Skip to content
This repository has been archived by the owner on Apr 16, 2022. It is now read-only.

Latest commit

 

History

History
199 lines (143 loc) · 10 KB

export-format.md

File metadata and controls

199 lines (143 loc) · 10 KB
Raw

Briefcase export CSV format

Format parameters

Parameter Value Description
Separator , Comma character
Quoted " Double quote character. See the Values block columns section to know which fields get quoted
Header Yes Every file will have a header with the respective column names
Empty columns Yes See the Values block columns section to know which fields get empty columns
Encoding of null values No Nulls are empty columns and, sometimes, empty strings

Main output file

The main output file has values from any non-repeating group field found at the topmost level of a form's model and other metadata.

Structure of the file:

Column name Required Quoted Description
SubmissionDate Yes Yes See the Values block columns section to learn how dates are encoded
Values block Yes Holds columns with the values of topmost model fields and non-repeating groups
KEY Yes No Holds the instanceID of each submission
isValidated No No It's only present if the form is encrypted. It holds true if the cryptographic signature has been validated, false otherwise

Repeat output file

Each repeat group will produce an output CSV file, and the filename will be the composition of the main csv filename and the repeat group's name.

  • If form X has a repeat group Y, then the output file for the repeat group will be named X-Y.csv

  • Middle groups (including repeats) are ignored. If form X has a non-repeat group Y, which has a repeat group Z, then the output file for the repeat group will be named X-Z.csv

  • If form X has two repeat groups called Y (in any group arrangement), then a sequence number suffix is added for disambiguation and you will get output files X-Y~1.csv, and X.Y~2.csv

    The sequence follows a depth-first ordering. In other words, files will have the same order as the repeat groups are defined in the instance's model.

    Some examples:

    outer-group
  dupe-repeat     << gets the number 1
    inner-group
      dupe-repeat << gets the number 2

    group1
  dupe-repeat << gets the number 1
group2
  dupe-repeat << gets the number 2

Structure of the file:

Column name Required Description
Values block Yes Holds columns with the values of the repeat group's fields
PARENT_KEY Yes Identifies the parent row
KEY Yes Identifies this row
SET-OF-{GROUP NAME} Yes Identifies the group this row belongs to

PARENT_KEY

Pattern Example
Top group {INSTANCE ID} uuid:00000000-0000-0000-0000-000000000000
Descendant group {PARENT KEY} uuid:00000000-0000-0000-0000-000000000000/group1[1]

Note: non-repeat groups in the chain of ancestors are always ignored

KEY

Pattern Example
Top group {INSTANCE ID}/{GROUP NAME}[{ORDERING}] uuid:00000000-0000-0000-0000-000000000000/group1[1]
Descendant group {PARENT KEY}/{GROUP NAME}[{ORDERING}] uuid:00000000-0000-0000-0000-000000000000/group1[1]/group2[1]

Note: the [{ORDERING}] part is 1-indexed Note: non-repeat groups in the chain of ancestors are always ignored

SET-OF-{GROUP NAME}

Pattern Example
Top group {INSTANCE ID}/{GROUP NAME} uuid:00000000-0000-0000-0000-000000000000/group1
Descendant group {PARENT KEY}/{GROUP NAME} uuid:00000000-0000-0000-0000-000000000000/group1[1]/group2

Note: non-repeat groups in the chain of ancestors are always ignored

The SET-OF columns may be named differently in the two output files. For example, if form X contains a non-repeat group Y, which contains a repeat group Z, then:

  • The main output file will have a column named SET-OF-Y-Z (the long name of the repeat group).
  • The output file for the repeat group will have a column named SET-OF-Z (only the short name of the repeat group, without the name of its parent group).

Values block columns

The values block has a column for each non-repeating field.

Repeating groups

Repeating groups will produce a column named using the format SET-OF-{GROUP NAME} and their value will correspond to the same column in its own output file.

Non-repeating groups

The fields of a non-repeating group replace their parent's column in a CSV file. Their column names will be composed to be able to track the group to which they belong. Example:

  • Lets consider the following model:

    <model>
  <field1/>
  <field2>
    <field3/>
    <field4/>
  </field2>
</model>

  • The output CSV file will have these columns: field1,field2-field3,field2-field4

  • The field3 and field4 show up in place of the field2, which is not present in the file.

  • These fields get a column name composed of the name of their parent field2 and their own name, separated by a - (hyphen character).

This will happen in a cascading manner, if there are non-repeating groups inside non-repeating groups.

A very common example of this is the <meta> block, which is often used to add the instanceID to the submission. This field will get into the CSV inside a column named meta-instanceID.

Column headers do not contain the name of the group that contains the repeat group, even if the headers do contain the name of groups within the repeat group.

  • Lets consider the following model:

    <model>
  <group1>
    <repeat1>
      <group2>
        <text1/>
      </group2>
    </repeat1>
  </group1>
</model>

  • The column header for text1 will be group2-text1, not group1-repeat1-group2-text1.

Geopoint fields

The fields of type GEOPOINT will produce a column for each one of their Latitude, Longitude, Altitude, and Accuracy components. The column names will be composed like a non-repeating group, with the name of the model field and each component name, separated by a - (hyphen character). Example: some_place-Latitude.

Date and time fields

These fields get encoded following the user's locale settings and Java's MEDIUM style pattern.

Non-empty value codification

Type Quoted Example csv
text Yes something,"some text",something
int No something,42,something
decimal No something,42.33,something
date Yes something,"Apr 26, 2018",something
time No something,8:56:00 AM,something
date_time Yes something,"Apr 26, 2018 8:56:00 AM",something
geopoint No something,23.314925,-3.9869671,21.800003,15.478,something
geotrace No something,23.314926 -3.9869713 21.800003 10.0;23.314926 -3.9869713 21.800003 10.0;,something
geoshape No something,23.314926 -3.9869713 0.0 0.0;23.314926 -3.9869713 0.0 0.0;,something
barcode No something,000049499094,something

Empty value codification

Type Quoted Output Example csv
text Yes "" something,"",something
int Yes "" something,"",something
decimal Yes "" something,"",something
date No empty column something,,something
time No empty column something,,something
date_time No empty column something,,something
geopoint No 4 empty columns something,,,,,something
geotrace Yes "" something,"",something
geoshape Yes "" something,"",something
barcode Yes "" something,"",something

Notice that the quotation rules change respect to the non-empty value codifications.

Notes on empty values

All fields from the following example submission are considered to be empty:

<data>
  <field1></field1>
  <field2/>
  <field3> </field3>
  <field4>        </field4>
</data>

Notice that a field holding only whitespaces will get trimmed and be effectively empty.