suyiiyii/yq
1
0
Fork 0
mirror of https://github.com/mikefarah/yq.git synced 2025-01-29 18:45:39 +00:00
Code Issues Packages Projects Releases Wiki Activity

wip

Browse Source
This commit is contained in:
Mike Farah 2022-01-02 12:51:30 +11:00
parent 4b0cd5335b
commit eaeeffe417
3 changed files with 117 additions and 356 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
pkg/yqlib/decoder_xml.go
View File

@ -233,6 +233,7 @@ func (dec *xmlDecoder) decodeXml(root *xmlNode) error {
elem.n.Data = trimNonGraphic(string(se))
if elem.n.Data != "" {
elem.state = "chardata"
log.Debug("chardata [%v] for %v", elem.n.Data, elem.label)
}
case xml.EndElement:
log.Debug("end element %v", elem.label)
@ -249,18 +250,7 @@ func (dec *xmlDecoder) decodeXml(root *xmlNode) error {
commentStr := string(xml.CharData(se))
if elem.state == "started" {
log.Debug("got a foot comment for %v: [%v]", elem.label, commentStr)
// elem.n.FootComment = elem.n.FootComment + commentStr
// put the comment on the foot of the last child
if len(elem.n.Children) > 0 {
child := elem.n.Children[len(elem.n.Children)-1]
log.Debug("putting it here: %v", child.K)
child.V[len(child.V)-1].FootComment = joinFilter([]string{child.V[len(child.V)-1].FootComment, commentStr})
} else {
log.Debug("putting it on the element")
elem.n.FootComment = joinFilter([]string{elem.n.FootComment, commentStr})
}
dec.placeFootCommentOnLastChild(elem.n, commentStr)
} else if elem.state == "chardata" {
log.Debug("got a line comment for (%v) %v: [%v]", elem.state, elem.label, commentStr)
elem.n.LineComment = joinFilter([]string{elem.n.LineComment, commentStr})
@ -275,6 +265,21 @@ func (dec *xmlDecoder) decodeXml(root *xmlNode) error {
return nil
}
func (dec *xmlDecoder) placeFootCommentOnLastChild(n *xmlNode, commentStr string) {
if len(n.Children) > 0 {
child := n.Children[len(n.Children)-1]
log.Debug("putting it here: %v", child.K)
dec.placeFootCommentOnLastChild(child.V[len(child.V)-1], commentStr)
} else {
log.Debug("putting it on the element")
if n.FootComment != "" {
n.FootComment = n.FootComment + "\n" + strings.TrimSpace(commentStr)
} else {
n.FootComment = commentStr
}
}
}
func joinFilter(rawStrings []string) string {
stringsToJoin := make([]string, 0)
for _, str := range rawStrings {

270
pkg/yqlib/doc/usage/xml.md
View File

@ -22,273 +22,3 @@ XML nodes that have attributes then plain content, e.g:
The content of the node will be set as a field in the map with the key "+content". Use the `--xml-content-name` flag to change this.
## Parse xml: simple
Given a sample.xml file of:
```xml
<?xml version="1.0" encoding="UTF-8"?>
<cat>meow</cat>
```
then
```bash
yq e -p=xml '.' sample.xml
```
will output
```yaml
cat: meow
```
## Parse xml: array
Consecutive nodes with identical xml names are assumed to be arrays.
Given a sample.xml file of:
```xml
<?xml version="1.0" encoding="UTF-8"?>
<animal>1</animal>
<animal>2</animal>
```
then
```bash
yq e -p=xml '.' sample.xml
```
will output
```yaml
animal:
- "1"
- "2"
```
## Parse xml: attributes
Attributes are converted to fields, with the attribute prefix.
Given a sample.xml file of:
```xml
<?xml version="1.0" encoding="UTF-8"?>
<cat legs="4">
<legs>7</legs>
</cat>
```
then
```bash
yq e -p=xml '.' sample.xml
```
will output
```yaml
cat:
+legs: "4"
legs: "7"
```
## Parse xml: attributes with content
Content is added as a field, using the content name
Given a sample.xml file of:
```xml
<?xml version="1.0" encoding="UTF-8"?>
<cat legs="4">meow</cat>
```
then
```bash
yq e -p=xml '.' sample.xml
```
will output
```yaml
cat:
+content: meow
+legs: "4"
```
## Parse xml: with comments
A best attempt is made to preserve comments.
Given a sample.xml file of:
```xml
<!-- before cat -->
<cat>
<!-- in cat before -->
<x>3<!-- multi
line comment
for x --></x>
<!-- before y -->
<y>
<!-- in y before -->
<d><!-- in d before -->z<!-- in d after --></d>
<!-- in y after -->
</y>
<!-- in_cat_after -->
</cat>
<!-- after cat -->
```
then
```bash
yq e -p=xml '.' sample.xml
```
will output
```yaml
# before cat
cat:
# in cat before
x: "3" # multi
# line comment
# for x
# before y
y:
# in y before
# in d before
d: z # in d after
# in y after
# after cat
```
## Encode xml: simple
Given a sample.yml file of:
```yaml
cat: purrs
```
then
```bash
yq e -o=xml '.' sample.yml
```
will output
```xml
<cat>purrs</cat>
```
## Encode xml: array
Given a sample.yml file of:
```yaml
pets:
cat:
- purrs
- meows
```
then
```bash
yq e -o=xml '.' sample.yml
```
will output
```xml
<pets>
<cat>purrs</cat>
<cat>meows</cat>
</pets>
```
## Encode xml: attributes
Fields with the matching xml-attribute-prefix are assumed to be attributes.
Given a sample.yml file of:
```yaml
cat:
+name: tiger
meows: true
```
then
```bash
yq e -o=xml '.' sample.yml
```
will output
```xml
<cat name="tiger">
<meows>true</meows>
</cat>
```
## Encode xml: attributes with content
Fields with the matching xml-content-name is assumed to be content.
Given a sample.yml file of:
```yaml
cat:
+name: tiger
+content: cool
```
then
```bash
yq e -o=xml '.' sample.yml
```
will output
```xml
<cat name="tiger">cool</cat>
```
## Encode xml: comments
A best attempt is made to copy comments to xml.
Given a sample.yml file of:
```yaml
# above_cat
cat: # inline_cat
# above_array
array: # inline_array
- val1 # inline_val1
# above_val2
- val2 # inline_val2
# below_cat
```
then
```bash
yq e -o=xml '.' sample.yml
```
will output
```xml
<!-- above_cat inline_cat --><cat><!-- above_array inline_array -->
<array>val1<!-- inline_val1 --></array>
<array><!-- above_val2 -->val2<!-- inline_val2 --></array>
</cat><!-- below_cat -->
```
## Round trip: with comments
A best effort is made, but comment positions and white space are not preserved perfectly.
Given a sample.xml file of:
```xml
<!-- before cat -->
<cat>
<!-- in cat before -->
<x>3<!-- multi
line comment
for x --></x>
<!-- before y -->
<y>
<!-- in y before -->
<d><!-- in d before -->z<!-- in d after --></d>
<!-- in y after -->
</y>
<!-- in_cat_after -->
</cat>
<!-- after cat -->
```
then
```bash
yq e -p=xml '.' sample.xml
```
will output
```yaml
# before cat
cat:
# in cat before
x: "3" # multi
# line comment
# for x
# before y
y:
# in y before
# in d before
d: z # in d after
# in y after
# after cat
```

174
pkg/yqlib/xml_test.go
View File

@ -86,7 +86,7 @@ for x --></x>
<!-- before y -->
<y>
<!-- in y before -->
<d><!-- in d before --><z/><!-- in d after --></d>
<d><!-- in d before --><z sweet="cool"/><!-- in d after --></d>
<!-- in y after -->
</y>
@ -95,6 +95,25 @@ for x --></x>
<!-- after cat -->
`
var expectedDecodeYamlWithSubChild = `D0, P[], (doc)::# before cat
cat:
# in cat before
x: "3" # multi
# line comment
# for x
# before y
y:
# in y before
# in d before
d:
z:
sweet: cool
# in d after
# in y after
# after cat
`
var inputXmlWithCommentsWithArray = `
<!-- before cat -->
<cat>
@ -159,82 +178,89 @@ var expectedXmlWithComments = `<!-- above_cat inline_cat --><cat><!-- above_arra
`
var xmlScenarios = []xmlScenario{
// {
// description: "Parse xml: simple",
// input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<cat>meow</cat>",
// expected: "D0, P[], (doc)::cat: meow\n",
// },
// {
// description: "Parse xml: array",
// subdescription: "Consecutive nodes with identical xml names are assumed to be arrays.",
// input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<animal>1</animal>\n<animal>2</animal>",
// expected: "D0, P[], (doc)::animal:\n - \"1\"\n - \"2\"\n",
// },
// {
// description: "Parse xml: attributes",
// subdescription: "Attributes are converted to fields, with the attribute prefix.",
// input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<cat legs=\"4\">\n <legs>7</legs>\n</cat>",
// expected: "D0, P[], (doc)::cat:\n +legs: \"4\"\n legs: \"7\"\n",
// },
// {
// description: "Parse xml: attributes with content",
// subdescription: "Content is added as a field, using the content name",
// input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<cat legs=\"4\">meow</cat>",
// expected: "D0, P[], (doc)::cat:\n +content: meow\n +legs: \"4\"\n",
// },
// {
// description: "Parse xml: with comments",
// subdescription: "A best attempt is made to preserve comments.",
// input: inputXmlWithComments,
// expected: expectedDecodeYamlWithComments,
// scenarioType: "decode",
// },
{
description: "Parse xml: simple",
input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<cat>meow</cat>",
expected: "D0, P[], (doc)::cat: meow\n",
},
{
description: "Parse xml: array",
subdescription: "Consecutive nodes with identical xml names are assumed to be arrays.",
input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<animal>1</animal>\n<animal>2</animal>",
expected: "D0, P[], (doc)::animal:\n - \"1\"\n - \"2\"\n",
},
{
description: "Parse xml: attributes",
subdescription: "Attributes are converted to fields, with the attribute prefix.",
input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<cat legs=\"4\">\n <legs>7</legs>\n</cat>",
expected: "D0, P[], (doc)::cat:\n +legs: \"4\"\n legs: \"7\"\n",
},
{
description: "Parse xml: attributes with content",
subdescription: "Content is added as a field, using the content name",
input: "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<cat legs=\"4\">meow</cat>",
expected: "D0, P[], (doc)::cat:\n +content: meow\n +legs: \"4\"\n",
},
{
description: "Parse xml: with comments",
subdescription: "A best attempt is made to preserve comments.",
input: inputXmlWithComments,
expected: expectedDecodeYamlWithComments,
scenarioType: "decode",
},
{
description: "Encode xml: simple",
input: "cat: purrs",
expected: "<cat>purrs</cat>\n",
scenarioType: "encode",
},
{
description: "Encode xml: array",
input: "pets:\n cat:\n - purrs\n - meows",
expected: "<pets>\n <cat>purrs</cat>\n <cat>meows</cat>\n</pets>\n",
scenarioType: "encode",
},
{
description: "Encode xml: attributes",
subdescription: "Fields with the matching xml-attribute-prefix are assumed to be attributes.",
input: "cat:\n +name: tiger\n meows: true\n",
expected: "<cat name=\"tiger\">\n <meows>true</meows>\n</cat>\n",
scenarioType: "encode",
},
{
description: "Parse xml: with comments subchild",
skipDoc: true,
input: "cat:\n ++name: tiger\n meows: true\n",
expected: "<cat +name=\"tiger\">\n <meows>true</meows>\n</cat>\n",
scenarioType: "encode",
},
{
description: "Encode xml: attributes with content",
subdescription: "Fields with the matching xml-content-name is assumed to be content.",
input: "cat:\n +name: tiger\n +content: cool\n",
expected: "<cat name=\"tiger\">cool</cat>\n",
scenarioType: "encode",
},
{
description: "Encode xml: comments",
subdescription: "A best attempt is made to copy comments to xml.",
input: yamlWithComments,
expected: expectedXmlWithComments,
scenarioType: "encode",
},
{
description: "Round trip: with comments",
subdescription: "A best effort is made, but comment positions and white space are not preserved perfectly.",
input: inputXmlWithComments,
expected: expectedRoundtripXmlWithComments,
scenarioType: "roundtrip",
input: inputXmlWithCommentsWithSubChild,
expected: expectedDecodeYamlWithSubChild,
scenarioType: "decode",
},
// {
// description: "Encode xml: simple",
// input: "cat: purrs",
// expected: "<cat>purrs</cat>\n",
// scenarioType: "encode",
// },
// {
// description: "Encode xml: array",
// input: "pets:\n cat:\n - purrs\n - meows",
// expected: "<pets>\n <cat>purrs</cat>\n <cat>meows</cat>\n</pets>\n",
// scenarioType: "encode",
// },
// {
// description: "Encode xml: attributes",
// subdescription: "Fields with the matching xml-attribute-prefix are assumed to be attributes.",
// input: "cat:\n +name: tiger\n meows: true\n",
// expected: "<cat name=\"tiger\">\n <meows>true</meows>\n</cat>\n",
// scenarioType: "encode",
// },
// {
// skipDoc: true,
// input: "cat:\n ++name: tiger\n meows: true\n",
// expected: "<cat +name=\"tiger\">\n <meows>true</meows>\n</cat>\n",
// scenarioType: "encode",
// },
// {
// description: "Encode xml: attributes with content",
// subdescription: "Fields with the matching xml-content-name is assumed to be content.",
// input: "cat:\n +name: tiger\n +content: cool\n",
// expected: "<cat name=\"tiger\">cool</cat>\n",
// scenarioType: "encode",
// },
// {
// description: "Encode xml: comments",
// subdescription: "A best attempt is made to copy comments to xml.",
// input: yamlWithComments,
// expected: expectedXmlWithComments,
// scenarioType: "encode",
// },
// {
// description: "Round trip: with comments",
// subdescription: "A best effort is made, but comment positions and white space are not preserved perfectly.",
// input: inputXmlWithComments,
// expected: expectedRoundtripXmlWithComments,
// scenarioType: "roundtrip",
// },
}
func testXmlScenario(t *testing.T, s xmlScenario) {