diff options
author | Bjørn Erik Pedersen <bjorn.erik.pedersen@gmail.com> | 2023-12-24 19:11:05 +0100 |
---|---|---|
committer | Bjørn Erik Pedersen <bjorn.erik.pedersen@gmail.com> | 2024-01-27 16:28:14 +0100 |
commit | 7285e74090852b5d52f25e577850fa75f4aa8573 (patch) | |
tree | 54d07cb4a7de2db5c89f2590266595f0aca6cbd6 /resources/resource.go | |
parent | 5fd1e7490305570872d3899f5edda950903c5213 (diff) | |
download | hugo-develop2024.tar.gz hugo-develop2024.zip |
all: Rework page store, add a dynacache, improve partial rebuilds, and some general spring cleaningdevelop2024
There are some breaking changes in this commit, see #11455.
Closes #11455
Closes #11549
This fixes a set of bugs (see issue list) and it is also paying some technical debt accumulated over the years. We now build with Staticcheck enabled in the CI build.
The performance should be about the same as before for regular sized Hugo sites, but it should perform and scale much better to larger data sets, as objects that uses lots of memory (e.g. rendered Markdown, big JSON files read into maps with transform.Unmarshal etc.) will now get automatically garbage collected if needed. Performance on partial rebuilds when running the server in fast render mode should be the same, but the change detection should be much more accurate.
A list of the notable new features:
* A new dependency tracker that covers (almost) all of Hugo's API and is used to do fine grained partial rebuilds when running the server.
* A new and simpler tree document store which allows fast lookups and prefix-walking in all dimensions (e.g. language) concurrently.
* You can now configure an upper memory limit allowing for much larger data sets and/or running on lower specced PCs.
We have lifted the "no resources in sub folders" restriction for branch bundles (e.g. sections).
Memory Limit
* Hugos will, by default, set aside a quarter of the total system memory, but you can set this via the OS environment variable HUGO_MEMORYLIMIT (in gigabytes). This is backed by a partitioned LRU cache used throughout Hugo. A cache that gets dynamically resized in low memory situations, allowing Go's Garbage Collector to free the memory.
New Dependency Tracker: Hugo has had a rule based coarse grained approach to server rebuilds that has worked mostly pretty well, but there have been some surprises (e.g. stale content). This is now revamped with a new dependency tracker that can quickly calculate the delta given a changed resource (e.g. a content file, template, JS file etc.). This handles transitive relations, e.g. $page -> js.Build -> JS import, or $page1.Content -> render hook -> site.GetPage -> $page2.Title, or $page1.Content -> shortcode -> partial -> site.RegularPages -> $page2.Content -> shortcode ..., and should also handle changes to aggregated values (e.g. site.Lastmod) effectively.
This covers all of Hugo's API with 2 known exceptions (a list that may not be fully exhaustive):
Changes to files loaded with template func os.ReadFile may not be handled correctly. We recommend loading resources with resources.Get
Changes to Hugo objects (e.g. Page) passed in the template context to lang.Translate may not be detected correctly. We recommend having simple i18n templates without too much data context passed in other than simple types such as strings and numbers.
Note that the cachebuster configuration (when A changes then rebuild B) works well with the above, but we recommend that you revise that configuration, as it in most situations should not be needed. One example where it is still needed is with TailwindCSS and using changes to hugo_stats.json to trigger new CSS rebuilds.
Document Store: Previously, a little simplified, we split the document store (where we store pages and resources) in a tree per language. This worked pretty well, but the structure made some operations harder than they needed to be. We have now restructured it into one Radix tree for all languages. Internally the language is considered to be a dimension of that tree, and the tree can be viewed in all dimensions concurrently. This makes some operations re. language simpler (e.g. finding translations is just a slice range), but the idea is that it should also be relatively inexpensive to add more dimensions if needed (e.g. role).
Fixes #10169
Fixes #10364
Fixes #10482
Fixes #10630
Fixes #10656
Fixes #10694
Fixes #10918
Fixes #11262
Fixes #11439
Fixes #11453
Fixes #11457
Fixes #11466
Fixes #11540
Fixes #11551
Fixes #11556
Fixes #11654
Fixes #11661
Fixes #11663
Fixes #11664
Fixes #11669
Fixes #11671
Fixes #11807
Fixes #11808
Fixes #11809
Fixes #11815
Fixes #11840
Fixes #11853
Fixes #11860
Fixes #11883
Fixes #11904
Fixes #7388
Fixes #7425
Fixes #7436
Fixes #7544
Fixes #7882
Fixes #7960
Fixes #8255
Fixes #8307
Fixes #8863
Fixes #8927
Fixes #9192
Fixes #9324
Diffstat (limited to 'resources/resource.go')
-rw-r--r-- | resources/resource.go | 626 |
1 files changed, 294 insertions, 332 deletions
diff --git a/resources/resource.go b/resources/resource.go index b7e6b65a8..e78dd12cb 100644 --- a/resources/resource.go +++ b/resources/resource.go @@ -1,4 +1,4 @@ -// Copyright 2022 The Hugo Authors. All rights reserved. +// Copyright 2024 The Hugo Authors. All rights reserved. // // Licensed under the Apache License, Version 2.0 (the "License"); // you may not use this file except in compliance with the License. @@ -15,68 +15,55 @@ package resources import ( "context" + "errors" "fmt" "io" - "os" - "path" - "path/filepath" + "mime" "strings" "sync" + "sync/atomic" + "github.com/gohugoio/hugo/identity" "github.com/gohugoio/hugo/resources/internal" "github.com/gohugoio/hugo/common/herrors" - - "github.com/gohugoio/hugo/hugofs" + "github.com/gohugoio/hugo/common/paths" "github.com/gohugoio/hugo/media" - "github.com/gohugoio/hugo/source" - - "errors" "github.com/gohugoio/hugo/common/hugio" "github.com/gohugoio/hugo/common/maps" - "github.com/gohugoio/hugo/resources/page" "github.com/gohugoio/hugo/resources/resource" - "github.com/spf13/afero" "github.com/gohugoio/hugo/helpers" ) var ( - _ resource.ContentResource = (*genericResource)(nil) - _ resource.ReadSeekCloserResource = (*genericResource)(nil) - _ resource.Resource = (*genericResource)(nil) - _ resource.Source = (*genericResource)(nil) - _ resource.Cloner = (*genericResource)(nil) - _ resource.ResourcesLanguageMerger = (*resource.Resources)(nil) - _ permalinker = (*genericResource)(nil) - _ resource.Identifier = (*genericResource)(nil) - _ fileInfo = (*genericResource)(nil) + _ resource.ContentResource = (*genericResource)(nil) + _ resource.ReadSeekCloserResource = (*genericResource)(nil) + _ resource.Resource = (*genericResource)(nil) + _ resource.Source = (*genericResource)(nil) + _ resource.Cloner = (*genericResource)(nil) + _ resource.ResourcesLanguageMerger = (*resource.Resources)(nil) + _ resource.Identifier = (*genericResource)(nil) + _ identity.IdentityGroupProvider = (*genericResource)(nil) + _ identity.DependencyManagerProvider = (*genericResource)(nil) + _ identity.Identity = (*genericResource)(nil) + _ fileInfo = (*genericResource)(nil) ) type ResourceSourceDescriptor struct { - // TargetPaths is a callback to fetch paths's relative to its owner. - TargetPaths func() page.TargetPaths - - // Need one of these to load the resource content. - SourceFile source.File - OpenReadSeekCloser resource.OpenReadSeekCloser - - FileInfo os.FileInfo - - // If OpenReadSeekerCloser is not set, we use this to open the file. - SourceFilename string + // The source content. + OpenReadSeekCloser hugio.OpenReadSeekCloser - Fs afero.Fs + // The canonical source path. + Path *paths.Path - Data map[string]any - - // Set when its known up front, else it's resolved from the target filename. - MediaType media.Type + // The name of the resource. + Name string - // The relative target filename without any language code. - RelTargetFilename string + // The name of the resource as it was read from the source. + NameOriginal string // Any base paths prepended to the target path. This will also typically be the // language code, but setting it here means that it should not have any effect on @@ -85,15 +72,109 @@ type ResourceSourceDescriptor struct { // multiple targets. TargetBasePaths []string + TargetPath string + BasePathRelPermalink string + BasePathTargetPath string + + // The Data to associate with this resource. + Data map[string]any + // Delay publishing until either Permalink or RelPermalink is called. Maybe never. LazyPublish bool + + // Set when its known up front, else it's resolved from the target filename. + MediaType media.Type + + // Used to track depenencies (e.g. imports). May be nil if that's of no concern. + DependencyManager identity.Manager + + // A shared identity for this resource and all its clones. + // If this is not set, an Identity is created. + GroupIdentity identity.Identity } -func (r ResourceSourceDescriptor) Filename() string { - if r.SourceFile != nil { - return r.SourceFile.Filename() +func (fd *ResourceSourceDescriptor) init(r *Spec) error { + if len(fd.TargetBasePaths) == 0 { + // If not set, we publish the same resource to all hosts. + fd.TargetBasePaths = r.MultihostTargetBasePaths + } + + if fd.OpenReadSeekCloser == nil { + panic(errors.New("OpenReadSeekCloser is nil")) + } + + if fd.TargetPath == "" { + panic(errors.New("RelPath is empty")) + } + + if fd.Path == nil { + fd.Path = paths.Parse("", fd.TargetPath) + } + + if fd.TargetPath == "" { + fd.TargetPath = fd.Path.Path() + } else { + fd.TargetPath = paths.ToSlashPreserveLeading(fd.TargetPath) } - return r.SourceFilename + + fd.BasePathRelPermalink = paths.ToSlashPreserveLeading(fd.BasePathRelPermalink) + if fd.BasePathRelPermalink == "/" { + fd.BasePathRelPermalink = "" + } + fd.BasePathTargetPath = paths.ToSlashPreserveLeading(fd.BasePathTargetPath) + if fd.BasePathTargetPath == "/" { + fd.BasePathTargetPath = "" + } + + fd.TargetPath = paths.ToSlashPreserveLeading(fd.TargetPath) + for i, base := range fd.TargetBasePaths { + dir := paths.ToSlashPreserveLeading(base) + if dir == "/" { + dir = "" + } + fd.TargetBasePaths[i] = dir + } + + if fd.Name == "" { + fd.Name = fd.TargetPath + } + + if fd.NameOriginal == "" { + fd.NameOriginal = fd.Name + } + + mediaType := fd.MediaType + if mediaType.IsZero() { + ext := fd.Path.Ext() + var ( + found bool + suffixInfo media.SuffixInfo + ) + mediaType, suffixInfo, found = r.MediaTypes().GetFirstBySuffix(ext) + // TODO(bep) we need to handle these ambiguous types better, but in this context + // we most likely want the application/xml type. + if suffixInfo.Suffix == "xml" && mediaType.SubType == "rss" { + mediaType, found = r.MediaTypes().GetByType("application/xml") + } + + if !found { + // A fallback. Note that mime.TypeByExtension is slow by Hugo standards, + // so we should configure media types to avoid this lookup for most + // situations. + mimeStr := mime.TypeByExtension("." + ext) + if mimeStr != "" { + mediaType, _ = media.FromStringAndExt(mimeStr, ext) + } + } + } + + fd.MediaType = mediaType + + if fd.DependencyManager == nil { + fd.DependencyManager = identity.NopManager + } + + return nil } type ResourceTransformer interface { @@ -147,23 +228,25 @@ type baseResourceResource interface { type baseResourceInternal interface { resource.Source + resource.NameOriginalProvider fileInfo - metaAssigner + mediaTypeAssigner targetPather ReadSeekCloser() (hugio.ReadSeekCloser, error) + identity.IdentityGroupProvider + identity.DependencyManagerProvider + // For internal use. cloneWithUpdates(*transformationUpdate) (baseResource, error) tryTransformedFileCache(key string, u *transformationUpdate) io.ReadCloser + getResourcePaths() internal.ResourcePaths + specProvider - getResourcePaths() *resourcePathDescriptor - getTargetFilenames() []string openPublishFileForWriting(relTargetPath string) (io.WriteCloser, error) - - relTargetPathForRel(rel string, addBaseTargetPath, isAbs, isURL bool) string } type specProvider interface { @@ -173,10 +256,10 @@ type specProvider interface { type baseResource interface { baseResourceResource baseResourceInternal + resource.Staler } -type commonResource struct { -} +type commonResource struct{} // Slice is for internal use. // for the template functions. See collections.Slice. @@ -201,60 +284,131 @@ func (commonResource) Slice(in any) (any, error) { } } -type dirFile struct { - // This is the directory component with Unix-style slashes. - dir string - // This is the file component. - file string +type fileInfo interface { + setOpenSource(hugio.OpenReadSeekCloser) + setSourceFilenameIsHash(bool) + setTargetPath(internal.ResourcePaths) + size() int64 + hashProvider +} + +type hashProvider interface { + hash() string } -func (d dirFile) path() string { - return path.Join(d.dir, d.file) +type StaleValue[V any] struct { + // The value. + Value V + + // IsStaleFunc reports whether the value is stale. + IsStaleFunc func() bool } -type fileInfo interface { - getSourceFilename() string - setSourceFilename(string) - setSourceFilenameIsHash(bool) - setSourceFs(afero.Fs) - getFileInfo() hugofs.FileMetaInfo - hash() (string, error) - size() int +func (s *StaleValue[V]) IsStale() bool { + return s.IsStaleFunc() +} + +type AtomicStaler struct { + stale uint32 +} + +func (s *AtomicStaler) MarkStale() { + atomic.StoreUint32(&s.stale, 1) +} + +func (s *AtomicStaler) IsStale() bool { + return atomic.LoadUint32(&(s.stale)) > 0 +} + +// For internal use. +type GenericResourceTestInfo struct { + Paths internal.ResourcePaths +} + +// For internal use. +func GetTestInfoForResource(r resource.Resource) GenericResourceTestInfo { + var gr *genericResource + switch v := r.(type) { + case *genericResource: + gr = v + case *resourceAdapter: + gr = v.target.(*genericResource) + default: + panic(fmt.Sprintf("unknown resource type: %T", r)) + } + return GenericResourceTestInfo{ + Paths: gr.paths, + } } // genericResource represents a generic linkable resource. type genericResource struct { - *resourcePathDescriptor - *resourceFileInfo *resourceContent - spec *Spec + sd ResourceSourceDescriptor + paths internal.ResourcePaths + + sourceFilenameIsHash bool + + h *resourceHash // A hash of the source content. Is only calculated in caching situations. + + resource.Staler title string name string params map[string]any - data map[string]any - resourceType string - mediaType media.Type + spec *Spec +} + +func (l *genericResource) IdentifierBase() string { + return l.sd.Path.IdentifierBase() +} + +func (l *genericResource) GetIdentityGroup() identity.Identity { + return l.sd.GroupIdentity +} + +func (l *genericResource) GetDependencyManager() identity.Manager { + return l.sd.DependencyManager +} + +func (l *genericResource) ReadSeekCloser() (hugio.ReadSeekCloser, error) { + return l.sd.OpenReadSeekCloser() } func (l *genericResource) Clone() resource.Resource { return l.clone() } -func (l *genericResource) cloneTo(targetPath string) resource.Resource { - c := l.clone() - - targetPath = helpers.ToSlashTrimLeading(targetPath) - dir, file := path.Split(targetPath) +func (l *genericResource) size() int64 { + l.hash() + return l.h.size +} - c.resourcePathDescriptor = &resourcePathDescriptor{ - relTargetDirFile: dirFile{dir: dir, file: file}, +func (l *genericResource) hash() string { + if err := l.h.init(l); err != nil { + panic(err) } + return l.h.value +} - return c +func (l *genericResource) setOpenSource(openSource hugio.OpenReadSeekCloser) { + l.sd.OpenReadSeekCloser = openSource +} +func (l *genericResource) setSourceFilenameIsHash(b bool) { + l.sourceFilenameIsHash = b +} + +func (l *genericResource) setTargetPath(d internal.ResourcePaths) { + l.paths = d +} + +func (l *genericResource) cloneTo(targetPath string) resource.Resource { + c := l.clone() + c.paths = c.paths.FromTargetPath(targetPath) + return c } func (l *genericResource) Content(context.Context) (any, error) { @@ -270,41 +424,50 @@ func (r *genericResource) Err() resource.ResourceError { } func (l *genericResource) Data() any { - return l.data + return l.sd.Data } func (l *genericResource) Key() string { - basePath := l.spec.Cfg.BaseURL().BasePath + basePath := l.spec.Cfg.BaseURL().BasePathNoTrailingSlash + var key string if basePath == "" { - return l.RelPermalink() + key = l.RelPermalink() + } else { + key = strings.TrimPrefix(l.RelPermalink(), basePath) } - return strings.TrimPrefix(l.RelPermalink(), basePath) + + if l.spec.Cfg.IsMultihost() { + key = l.spec.Lang() + key + } + + return key } func (l *genericResource) MediaType() media.Type { - return l.mediaType + return l.sd.MediaType } func (l *genericResource) setMediaType(mediaType media.Type) { - l.mediaType = mediaType + l.sd.MediaType = mediaType } func (l *genericResource) Name() string { return l.name } -func (l *genericResource) Params() maps.Params { - return l.params +func (l *genericResource) NameOriginal() string { + return l.sd.NameOriginal } -func (l *genericResource) Permalink() string { - return l.spec.PermalinkForBaseURL(l.relPermalinkForRel(l.relTargetDirFile.path(), true), l.spec.Cfg.BaseURL().HostURL()) +func (l *genericResource) Params() maps.Params { + return l.params } func (l *genericResource) Publish() error { var err error l.publishInit.Do(func() { - targetFilenames := l.getTargetFilenames() + targetFilenames := l.getResourcePaths().TargetFilenames() + if l.sourceFilenameIsHash { // This is a processed image. We want to avoid copying it if it hasn't changed. var changedFilenames []string @@ -340,40 +503,30 @@ func (l *genericResource) Publish() error { } func (l *genericResource) RelPermalink() string { - return l.relPermalinkFor(l.relTargetDirFile.path()) + return l.spec.PathSpec.GetBasePath(false) + paths.PathEscape(l.paths.TargetLink()) +} + +func (l *genericResource) Permalink() string { + return l.spec.Cfg.BaseURL().WithPathNoTrailingSlash + paths.PathEscape(l.paths.TargetPath()) } func (l *genericResource) ResourceType() string { - return l.resourceType + return l.MediaType().MainType } func (l *genericResource) String() string { - return fmt.Sprintf("Resource(%s: %s)", l.resourceType, l.name) + return fmt.Sprintf("Resource(%s: %s)", l.ResourceType(), l.name) } // Path is stored with Unix style slashes. func (l *genericResource) TargetPath() string { - return l.relTargetDirFile.path() + return l.paths.TargetPath() } func (l *genericResource) Title() string { return l.title } -func (l *genericResource) createBasePath(rel string, isURL bool) string { - if l.targetPathBuilder == nil { - return rel - } - tp := l.targetPathBuilder() - - if isURL { - return path.Join(tp.SubResourceBaseLink, rel) - } - - // TODO(bep) path - return path.Join(filepath.ToSlash(tp.SubResourceBaseTarget), rel) -} - func (l *genericResource) initContent() error { var err error l.contentInit.Do(func() { @@ -396,28 +549,12 @@ func (l *genericResource) initContent() error { return err } -func (l *genericResource) setName(name string) { - l.name = name -} - -func (l *genericResource) getResourcePaths() *resourcePathDescriptor { - return l.resourcePathDescriptor -} - func (l *genericResource) getSpec() *Spec { return l.spec } -func (l *genericResource) getTargetFilenames() []string { - paths := l.relTargetPaths() - for i, p := range paths { - paths[i] = filepath.Clean(p) - } - return paths -} - -func (l *genericResource) setTitle(title string) { - l.title = title +func (l *genericResource) getResourcePaths() internal.ResourcePaths { + return l.paths } func (r *genericResource) tryTransformedFileCache(key string, u *transformationUpdate) io.ReadCloser { @@ -437,12 +574,12 @@ func (r *genericResource) mergeData(in map[string]any) { if len(in) == 0 { return } - if r.data == nil { - r.data = make(map[string]any) + if r.sd.Data == nil { + r.sd.Data = make(map[string]any) } for k, v := range in { - if _, found := r.data[k]; !found { - r.data[k] = v + if _, found := r.sd.Data[k]; !found { + r.sd.Data[k] = v } } } @@ -453,142 +590,49 @@ func (rc *genericResource) cloneWithUpdates(u *transformationUpdate) (baseResour if u.content != nil { r.contentInit.Do(func() { r.content = *u.content - r.openReadSeekerCloser = func() (hugio.ReadSeekCloser, error) { + r.sd.OpenReadSeekCloser = func() (hugio.ReadSeekCloser, error) { return hugio.NewReadSeekerNoOpCloserFromString(r.content), nil } }) } - r.mediaType = u.mediaType + r.sd.MediaType = u.mediaType if u.sourceFilename != nil { - r.setSourceFilename(*u.sourceFilename) - } - - if u.sourceFs != nil { - r.setSourceFs(u.sourceFs) + if u.sourceFs == nil { + return nil, errors.New("sourceFs is nil") + } + r.setOpenSource(func() (hugio.ReadSeekCloser, error) { + return u.sourceFs.Open(*u.sourceFilename) + }) + } else if u.sourceFs != nil { + return nil, errors.New("sourceFs is set without sourceFilename") } if u.targetPath == "" { return nil, errors.New("missing targetPath") } - fpath, fname := path.Split(u.targetPath) - r.resourcePathDescriptor.relTargetDirFile = dirFile{dir: fpath, file: fname} - + r.setTargetPath(r.paths.FromTargetPath(u.targetPath)) r.mergeData(u.data) return r, nil } func (l genericResource) clone() *genericResource { - gi := *l.resourceFileInfo - rp := *l.resourcePathDescriptor - l.resourceFileInfo = &gi - l.resourcePathDescriptor = &rp l.resourceContent = &resourceContent{} return &l } func (r *genericResource) openPublishFileForWriting(relTargetPath string) (io.WriteCloser, error) { - return helpers.OpenFilesForWriting(r.spec.BaseFs.PublishFs, r.relTargetPathsFor(relTargetPath)...) -} - -func (l *genericResource) permalinkFor(target string) string { - return l.spec.PermalinkForBaseURL(l.relPermalinkForRel(target, true), l.spec.Cfg.BaseURL().HostURL()) -} - -func (l *genericResource) relPermalinkFor(target string) string { - return l.relPermalinkForRel(target, false) -} - -func (l *genericResource) relPermalinkForRel(rel string, isAbs bool) string { - return l.spec.PathSpec.URLizeFilename(l.relTargetPathForRel(rel, false, isAbs, true)) -} - -func (l *genericResource) relTargetPathForRel(rel string, addBaseTargetPath, isAbs, isURL bool) string { - if addBaseTargetPath && len(l.baseTargetPathDirs) > 1 { - panic("multiple baseTargetPathDirs") - } - var basePath string - if addBaseTargetPath && len(l.baseTargetPathDirs) > 0 { - basePath = l.baseTargetPathDirs[0] - } - - return l.relTargetPathForRelAndBasePath(rel, basePath, isAbs, isURL) -} - -func (l *genericResource) relTargetPathForRelAndBasePath(rel, basePath string, isAbs, isURL bool) string { - rel = l.createBasePath(rel, isURL) - - if basePath != "" { - rel = path.Join(basePath, rel) - } - - if l.baseOffset != "" { - rel = path.Join(l.baseOffset, rel) - } - - if isURL { - bp := l.spec.PathSpec.GetBasePath(!isAbs) - if bp != "" { - rel = path.Join(bp, rel) - } - } - - if len(rel) == 0 || rel[0] != '/' { - rel = "/" + rel - } - - return rel -} - -func (l *genericResource) relTargetPaths() []string { - return l.relTargetPathsForRel(l.TargetPath()) -} - -func (l *genericResource) relTargetPathsFor(target string) []string { - return l.relTargetPathsForRel(target) -} - -func (l *genericResource) relTargetPathsForRel(rel string) []string { - if len(l.baseTargetPathDirs) == 0 { - return []string{l.relTargetPathForRelAndBasePath(rel, "", false, false)} - } - - targetPaths := make([]string, len(l.baseTargetPathDirs)) - for i, dir := range l.baseTargetPathDirs { - targetPaths[i] = l.relTargetPathForRelAndBasePath(rel, dir, false, false) - } - return targetPaths -} - -func (l *genericResource) updateParams(params map[string]any) { - if l.params == nil { - l.params = params - return - } - - // Sets the params not already set - for k, v := range params { - if _, found := l.params[k]; !found { - l.params[k] = v - } - } + filenames := r.paths.FromTargetPath(relTargetPath).TargetFilenames() + return helpers.OpenFilesForWriting(r.spec.BaseFs.PublishFs, filenames...) } type targetPather interface { TargetPath() string } -type permalinker interface { - targetPather - permalinkFor(target string) string - relPermalinkFor(target string) string - relTargetPaths() []string - relTargetPathsFor(target string) []string -} - type resourceContent struct { content string contentInit sync.Once @@ -596,113 +640,31 @@ type resourceContent struct { publishInit sync.Once } -type resourceFileInfo struct { - // Will be set if this resource is backed by something other than a file. - openReadSeekerCloser resource.OpenReadSeekCloser - - // This may be set to tell us to look in another filesystem for this resource. - // We, by default, use the sourceFs filesystem in the spec below. - sourceFs afero.Fs - - // Absolute filename to the source, including any content folder path. - // Note that this is absolute in relation to the filesystem it is stored in. - // It can be a base path filesystem, and then this filename will not match - // the path to the file on the real filesystem. - sourceFilename string - - // For performance. This means that whenever the content changes, the filename changes. - sourceFilenameIsHash bool - - fi hugofs.FileMetaInfo - - // A hash of the source content. Is only calculated in caching situations. - h *resourceHash -} - -func (fi *resourceFileInfo) ReadSeekCloser() (hugio.ReadSeekCloser, error) { - if fi.openReadSeekerCloser != nil { - return fi.openReadSeekerCloser() - } - - f, err := fi.getSourceFs().Open(fi.getSourceFilename()) - if err != nil { - return nil, err - } - return f, nil -} - -func (fi *resourceFileInfo) getFileInfo() hugofs.FileMetaInfo { - return fi.fi -} - -func (fi *resourceFileInfo) getSourceFilename() string { - return fi.sourceFilename -} - -func (fi *resourceFileInfo) setSourceFilename(s string) { - // Make sure it's always loaded by sourceFilename. - fi.openReadSeekerCloser = nil - fi.sourceFilename = s -} - -func (fi *resourceFileInfo) setSourceFilenameIsHash(b bool) { - fi.sourceFilenameIsHash = b -} - -func (fi *resourceFileInfo) getSourceFs() afero.Fs { - return fi.sourceFs -} - -func (fi *resourceFileInfo) setSourceFs(fs afero.Fs) { - fi.sourceFs = fs +type resourceHash struct { + value string + size int64 + initOnce sync.Once } -func (fi *resourceFileInfo) hash() (string, error) { - var err error - fi.h.init.Do(func() { +func (r *resourceHash) init(l hugio.ReadSeekCloserProvider) error { + var initErr error + r.initOnce.Do(func() { var hash string - var f hugio.ReadSeekCloser - f, err = fi.ReadSeekCloser() + var size int64 + f, err := l.ReadSeekCloser() if err != nil { - err = fmt.Errorf("failed to open source file: %w", err) + initErr = fmt.Errorf("failed to open source: %w", err) return } defer f.Close() - - hash, err = helpers.MD5FromFileFast(f) + hash, size, err = helpers.MD5FromReaderFast(f) if err != nil { + initErr = fmt.Errorf("failed to calculate hash: %w", err) return } - fi.h.value = hash + r.value = hash + r.size = size }) - return fi.h.value, err -} - -func (fi *resourceFileInfo) size() int { - if fi.fi == nil { - return 0 - } - - return int(fi.fi.Size()) -} - -type resourceHash struct { - value string - init sync.Once -} - -type resourcePathDescriptor struct { - // The relative target directory and filename. - relTargetDirFile dirFile - - // Callback used to construct a target path relative to its owner. - targetPathBuilder func() page.TargetPaths - - // This will normally be the same as above, but this will only apply to publishing - // of resources. It may be multiple values when in multihost mode. - baseTargetPathDirs []string - - // baseOffset is set when the output format's path has a offset, e.g. for AMP. - baseOffset string + return initErr } |