hugo/helpers/path.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

429 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 helpers
import (
"errors"
"fmt"
"io"
"os"
"path"
"path/filepath"
"regexp"
"sort"
"strings"
"github.com/gohugoio/hugo/common/herrors"
"github.com/gohugoio/hugo/common/text"
"github.com/gohugoio/hugo/htesting"
"github.com/gohugoio/hugo/common/paths"
"github.com/gohugoio/hugo/hugofs"
"github.com/gohugoio/hugo/common/hugio"
"github.com/spf13/afero"
)
// MakePath takes a string with any characters and replace it
// so the string could be used in a path.
// It does so by creating a Unicode-sanitized string, with the spaces replaced,
// whilst preserving the original casing of the string.
// E.g. Social Media -> Social-Media
func (p *PathSpec) MakePath(s string) string {
s = paths.Sanitize(s)
if p.Cfg.RemovePathAccents() {
s = text.RemoveAccentsString(s)
}
return s
}
// MakePathsSanitized applies MakePathSanitized on every item in the slice
func (p *PathSpec) MakePathsSanitized(paths []string) {
for i, path := range paths {
paths[i] = p.MakePathSanitized(path)
}
}
// MakePathSanitized creates a Unicode-sanitized string, with the spaces replaced
func (p *PathSpec) MakePathSanitized(s string) string {
if p.Cfg.DisablePathToLower() {
return p.MakePath(s)
}
return strings.ToLower(p.MakePath(s))
}
// MakeTitle converts the path given to a suitable title, trimming whitespace
// and replacing hyphens with whitespace.
func MakeTitle(inpath string) string {
return strings.Replace(strings.TrimSpace(inpath), "-", " ", -1)
}
// MakeTitleInPath converts the path given to a suitable title, trimming whitespace
func MakePathRelative(inPath string, possibleDirectories ...string) (string, error) {
for _, currentPath := range possibleDirectories {
if strings.HasPrefix(inPath, currentPath) {
return strings.TrimPrefix(inPath, currentPath), nil
}
}
return inPath, errors.New("can't extract relative path, unknown prefix")
}
// Should be good enough for Hugo.
var isFileRe = regexp.MustCompile(`.*\..{1,6}$`)
// GetDottedRelativePath expects a relative path starting after the content directory.
// It returns a relative path with dots ("..") navigating up the path structure.
func GetDottedRelativePath(inPath string) string {
inPath = path.Clean(filepath.ToSlash(inPath))
if inPath == "." {
return "./"
}
if !isFileRe.MatchString(inPath) && !strings.HasSuffix(inPath, "/") {
inPath += "/"
}
if !strings.HasPrefix(inPath, "/") {
inPath = "/" + inPath
}
dir, _ := path.Split(inPath)
sectionCount := strings.Count(dir, "/")
if sectionCount == 0 || dir == "/" {
return "./"
}
var dottedPath string
for i := 1; i < sectionCount; i++ {
dottedPath += "../"
}
return dottedPath
}
type NamedSlice struct {
Name string
Slice []string
}
func (n NamedSlice) String() string {
if len(n.Slice) == 0 {
return n.Name
}
return fmt.Sprintf("%s%s{%s}", n.Name, FilePathSeparator, strings.Join(n.Slice, ","))
}
func ExtractAndGroupRootPaths(paths []string) []NamedSlice {
if len(paths) == 0 {
return nil
}
pathsCopy := make([]string, len(paths))
hadSlashPrefix := strings.HasPrefix(paths[0], FilePathSeparator)
for i, p := range paths {
pathsCopy[i] = strings.Trim(filepath.ToSlash(p), "/")
}
sort.Strings(pathsCopy)
pathsParts := make([][]string, len(pathsCopy))
for i, p := range pathsCopy {
pathsParts[i] = strings.Split(p, "/")
}
var groups [][]string
for i, p1 := range pathsParts {
c1 := -1
for j, p2 := range pathsParts {
if i == j {
continue
}
c2 := -1
for i, v := range p1 {
if i >= len(p2) {
break
}
if v != p2[i] {
break
}
c2 = i
}
if c1 == -1 || (c2 != -1 && c2 < c1) {
c1 = c2
}
}
if c1 != -1 {
groups = append(groups, p1[:c1+1])
} else {
groups = append(groups, p1)
}
}
groupsStr := make([]string, len(groups))
for i, g := range groups {
groupsStr[i] = strings.Join(g, "/")
}
groupsStr = UniqueStringsSorted(groupsStr)
var result []NamedSlice
for _, g := range groupsStr {
name := filepath.FromSlash(g)
if hadSlashPrefix {
name = FilePathSeparator + name
}
ns := NamedSlice{Name: name}
for _, p := range pathsCopy {
if !strings.HasPrefix(p, g) {
continue
}
p = strings.TrimPrefix(p, g)
if p != "" {
ns.Slice = append(ns.Slice, p)
}
}
ns.Slice = UniqueStrings(ExtractRootPaths(ns.Slice))
result = append(result, ns)
}
return result
}
// ExtractRootPaths extracts the root paths from the supplied list of paths.
// The resulting root path will not contain any file separators, but there
// may be duplicates.
// So "/content/section/" becomes "content"
func ExtractRootPaths(paths []string) []string {
r := make([]string, len(paths))
for i, p := range paths {
root := filepath.ToSlash(p)
sections := strings.Split(root, "/")
for _, section := range sections {
if section != "" {
root = section
break
}
}
r[i] = root
}
return r
}
// FindCWD returns the current working directory from where the Hugo
// executable is run.
func FindCWD() (string, error) {
serverFile, err := filepath.Abs(os.Args[0])
if err != nil {
return "", fmt.Errorf("can't get absolute path for executable: %v", err)
}
path := filepath.Dir(serverFile)
realFile, err := filepath.EvalSymlinks(serverFile)
if err != nil {
if _, err = os.Stat(serverFile + ".exe"); err == nil {
realFile = filepath.Clean(serverFile + ".exe")
}
}
if err == nil && realFile != serverFile {
path = filepath.Dir(realFile)
}
return path, nil
}
// Walk walks the file tree rooted at root, calling walkFn for each file or
// directory in the tree, including root.
func Walk(fs afero.Fs, root string, walker hugofs.WalkFunc) error {
if _, isOs := fs.(*afero.OsFs); isOs {
fs = hugofs.NewBaseFileDecorator(fs)
}
w := hugofs.NewWalkway(hugofs.WalkwayConfig{
Fs: fs,
Root: root,
WalkFn: walker,
})
return w.Walk()
}
// SafeWriteToDisk is the same as WriteToDisk
// but it also checks to see if file/directory already exists.
func SafeWriteToDisk(inpath string, r io.Reader, fs afero.Fs) (err error) {
return afero.SafeWriteReader(fs, inpath, r)
}
// WriteToDisk writes content to disk.
func WriteToDisk(inpath string, r io.Reader, fs afero.Fs) (err error) {
return afero.WriteReader(fs, inpath, r)
}
// OpenFilesForWriting opens all the given filenames for writing.
func OpenFilesForWriting(fs afero.Fs, filenames ...string) (io.WriteCloser, error) {
var writeClosers []io.WriteCloser
for _, filename := range filenames {
f, err := OpenFileForWriting(fs, filename)
if err != nil {
for _, wc := range writeClosers {
wc.Close()
}
return nil, err
}
writeClosers = append(writeClosers, f)
}
return hugio.NewMultiWriteCloser(writeClosers...), nil
}
// OpenFileForWriting opens or creates the given file. If the target directory
// does not exist, it gets created.
func OpenFileForWriting(fs afero.Fs, filename string) (afero.File, error) {
filename = filepath.Clean(filename)
// Create will truncate if file already exists.
// os.Create will create any new files with mode 0666 (before umask).
f, err := fs.Create(filename)
if err != nil {
if !herrors.IsNotExist(err) {
return nil, err
}
if err = fs.MkdirAll(filepath.Dir(filename), 0o777); err != nil { // before umask
return nil, err
}
f, err = fs.Create(filename)
}
return f, err
}
// GetCacheDir returns a cache dir from the given filesystem and config.
// The dir will be created if it does not exist.
func GetCacheDir(fs afero.Fs, cacheDir string) (string, error) {
cacheDir = cacheDirDefault(cacheDir)
if cacheDir != "" {
exists, err := DirExists(cacheDir, fs)
if err != nil {
return "", err
}
if !exists {
err := fs.MkdirAll(cacheDir, 0o777) // Before umask
if err != nil {
return "", fmt.Errorf("failed to create cache dir: %w", err)
}
}
return cacheDir, nil
}
const hugoCacheBase = "hugo_cache"
// Avoid filling up the home dir with Hugo cache dirs from development.
if !htesting.IsTest {
userCacheDir, err := os.UserCacheDir()
if err == nil {
cacheDir := filepath.Join(userCacheDir, hugoCacheBase)
if err := fs.Mkdir(cacheDir, 0o777); err == nil || os.IsExist(err) {
return cacheDir, nil
}
}
}
// Fall back to a cache in /tmp.
userName := os.Getenv("USER")
if userName != "" {
return GetTempDir(hugoCacheBase+"_"+userName, fs), nil
} else {
return GetTempDir(hugoCacheBase, fs), nil
}
}
func cacheDirDefault(cacheDir string) string {
// Always use the cacheDir config if set.
if len(cacheDir) > 1 {
return addTrailingFileSeparator(cacheDir)
}
// See Issue #8714.
// Turns out that Cloudflare also sets NETLIFY=true in its build environment,
// but all of these 3 should not give any false positives.
if os.Getenv("NETLIFY") == "true" && os.Getenv("PULL_REQUEST") != "" && os.Getenv("DEPLOY_PRIME_URL") != "" {
// Netlify's cache behaviour is not documented, the currently best example
// is this project:
// https://github.com/philhawksworth/content-shards/blob/master/gulpfile.js
return "/opt/build/cache/hugo_cache/"
}
// This will fall back to an hugo_cache folder in either os.UserCacheDir or the tmp dir, which should work fine for most CI
// providers. See this for a working CircleCI setup:
// https://github.com/bep/hugo-sass-test/blob/6c3960a8f4b90e8938228688bc49bdcdd6b2d99e/.circleci/config.yml
// If not, they can set the HUGO_CACHEDIR environment variable or cacheDir config key.
return ""
}
func addTrailingFileSeparator(s string) string {
if !strings.HasSuffix(s, FilePathSeparator) {
s = s + FilePathSeparator
}
return s
}
// GetTempDir returns a temporary directory with the given sub path.
func GetTempDir(subPath string, fs afero.Fs) string {
return afero.GetTempDir(fs, subPath)
}
// DirExists checks if a path exists and is a directory.
func DirExists(path string, fs afero.Fs) (bool, error) {
return afero.DirExists(fs, path)
}
// IsDir checks if a given path is a directory.
func IsDir(path string, fs afero.Fs) (bool, error) {
return afero.IsDir(fs, path)
}
// IsEmpty checks if a given path is empty, meaning it doesn't contain any regular files.
func IsEmpty(path string, fs afero.Fs) (bool, error) {
var hasFile bool
err := afero.Walk(fs, path, func(path string, info os.FileInfo, err error) error {
if info.IsDir() {
return nil
}
hasFile = true
return filepath.SkipDir
})
return !hasFile, err
}
// Exists checks if a file or directory exists.
func Exists(path string, fs afero.Fs) (bool, error) {
return afero.Exists(fs, path)
}