Raw input from legacy documentation
‘Tsv’ files are flat files that include a ‘tab delimited’ sequence of values in each line
instead of one value per line/record.
– In most files the sequences of values are time series.
– For datasets without the dimension time (e.g. ‘area of the regions’), or that cover only
one period of time, the sequences of values are not time series but another dimension,
e.g. geographical series.
– The columns (or fields or cells) of the records are ‘tab delimited’.
– Time series are sorted in descending order (for explanation see the chapter ‘Hints for
Excel users’).
– ATTENTION: cells for which there is "no data available" are NOT stored in the tsv
files on the bulk download, doing so would explode the size of the tsv files. When using
the on-line extraction tools like Tables, Graphs and Maps or Data Explorer, such cells
contain the symbol ":"
EXAMPLE
Dataset with time series (with made-up values)
unit,s_adj,partner,flow,indic,geo\time 2004m05 2004m04 2004m03 2004m02
mio-eur,nsa,ext_eurozone,net,bp-100,eurozone 11148 10660 13398 9437
mio-eur,nsa,ext_eurozone,net,bp-200,eurozone 3386e 539 -185 -432
mio-eur,nsa,ext_eurozone,net,bp-300,eurozone -5626e -6696i 1902 919
mio-eur,nsa,ext_eurozone,net,bp-379,eurozone -5758e -4165 -3970 -4703
mio-eur,nsa,ext_eurozone,net,bp-993,eurozone 3151.5e 338.7i 11146.1 5221.0
mio-eur,nsa,ext_eurozone,net,bp-994,eurozone 2314 669 543 2113
mio-eur,nsa,ext_eurozone,net,bp-010,eurozone 5465.1 1006.5 11689.0 7334.3
– First line: header.
– Other lines: records with the sequence of values.
– First column — first line: sequence of codes separated by a comma followed by a
code separated by a back slash ‘\’
The codes separated by a comma ‘,’ are the ‘names’ of the dimensions used for
identifying each (time) series.
For each of these codes there is a file (with the same name plus the extension
‘dic’) in the directory dic.
The code separated by a back slash ‘\’ is the ‘name’ of the dimension of the
sequence of values, e.g. ‘time’ (if this is a time series) or ‘geo’ (in the case of a
geographical series).
– First column except the first line: sequence of codes separated by a comma ‘,’
that represent the ‘names’ of the items (or instances or positions) of the
dimensions. The label/title of these codes can be found in the ‘dic’ file that has
the same name of the corresponding dimension.
– Other columns of the first line: sequence of codes corresponding to the items of
the dimension.
– All other columns but the first line: sequence of values.
Where available, flags are attached to values. The separator used between values
and flags is a blank. If there are no flags, the value is followed by a blank.
– The decimal symbol used in the files is the dot ‘.’.
Note for Excel users: these files can be straightforwardly opened in Excel (see chapter
Hints for Excel users).
Should copy and refresh the section 3. HINTS FOR EXCEL USERS
Raw input from Migrating to API TSV
Should be converted to a confluence page with minor adaptation
- case of dimension header
- why it is TIME_PERIOD uppercase