings.TrimSpace(listRule.KeyField) == "" {
failures = append(failures, fmt.Errorf("%s.key_field must not be empty", listPrefix))
}
switch listRule.Strategy {
case "preserve_existing_order_append_new":
case "":
failures = append(failures, fmt.Errorf("%s.strategy must not be empty", listPrefix))
default:
failures = append(failures, fmt.Errorf("%s.strategy %q is not supported", listPrefix, listRule.Strategy))
}
}
}
return failures
}
func (p *Project) Scan() error {
var sourceFiles []string
var sourceExts []string
@@ -1100,6 +1161,19 @@ func normalizeConfig(cfg *Config) {
}
cfg.Extract.IgnoreExtensions = normalizeStringSlice(cfg.Extract.IgnoreExtensions)
cfg.Extract.Archives = normalizeStringSlice(cfg.Extract.Archives)
for i := range cfg.Extract.Merge.GFFJSON {
target := strings.TrimSpace(cfg.Extract.Merge.GFFJSON[i].Target)
if target != "" {
target = filepath.ToSlash(filepath.Clean(filepath.FromSlash(target)))
}
cfg.Extract.Merge.GFFJSON[i].Target = target
cfg.Extract.Merge.GFFJSON[i].PreserveFields = normalizeStringSlice(cfg.Extract.Merge.GFFJSON[i].PreserveFields)
for j := range cfg.Extract.Merge.GFFJSON[i].MergeLists {
cfg.Extract.Merge.GFFJSON[i].MergeLists[j].Field = strings.TrimSpace(cfg.Extract.Merge.GFFJSON[i].MergeLists[j].Field)
cfg.Extract.Merge.GFFJSON[i].MergeLists[j].KeyField = strings.TrimSpace(cfg.Extract.Merge.GFFJSON[i].MergeLists[j].KeyField)
cfg.Extract.Merge.GFFJSON[i].MergeLists[j].Strategy = strings.TrimSpace(cfg.Extract.Merge.GFFJSON[i].MergeLists[j].Strategy)
}
}
for i := range cfg.Autogen.Producers {
cfg.Autogen.Producers[i].Include = normalizeStringSlice(cfg.Autogen.Producers[i].Include)
}