Love Gnucash -- A few questions though . .

[Jon Hamkins]

On Thu, 2004-08-19 at 10:14, David Frascone wrote:
> On Thursday, 19 Aug 2004, Derek Atkins wrote:
> > David Frascone <dave at frascone.com> writes:
> > 

> I guess my question was more generic, and slightly off-topic:  What's
> the easiest scripting language for a complex web query.  A login (form),
> with cookies, then a form to select my download type, then a scrape to
> grab the url to download.  I've never done a web script that was this
> complex before.  And, of course, it's only complex, since my bank does
> not support DirectConnect.

I don't know what is the easiest, but I've been writing a set of perl
scripts to navigate web sites to download OFX files for institutions
that don't seem to work with ofx.py that is at
http://www.jongsma.org/gc/.

Although perl has spiffy modules for navigating web sites
(WWW::Mecanize), including automatically storing cookies, it doesn't
handle javascript, so one has to read the returned HTML files carefully
and manually implement anything that happens in the javascript (e.g.,
field and cookie settings).

I've written three such OFX web download scripts, but I've only debugged
one to the point where it works.  I'm attaching it here.  If there is a
better solution, let me know.  In particular, if anyone knows of
solutions for Bank of America, US Bank, and TIAA-CREF, I'd be
interested.  The bankinfo for US Bank doesn't seem to work for me with
ofx.py-- when I download from the web site, the fid in the OFX file
1402, while bankinfo says it should be 1401.

     ----Jon
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fetch_cefcu_ofx
Type: text/x-perl
Size: 3606 bytes
Desc: not available
Url : http://lists.gnucash.org/pipermail/gnucash-user/attachments/20040918/48e1c05d/fetch_cefcu_ofx.bin