As stated previously, I have begun to track personal finances via GnuCash. It really has been very empowering, since I typically use a spreadsheet to reconcile my budget against monthly receipts. Using the old method it was not possible to capture the cost of utilities, mortgage, and the other bills paid electronically. Despite the fact that I P-Cash application installed on my Treo650, I am unable to sync the data to my desktop financial software application. Sigh, perhaps one day this will become very simple to accomplish. I digress..
Now to the point.. I have noticed that .qif files have document headers which describe the intent of the bank transaction. For instance, if you're downloading data from subsequent debit card or check card you should see !Type:CCard in the header.
The problem arises when your bank delivers files which do _not_ have the required header file. GnuCash is unable to read files that do not include the required QIF header. While I'm very happy that GnuCash adheres to the QIF standard, the only workaround is to manipulate the file. What a royal PITA. After grabbing nearly ten files, I noticed this problem.
Sure I could contact the bank and complain about the issue. The reality is they probably would be clueless. Moreover, the script that they use to generate these .qif files is likely broken, and they will have to get the developer to fix the script. Perhaps costly, dunno.
Rather than re-invent the wheel, I found a decent Perl script which inserts
the a desired header after reading in a flat text file. I would prefer to use Python, but of course, I'm still learning. I would have thought that this sort of text manipulation would be a prime candidate for awk or sed.. I'll keep digging, simple shell scripting problem for sure.
Clearly I am not interested in manually changing each file, I am much too lazy ;)