hugo/common/paths/pathparser.go
Bjørn Erik Pedersen 7285e74090
all: Rework page store, add a dynacache, improve partial rebuilds, and some general spring cleaning
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
2024-01-27 16:28:14 +01:00

495 lines
11 KiB
Go

// 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.
// You may obtain a copy of the License at
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package paths
import (
"path"
"path/filepath"
"runtime"
"strings"
"github.com/gohugoio/hugo/common/types"
"github.com/gohugoio/hugo/hugofs/files"
)
var defaultPathParser PathParser
// PathParser parses a path into a Path.
type PathParser struct {
// Maps the language code to its index in the languages/sites slice.
LanguageIndex map[string]int
}
// Parse parses component c with path s into Path using the default path parser.
func Parse(c, s string) *Path {
return defaultPathParser.Parse(c, s)
}
// NormalizePathString returns a normalized path string using the very basic Hugo rules.
func NormalizePathStringBasic(s string) string {
// All lower case.
s = strings.ToLower(s)
// Replace spaces with hyphens.
s = strings.ReplaceAll(s, " ", "-")
return s
}
// Parse parses component c with path s into Path using Hugo's content path rules.
func (parser PathParser) Parse(c, s string) *Path {
p, err := parser.parse(c, s)
if err != nil {
panic(err)
}
return p
}
func (pp *PathParser) parse(component, s string) (*Path, error) {
ss := NormalizePathStringBasic(s)
p, err := pp.doParse(component, ss)
if err != nil {
return nil, err
}
if s != ss {
var err error
// Preserve the original case for titles etc.
p.unnormalized, err = pp.doParse(component, s)
if err != nil {
return nil, err
}
} else {
p.unnormalized = p
}
return p, nil
}
func (pp *PathParser) doParse(component, s string) (*Path, error) {
p := &Path{
component: component,
posContainerLow: -1,
posContainerHigh: -1,
posSectionHigh: -1,
posIdentifierLanguage: -1,
}
hasLang := pp.LanguageIndex != nil
hasLang = hasLang && (component == files.ComponentFolderContent || component == files.ComponentFolderLayouts)
if runtime.GOOS == "windows" {
s = path.Clean(filepath.ToSlash(s))
if s == "." {
s = ""
}
}
if s == "" {
s = "/"
}
// Leading slash, no trailing slash.
if !strings.HasPrefix(s, "/") {
s = "/" + s
}
if s != "/" && s[len(s)-1] == '/' {
s = s[:len(s)-1]
}
p.s = s
slashCount := 0
for i := len(s) - 1; i >= 0; i-- {
c := s[i]
switch c {
case '.':
if p.posContainerHigh == -1 {
var high int
if len(p.identifiers) > 0 {
high = p.identifiers[len(p.identifiers)-1].Low - 1
} else {
high = len(p.s)
}
id := types.LowHigh{Low: i + 1, High: high}
if len(p.identifiers) == 0 {
p.identifiers = append(p.identifiers, id)
} else if len(p.identifiers) == 1 {
// Check for a valid language.
s := p.s[id.Low:id.High]
if hasLang {
if _, found := pp.LanguageIndex[s]; found {
p.posIdentifierLanguage = 1
p.identifiers = append(p.identifiers, id)
}
}
}
}
case '/':
slashCount++
if p.posContainerHigh == -1 {
p.posContainerHigh = i + 1
} else if p.posContainerLow == -1 {
p.posContainerLow = i + 1
}
if i > 0 {
p.posSectionHigh = i
}
}
}
isContentComponent := p.component == files.ComponentFolderContent || p.component == files.ComponentFolderArchetypes
isContent := isContentComponent && files.IsContentExt(p.Ext())
if isContent {
id := p.identifiers[len(p.identifiers)-1]
b := p.s[p.posContainerHigh : id.Low-1]
switch b {
case "index":
p.bundleType = PathTypeLeaf
case "_index":
p.bundleType = PathTypeBranch
default:
p.bundleType = PathTypeContentSingle
}
if slashCount == 2 && p.IsLeafBundle() {
p.posSectionHigh = 0
}
}
return p, nil
}
func ModifyPathBundleTypeResource(p *Path) {
if p.IsContent() {
p.bundleType = PathTypeContentResource
} else {
p.bundleType = PathTypeFile
}
}
type PathType int
const (
// A generic resource, e.g. a JSON file.
PathTypeFile PathType = iota
// All below are content files.
// A resource of a content type with front matter.
PathTypeContentResource
// E.g. /blog/my-post.md
PathTypeContentSingle
// All bewlow are bundled content files.
// Leaf bundles, e.g. /blog/my-post/index.md
PathTypeLeaf
// Branch bundles, e.g. /blog/_index.md
PathTypeBranch
)
type Path struct {
s string
posContainerLow int
posContainerHigh int
posSectionHigh int
component string
bundleType PathType
identifiers []types.LowHigh
posIdentifierLanguage int
trimLeadingSlash bool
unnormalized *Path
}
// TrimLeadingSlash returns a copy of the Path with the leading slash removed.
func (p Path) TrimLeadingSlash() *Path {
p.trimLeadingSlash = true
return &p
}
func (p *Path) norm(s string) string {
if p.trimLeadingSlash {
s = strings.TrimPrefix(s, "/")
}
return s
}
// IdentifierBase satifies identity.Identity.
func (p *Path) IdentifierBase() string {
return p.Base()[1:]
}
// Component returns the component for this path (e.g. "content").
func (p *Path) Component() string {
return p.component
}
// Container returns the base name of the container directory for this path.
func (p *Path) Container() string {
if p.posContainerLow == -1 {
return ""
}
return p.norm(p.s[p.posContainerLow : p.posContainerHigh-1])
}
// ContainerDir returns the container directory for this path.
// For content bundles this will be the parent directory.
func (p *Path) ContainerDir() string {
if p.posContainerLow == -1 || !p.IsBundle() {
return p.Dir()
}
return p.norm(p.s[:p.posContainerLow-1])
}
// Section returns the first path element (section).
func (p *Path) Section() string {
if p.posSectionHigh <= 0 {
return ""
}
return p.norm(p.s[1:p.posSectionHigh])
}
// IsContent returns true if the path is a content file (e.g. mypost.md).
// Note that this will also return true for content files in a bundle.
func (p *Path) IsContent() bool {
return p.BundleType() >= PathTypeContentResource
}
// isContentPage returns true if the path is a content file (e.g. mypost.md),
// but nof if inside a leaf bundle.
func (p *Path) isContentPage() bool {
return p.BundleType() >= PathTypeContentSingle
}
// Name returns the last element of path.
func (p *Path) Name() string {
if p.posContainerHigh > 0 {
return p.s[p.posContainerHigh:]
}
return p.s
}
// Name returns the last element of path withhout any extension.
func (p *Path) NameNoExt() string {
if i := p.identifierIndex(0); i != -1 {
return p.s[p.posContainerHigh : p.identifiers[i].Low-1]
}
return p.s[p.posContainerHigh:]
}
// Name returns the last element of path withhout any language identifier.
func (p *Path) NameNoLang() string {
i := p.identifierIndex(p.posIdentifierLanguage)
if i == -1 {
return p.Name()
}
return p.s[p.posContainerHigh:p.identifiers[i].Low-1] + p.s[p.identifiers[i].High:]
}
// BaseNameNoIdentifier returns the logcical base name for a resource without any idenifier (e.g. no extension).
// For bundles this will be the containing directory's name, e.g. "blog".
func (p *Path) BaseNameNoIdentifier() string {
if p.IsBundle() {
return p.Container()
}
return p.NameNoIdentifier()
}
// NameNoIdentifier returns the last element of path withhout any identifier (e.g. no extension).
func (p *Path) NameNoIdentifier() string {
if len(p.identifiers) > 0 {
return p.s[p.posContainerHigh : p.identifiers[len(p.identifiers)-1].Low-1]
}
return p.s[p.posContainerHigh:]
}
// Dir returns all but the last element of path, typically the path's directory.
func (p *Path) Dir() (d string) {
if p.posContainerHigh > 0 {
d = p.s[:p.posContainerHigh-1]
}
if d == "" {
d = "/"
}
d = p.norm(d)
return
}
// Path returns the full path.
func (p *Path) Path() (d string) {
return p.norm(p.s)
}
// Unmormalized returns the Path with the original case preserved.
func (p *Path) Unmormalized() *Path {
return p.unnormalized
}
// PathNoLang returns the Path but with any language identifier removed.
func (p *Path) PathNoLang() string {
return p.base(true, false)
}
// PathNoIdentifier returns the Path but with any identifier (ext, lang) removed.
func (p *Path) PathNoIdentifier() string {
return p.base(false, false)
}
// PathRel returns the path relativeto the given owner.
func (p *Path) PathRel(owner *Path) string {
ob := owner.Base()
if !strings.HasSuffix(ob, "/") {
ob += "/"
}
return strings.TrimPrefix(p.Path(), ob)
}
// BaseRel returns the base path relative to the given owner.
func (p *Path) BaseRel(owner *Path) string {
ob := owner.Base()
if ob == "/" {
ob = ""
}
return p.Base()[len(ob)+1:]
}
// For content files, Base returns the path without any identifiers (extension, language code etc.).
// Any 'index' as the last path element is ignored.
//
// For other files (Resources), any extension is kept.
func (p *Path) Base() string {
return p.base(!p.isContentPage(), p.IsBundle())
}
// BaseNoLeadingSlash returns the base path without the leading slash.
func (p *Path) BaseNoLeadingSlash() string {
return p.Base()[1:]
}
func (p *Path) base(preserveExt, isBundle bool) string {
if len(p.identifiers) == 0 {
return p.norm(p.s)
}
if preserveExt && len(p.identifiers) == 1 {
// Preserve extension.
return p.norm(p.s)
}
id := p.identifiers[len(p.identifiers)-1]
high := id.Low - 1
if isBundle {
high = p.posContainerHigh - 1
}
if high == 0 {
high++
}
if !preserveExt {
return p.norm(p.s[:high])
}
// For txt files etc. we want to preserve the extension.
id = p.identifiers[0]
return p.norm(p.s[:high] + p.s[id.Low-1:id.High])
}
func (p *Path) Ext() string {
return p.identifierAsString(0)
}
func (p *Path) Lang() string {
return p.identifierAsString(1)
}
func (p *Path) Identifier(i int) string {
return p.identifierAsString(i)
}
func (p *Path) Identifiers() []string {
ids := make([]string, len(p.identifiers))
for i, id := range p.identifiers {
ids[i] = p.s[id.Low:id.High]
}
return ids
}
func (p *Path) IsHTML() bool {
return files.IsHTML(p.Ext())
}
func (p *Path) BundleType() PathType {
return p.bundleType
}
func (p *Path) IsBundle() bool {
return p.bundleType >= PathTypeLeaf
}
func (p *Path) IsBranchBundle() bool {
return p.bundleType == PathTypeBranch
}
func (p *Path) IsLeafBundle() bool {
return p.bundleType == PathTypeLeaf
}
func (p *Path) identifierAsString(i int) string {
i = p.identifierIndex(i)
if i == -1 {
return ""
}
id := p.identifiers[i]
return p.s[id.Low:id.High]
}
func (p *Path) identifierIndex(i int) int {
if i < 0 || i >= len(p.identifiers) {
return -1
}
return i
}
// HasExt returns true if the Unix styled path has an extension.
func HasExt(p string) bool {
for i := len(p) - 1; i >= 0; i-- {
if p[i] == '.' {
return true
}
if p[i] == '/' {
return false
}
}
return false
}