Closed bdcallen closed 4 years ago
@iangow As I've mentioned in an email to you, I've decided to leave the footnote_variable
as transactionCoding
in these cases for now. This is possible to do since none of its child node names (transactionCode
, transactionFormType
or equitySwapInvolved
) appear in footnote_variable
field, making these cases easy to fix if we decide to make a change. We can decide what to do here after we've downloaded the data.
@iangow I am going to close this for now.
@iangow It turns out that my function
get_full_footnotes_indices
sometimes assigns the name of a node which is a parent to several variables for which my functions get the data for inget_derivative_df
andget_nonDerivative_df
. An example is this filing, for whichget_full_footnotes_indices
assignstransactionCoding
tovariable
for several entries in its table as can be seen belowThe footnotes for this case are
transactionCoding
is the name of a common node which contains the information for the variablestransactionFormType
,transactionCode
andequitySwapInvolved
. For this filing, if you look at the xml file, the footnoteF1
is indeed written as a direct child oftransactionCoding
. This actually is not in accordance with the SEC's style guide for the xml documents for Forms 3, 4 and 5 (which I have largely based my scraping functions on). Rather, the footnote should be a child of one oftransactionFormType
,transactionCode
andequitySwapInvolved
in this case.I'm just wondering how we should approach treating these cases. Assign the footnote to all the parent node's children? Or keep the parent node as the name?