On Mon, 14 Jun 1999, you wrote: > I'm going to download it and experiment. > > I have a subsidiary requirement, which is to > generate blocks of transactions automatically. > Currently, this occurs as an ordinary database > which spits out a QIF file that can be imported. > It's a bit ugly, but works. It would be great > if there were a command-line way of doing this > sort of thing. Is there one? I take it you want to generate transactions based on some condition which can be easily specified on the command line, and on the data. If so then I don't see much problem provided it doesn't get too convoluted. It should be easy enough to use a batch approach: read the old file and write a new one containing the old plus the generated transactions. If you give me a specific example of what you had in mind then I'll give it a quick try. > At the moment, I'm seriously thinking about > building something analogous to xacc that runs > on top of a real SQL database and utilizes the > transaction and trigger stuff to enforce the > integrity of the financial record. The advantage > would be to be able to auto-generate transactions > and make reports easily; the database is operating > all the non-financial aspects of the business and > the tie-in would save considerable effort. What I like about xacc is that it reads the whole file into memory. This makes it easy for me to apply simple checkpointing, simply by keeping the old copy, the current copy, and the working copy which is in memory. I finish my tiny days transactions and save the file. If the reports tell me its not corrupted then I back it up, otherwise I restore from back and repeat posting the transactions. Either way, it's easy for me to understand at the end of a long day at the office. The database approach is of course the best way in the medium - long term. Maybe I could produce a downmarket solution that will tide you over until your database was there. Thinking a little more big picture, I haven't been keeping too much track for what GnuCash plans to do with databases, but I would imagine that if they are not thinking database now then when they find that the front-end has the nice GUI and all the features they can think of then they will back-end a database to it to add scalability, reliability etc. If this is the case then it would be better to have one database development effort than many. In fact I think I'll copy this message to the GnuCash list so they can see it and correct my thinking if necessary. > > Your comments (as a real user of xacc) would be appreciated. > > Alex. -- Graham Chapman email: [EMAIL PROTECTED] web : http://www.zeta.org.au/~grahamc/ ----- %< -------------------------------------------- >% ------ The GnuCash / X-Accountant Mailing List To unsubscribe, send mail to [EMAIL PROTECTED] and put "unsubscribe gnucash-devel [EMAIL PROTECTED]" in the body
