While sorting on data sources with missing fields, a panic can occur in
pairList.Less if `Interface()` is called on a invalid `reflect.Value`.
This commit detects an invalid Value and replacing it with a zero value
for the comparison.
This also includes a refactor of the hugofs package and its usage.
The motivation for that is:
The Afero filesystems are brilliant. Hugo's way of adding a dozen of global variables for the different filesystems was a mistake. In readFile (and also in some other places in Hugo today) we need a way to restrict the access inside the working dir. We could use ioutil.ReadFile and implement the path checking, checking the base path and the dots ("..") etc. But it is obviously better to use an Afero BasePathFs combined witha ReadOnlyFs. We could create a use-once-filesystem and handle the initialization ourselves, but since this is also useful to others and the initialization depends on some other global state (which would mean to create a new file system on every invocation), we might as well do it properly and encapsulate the predefined set of filesystems. This change also leads the way, if needed, to encapsulate the file systems in a struct, making it possible to have several file system sets in action at once (parallel multilanguage site building? With Moore's law and all...)
Fixes#1551
To strip away any HTML. May be useful for the .Title in head etc.
People may shoot themself in the foot with this, maybe ...
The replacement function is pretty fast.
The `intersect` function uses `in` to avoid adding duplicates to the
resulting set. We were passing `reflect.Value` items when we should
have been using `Value.Interface()` to send the actual data structure.
This fixes that.
See #1952
This uses the Emoji map from https://github.com/kyokomi/emoji -- but with a custom replacement implementation.
The built-in are fine for most use cases, but in Hugo we do care about pure speed.
The benchmarks below are skewed in Hugo's direction as the source and result is a byte slice,
Kyokomi's implementation works best with strings.
Curious: The easy-to-use `strings.Replacer` is also plenty fast.
```
BenchmarkEmojiKyokomiFprint-4 20000 86038 ns/op 33960 B/op 117 allocs/op
BenchmarkEmojiKyokomiSprint-4 20000 83252 ns/op 38232 B/op 122 allocs/op
BenchmarkEmojiStringsReplacer-4 100000 21092 ns/op 17248 B/op 25 allocs/op
BenchmarkHugoEmoji-4 500000 5728 ns/op 624 B/op 13 allocs/op
```
Fixes#1891
This commit adds a custom index template function that deviates from the stdlib
simply by not returning an "index out of range" error if an array, slice or
string index is out of range. Instead, we just return nil values. This should
help make the new default function more useful for Hugo users.
Fixes#1949
This commit fixes a few things:
1. `given` is now a variadic parameter so that piping works properly
2. add separate template tests to make sure piping works
3. support time values
4. add more tests of the dfault function
The common is the `where` func and this:
```
panic: reflect: call of reflect.Value.Type on zero Value [recovered]
panic: reflect: call of reflect.Value.Type on zero Value
```
There is no good reason to export all the template funcs:
* They're not used outside the templates.
* If usable in other packages, they should be moved (to helpers?)
* They create too broad an interface;
users of the tpl package don't see the forest for all the trees.
Add humanize (inflect.Humanize) to the template funcMap. Documentation and
tests are included.
Various code cleanups of the template funcs:
- Break pluralize and singularize out into stand-alone funcs.
- Sort the list of funcMap entries.
- Add some minimal godoc comments to all public funcs.
- Fix some issues found by golint and grind.
This fixes a exported field check condition in a way described at Go
issue https://golang.org/issue/12367
According to the issue comments, this fix should be safe under Go 1.6.
`where` template function's internal condition check function doesn't
check boolean values and always returns `false` silently.
This adds missing boolean value comparison to the function.
`where Values ".Param.key" true` like clause can be used.
Only "=", "==", "eq", "!=", "<>", "ne" operators are allowed to be used
with a boolean value. If an other operator is passed with it, the
condition check function returns `false` like before.
'sort' template function used to accept only each element's struct field
name, method name and map key name as its second argument. This extends
it to accept a field/method/key chaining key string like
'Params.foo.bar' as the argument. It evaluates sub elements of each
array or map elements and sorts by them.
Typical use case would be sorting pages by user defined front matter
value. For example, sorting pages by 'Params.foo.bar' is possible by
writing the following template code
{{ range sort .Data.Pages "Params.foo.bar" }}
{{ .Content }}
{{ end }}
It ignores all leading and trailing dots so "Params.foo.bar" can be
written in ".Params.foo.bar"
This also fixes the issue that 'sort' cannot evaluate a pointer value.
Fix#1330
sort template function returns `[]interface{}` type slice value
regardless of its original element type.
This fixes it to keep the original element type. For example, if it
sorts `map[string]int` type value, it returns `[]int` slice value
instead of `[]interface{}` slice value.
