Repository navigation
Expand file tree
/
Copy pathfocus.go
More file actions
122 lines (100 loc) · 4.7 KB
/
Copy pathfocus.go
File metadata and controls
122 lines (100 loc) · 4.7 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
package internal
import (
"regexp"
"strings"
"github.com/onsi/ginkgo/v2/types"
)
/*
If a container marked as focus has a descendant that is also marked as focus, Ginkgo's policy is to
unmark the container's focus. This gives developers a more intuitive experience when debugging specs.
It is common to focus a container to just run a subset of specs, then identify the specific specs within the container to focus -
this policy allows the developer to simply focus those specific specs and not need to go back and turn the focus off of the container:
As a common example, consider:
FDescribe("something to debug", function() {
It("works", function() {...})
It("works", function() {...})
FIt("doesn't work", function() {...})
It("works", function() {...})
})
here the developer's intent is to focus in on the `"doesn't work"` spec and not to run the adjacent specs in the focused `"something to debug"` container.
The nested policy applied by this function enables this behavior.
*/
func ApplyNestedFocusPolicyToTree(tree *TreeNode) {
var walkTree func(tree *TreeNode) bool
walkTree = func(tree *TreeNode) bool {
if tree.Node.MarkedPending {
return false
}
hasFocusedDescendant := false
for _, child := range tree.Children {
childHasFocus := walkTree(child)
hasFocusedDescendant = hasFocusedDescendant || childHasFocus
}
tree.Node.MarkedFocus = tree.Node.MarkedFocus && !hasFocusedDescendant
return tree.Node.MarkedFocus || hasFocusedDescendant
}
walkTree(tree)
}
/*
Ginkgo supports focussing specs using `FIt`, `FDescribe`, etc. - this is called "programmatic focus"
It also supports focussing specs using regular expressions on the command line (`-focus=`, `-skip=`) that match against spec text and file filters (`-focus-files=`, `-skip-files=`) that match against code locations for nodes in specs.
When both programmatic and file filters are provided their results are ANDed together. If multiple kinds of filters are provided, the file filters run first followed by the regex filters.
This function sets the `Skip` property on specs by applying Ginkgo's focus policy:
- If there are no CLI arguments and no programmatic focus, do nothing.
- If a spec somewhere has programmatic focus skip any specs that have no programmatic focus.
- If there are CLI arguments parse them and skip any specs that either don't match the focus filters or do match the skip filters.
*Note:* specs with pending nodes are Skipped when created by NewSpec.
*/
func ApplyFocusToSpecs(specs Specs, description string, suiteLabels Labels, suiteConfig types.SuiteConfig) (Specs, bool) {
focusString := strings.Join(suiteConfig.FocusStrings, "|")
skipString := strings.Join(suiteConfig.SkipStrings, "|")
type SkipCheck func(spec Spec) bool
// by default, skip any specs marked pending
skipChecks := []SkipCheck{func(spec Spec) bool { return spec.Nodes.HasNodeMarkedPending() }}
hasProgrammaticFocus := false
for _, spec := range specs {
if spec.Nodes.HasNodeMarkedFocus() && !spec.Nodes.HasNodeMarkedPending() {
hasProgrammaticFocus = true
break
}
}
if hasProgrammaticFocus {
skipChecks = append(skipChecks, func(spec Spec) bool { return !spec.Nodes.HasNodeMarkedFocus() })
}
if suiteConfig.LabelFilter != "" {
labelFilter, _ := types.ParseLabelFilter(suiteConfig.LabelFilter)
skipChecks = append(skipChecks, func(spec Spec) bool {
return !labelFilter(UnionOfLabels(suiteLabels, spec.Nodes.UnionOfLabels()))
})
}
if len(suiteConfig.FocusFiles) > 0 {
focusFilters, _ := types.ParseFileFilters(suiteConfig.FocusFiles)
skipChecks = append(skipChecks, func(spec Spec) bool { return !focusFilters.Matches(spec.Nodes.CodeLocations()) })
}
if len(suiteConfig.SkipFiles) > 0 {
skipFilters, _ := types.ParseFileFilters(suiteConfig.SkipFiles)
skipChecks = append(skipChecks, func(spec Spec) bool { return skipFilters.Matches(spec.Nodes.CodeLocations()) })
}
if focusString != "" {
// skip specs that don't match the focus string
re := regexp.MustCompile(focusString)
skipChecks = append(skipChecks, func(spec Spec) bool { return !re.MatchString(description + " " + spec.Text()) })
}
if skipString != "" {
// skip specs that match the skip string
re := regexp.MustCompile(skipString)
skipChecks = append(skipChecks, func(spec Spec) bool { return re.MatchString(description + " " + spec.Text()) })
}
// skip specs if shouldSkip() is true. note that we do nothing if shouldSkip() is false to avoid overwriting skip status established by the node's pending status
processedSpecs := Specs{}
for _, spec := range specs {
for _, skipCheck := range skipChecks {
if skipCheck(spec) {
spec.Skip = true
break
}
}
processedSpecs = append(processedSpecs, spec)
}
return processedSpecs, hasProgrammaticFocus
}