grafana/pkg/services/live/pipeline/tree/tree.go

// Copyright 2013 Julien Schmidt. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be found
// at https://github.com/julienschmidt/httprouter/blob/master/LICENSE

package tree

import (
	"bytes"
	"net/url"
	"strings"
	"unicode"
	"unicode/utf8"
)

var (
	strColon = []byte(":")
	strStar  = []byte("*")
)

func min(a, b int) int {
	if a <= b {
		return a
	}
	return b
}

func longestCommonPrefix(a, b string) int {
	i := 0
	max := min(len(a), len(b))
	for i < max && a[i] == b[i] {
		i++
	}
	return i
}

// addChild will add a child Node, keeping wildcards at the end
func (n *Node) addChild(child *Node) {
	if n.wildChild && len(n.children) > 0 {
		wildcardChild := n.children[len(n.children)-1]
		n.children = append(n.children[:len(n.children)-1], child, wildcardChild)
	} else {
		n.children = append(n.children, child)
	}
}

func countParams(path string) uint16 {
	var n uint16
	s := StringToBytes(path)
	n += uint16(bytes.Count(s, strColon))
	n += uint16(bytes.Count(s, strStar))
	return n
}

type nodeType uint8

const (
	root nodeType = iota + 1
	param
	catchAll
)

type Handler interface{}

func New() *Node {
	return new(Node)
}

type Node struct {
	path      string
	indices   string
	wildChild bool
	nType     nodeType
	priority  uint32
	children  []*Node // child nodes, at most 1 :param style Node at the end of the array
	handler   Handler
	fullPath  string
}

// Increments priority of the given child and reorders if necessary
func (n *Node) incrementChildPrio(pos int) int {
	cs := n.children
	cs[pos].priority++
	prio := cs[pos].priority

	// Adjust position (move to front)
	newPos := pos
	for ; newPos > 0 && cs[newPos-1].priority < prio; newPos-- {
		// Swap Node positions
		cs[newPos-1], cs[newPos] = cs[newPos], cs[newPos-1]
	}

	// Build new index char string
	if newPos != pos {
		n.indices = n.indices[:newPos] + // Unchanged prefix, might be empty
			n.indices[pos:pos+1] + // The index char we move
			n.indices[newPos:pos] + n.indices[pos+1:] // Rest without char at 'pos'
	}

	return newPos
}

func (n *Node) AddRoute(path string, handlers Handler) {
	n.addRoute(path, handlers)
}

// addRoute adds a Node with the given handle to the path.
// Not concurrency-safe!
func (n *Node) addRoute(path string, handlers Handler) {
	fullPath := path
	n.priority++

	// Empty tree
	if len(n.path) == 0 && len(n.children) == 0 {
		n.insertChild(path, fullPath, handlers)
		n.nType = root
		return
	}

	parentFullPathIndex := 0

walk:
	for {
		// Find the longest common prefix.
		// This also implies that the common prefix contains no ':' or '*'
		// since the existing key can't contain those chars.
		i := longestCommonPrefix(path, n.path)

		// Split edge
		if i < len(n.path) {
			child := Node{
				path:      n.path[i:],
				wildChild: n.wildChild,
				indices:   n.indices,
				children:  n.children,
				handler:   n.handler,
				priority:  n.priority - 1,
				fullPath:  n.fullPath,
			}

			n.children = []*Node{&child}
			// []byte for proper unicode char conversion, see #65
			n.indices = BytesToString([]byte{n.path[i]})
			n.path = path[:i]
			n.handler = nil
			n.wildChild = false
			n.fullPath = fullPath[:parentFullPathIndex+i]
		}

		// Make new Node a child of this Node
		if i < len(path) {
			path = path[i:]
			c := path[0]

			// '/' after param
			if n.nType == param && c == '/' && len(n.children) == 1 {
				parentFullPathIndex += len(n.path)
				n = n.children[0]
				n.priority++
				continue walk
			}

			// Check if a child with the next path byte exists
			for i, max := 0, len(n.indices); i < max; i++ {
				if c == n.indices[i] {
					parentFullPathIndex += len(n.path)
					i = n.incrementChildPrio(i)
					n = n.children[i]
					continue walk
				}
			}

			// Otherwise insert it
			if c != ':' && c != '*' && n.nType != catchAll {
				// []byte for proper unicode char conversion, see #65
				n.indices += BytesToString([]byte{c})
				child := &Node{
					fullPath: fullPath,
				}
				n.addChild(child)
				n.incrementChildPrio(len(n.indices) - 1)
				n = child
			} else if n.wildChild {
				// inserting a wildcard Node, need to check if it conflicts with the existing wildcard
				n = n.children[len(n.children)-1]
				n.priority++

				// Check if the wildcard matches
				if len(path) >= len(n.path) && n.path == path[:len(n.path)] &&
					// Adding a child to a catchAll is not possible
					n.nType != catchAll &&
					// Check for longer wildcard, e.g. :name and :names
					(len(n.path) >= len(path) || path[len(n.path)] == '/') {
					continue walk
				}

				// Wildcard conflict
				pathSeg := path
				if n.nType != catchAll {
					pathSeg = strings.SplitN(pathSeg, "/", 2)[0]
				}
				prefix := fullPath[:strings.Index(fullPath, pathSeg)] + n.path
				panic("'" + pathSeg +
					"' in new path '" + fullPath +
					"' conflicts with existing wildcard '" + n.path +
					"' in existing prefix '" + prefix +
					"'")
			}

			n.insertChild(path, fullPath, handlers)
			return
		}

		// Otherwise add handle to current Node
		if n.handler != nil {
			panic("handler are already registered for path '" + fullPath + "'")
		}
		n.handler = handlers
		n.fullPath = fullPath
		return
	}
}

// Search for a wildcard segment and check the name for invalid characters.
// Returns -1 as index, if no wildcard was found.
func findWildcard(path string) (wildcard string, i int, valid bool) {
	// Find start
	for start, c := range []byte(path) {
		// A wildcard starts with ':' (param) or '*' (catch-all)
		if c != ':' && c != '*' {
			continue
		}

		// Find end and check for invalid characters
		valid = true
		for end, c := range []byte(path[start+1:]) {
			switch c {
			case '/':
				return path[start : start+1+end], start, valid
			case ':', '*':
				valid = false
			}
		}
		return path[start:], start, valid
	}
	return "", -1, false
}

func (n *Node) insertChild(path string, fullPath string, handlers Handler) {
	for {
		// Find prefix until first wildcard
		wildcard, i, valid := findWildcard(path)
		if i < 0 { // No wildcard found
			break
		}

		// The wildcard name must not contain ':' and '*'
		if !valid {
			panic("only one wildcard per path segment is allowed, has: '" +
				wildcard + "' in path '" + fullPath + "'")
		}

		// check if the wildcard has a name
		if len(wildcard) < 2 {
			panic("wildcards must be named with a non-empty name in path '" + fullPath + "'")
		}

		if wildcard[0] == ':' { // param
			if i > 0 {
				// Insert prefix before the current wildcard
				n.path = path[:i]
				path = path[i:]
			}

			child := &Node{
				nType:    param,
				path:     wildcard,
				fullPath: fullPath,
			}
			n.addChild(child)
			n.wildChild = true
			n = child
			n.priority++

			// if the path doesn't end with the wildcard, then there
			// will be another non-wildcard subpath starting with '/'
			if len(wildcard) < len(path) {
				path = path[len(wildcard):]

				child := &Node{
					priority: 1,
					fullPath: fullPath,
				}
				n.addChild(child)
				n = child
				continue
			}

			// Otherwise we're done. Insert the handle in the new leaf
			n.handler = handlers
			return
		}

		// catchAll
		if i+len(wildcard) != len(path) {
			panic("catch-all routes are only allowed at the end of the path in path '" + fullPath + "'")
		}

		if len(n.path) > 0 && n.path[len(n.path)-1] == '/' {
			panic("catch-all conflicts with existing handle for the path segment root in path '" + fullPath + "'")
		}

		// currently fixed width 1 for '/'
		i--
		if path[i] != '/' {
			panic("no / before catch-all in path '" + fullPath + "'")
		}

		n.path = path[:i]

		// First Node: catchAll Node with empty path
		child := &Node{
			wildChild: true,
			nType:     catchAll,
			fullPath:  fullPath,
		}

		n.addChild(child)
		n.indices = string('/')
		n = child
		n.priority++

		// second Node: Node holding the variable
		child = &Node{
			path:     path[i:],
			nType:    catchAll,
			handler:  handlers,
			priority: 1,
			fullPath: fullPath,
		}
		n.children = []*Node{child}

		return
	}

	// If no wildcard was found, simply insert the path and handle
	n.path = path
	n.handler = handlers
	n.fullPath = fullPath
}

// NodeValue holds return values of (*Node).getValue method
type NodeValue struct {
	Handler  Handler
	Params   *Params
	Tsr      bool
	FullPath string
}

func (n *Node) GetValue(path string, params *Params, unescape bool) (value NodeValue) {
	return n.getValue(path, params, unescape)
}

// Returns the handle registered with the given path (key). The values of
// wildcards are saved to a map.
// If no handle can be found, a TSR (trailing slash redirect) recommendation is
// made if a handle exists with an extra (without the) trailing slash for the
// given path.
// nolint:gocyclo
func (n *Node) getValue(path string, params *Params, unescape bool) (value NodeValue) {
	var (
		skippedPath string
		latestNode  = n // Caching the latest Node
	)

walk: // Outer loop for walking the tree
	for {
		prefix := n.path
		if len(path) > len(prefix) {
			if path[:len(prefix)] == prefix {
				path = path[len(prefix):]

				// Try all the non-wildcard children first by matching the indices
				idxc := path[0]
				for i, c := range []byte(n.indices) {
					if c == idxc {
						//  strings.HasPrefix(n.children[len(n.children)-1].path, ":") == n.wildChild
						if n.wildChild {
							skippedPath = prefix + path
							latestNode = &Node{
								path:      n.path,
								wildChild: n.wildChild,
								nType:     n.nType,
								priority:  n.priority,
								children:  n.children,
								handler:   n.handler,
								fullPath:  n.fullPath,
							}
						}

						n = n.children[i]
						continue walk
					}
				}
				// If the path at the end of the loop is not equal to '/' and the current Node has no child nodes
				// the current Node needs to be equal to the latest matching Node
				matched := path != "/" && !n.wildChild
				if matched {
					n = latestNode
				}

				// If there is no wildcard pattern, recommend a redirection
				if !n.wildChild {
					// Nothing found.
					// We can recommend to redirect to the same URL without a
					// trailing slash if a leaf exists for that path.
					value.Tsr = path == "/" && n.handler != nil
					return
				}

				// Handle wildcard child, which is always at the end of the array
				n = n.children[len(n.children)-1]

				switch n.nType {
				case param:
					// fix truncate the parameter
					// tree_test.go  line: 204
					if matched {
						path = prefix + path
						// The saved path is used after the prefix route is intercepted by matching
						if n.indices == "/" {
							path = skippedPath[1:]
						}
					}

					// Find param end (either '/' or path end)
					end := 0
					for end < len(path) && path[end] != '/' {
						end++
					}

					// Save param value
					if params != nil && cap(*params) > 0 {
						if value.Params == nil {
							value.Params = params
						}
						// Expand slice within preallocated capacity
						i := len(*value.Params)
						*value.Params = (*value.Params)[:i+1]
						val := path[:end]
						if unescape {
							if v, err := url.QueryUnescape(val); err == nil {
								val = v
							}
						}
						(*value.Params)[i] = Param{
							Key:   n.path[1:],
							Value: val,
						}
					}

					// we need to go deeper!
					if end < len(path) {
						if len(n.children) > 0 {
							path = path[end:]
							n = n.children[0]
							continue walk
						}

						// ... but we can't
						value.Tsr = len(path) == end+1
						return
					}

					if value.Handler = n.handler; value.Handler != nil {
						value.FullPath = n.fullPath
						return
					}
					if len(n.children) == 1 {
						// No handle found. Check if a handle for this path + a
						// trailing slash exists for TSR recommendation
						n = n.children[0]
						value.Tsr = n.path == "/" && n.handler != nil
					}
					return

				case catchAll:
					// Save param value
					if params != nil {
						if value.Params == nil {
							value.Params = params
						}
						// Expand slice within preallocated capacity
						i := len(*value.Params)
						*value.Params = (*value.Params)[:i+1]
						val := path
						if unescape {
							if v, err := url.QueryUnescape(path); err == nil {
								val = v
							}
						}
						(*value.Params)[i] = Param{
							Key:   n.path[2:],
							Value: val,
						}
					}

					value.Handler = n.handler
					value.FullPath = n.fullPath
					return

				default:
					panic("invalid Node type")
				}
			}
		}

		if path == prefix {
			// If the current path does not equal '/' and the Node does not have a registered handle and the most recently matched Node has a child Node
			// the current Node needs to be equal to the latest matching Node
			if latestNode.wildChild && n.handler == nil && path != "/" {
				n = latestNode.children[len(latestNode.children)-1]
			}
			// We should have reached the Node containing the handle.
			// Check if this Node has a handle registered.
			if value.Handler = n.handler; value.Handler != nil {
				value.FullPath = n.fullPath
				return
			}

			// If there is no handle for this route, but this route has a
			// wildcard child, there must be a handle for this path with an
			// additional trailing slash
			if path == "/" && n.wildChild && n.nType != root {
				value.Tsr = true
				return
			}

			// No handle found. Check if a handle for this path + a
			// trailing slash exists for trailing slash recommendation
			for i, c := range []byte(n.indices) {
				if c == '/' {
					n = n.children[i]
					value.Tsr = (len(n.path) == 1 && n.handler != nil) ||
						(n.nType == catchAll && n.children[0].handler != nil)
					return
				}
			}

			return
		}

		if path != "/" && len(skippedPath) > 0 && strings.HasSuffix(skippedPath, path) {
			path = skippedPath
			// Reduce the number of cycles
			n, latestNode = latestNode, n
			// skippedPath cannot execute
			// example:
			// * /:cc/cc
			// call /a/cc 	     expectations:match/200      Actual:match/200
			// call /a/dd 	     expectations:unmatch/404    Actual: panic
			// call /addr/dd/aa  expectations:unmatch/404    Actual: panic
			// skippedPath: It can only be executed if the secondary route is not found
			skippedPath = ""
			continue walk
		}

		// Nothing found. We can recommend to redirect to the same URL with an
		// extra trailing slash if a leaf exists for that path
		value.Tsr = path == "/" ||
			(len(prefix) == len(path)+1 && n.handler != nil)
		return
	}
}

// Makes a case-insensitive lookup of the given path and tries to find a handler.
// It can optionally also fix trailing slashes.
// It returns the case-corrected path and a bool indicating whether the lookup
// was successful.
func (n *Node) findCaseInsensitivePath(path string, fixTrailingSlash bool) ([]byte, bool) {
	const stackBufSize = 128

	// Use a static sized buffer on the stack in the common case.
	// If the path is too long, allocate a buffer on the heap instead.
	buf := make([]byte, 0, stackBufSize)
	if length := len(path) + 1; length > stackBufSize {
		buf = make([]byte, 0, length)
	}

	ciPath := n.findCaseInsensitivePathRec(
		path,
		buf,       // Preallocate enough memory for new path
		[4]byte{}, // Empty rune buffer
		fixTrailingSlash,
	)

	return ciPath, ciPath != nil
}

// Shift bytes in array by n bytes left
func shiftNRuneBytes(rb [4]byte, n int) [4]byte {
	switch n {
	case 0:
		return rb
	case 1:
		return [4]byte{rb[1], rb[2], rb[3], 0}
	case 2:
		return [4]byte{rb[2], rb[3]}
	case 3:
		return [4]byte{rb[3]}
	default:
		return [4]byte{}
	}
}

// Recursive case-insensitive lookup function used by n.findCaseInsensitivePath
// nolint:gocyclo
func (n *Node) findCaseInsensitivePathRec(path string, ciPath []byte, rb [4]byte, fixTrailingSlash bool) []byte {
	npLen := len(n.path)

walk: // Outer loop for walking the tree
	for len(path) >= npLen && (npLen == 0 || strings.EqualFold(path[1:npLen], n.path[1:])) {
		// Add common prefix to result
		oldPath := path
		path = path[npLen:]
		ciPath = append(ciPath, n.path...)

		if len(path) == 0 {
			// We should have reached the Node containing the handle.
			// Check if this Node has a handle registered.
			if n.handler != nil {
				return ciPath
			}

			// No handle found.
			// Try to fix the path by adding a trailing slash
			if fixTrailingSlash {
				for i, c := range []byte(n.indices) {
					if c == '/' {
						n = n.children[i]
						if (len(n.path) == 1 && n.handler != nil) ||
							(n.nType == catchAll && n.children[0].handler != nil) {
							return append(ciPath, '/')
						}
						return nil
					}
				}
			}
			return nil
		}

		// If this Node does not have a wildcard (param or catchAll) child,
		// we can just look up the next child Node and continue to walk down
		// the tree
		if !n.wildChild {
			// Skip rune bytes already processed
			rb = shiftNRuneBytes(rb, npLen)

			if rb[0] != 0 {
				// Old rune not finished
				idxc := rb[0]
				for i, c := range []byte(n.indices) {
					if c == idxc {
						// continue with child Node
						n = n.children[i]
						npLen = len(n.path)
						continue walk
					}
				}
			} else {
				// Process a new rune
				var rv rune

				// Find rune start.
				// Runes are up to 4 byte long,
				// -4 would definitely be another rune.
				var off int
				for max := min(npLen, 3); off < max; off++ {
					if i := npLen - off; utf8.RuneStart(oldPath[i]) {
						// read rune from cached path
						rv, _ = utf8.DecodeRuneInString(oldPath[i:])
						break
					}
				}

				// Calculate lowercase bytes of current rune
				lo := unicode.ToLower(rv)
				utf8.EncodeRune(rb[:], lo)

				// Skip already processed bytes
				rb = shiftNRuneBytes(rb, off)

				idxc := rb[0]
				for i, c := range []byte(n.indices) {
					// Lowercase matches
					if c == idxc {
						// must use a recursive approach since both the
						// uppercase byte and the lowercase byte might exist
						// as an index
						if out := n.children[i].findCaseInsensitivePathRec(
							path, ciPath, rb, fixTrailingSlash,
						); out != nil {
							return out
						}
						break
					}
				}

				// If we found no match, the same for the uppercase rune,
				// if it differs
				if up := unicode.ToUpper(rv); up != lo {
					utf8.EncodeRune(rb[:], up)
					rb = shiftNRuneBytes(rb, off)

					idxc := rb[0]
					for i, c := range []byte(n.indices) {
						// Uppercase matches
						if c == idxc {
							// Continue with child Node
							n = n.children[i]
							npLen = len(n.path)
							continue walk
						}
					}
				}
			}

			// Nothing found. We can recommend to redirect to the same URL
			// without a trailing slash if a leaf exists for that path
			if fixTrailingSlash && path == "/" && n.handler != nil {
				return ciPath
			}
			return nil
		}

		n = n.children[0]
		switch n.nType {
		case param:
			// Find param end (either '/' or path end)
			end := 0
			for end < len(path) && path[end] != '/' {
				end++
			}

			// Add param value to case insensitive path
			ciPath = append(ciPath, path[:end]...)

			// We need to go deeper!
			if end < len(path) {
				if len(n.children) > 0 {
					// Continue with child Node
					n = n.children[0]
					npLen = len(n.path)
					path = path[end:]
					continue
				}

				// ... but we can't
				if fixTrailingSlash && len(path) == end+1 {
					return ciPath
				}
				return nil
			}

			if n.handler != nil {
				return ciPath
			}

			if fixTrailingSlash && len(n.children) == 1 {
				// No handle found. Check if a handle for this path + a
				// trailing slash exists
				n = n.children[0]
				if n.path == "/" && n.handler != nil {
					return append(ciPath, '/')
				}
			}

			return nil

		case catchAll:
			return append(ciPath, path...)

		default:
			panic("invalid Node type")
		}
	}

	// Nothing found.
	// Try to fix the path by adding / removing a trailing slash
	if fixTrailingSlash {
		if path == "/" {
			return ciPath
		}
		if len(path)+1 == npLen && n.path[len(path)] == '/' &&
			strings.EqualFold(path[1:], n.path[1:len(path)]) && n.handler != nil {
			return append(ciPath, n.path...)
		}
	}
	return nil
}