There are some breaking changes in this commit, see #11455.
Closes#11455Closes#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#10169Fixes#10364Fixes#10482Fixes#10630Fixes#10656Fixes#10694Fixes#10918Fixes#11262Fixes#11439Fixes#11453Fixes#11457Fixes#11466Fixes#11540Fixes#11551Fixes#11556Fixes#11654Fixes#11661Fixes#11663Fixes#11664Fixes#11669Fixes#11671Fixes#11807Fixes#11808Fixes#11809Fixes#11815Fixes#11840Fixes#11853Fixes#11860Fixes#11883Fixes#11904Fixes#7388Fixes#7425Fixes#7436Fixes#7544Fixes#7882Fixes#7960Fixes#8255Fixes#8307Fixes#8863Fixes#8927Fixes#9192Fixes#9324
Primary motivation is documentation, but it will also hopefully simplify the code.
Also,
* Lower case the default output format names; this is in line with the custom ones (map keys) and how
it's treated all the places. This avoids doing `stringds.EqualFold` everywhere.
Closes#10896Closes#10620
This commit replaces the main part of `helpers.StripHTML` with Go's implementation in its html/template package.
It's a little slower, but correctness is more important:
```bash
BenchmarkStripHTMLOld-10 680316 1764 ns/op 728 B/op 4 allocs/op
BenchmarkStripHTMLNew-10 384520 3099 ns/op 2089 B/op 10 allocs/op
```
Fixes#9199Fixes#9909Closes#9410
This ommmit contains some security hardening measures for the Hugo build runtime.
There are some rarely used features in Hugo that would be good to have disabled by default. One example would be the "external helpers".
For `asciidoctor` and some others we use Go's `os/exec` package to start a new process.
These are a predefined set of binary names, all loaded from `PATH` and with a predefined set of arguments. Still, if you don't use `asciidoctor` in your project, you might as well have it turned off.
You can configure your own in the new `security` configuration section, but the defaults are configured to create a minimal amount of site breakage. And if that do happen, you will get clear instructions in the loa about what to do.
The default configuration is listed below. Note that almost all of these options are regular expression _whitelists_ (a string or a slice); the value `none` will block all.
```toml
[security]
enableInlineShortcodes = false
[security.exec]
allow = ['^dart-sass-embedded$', '^go$', '^npx$', '^postcss$']
osEnv = ['(?i)^(PATH|PATHEXT|APPDATA|TMP|TEMP|TERM)$']
[security.funcs]
getenv = ['^HUGO_']
[security.http]
methods = ['(?i)GET|POST']
urls = ['.*']
```
The main motivation behind this is simplicity and correctnes, but the new small config library is also faster:
```
BenchmarkDefaultConfigProvider/Viper-16 252418 4546 ns/op 2720 B/op 30 allocs/op
BenchmarkDefaultConfigProvider/Custom-16 450756 2651 ns/op 1008 B/op 6 allocs/op
```
Fixes#8633Fixes#8618Fixes#8630
Updates #8591Closes#6680Closes#5192
This commmit prepares for the addition of Goldmark as the new Markdown renderer in Hugo.
This introduces a new `markup` package with some common interfaces and each implementation in its own package.
See #5963
This commit implements Hugo Modules.
This is a broad subject, but some keywords include:
* A new `module` configuration section where you can import almost anything. You can configure both your own file mounts nd the file mounts of the modules you import. This is the new recommended way of configuring what you earlier put in `configDir`, `staticDir` etc. And it also allows you to mount folders in non-Hugo-projects, e.g. the `SCSS` folder in the Bootstrap GitHub project.
* A module consists of a set of mounts to the standard 7 component types in Hugo: `static`, `content`, `layouts`, `data`, `assets`, `i18n`, and `archetypes`. Yes, Theme Components can now include content, which should be very useful, especially in bigger multilingual projects.
* Modules not in your local file cache will be downloaded automatically and even "hot replaced" while the server is running.
* Hugo Modules supports and encourages semver versioned modules, and uses the minimal version selection algorithm to resolve versions.
* A new set of CLI commands are provided to manage all of this: `hugo mod init`, `hugo mod get`, `hugo mod graph`, `hugo mod tidy`, and `hugo mod vendor`.
All of the above is backed by Go Modules.
Fixes#5973Fixes#5996Fixes#6010Fixes#5911Fixes#5940Fixes#6074Fixes#6082Fixes#6092
Add the ability to have a `summary` page variable that overrides
the auto-generated summary. Logic for obtaining summary becomes:
* if summary divider is present in content, use the text above it
* if summary variables is present in page metadata, use that
* auto-generate summary from first _x_ words of the content
Fixes#5800
The main motivation of this commit is to add a `page.Page` interface to replace the very file-oriented `hugolib.Page` struct.
This is all a preparation step for issue #5074, "pages from other data sources".
But this also fixes a set of annoying limitations, especially related to custom output formats, and shortcodes.
Most notable changes:
* The inner content of shortcodes using the `{{%` as the outer-most delimiter will now be sent to the content renderer, e.g. Blackfriday.
This means that any markdown will partake in the global ToC and footnote context etc.
* The Custom Output formats are now "fully virtualized". This removes many of the current limitations.
* The taxonomy list type now has a reference to the `Page` object.
This improves the taxonomy template `.Title` situation and make common template constructs much simpler.
See #5074Fixes#5763Fixes#5758Fixes#5090Fixes#5204Fixes#4695Fixes#5607Fixes#5707Fixes#5719Fixes#3113Fixes#5706Fixes#5767Fixes#5723Fixes#5769Fixes#5770Fixes#5771Fixes#5759Fixes#5776Fixes#5777Fixes#5778
Add a configuration option "noreferrerLinks". When set to "true" the "HTML_NOREFERRER_LINKS" flag is being passed to Blackfriday. Thereby all *absolute* links will get a "noreferrer" value for their "rel" attribute.
See #4722
Add a configuration option "nofollowLinks". When set to "true" the "HTML_NOFOLLOW_LINKS" flag is being passed to Blackfriday. Thereby all *absolute* links will get a "nofollow" value for the "rel" attribute.
Fixes#4722
This commit is not the smallest in Hugo's history.
Some hightlights include:
* Page bundles (for complete articles, keeping images and content together etc.).
* Bundled images can be processed in as many versions/sizes as you need with the three methods `Resize`, `Fill` and `Fit`.
* Processed images are cached inside `resources/_gen/images` (default) in your project.
* Symbolic links (both files and dirs) are now allowed anywhere inside /content
* A new table based build summary
* The "Total in nn ms" now reports the total including the handling of the files inside /static. So if it now reports more than you're used to, it is just **more real** and probably faster than before (see below).
A site building benchmark run compared to `v0.31.1` shows that this should be slightly faster and use less memory:
```bash
▶ ./benchSite.sh "TOML,num_langs=.*,num_root_sections=5,num_pages=(500|1000),tags_per_page=5,shortcodes,render"
benchmark old ns/op new ns/op delta
BenchmarkSiteBuilding/TOML,num_langs=1,num_root_sections=5,num_pages=500,tags_per_page=5,shortcodes,render-4 101785785 78067944 -23.30%
BenchmarkSiteBuilding/TOML,num_langs=1,num_root_sections=5,num_pages=1000,tags_per_page=5,shortcodes,render-4 185481057 149159919 -19.58%
BenchmarkSiteBuilding/TOML,num_langs=3,num_root_sections=5,num_pages=500,tags_per_page=5,shortcodes,render-4 103149918 85679409 -16.94%
BenchmarkSiteBuilding/TOML,num_langs=3,num_root_sections=5,num_pages=1000,tags_per_page=5,shortcodes,render-4 203515478 169208775 -16.86%
benchmark old allocs new allocs delta
BenchmarkSiteBuilding/TOML,num_langs=1,num_root_sections=5,num_pages=500,tags_per_page=5,shortcodes,render-4 532464 391539 -26.47%
BenchmarkSiteBuilding/TOML,num_langs=1,num_root_sections=5,num_pages=1000,tags_per_page=5,shortcodes,render-4 1056549 772702 -26.87%
BenchmarkSiteBuilding/TOML,num_langs=3,num_root_sections=5,num_pages=500,tags_per_page=5,shortcodes,render-4 555974 406630 -26.86%
BenchmarkSiteBuilding/TOML,num_langs=3,num_root_sections=5,num_pages=1000,tags_per_page=5,shortcodes,render-4 1086545 789922 -27.30%
benchmark old bytes new bytes delta
BenchmarkSiteBuilding/TOML,num_langs=1,num_root_sections=5,num_pages=500,tags_per_page=5,shortcodes,render-4 53243246 43598155 -18.12%
BenchmarkSiteBuilding/TOML,num_langs=1,num_root_sections=5,num_pages=1000,tags_per_page=5,shortcodes,render-4 105811617 86087116 -18.64%
BenchmarkSiteBuilding/TOML,num_langs=3,num_root_sections=5,num_pages=500,tags_per_page=5,shortcodes,render-4 54558852 44545097 -18.35%
BenchmarkSiteBuilding/TOML,num_langs=3,num_root_sections=5,num_pages=1000,tags_per_page=5,shortcodes,render-4 106903858 86978413 -18.64%
```
Fixes#3651Closes#3158Fixes#1014Closes#2021Fixes#1240
Updates #3757
This is in heavy use in rendering, so this makes a difference:
```bash
benchmark old ns/op new ns/op delta
BenchmarkSiteBuilding/TOML,num_langs=1,num_pages=500,tags_per_page=5,shortcodes,render-4 124551144 107743429 -13.49%
benchmark old allocs new allocs delta
BenchmarkSiteBuilding/TOML,num_langs=1,num_pages=500,tags_per_page=5,shortcodes,render-4 528684 435118 -17.70%
benchmark old bytes new bytes delta
BenchmarkSiteBuilding/TOML,num_langs=1,num_pages=500,tags_per_page=5,shortcodes,render-4 53306848 45147832 -15.31%
```
Move SummaryLength into the ContentSpec struct and refactor the
relevant summary functions to be methods of ContentSpec. The new
summaryLength struct member is configurable by the summaryLength config
value, and the default remains 70. Also updates hugolib/page to use the
refactored methods.
Resolves#3734
This avoids having to execute these expensive operations for sites not using these values.
This commit sums up a set of wordcounting and autosummary related performance improvements.
The effect of these kind of depends on what features your site use, but a benchmark from 4 Hugo sites in the wild shows promise:
```
benchmark old ns/op new ns/op delta
BenchmarkHugo-4 21293005843 20032857342 -5.92%
benchmark old allocs new allocs delta
BenchmarkHugo-4 65290922 65186032 -0.16%
benchmark old bytes new bytes delta
BenchmarkHugo-4 9771213416 9681866464 -0.91%
```
Closes#2378
It is obviously more efficient when we do not care about the actual words.
```
BenchmarkTotalWords-4 100000 18795 ns/op 0 B/op 0 allocs/op
BenchmarkTotalWordsOld-4 30000 46751 ns/op 6400 B/op 1 allocs/op
```
For people using autogenerated summaries, this is one of the hot spots in the memory department.
We don't need to split al the content into words to do proper summary truncation.
This is obviously more effective:
```
BenchmarkTestTruncateWordsToWholeSentence-4 300000 4720 ns/op 0 B/op 0 allocs/op
BenchmarkTestTruncateWordsToWholeSentenceOld-4 100000 17699 ns/op 3072 B/op 3 allocs/op
```
Started to increase coverage in helpers package, now at 74.9% of statements.
In the process, also a few minor changes have been applied to content.go.
* Content.go has undergone a formatting refactor regarding comments
* Unused function TruncateWords has been removed
* RenderingContext's "mmark" has been changed to use MmarkRender
* Content_test.go added to cover content.go's functionality
* add global `hasCJKLanguage` flag, if true, turn on auto-detecting CJKLanguage
* add `isCJKLanguage` frontmatter to force specify whether is CJKLanguage or not
* For .Summary: If isCJKLanguage is true, use the runes as basis for truncation, else keep as today.
* For WordCount: If isCJKLanguage is true, use the runes as basis for calculation, else keep as today.
* Unexport RuneCount
Fixes#1377
First step to use initialisms that golint suggests,
for example:
Line 116: func GetHtmlRenderer should be GetHTMLRenderer
as see on http://goreportcard.com/report/spf13/hugo
Thanks to @bep for the idea!
Note that command-line flags (cobra and pflag)
as well as struct fields like .BaseUrl and .Url
that are used in Go HTML templates need more work
to maintain backward-compatibility, and thus
are NOT yet dealt with in this commit.
First step in fixing #959.
go test -test.run=NONE -bench=".*" -test.benchmem=true ./helpers
Old vs new impl (string.Replace vs string.Replacer):
benchmark old ns/op new ns/op delta
BenchmarkStripHTML 10210 6572 -35.63%
benchmark old allocs new allocs delta
BenchmarkStripHTML 6 5 -16.67%
benchmark old bytes new bytes delta
BenchmarkStripHTML 1456 848 -41.76%
