forgejo/modules/templates/htmlrenderer.go

288 lines
8.7 KiB
Go
Raw Normal View History

// Copyright 2022 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package templates
import (
"bufio"
"bytes"
"context"
"errors"
"fmt"
"io"
"net/http"
"path/filepath"
"regexp"
"strconv"
"strings"
"sync"
"sync/atomic"
texttemplate "text/template"
"code.gitea.io/gitea/modules/assetfs"
"code.gitea.io/gitea/modules/graceful"
"code.gitea.io/gitea/modules/log"
"code.gitea.io/gitea/modules/setting"
Make HTML template functions support context (#24056) # Background Golang template is not friendly for large projects, and Golang template team is quite slow, related: * `https://github.com/golang/go/issues/54450` Without upstream support, we can also have our solution to make HTML template functions support context. It helps a lot, the above Golang template issue `#54450` explains a lot: 1. It makes `{{Locale.Tr}}` could be used in any template, without passing unclear `(dict "root" . )` anymore. 2. More and more functions need `context`, like `avatar`, etc, we do not need to do `(dict "Context" $.Context)` anymore. 3. Many request-related functions could be shared by parent&children templates, like "user setting" / "system setting" See the test `TestScopedTemplateSetFuncMap`, one template set, two `Execute` calls with different `CtxFunc`. # The Solution Instead of waiting for upstream, this PR re-uses the escaped HTML template trees, use `AddParseTree` to add related templates/trees to a new template instance, then the new template instance can have its own FuncMap , the function calls in the template trees will always use the new template's FuncMap. `template.New` / `template.AddParseTree` / `adding-FuncMap` are all quite fast, so the performance is not affected. The details: 1. Make a new `html/template/Template` for `all` templates 2. Add template code to the `all` template 3. Freeze the `all` template, reset its exec func map, it shouldn't execute any template. 4. When a router wants to render a template by its `name` 1. Find the `name` in `all` 2. Find all its related sub templates 3. Escape all related templates (just like what the html template package does) 4. Add the escaped parse-trees of related templates into a new (scoped) `text/template/Template` 5. Add context-related func map into the new (scoped) text template 6. Execute the new (scoped) text template 7. To improve performance, the escaped templates are cached to `template sets` # FAQ ## There is a `unsafe` call, is this PR unsafe? This PR is safe. Golang has strict language definition, it's safe to do so: https://pkg.go.dev/unsafe#Pointer (1) Conversion of a *T1 to Pointer to *T2 ## What if Golang template supports such feature in the future? The public structs/interfaces/functions introduced by this PR is quite simple, the code of `HTMLRender` is not changed too much. It's very easy to switch to the official mechanism if there would be one. ## Does this PR change the template execution behavior? No, see the tests (welcome to design more tests if it's necessary) --------- Co-authored-by: silverwind <me@silverwind.io> Co-authored-by: Jason Song <i@wolfogre.com> Co-authored-by: Giteabot <teabot@gitea.io>
2023-04-20 10:08:58 +02:00
"code.gitea.io/gitea/modules/templates/scopedtmpl"
"code.gitea.io/gitea/modules/util"
)
Make HTML template functions support context (#24056) # Background Golang template is not friendly for large projects, and Golang template team is quite slow, related: * `https://github.com/golang/go/issues/54450` Without upstream support, we can also have our solution to make HTML template functions support context. It helps a lot, the above Golang template issue `#54450` explains a lot: 1. It makes `{{Locale.Tr}}` could be used in any template, without passing unclear `(dict "root" . )` anymore. 2. More and more functions need `context`, like `avatar`, etc, we do not need to do `(dict "Context" $.Context)` anymore. 3. Many request-related functions could be shared by parent&children templates, like "user setting" / "system setting" See the test `TestScopedTemplateSetFuncMap`, one template set, two `Execute` calls with different `CtxFunc`. # The Solution Instead of waiting for upstream, this PR re-uses the escaped HTML template trees, use `AddParseTree` to add related templates/trees to a new template instance, then the new template instance can have its own FuncMap , the function calls in the template trees will always use the new template's FuncMap. `template.New` / `template.AddParseTree` / `adding-FuncMap` are all quite fast, so the performance is not affected. The details: 1. Make a new `html/template/Template` for `all` templates 2. Add template code to the `all` template 3. Freeze the `all` template, reset its exec func map, it shouldn't execute any template. 4. When a router wants to render a template by its `name` 1. Find the `name` in `all` 2. Find all its related sub templates 3. Escape all related templates (just like what the html template package does) 4. Add the escaped parse-trees of related templates into a new (scoped) `text/template/Template` 5. Add context-related func map into the new (scoped) text template 6. Execute the new (scoped) text template 7. To improve performance, the escaped templates are cached to `template sets` # FAQ ## There is a `unsafe` call, is this PR unsafe? This PR is safe. Golang has strict language definition, it's safe to do so: https://pkg.go.dev/unsafe#Pointer (1) Conversion of a *T1 to Pointer to *T2 ## What if Golang template supports such feature in the future? The public structs/interfaces/functions introduced by this PR is quite simple, the code of `HTMLRender` is not changed too much. It's very easy to switch to the official mechanism if there would be one. ## Does this PR change the template execution behavior? No, see the tests (welcome to design more tests if it's necessary) --------- Co-authored-by: silverwind <me@silverwind.io> Co-authored-by: Jason Song <i@wolfogre.com> Co-authored-by: Giteabot <teabot@gitea.io>
2023-04-20 10:08:58 +02:00
type TemplateExecutor scopedtmpl.TemplateExecutor
type HTMLRender struct {
Make HTML template functions support context (#24056) # Background Golang template is not friendly for large projects, and Golang template team is quite slow, related: * `https://github.com/golang/go/issues/54450` Without upstream support, we can also have our solution to make HTML template functions support context. It helps a lot, the above Golang template issue `#54450` explains a lot: 1. It makes `{{Locale.Tr}}` could be used in any template, without passing unclear `(dict "root" . )` anymore. 2. More and more functions need `context`, like `avatar`, etc, we do not need to do `(dict "Context" $.Context)` anymore. 3. Many request-related functions could be shared by parent&children templates, like "user setting" / "system setting" See the test `TestScopedTemplateSetFuncMap`, one template set, two `Execute` calls with different `CtxFunc`. # The Solution Instead of waiting for upstream, this PR re-uses the escaped HTML template trees, use `AddParseTree` to add related templates/trees to a new template instance, then the new template instance can have its own FuncMap , the function calls in the template trees will always use the new template's FuncMap. `template.New` / `template.AddParseTree` / `adding-FuncMap` are all quite fast, so the performance is not affected. The details: 1. Make a new `html/template/Template` for `all` templates 2. Add template code to the `all` template 3. Freeze the `all` template, reset its exec func map, it shouldn't execute any template. 4. When a router wants to render a template by its `name` 1. Find the `name` in `all` 2. Find all its related sub templates 3. Escape all related templates (just like what the html template package does) 4. Add the escaped parse-trees of related templates into a new (scoped) `text/template/Template` 5. Add context-related func map into the new (scoped) text template 6. Execute the new (scoped) text template 7. To improve performance, the escaped templates are cached to `template sets` # FAQ ## There is a `unsafe` call, is this PR unsafe? This PR is safe. Golang has strict language definition, it's safe to do so: https://pkg.go.dev/unsafe#Pointer (1) Conversion of a *T1 to Pointer to *T2 ## What if Golang template supports such feature in the future? The public structs/interfaces/functions introduced by this PR is quite simple, the code of `HTMLRender` is not changed too much. It's very easy to switch to the official mechanism if there would be one. ## Does this PR change the template execution behavior? No, see the tests (welcome to design more tests if it's necessary) --------- Co-authored-by: silverwind <me@silverwind.io> Co-authored-by: Jason Song <i@wolfogre.com> Co-authored-by: Giteabot <teabot@gitea.io>
2023-04-20 10:08:58 +02:00
templates atomic.Pointer[scopedtmpl.ScopedTemplate]
}
var (
htmlRender *HTMLRender
htmlRenderOnce sync.Once
)
var ErrTemplateNotInitialized = errors.New("template system is not initialized, check your log for errors")
func (h *HTMLRender) HTML(w io.Writer, status int, name string, data any, ctx context.Context) error { //nolint:revive
if respWriter, ok := w.(http.ResponseWriter); ok {
if respWriter.Header().Get("Content-Type") == "" {
respWriter.Header().Set("Content-Type", "text/html; charset=utf-8")
}
respWriter.WriteHeader(status)
}
t, err := h.TemplateLookup(name, ctx)
if err != nil {
return texttemplate.ExecError{Name: name, Err: err}
}
return t.Execute(w, data)
}
func (h *HTMLRender) TemplateLookup(name string, ctx context.Context) (TemplateExecutor, error) { //nolint:revive
tmpls := h.templates.Load()
if tmpls == nil {
return nil, ErrTemplateNotInitialized
}
m := NewFuncMap()
m["ctx"] = func() any { return ctx }
return tmpls.Executor(name, m)
}
func (h *HTMLRender) CompileTemplates() error {
assets := AssetFS()
Make HTML template functions support context (#24056) # Background Golang template is not friendly for large projects, and Golang template team is quite slow, related: * `https://github.com/golang/go/issues/54450` Without upstream support, we can also have our solution to make HTML template functions support context. It helps a lot, the above Golang template issue `#54450` explains a lot: 1. It makes `{{Locale.Tr}}` could be used in any template, without passing unclear `(dict "root" . )` anymore. 2. More and more functions need `context`, like `avatar`, etc, we do not need to do `(dict "Context" $.Context)` anymore. 3. Many request-related functions could be shared by parent&children templates, like "user setting" / "system setting" See the test `TestScopedTemplateSetFuncMap`, one template set, two `Execute` calls with different `CtxFunc`. # The Solution Instead of waiting for upstream, this PR re-uses the escaped HTML template trees, use `AddParseTree` to add related templates/trees to a new template instance, then the new template instance can have its own FuncMap , the function calls in the template trees will always use the new template's FuncMap. `template.New` / `template.AddParseTree` / `adding-FuncMap` are all quite fast, so the performance is not affected. The details: 1. Make a new `html/template/Template` for `all` templates 2. Add template code to the `all` template 3. Freeze the `all` template, reset its exec func map, it shouldn't execute any template. 4. When a router wants to render a template by its `name` 1. Find the `name` in `all` 2. Find all its related sub templates 3. Escape all related templates (just like what the html template package does) 4. Add the escaped parse-trees of related templates into a new (scoped) `text/template/Template` 5. Add context-related func map into the new (scoped) text template 6. Execute the new (scoped) text template 7. To improve performance, the escaped templates are cached to `template sets` # FAQ ## There is a `unsafe` call, is this PR unsafe? This PR is safe. Golang has strict language definition, it's safe to do so: https://pkg.go.dev/unsafe#Pointer (1) Conversion of a *T1 to Pointer to *T2 ## What if Golang template supports such feature in the future? The public structs/interfaces/functions introduced by this PR is quite simple, the code of `HTMLRender` is not changed too much. It's very easy to switch to the official mechanism if there would be one. ## Does this PR change the template execution behavior? No, see the tests (welcome to design more tests if it's necessary) --------- Co-authored-by: silverwind <me@silverwind.io> Co-authored-by: Jason Song <i@wolfogre.com> Co-authored-by: Giteabot <teabot@gitea.io>
2023-04-20 10:08:58 +02:00
extSuffix := ".tmpl"
tmpls := scopedtmpl.NewScopedTemplate()
tmpls.Funcs(NewFuncMap())
files, err := ListWebTemplateAssetNames(assets)
if err != nil {
return nil
}
for _, file := range files {
if !strings.HasSuffix(file, extSuffix) {
continue
}
name := strings.TrimSuffix(file, extSuffix)
tmpl := tmpls.New(filepath.ToSlash(name))
buf, err := assets.ReadFile(file)
if err != nil {
return err
}
if _, err = tmpl.Parse(string(buf)); err != nil {
return err
}
}
Make HTML template functions support context (#24056) # Background Golang template is not friendly for large projects, and Golang template team is quite slow, related: * `https://github.com/golang/go/issues/54450` Without upstream support, we can also have our solution to make HTML template functions support context. It helps a lot, the above Golang template issue `#54450` explains a lot: 1. It makes `{{Locale.Tr}}` could be used in any template, without passing unclear `(dict "root" . )` anymore. 2. More and more functions need `context`, like `avatar`, etc, we do not need to do `(dict "Context" $.Context)` anymore. 3. Many request-related functions could be shared by parent&children templates, like "user setting" / "system setting" See the test `TestScopedTemplateSetFuncMap`, one template set, two `Execute` calls with different `CtxFunc`. # The Solution Instead of waiting for upstream, this PR re-uses the escaped HTML template trees, use `AddParseTree` to add related templates/trees to a new template instance, then the new template instance can have its own FuncMap , the function calls in the template trees will always use the new template's FuncMap. `template.New` / `template.AddParseTree` / `adding-FuncMap` are all quite fast, so the performance is not affected. The details: 1. Make a new `html/template/Template` for `all` templates 2. Add template code to the `all` template 3. Freeze the `all` template, reset its exec func map, it shouldn't execute any template. 4. When a router wants to render a template by its `name` 1. Find the `name` in `all` 2. Find all its related sub templates 3. Escape all related templates (just like what the html template package does) 4. Add the escaped parse-trees of related templates into a new (scoped) `text/template/Template` 5. Add context-related func map into the new (scoped) text template 6. Execute the new (scoped) text template 7. To improve performance, the escaped templates are cached to `template sets` # FAQ ## There is a `unsafe` call, is this PR unsafe? This PR is safe. Golang has strict language definition, it's safe to do so: https://pkg.go.dev/unsafe#Pointer (1) Conversion of a *T1 to Pointer to *T2 ## What if Golang template supports such feature in the future? The public structs/interfaces/functions introduced by this PR is quite simple, the code of `HTMLRender` is not changed too much. It's very easy to switch to the official mechanism if there would be one. ## Does this PR change the template execution behavior? No, see the tests (welcome to design more tests if it's necessary) --------- Co-authored-by: silverwind <me@silverwind.io> Co-authored-by: Jason Song <i@wolfogre.com> Co-authored-by: Giteabot <teabot@gitea.io>
2023-04-20 10:08:58 +02:00
tmpls.Freeze()
h.templates.Store(tmpls)
return nil
}
// HTMLRenderer init once and returns the globally shared html renderer
func HTMLRenderer() *HTMLRender {
htmlRenderOnce.Do(initHTMLRenderer)
return htmlRender
}
func ReloadHTMLTemplates() error {
log.Trace("Reloading HTML templates")
if err := htmlRender.CompileTemplates(); err != nil {
log.Error("Template error: %v\n%s", err, log.Stack(2))
return err
}
return nil
}
func initHTMLRenderer() {
rendererType := "static"
if !setting.IsProd {
rendererType = "auto-reloading"
}
log.Debug("Creating %s HTML Renderer", rendererType)
htmlRender = &HTMLRender{}
if err := htmlRender.CompileTemplates(); err != nil {
p := &templateErrorPrettier{assets: AssetFS()}
wrapTmplErrMsg(p.handleFuncNotDefinedError(err))
wrapTmplErrMsg(p.handleUnexpectedOperandError(err))
wrapTmplErrMsg(p.handleExpectedEndError(err))
wrapTmplErrMsg(p.handleGenericTemplateError(err))
wrapTmplErrMsg(fmt.Sprintf("CompileTemplates error: %v", err))
}
if !setting.IsProd {
go AssetFS().WatchLocalChanges(graceful.GetManager().ShutdownContext(), func() {
_ = ReloadHTMLTemplates()
})
}
}
func wrapTmplErrMsg(msg string) {
if msg == "" {
return
}
if setting.IsProd {
// in prod mode, Forgejo must have correct templates to run
log.Fatal("Forgejo can't run with template errors: %s", msg)
}
// in dev mode, do not need to really exit, because the template errors could be fixed by developer soon and the templates get reloaded
log.Error("There are template errors but Forgejo continues to run in dev mode: %s", msg)
}
type templateErrorPrettier struct {
assets *assetfs.LayeredFS
}
var reGenericTemplateError = regexp.MustCompile(`^template: (.*):([0-9]+): (.*)`)
func (p *templateErrorPrettier) handleGenericTemplateError(err error) string {
groups := reGenericTemplateError.FindStringSubmatch(err.Error())
if len(groups) != 4 {
return ""
}
tmplName, lineStr, message := groups[1], groups[2], groups[3]
return p.makeDetailedError(message, tmplName, lineStr, -1, "")
}
var reFuncNotDefinedError = regexp.MustCompile(`^template: (.*):([0-9]+): (function "(.*)" not defined)`)
func (p *templateErrorPrettier) handleFuncNotDefinedError(err error) string {
groups := reFuncNotDefinedError.FindStringSubmatch(err.Error())
if len(groups) != 5 {
return ""
}
tmplName, lineStr, message, funcName := groups[1], groups[2], groups[3], groups[4]
funcName, _ = strconv.Unquote(`"` + funcName + `"`)
return p.makeDetailedError(message, tmplName, lineStr, -1, funcName)
}
var reUnexpectedOperandError = regexp.MustCompile(`^template: (.*):([0-9]+): (unexpected "(.*)" in operand)`)
func (p *templateErrorPrettier) handleUnexpectedOperandError(err error) string {
groups := reUnexpectedOperandError.FindStringSubmatch(err.Error())
if len(groups) != 5 {
return ""
}
tmplName, lineStr, message, unexpected := groups[1], groups[2], groups[3], groups[4]
unexpected, _ = strconv.Unquote(`"` + unexpected + `"`)
return p.makeDetailedError(message, tmplName, lineStr, -1, unexpected)
}
var reExpectedEndError = regexp.MustCompile(`^template: (.*):([0-9]+): (expected end; found (.*))`)
func (p *templateErrorPrettier) handleExpectedEndError(err error) string {
groups := reExpectedEndError.FindStringSubmatch(err.Error())
if len(groups) != 5 {
return ""
}
tmplName, lineStr, message, unexpected := groups[1], groups[2], groups[3], groups[4]
return p.makeDetailedError(message, tmplName, lineStr, -1, unexpected)
}
var (
reTemplateExecutingError = regexp.MustCompile(`^template: (.*):([1-9][0-9]*):([1-9][0-9]*): (executing .*)`)
reTemplateExecutingErrorMsg = regexp.MustCompile(`^executing "(.*)" at <(.*)>: `)
)
func (p *templateErrorPrettier) handleTemplateRenderingError(err error) string {
if groups := reTemplateExecutingError.FindStringSubmatch(err.Error()); len(groups) > 0 {
tmplName, lineStr, posStr, msgPart := groups[1], groups[2], groups[3], groups[4]
target := ""
if groups = reTemplateExecutingErrorMsg.FindStringSubmatch(msgPart); len(groups) > 0 {
target = groups[2]
}
return p.makeDetailedError(msgPart, tmplName, lineStr, posStr, target)
} else if execErr, ok := err.(texttemplate.ExecError); ok {
layerName := p.assets.GetFileLayerName(execErr.Name + ".tmpl")
return fmt.Sprintf("asset from: %s, %s", layerName, err.Error())
}
return err.Error()
}
func HandleTemplateRenderingError(err error) string {
p := &templateErrorPrettier{assets: AssetFS()}
return p.handleTemplateRenderingError(err)
}
const dashSeparator = "----------------------------------------------------------------------"
func (p *templateErrorPrettier) makeDetailedError(errMsg, tmplName string, lineNum, posNum any, target string) string {
code, layer, err := p.assets.ReadLayeredFile(tmplName + ".tmpl")
if err != nil {
return fmt.Sprintf("template error: %s, and unable to find template file %q", errMsg, tmplName)
}
line, err := util.ToInt64(lineNum)
if err != nil {
return fmt.Sprintf("template error: %s, unable to parse template %q line number %q", errMsg, tmplName, lineNum)
}
pos, err := util.ToInt64(posNum)
if err != nil {
return fmt.Sprintf("template error: %s, unable to parse template %q pos number %q", errMsg, tmplName, posNum)
}
detail := extractErrorLine(code, int(line), int(pos), target)
var msg string
if pos >= 0 {
msg = fmt.Sprintf("template error: %s:%s:%d:%d : %s", layer, tmplName, line, pos, errMsg)
} else {
msg = fmt.Sprintf("template error: %s:%s:%d : %s", layer, tmplName, line, errMsg)
}
return msg + "\n" + dashSeparator + "\n" + detail + "\n" + dashSeparator
}
func extractErrorLine(code []byte, lineNum, posNum int, target string) string {
b := bufio.NewReader(bytes.NewReader(code))
var line []byte
var err error
for i := 0; i < lineNum; i++ {
if line, err = b.ReadBytes('\n'); err != nil {
if i == lineNum-1 && errors.Is(err, io.EOF) {
err = nil
}
break
}
}
if err != nil {
return fmt.Sprintf("unable to find target line %d", lineNum)
}
line = bytes.TrimRight(line, "\r\n")
var indicatorLine []byte
targetBytes := []byte(target)
targetLen := len(targetBytes)
for i := 0; i < len(line); {
if posNum == -1 && target != "" && bytes.HasPrefix(line[i:], targetBytes) {
for j := 0; j < targetLen && i < len(line); j++ {
indicatorLine = append(indicatorLine, '^')
i++
}
} else if i == posNum {
indicatorLine = append(indicatorLine, '^')
i++
} else {
if line[i] == '\t' {
indicatorLine = append(indicatorLine, '\t')
} else {
indicatorLine = append(indicatorLine, ' ')
}
i++
}
}
// if the indicatorLine only contains spaces, trim it together
return strings.TrimRight(string(line)+"\n"+string(indicatorLine), " \t\r\n")
}