Output formatting is not right in some arrays

[piotrwiniarczyk-silvair]

Describe the bug
The array is wrongly output. See repro.

To Reproduce

Output -

echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w - 'u.all()'
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w - 'u'
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w - 

Output xml

echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w xml 'u'
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w xml

while
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w xml 'u.all()'
seems to be better

Output csv

echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w csv 'u.all()'
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w csv 'u.all().n'
Error: CSVParser.toBytes cannot handle type string

Expected behavior

1. Better output, i.e. not showing memory addresses
2. Maybe a page in the documentation how to extract an array to csv but using only some fields:
   echo '{"u":[{"n":"A"},{"n":"B","a":"1"}]}' | dasel -r json -w csv 'u'
   works, but trying to get a csv with column "n" only is too hard for me. Maybe this is related to above problems.

This also should not be like this:
echo '{"u":[{"n":"A"},{"n":"B","a":"1"}]}' | dasel -r json -w xml 'u.all()

Desktop (please complete the following information):

• OS: Windows 11
• dasel version v2.3.6

[piotrwiniarczyk-silvair]

More on these.

One would expect this will work:

echo -e 'A,B,C\na,b,c\nd,e,f' | dasel -r csv -w csv
Error: CSVParser.toBytes cannot handle type []map[string]interface {}

The json seems to be reasonable.

echo -e 'A,B,C\na,b,c\nd,e,f' | dasel -r csv -w json
[
  {
    "A": "a",
    "B": "b",
    "C": "c"
  },
  {
    "A": "d",
    "B": "e",
    "C": "f"
  }
]

Same json output.

echo -e 'A,B,C\na,b,c\nd,e,f' | dasel -r csv -w json 'all().merge()'
[
  {
    "A": "a",
    "B": "b",
    "C": "c"
  },
  {
    "A": "d",
    "B": "e",
    "C": "f"
  }
]

Now csv to csv works.

echo -e 'A,B,C\na,b,c\nd,e,f' | dasel -r csv -w csv 'all().merge()'
A,B,C
a,b,c
d,e,f

[TomWright]

Thank you for the detailed examples @piotrwiniarczyk-silvair. I will take a look at these cases to fully understand them ASAP.

[TomWright]

Hey @piotrwiniarczyk-silvair , I have fixed this issue. To access the fix immediately please use the development version of dasel.
The fix will be included in the next release.

[TomWright]

This has been released under v2.4.0.

[piotrwiniarczyk-silvair]

The csv is case fixed. The xml and - cases are not.

$ dasel --version
dasel version v2.4.0

echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w - 'u.all()'
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w - 'u'
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w -
&{[n] map[n:A]}
&{[n] map[n:B]}
[0xc00017bb20 0xc00017bb60]
&{[u] map[u:[0xc00020e4e0 0xc00020e500]]}

echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w xml 'u'
echo '{"u":[{"n":"A"},{"n":"B"}]}' | dasel -r json -w xml
[0xc00026b4e0 0xc00026b500]
<doc>
  <u>&{[n] map[n:A]}</u>
  <u>&{[n] map[n:B]}</u>
</doc>

Not sure about this one.

echo '{"u":[{"n":"A"},{"n":"B","a":"1"}]}' | dasel -r json -w xml 'u.all()'
<n>A</n>
<doc>
  <a>1</a>
  <n>B</n>
</doc>

[TomWright]

Sorry I forgot about XML - my mistake.

I will revisit that.