This program is free software; you can redistribute
it and/or modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2 of the
License, or (at your option) any later version.
This program is distributed in the hope that it will
be useful, but WITHOUT ANY WARRANTY; without even the implied warranty
of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.
You should have received a copy of the GNU General
Public License along with this program; if not, write to the Free
Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
CBB is a personal finance management utility for Unix/X11. It is
written entirely in Perl and Tcl/Tk so it is very portable and very
extendable. It is a program for anyone who would like to balance
their checkbook and manage their money using free software under Unix
and X11.
CBB is intended to be an open, extensible program. It utilizes a
simple, tab-delimited data file format and, because it is written
entirely in Perl and Tcl/Tk, it is very modifiable. In addition, it
provides a simple interface for users to add their own reports,
graphs, and external modules without modifying any of the CBB source.
The name, CBB, stands for the Check Book Balancer. It is intended to
fit well in the Unix naming scheme where commands should be short and
easy to type. Every once in a while I try to think of something a
little more creating, but as of yet, nothing has come to mind.
CBB has now existed for more than two and a half years. Version 0.60
represented a fairly substantial reworking/reorganization of version
0.53--although this may not always be immediately visible on the
surface. Version 0.62 represented a substantial amount of rework to
better support tk4.0. Now, versions 0.70 and later look much spiffier
and sport completely reworked internals, plus many new features,
tweaks, and fixes.
I sincerely hope you can find CBB useful in your day to day and year
to year money management. I am always working to improve this program
so please send in your comments, suggestions, and complaints.
If you would like to contact me, you may send email to curt@infoplane.com. Although sometimes the business of life gets the
best of me, I will do my best to respond in a timely manner and
address your issue or concern.
The address for the CBB mailing list is cbbers@infoplane.com.
To subscribe, send a message to cbbers-request@infoplane.com
with a message body of subscribe. To unsubscribe, send a
message to cbbers-request@infoplane.com with a message body of
unsubscribe. Recent messages to this list are archived at:
CBB is written in Perl and Tcl/Tk. These need to be installed on your
system before you can can continue. Note, CBB now requires tk4.0 or
higher and perl5.002 or higher. If you notice any problems or
incompatibilities please let me know.
Tcl and Tk are available via anonymous ftp from:
ftp://ftp.smli.com/pub/tcl/
Note, the Tcl FAQ has a list of mirror sites. You can read the FAQ
online at:
Periodically, new version of CBB are released. Most of the general
laws of computer upgrades apply to CBB. Although I go to great
efforts to ensure new versions work better than the previous version,
occasionally problems creep in. Use a good measure of caution (such
as backing up your data files, and keeping a copy of the previous
version of CBB) when upgrading this or any other component of your
system.
Once you have CBB installed and running, and if you are ``net''
connected, upgrading to the latest version is trivial. Just follow
the simple instructions in section 5.1.
You can manually upgrade CBB by simply installing the new version over
the top of the old. Be aware that I occassionally add, remove, or
change the names of files in the distribution. Over time, old,
outdated, useless files could accumulate in the CBB lib directory.
The environments and needs of users are often widely varied. The CBB
install script supports several additional options. If you are having
trouble installing CBB, you may want to read through this section for
ideas.
If the copy of perl that comes first in your path is an older version
(such as perl4) and you have perl5 located someplace else on your
system, you can use the following command to install CBB:
The CBB install script by default is interactive and askes several
questions. Running make install simply runs the install
script, and nothing else. If you run the install script by hand,
./install.pl you can pass it several command line options.
The following options are available:
--help: Display the usage message.
--perlpath <path>: Full pathname to perl interpreter.
--wishpath <path>: Full pathname to wish interpreter.
--bindir <dir>: Directory to install executables.
--libdir <dir>: Directory to install support files.
If you specify any one of the above paths from the command line, the
install script will not prompt for it.
The CBB install script has two addtional options to assist package
maintainers. Most Unix operating systems have a ``package'' format
for installing packages. Packages are convenient because they can be
quickly and easily installed, upgraded, and deinstalled. The software
come preconfigured in the package. Unfortunately most versions of
Unix have their own incompatible package format.
However, if you are a package maintainer for some brand of Unix, the
following options will interest you:
--prefix <dir>: Root directory where CBB will eventually be
installed and run from ... this is where CBB
thinks it lives.
--destdir <dir>: Root directory where CBB will actually be
installed right now for package building
purposes. It is assumed that once the package
is installed, CBB will reside in the --prefix
directory.
These options allow you to install CBB in a one directory, but modify
the internal variables and pointers to make CBB think it has been
installed in a completely different directory tree.
2.7 Migrating Data Files to Version 0.60 or Later
If you are upgrading from a pre-0.60 version, you will need to
upgrade your data files as well.
If you are not upgrading from a pre 0.60 version, you can safely skip
this section. This section describes the procedure to migrate your
data files from the pre-0.60 format to the current format.
The initial versions of CBB saved its data in an ASCII text file with
the fields delimited by the `:' character. However, not too far into
this CBB project (at version 0.50a I believe) I switched my thinking
and decided to take advantage of perl's dbm support. This has one
primary advantage. Any changes made to the data file are immediately
saved. This also has some disadvantages. It tends to close off the
data file so it can only be manipulated from within CBB. This is good
from a ``data encapsulation'' perspective, but it suddenly becomes a
problem if the data file needs to be manipulated in a way which CBB
doesn't support. Another point to note is that because perl is so
good at slurping in text files, and because of certain constraints
imposed by the Tk front end, using the dbm format did not provide any
speed advantage. In fact, for certain operations I noted a speed
decrease.
With the above observations in mind, I decided to come full circle and
return to saving CBB data files in an open ASCII format by default.
The astute among you may observe that ASCII files have one major
disadvantage. They are only saved at the user's request. So, one
hapless user working all day without saving + one unexpected power
outage, one press of the reset button, or one of any other creative
ways people invent to destroy their data = trouble. To counter this
problem, I created an auto save feature. At regular intervals,
if the current data file has been modified, it is saved to a temporary
file. This greatly reduces the risk of data loss stupidity or acts of
God--or acts of God in response to our stupidity.
The ASCII data file format is defined to be one record per line with
the fields delimited by the <tab> character. CBB saves the
following fields in this order: date, check number, description, debit
amount, credit amount, category, comment, and the cleared flag.
Please feel free (in fact, I encourage you) to view a CBB data file
with any text viewer (more or less.) The format is mostly self
explanitory.
REMEMBER to always backup up your data before doing anything
radical to it.
The easiest and simplest way convert your data files is to export them
to the ``CBB'' format using your pre-0.60 version of CBB. This is a
slightly different format than current versions of CBB use, but
current versions of CBB will be able to detect this slightly different
format and load the files correctly. When you save the file from the
new version of CBB is will be save in the new format.
This section describes a perl script which automates the
conversion of data file formats. However it is rumored to only work
with perl4 and not perl5. All other parts of CBB require perl5.002 or
greater. Please be careful with this script and always back up your
data before you do anything radical. ``Your millage may vary.''
I have provided a perl script to automate the data file migration,
migrate-to-0.60a.pl. It is located in the contrib
subdirectory of the distribution-dir directory.
1.
Backup your data file directory. Do it now!!!
2.
cd $CBB_LIB_DIR/contrib
3.
migrate-to-0.60a.pl <data-dir>
4.
Choose whether you want to migrain (oops I mean migrate) to
ASCII or dbm format. Type ascii or dbm. I personally
recommend using the ASCII format, unless you have strong
convictions otherwise.
The migrate script will perform the following tasks:
1.
Convert categories file from `:' delimited to <tab>
delimited, and save it in an ASCII format.
2.
Remove any *.bal.dir and *.bal.pag. These have
always been redundant from CBB's perspective.
3.
Rename all $file.txn.dir to $file.dir, and all
$file.txn.pag to $file.pag. This makes for a
slightly more pleasant naming scheme.
4.
Open each $file.dir and convert all transactions from
`:' delimited to <tab> delimited.
5.
If migrating to ASCII, save file to ASCII CBB format and
delete dbm file.
When the script has finished, your data files should be all set for
your new version of CBB. Start up CBB and poke around your files a
bit. If you notice any problems, let me know immediately--and be
glad you backed up all your data like I suggested earlier.
So, you want to go for a little test drive? Want to see how or if
this thing works? Want to send me a 21 inch monitor? Just checking.
:-) Well, read on ...
The following procedure will lead you through the process of creating
a new account, importing some data, editing transactions, and
balancing your account.
First, here is the ``one-paragraph'' version of this manual. To use
CBB, first make a directory where you would like to keep your group of
accounts. Then, go to this directory, run cbb, and make your
account(s). Optionally you could choose to import the default
categories, but this can be done at any time. Next, load the desired
account (if it is not already loaded) and create, delete, and edit
transactions to your hearts content. When your statement arrives in
the mail, balance your account. Meanwhile, you have been printing
some reports, viewing some graphs, maybe setting up some recurring
transactions, and hopefully managing your money better than before you
started using CBB!
Ok, so you've just installed CBB. What now? Well, if you are like
me, you will have already run it a few times before you cracked open
the manual and read all the way down to here. So go ahead and lauch
CBB again. The first thing you need to do is create an account.
1.
Run CBB by typing cbb.
2.
Select Make New Account ... from the File menu.
3.
Enter an account name (i.e. my-demo.cbb) and an account
description (i.e. My Demo Account).
4.
Click on the Create Account button. Your new account
will be created, and added to the master account list at the
bottom of the main window. The name of the account will also be
added to the category list (i.e. [my-demo]). This category
is used to specify transfers between accounts.
5.
You will be warned about not having a categories file. This
is perfectly normal at this point. An empty categories file will
be created for you.
6.
If you like, pull down the Functions menu and select
Categories -> Add Default Categories to create a bunch of
default categories.
Now that you have created an account it is time to enter a few
transactions. If you like, you can import some sample data to save
your fingers from the brutalities of typing. Otherwise, feel free to
skip this section and enter your own transactions. The CBB
distribution comes with some sample data just for this tutorial.
1.
Select Import QIF File ... from the File menu.
2.
You will be presented with a file selection dialog box.
Navigate your way to the CBB distribution directory. Beneath the
distribution directory is a demo directory. Go to the demo
directory and find the file named demo.qif. Double click on
this file to select it and import it.
3.4 Edit Transactions
Now, that you have some data to play with, try editing a transaction.
Click the ``!'' box or hit enter to commit the transaction. Click the
``X'' box or type <Meta-N> to clear the entry boxes and start
over.
Now try creating new transactions.
Try playing around with ``splits'' to specify more than one category
for a transaction. Feel free to explore the menus and buttons until
you get the hang of things. You can refer to Section
4.2 for a more detailed description of transaction
editing.
Now, lets pretend you just received your bank statement in the mail
and you want to reconcile your new account. Lets also pretend that
you didn't mess thing up too bad back in Section 3.4.
1.
Find the Balance button towards the bottom right hand
corner of the window and click on it. This will bring up a list of
all ``uncleared'' transactions.
2.
Enter a statement ending balance of 1740.00. (Leave the
statement beginning balance as 0.00.)
3.
Select the first four transactions as well as the sixth
transaction. (Just pretend these were the ones that showed up on
your statement.) Note, as you select transactions, watch the Debits = n, Credits = n, Difference = n line.
4.
When the starting balance - withdrawals + deposits = ending
balance, click on the Update button to mark all the selected
transactions as cleared.
5.
Congratulations: your account is (hopefully) balanced.
When your next statement arrives and you run the balance routine
again, you will only be presented with ``uncleared'' transactions to
select. This is a good way to spot old checks that never were cashed
...such as your mortgage payment that got ``lost in the mail.''
Note: by balancing your checkbook in this manner, two good things
happen. The first is, when your brother finally cashes that check for
$100.00 - many months after you wrote it - your bank balance
doesn't suddenly drop -$100.00 from where you think it should be.
That transaction had always been entered and subtracted out of your
bank balance. The second good thing is that it is easy to spot these
sorts of situations so that you can call up your brother and pester
him to cash the check.
Now that you have entered a bit of data, you may want to
``understand'' your data at a deeper level. CBB comes with several
reports and graphs which can help you get a better idea of where and
how your money is being spent. Feel free to look at a few reports and
graphs at this point.
Wow! You've been slaving away for the last 10 minutes perfecting your
demo account. Great job! That is about all there is to it. CBB
isn't rocket science. It just boils down ``plus and minus''. Since
this is not real data, you probably don't care to save it. However,
if this had been an actual account, you most definitely would want to
save your hard work. CBB stores all your changes in memory, so you
must save the account before you quit. If your forget to save your
work before you quit, CBB will remind you to do this. If you do
something awful, like reboot your machine or log out without saving,
you are not completely out of luck. CBB periodically saves a backup
copy of your account with a file name of #account.cbb#. When
you reload your account, CBB will notice the autosave file and ask if
you want to load it instead. Under normal circumstances you will want
to answer yes. If you aren't sure, don't do anything. Exit CBB, go
to your data directory and look at the files manually to make sure
which version is the one you want.
To create a new account from scratch, choose Make New Account
... from the File menu. Enter the account name without
any extension. (CBB will automatically add a .cbb to the name
you provide.) This account name will become a category of the form
[acct-name] description for use in transfers between accounts.
Section 4.4 explains how transfers work. Next enter a
description for this account. When you satisfied with your name and
description, click on the Create Account button and your new
empty account will be created and loaded.
To create a new account based on exported Quicken data, create a new
account from scratch as described in the previous subsection. Once
the account has been created, Quicken data can be imported into it.
Importing data from Quicken is explained in Section
4.10.1.
At all times in CBB (except for those times when you are doing other
things) you are either creating new transactions, or editing existing
transactions. Other operations such as balancing may temporarily
suspend the entry, but when the chosen function is completed, you are
returned to your current edit or insert operation.
At any time you can abort the current edit or insert operation by
initiating a new edit or insert or by clicking the ``X'' button to the
right of the entry area.
When you are satisfied with the current contents of the edit area, hit
Return or click the ``!'' button to the right of the entry area
and the transaction will be updated. If you are creating a new
transaction, it will be inserted. If if you are editing an existing
transaction, it will be updated.
Tk4.x provides several standard key bindings to facilitate data entry.
These tend to mimic emacs key bindings. Figure 1
shows a list of some of the more useful of these key bindings. For
those hackers among us: look in $(TK_LIB_DIR)/entry.tcl for a
complete list of key/mouse bindings.
When the focus is in the check number field you can use the +
and - keys to increment and decrement the number. CBB remembers
the last check number you used, so when you are creating a new
transaction, pressing + will insert the next check number.
Figure 2 shows a list of these key bindings.
The + and - keys work as you expect when the focus is in
the date field. They will increment or decrement the date. CBB tries
to keep track of things like leap year or even the number of days in a
specific month. So, if CBB generates an illegal date such as 2/30/97
please report this as a bug. Figure 3 shows a list of
these key bindings.
The Splits function allows you to ``split'' the value of a
transaction among several different categories. Caution: Even
though you may click Dismiss in the Splits window, the
transaction will not be updated. You need to hit Return or
click the ``!'' button again once the splits window is closed. Note:
If the splits window does not provide enough lines, you can increase
the maximum number of splits in your .cbbrc.tcl file.
CBB keeps a list of transactions sorted and indexed by the
description. When entering a typical transaction you would normally
first specify the check number, then specify the date, then specify
the description. If you have entered a similar transaction in the
past, then you only need to type the first few characters of the
description, hit Tab and the rest of the transaction is filled
in for you. At this point you should make any necessary changes.
Finally, you can update the transaction by hitting Return.
If you don't want your transaction completed (i.e. your transaction is
being completed in an undesired fashion) you can type Control-Tab in the description field to avoid the completion feature.
The transaction is auto-completed only once per transaction. In other
words, your changes won't keep getting overwritten each time you tab
through the description field. Also, this feature is only activated
by Tabbing from the description field. Shift-Tabbing or
pressing Return have the same effect as they do in any other
field.
Note: You can turn this feature on and off from the File ->
Preferences menu.
Categories are used to give each transaction a ``type.'' This is
useful for reporting, since the report can be organized by category.
4.3.1 Viewing Categories
Categories can be viewed by typeing <Meta-C> or selecting Categories -> Category List ... from the Functions menu.
This brings up a list of categories. When the category list is open,
double clicking on an category will paste it into the current entry.
Alternatively, you can select a category and click on the Paste
button.
Adding a category is as easy as using it in a transaction. When the
transaction is updated, and the category is unknown, you will be asked
if you wish to add the new category. Simply fill in the category
description and check the Tax Related box (if it is tax
related.) Once you are satisfied, click on the Add button.
Alternatively, you can bring up the category list and click on the
Add button.
As a last resort, you can manually edit the categories file.
CBB stores the category file in the same subdirectory as the
associated account file. Note, the fields are delimited by tabs.
Have fun and be careful if you try this!
Once the category list window has been opened (see Section
4.3.1) simply select a category and click on the Delete button and the category will be deleted.
4.4 Transfers Between Accounts
When an account is created, a category of the form [acct-name]
is also created. To create a transfer to another account, use the
destination account name (enclosed in [ and ]) to
specify the account being transfered to. When the transaction
is inserted into the current account a corresponding transaction will
be inserted in destination account.
WARNING: currently when a transfer transaction is edited or
deleted, its corresponding transaction in the other account should
be automatically changed as well. This code is not well tested at
this time, so keep your eyes open if you edit transfer transactions
and make sure the corresponding changes get made on the other end.
CBB will warn you whenever you edit or delete a transfer
transaction.
Clicking on the Balance button will bring up a list of all
``uncleared'' transactions in a window. CBB already knows your
statement beginning balance. It is simply the sum of all the cleared
transactions. Verify your statement's beginning balance then enter
your ending balance. Then go through all your ``uncleared''
transactions and check off all that are listed on your statement. CBB
will keep a running total of the beginning balance + the transactions.
When you are all done, this should equal the ending balance. If it
doesn't, there is a discrepancy someplace ...which will hopefully
be not too hard to track down. When everything matches, click on the
Update button. This will ``clear'' all the selected
transactions. When you balance your checkbook next month, only the
``uncleared'' transactions will be presented to you.
This technique makes it easy to spot and handle situations when some
``individual'' doesn't cash your check for 3 months ...it is still
in the system and easy to spot. This way your idea of your balance
stays in sync with the bank's idea of your balance. And since it is
entered into CBB and subtracted from your running balance, you won't
get burned when they finally do cash it.
If you notice a discrepancy that needs to be fixed while balancing,
you can bring the main CBB window forward and make your changes.
Then, bring the balance window forward and click the Refresh
button. This will bring the balance window back in sync with the main
window. Alternatively you could close and reopen the balance window.
As of version 0.60 of CBB, I have completely reworked the reporting
mechanism. Each report is actually a stand-alone, self-sufficient,
executable perl script (although there is nothing magical about the
use of perl, they could be written in any language.)
When you select a Reports ... you are presented with a dialog
box. In this dialog box you can select the report you would like to
print, the files you would like to include in the report, a date
range, and report destination. If you wish to include all
transactions up to a certain date, leave the Starting Date field
blank. If you wish to include all transactions from a certain date
on, leave the Ending Date field blank. To include all
transactions leave both fields blank. Dates should be in the
following format: mm/dd/[[yy]yy]. Valid dates are 5/19,
5/19/95, and 5/19/2095.
Reports can be printed to a variety of destinations. If you select
Send to Screen the report will be displayed in a window. If you
select Save to File or Pipe, enter a file name in the
appropriate field and the report will saved to that file. If you
specify a file name of the form | command, CBB will pipe the
output of the report through the specified command. For instance, if
you wanted to send the report to a printer, you could enter something
like:
This report is similar to the Transaction List By Category report
except it omits the individual transactions and only displays the sum
of the transactions for each category. It also properly handles
splits.
When you balance your account you mark all the transactions that are
listed on your bank statement as ``cleared.'' The uncleared
transactions are those that the bank has not yet reported on any
statement. This report lists all the uncleared
transactions. This list of transactions should be identical to the
list that is displayed when you select the Balance function.
This report will scan through all the transactions of the selected
accounts and find any breaks in the check number sequence. It will
also flag any duplicate check numbers.
Graphing is very similar to reporting. Each graph is actually a
stand-alone, self-sufficient, executable perl script which process the
input data and calls a second Tk script to display the result.
When you select a graph you are presented with a dialog box similar to
the one presented for report printing. You can select a graph, group
of input files, and a date range. If you wish to include all
transactions up to a certain date, leave the Starting Date field
blank. If you wish to include all transactions from a certain date
on, leave the Ending Date field blank. To include all
transactions leave both fields blank. Dates should be in the
following format: mm/dd/[[yy]yy]. Valid dates are 5/19,
5/19/95, and 5/19/2095.
Currently, preferences need much work ...However, you can set a
few things including fonts in your .cbbrc.tcl file.
If you have your own custom Tcl code you would like to include (such
as setting your favorite key bindings) you can created a file called
.wishrc in your home directory and place the code there. CBB
will source this file if it exists after it sources your
.cbbrc.tcl file.
All changes must be saved--or how could they be called changes.
CBB is similar to vi or emacs in that all your changes are
made in RAM. So, just like you must save your work in vi or
emacs, you must save your work in CBB. To do this, click on the
Save button or select Save Account or Save Account As
... from the File menu.
CBB has an ``auto save'' feature. Your proposed changes are saved to
a temporary file at regular intervals. (This interval defaults to 3
minutes and can be set in your .cbbrc.tcl file.) If CBB is killed or
crashes, the autosave file will be left intact. When you startup CBB
and load the account, CBB will notice the autosave file and ask you if
you would like to use it instead.
And, as my great grandfather always said, ``save early and save
often!''
CBB is capable of importing Quicken export files. These files are of
the form <name>.qif. The transfer process is relatively simple.
1.
Boot dos/windows. (dosemu?!?, wine?!? - anyone?) Run
Quicken and open up the desired account.
2.
From the File menu, choose Export->Export QIF
and specify the export file name. This should be <name>.qif
3.
Once the file is created, copy this file to someplace where you
can access it from Linux.
4.
Now, quickly exit from dos/windows before it begins to
corrupt your mind or your machine--you stinkin' Bill Gates
lover.
5.
Boot Linux! Whew ...much better!
6.
Run cbb.
7.
Make a new account. (Or load an existing account if you
wish to import the transaction into it.)
8.
Select Import QIF File ... from the File menu and
select the file you wish to import. Click on the Import button and the file will be imported into a New
account.
CBB is also capable of generating Quicken export files. These files
can be imported into Quicken. This process is also relatively simple.
1.
Run cbb.
2.
Load the account you would like to export.
3.
Select Export QIF File from the File menu. A
file called <acct-name>.qif will be created.
4.
Move this file to dos/windows side of the world. (Use your
favorite method.)
5.
Boot dos/windows, run Quicken. (At this point things begin
to get hazy since I've never done this ...use your
noggin ...it can't be that hard.)
6.
From the File menu, choose Import and specify the
Import file name. This should be <acct-name>.qif
7.
You are on your own. Enjoy your stay in dos/windows land.
Say ``Hi!'' to our buddy, Bill G. for me while you are
there. And while you are at it, please let me know what I
can do to win back your business. :-)
Computer security is a lot like other types of security. There is no
such thing as a perfect secure computer system or network. For every
protection you put in place, some clever person will be able to devise
a way to circumvent it. Usually when a cracker is discovered,
whatever hole they leveraged to gain their illegal access can be
closed. The cracker then proceeds to find a new hole. It is a never
ending cycle.
Having said that, there are steps you can take to protect your
personal privacy and make sure that any violation is intentional and
nonaccidental.
If you are the only user of a non-networked machine, this section will
probably not be your biggest concern, however, if you use CBB on a
shared machine, please read on. You may wish to protect your personal
financial data from being potentially viewed by others.
Starting with version 0.73, CBB sets it's umask to 0066. For those of
you who aren't brushed up on bitmasks and octal numbers, this means
that your data files are made to be read/write by only you. This is
equivalent to executing chmod 600 file.cbb from the Unix
command line. This means that only you or the ``super-user'' are
allowed to read or modify your data.
Making your data non-readable to everyone but yourself is usually
sufficient in most normal circumstances. However, the most vigilant
(or paranoid) users might like the additional security of keeping
their data encrypted. CBB is now capable of calling an external
encryption program such as ``crypt'' or ``pgp'' to encrypt and decrypt
your data as needed.
WARNING:The first rule of encryption says that using
cryptography programs without a proper level of understanding of
what is going on is usually worse than not using any encryption at
all!!!
There are many pitfalls here, so proceed only if understand what
encryption can and cannot do for you, and only if you know what you
are doing (in general), and make sure to keep good backups of your
data (in a secure location, otherwise what good is encryption.)
Basically you can use any program that follows the following criteria:
Accepts input from stdin (the data to be encrypted/decrypted).
Writes output to stdout (the data to be encrypted/decrypted).
Accepts the passing of the key (cryptcode) as the last
argument in the commandline.
It should be able to identify unencrypted data and pass it
along when decrypting (if you want to be able to load unencrypted
when in encryption mode - this is very convenient for converting
data.)
PGP fulfills these requirements if you use the following options:
Encryption:
``pgp -fc -z''
Decryption:
``pgp -f -z''
If you specify the encryption code of ``secret,'' CBB will then
invoke pgp with pgp -fc -z secret to encrypt and pgp
-f -z secret to decrypt the data.
Note that this does not use public key cryptography, but simply uses
the IDEA algorithm to encrypt the data.
This way of passing the code is also a possible security leak on some
systems, because other users might be able to snoop the command line
arguments to pgp (Linux allows pgp to blank out this particular
argument, so you cannot see it with ps -e).
There might be other security risks in this way of handling encryption,
so please check the pgp manual.
After having entered the encryption and decryption progam, you will be
asked for the ``cryptcode'' (the key). You have to enter it twice (you
don't want to lose data because you encrypted it with a mistyped key),
and instead of letters, there should appear only "*" in the entry box.
If you now save an account, it will be encrypted. Also the
preferences will be saved (they get saved each time you save an
account - and only then, so save an account after changing them.)
This does save the ``encrypted mode'' flag, and the encryption and
decryption commands. It does not save the cryptcode (you
don't want to read it in your preferences file, do you).
When ``encrypted mode'' is entered, logging is turned off (no sense in
having encrypted data files, if your transaction log is unencrypted).
It is turned back on, if you leave ``encrypted mode.''
If you are in ``non-encrypted mode,'' you will notice that your
encrypted accounts don't show any balances in the account-list
anymore.
If cbb gets started in ``encrypted mode'' it will ask you
automatically for your ``cryptcode'' before starting.
Pressing cancel in the different
crypto-windows might not reset non-encrypted mode.
Add-ons (or external scripts) that rely on being able to read
the files directly will not work anymore.
No error is reported when opening an account with a bad
cryptcode (it just does not show you any entries, and allows you
to overwrite the account with a blank account. This is not really
a security risk, since someone who can overwrite an account from
within cbb can do the same from the command line. But it is a
data loss risk.
This section is contributed by Michel Verdier
(mverdier@chez.com). It describes some techniques you can
use to share cbb accounts between multiple users. Please be aware
that some of the details in this section are Linux specific.
Also a word of WARNING: CBB works like an editor in that it reads a
copy of your account into memory. Your changes happen in RAM. Only
when you save the file are your changes written out to the disk.
CBB does not handle true concurrent usage! If two people are
working on the same account at the same, one person's changes
will get overwritten by the other persons changes.
But, with all due disclaimers, here is a description of something you
can do when you mix a little Unix and a little CBB.
Michel Verdier writes:
Your girlfriend wants to use CBB too! And of course you are sharing
some accounts. You can't just take some accounts and give her some
others. You wouldn't either take all accounts and give her your
password ...
Ok. Here is a solution using a common user and "super" to launch CBB.
First create a user and give it your accounts.
Create a group and a user cbb (with homedir set to /var/cbb).
For instance:
addgroup cbb; adduser cbb
Make a dir /var/cbb, or whatever you prefer (but use the same
for the cbb user homedir), readable only by the user cbb. For
instance:
mkdir /var/cbb; chmod 700 /var/cbb
Copy your accounts in this directory and correct access rights:
chown -R cbb:cbb /var/cbb; chmod 600 /var/cbb/*
Good! Now nobody except cbb would crash your accounts. :-) Let's
configure super. Hum, do you install it?
super will launch CBB with user cbb and group cbb. CBB will start in
/var/cbb and will find your accounts. Don't forget the display. Or
perhaps put in /etc/profile :
export DISPLAY=:0
If you still get a message like :
Xlib: connection to ":0.0" refused by server
Xlib: Client is not authorized to connect to Server
add somewhere after X starting (could be your window manager login) :
This section describes several scripts that are beyond the original
scope of CBB, yet may be useful in managing your money. These scripts
are installed in: $CBB_LIB_DIR/contrib
5.1 Fetching and Installing the Latest Greatest CBB
The script fetch-latest.pl will automatically fetch the latest
version of CBB from ftp.me.umn.edu. It will then untar and gunzip it.
Finally it will run the standard install procedure to install it.
You can run this script from within CBB by selecting Fetch &
Install Latest CBB from the External menu. Then, answer the
standard install questions when prompted. Don't forget to quit and
restart CBB once the install has completed successfully.
The script invest.pl is a simple hack to help keep track of
your investments. An alternative would be to use a real application
like ``Xinvest''. A pointer to Xinvest can be found on the CBB web
page.
The invest.pl accepts a simply formatted text file from stdin
and writes a simple report to stdout. A sample input file could
look like the following. Note: each field is separated by one or more
tabs.
# Date Description Shares Unit Price
#----- ----------- ------ ----------
19960101 Beginning of 1996 10.000 25.00
19960201 Updated value 0.000 26.56
19960301 Purchase more shares 10.00 25.87
19960401 Updated value 0.000 26.04
19961216 St Cap Gain Reinvest 1.157 26.43
19961216 Lt Cap Gain Reinvest 1.220 26.43
19961216 Service Fee -0.568 26.43
Executing the command: cat sample.inv | invest.pl would produce
the following output:
Price Total
New per Shares Total Total
Date Description Shares Share Owned Invstd Value
-------- ---------------------- ------ ------ ------ ------ ------
19960101 Beginning of 1996 10.000 25.00 10.000 250.00 250.00
19960201 Updated value 0.000 26.56 10.000 250.00 265.60
19960301 Purchase more shares 10.000 25.87 20.000 508.70 517.40
19960401 Updated value 0.000 26.04 20.000 508.70 520.80
19961216 St Cap Gain Reinvest 1.157 26.43 21.157 539.28 559.18
19961216 Lt Cap Gain Reinvest 1.220 26.43 22.377 571.52 591.42
19961216 Service Fee -0.568 26.43 21.809 556.51 576.41
Wouldn't you like your mortgage payment to entered automatically every
month? Would you like to have some idea of how much money you will
have in a month, six months, or a year? Then read on. This might be
just what you need.
Recurring transactions which have been automatically inserted are
denoted by placing a ``?'' in the ``cleared'' field. This is
changed to a ``!'' when the transaction date has passed. (Note:
this code is changed to a ``x'' when the transaction is
cleared.)
The first thing the script does is traverse the account
looking for any transactions with a ``?'' in the cleared
field. Of these transactions, those which have a date older than
today's date will have their cleared field changed to a `` !''. Those transactions with a future date will be deleted.
(They will be reinserted in the next step.)
2.
The last thing thing the script does is to traverse the .rcr file for the account and reinsert all future entries up
until the cutoff date. At this time the cutoff date is set for a
year in advance, but this can be changed by editing a variable at
the beginning of the recur.pl script.
The config (<account>.rcr) file is reminiscent of a crontab file. The format is slightly different, but the idea is
generally the same. Specify the days, months, and years that the
transaction will take place, then specify the transaction. One thing
to note is that all fields must be separated by one <Tab>
character. The script will then insert that transaction on those
days.
For instance, the following entry will cause an auto loan payment to
be inserted on the fifth of every month. Once the transaction has
been entered, you can make any changes you like to it, such as adding
# Days Months Years Description Debit Credit Comment Category
# ---- ------ ----- ----------- ----- ------ ------- --------
5 * * Loan's R Us 100.00 0.00 Auto-Loan
Another variation would be the following entry which will insert a
transaction every fourth of July.
4 7 * Fireworks 35.00 0.00 Bang Entertainment
The next entry will insert a transaction on the 15th and last day of
every month.
15,last * * Salary 0.00 3.14 Peanuts Salary Income
This entry will insert a transaction on the 1st day of January, April,
July, and October.
* 1,4,7,10 * Auto Insurance 200.00 0.00 Approx. Insurance
REMEMBER, separate each field by a <Tab> character and
not spaces.
Finally, a slightly different entry format will allow you to enter
transactions based on a regular interval such as every other week.
The first field specifies the start date of the interval. This can be
any date, but recur.pl will never enter past recurring
transactions. The transactions will kick in after the current date.
Then next field is the interval in days. So 14 would specify an
interval of every other week. The third field is ignored and not
used. The last five fields are identical to the previous types of
entries.
# Start Date Intrvl Not Used Description Debit Credit Comment Category
# ---------- ------ -------- ----------- ----- ------ ------- --------
19960119 14 * Salary 0.00 2.78 Peanuts Salary
You can make your recurring transaction happen automatically by
running recur.pl out of cron. The details of setting up a cron job
may vary between Unix systems. Please refer to your local
cron and crontab unix man pages.
WARNING: Be aware that if this cronjob runs while you are
actively editing a CBB account, you will likely loose the results of
the recur.pl job as soon as you save your current CBB session.
#!/bin/sh
# CBB recurrent transactions
# find all users files and apply recur.pl for each
CBB_PATH="/usr/X11R6/lib/X11/cbb"
CBB_DATA="/var/cbb"
if [ -x "$CBB_PATH/contrib/recur.pl" ]; then
for dir in $CBB_DATA; do
for fic in `find $dir -name "*.rcr" -print| sed s/\.rcr$//`; do
user=`ls -l $fic.cbb | awk '{print $3}'`
su -l $user -c "$CBB_PATH/contrib/recur.pl $fic.cbb > /dev/null"
done
done
fi
The script yearend.pl simply moves all uncleared transactions
from the specified account to a new account. This helps keep file
sizes smaller. Usage: yearend.pl <account>.cbb
<new-account>.cbb
A number of people have enquired about text interfaces to CBB.
Currently there is no complete solution, but there are a few options.
Note, rumor has it that someone is working on a dialog interface to
CBB, I look forward to seeing this when it is ready.
Christopher B. Browne (cbbrowne@hex.net) has
provided a perl script called txn. It will append records to
.cbb files, and do so from the command line. For instance:
[25%] txn checking -t '75' 'The Mansion in Dallas'
Lunch 75.00 'Gratuitously expensive luncheon'
Added to /home/cbbrowne/kwiken/checking.cbb
19970422 75 The Mansion in Dallas 75.00 0 Lunch Gratuitously
expensive luncheon
This script does have certain limitations:
It appears that ``splits'' have been changed somewhat; I haven't
checked yet to see if they are handled correctly.
Writing to a .cbb file whilst the TCL version is open and
using the file in question will have obviously questionable
results. If Curt implements his file locking idea (which probably
ought to put the PID of the main CBB process in either
/var/lock/LCK..cbb or do individual file locks like
/var/lock/LCK..checking.cbb), I could have txn look for and/or
lock these...
You'll have to modify the paths at the start of the script to
locate your favorite CBB directory.
This script assumes that if the category is not available that
you can look at a list of similar category names and pick one. If
you plan to write scripts that will create transactions as ``batch
jobs,'' then you'll have to ensure that the category is correct.
I'm not sure how to best deal with the error condition of not
having a correct category. Create ``ntxn'' (non-interactive txn)
that reports an error and dies upon receiving a bad category,
perhaps? An additional command line switch? Ideas anyone?
I have no further intentions to maintain the code that allowed
deletion/reconciliation from the command line. The cbbsh of 1994
did provide this functionality, at the cost of a fairly painful
user interface. It just makes too much sense to do transaction
editing with a somewhat ``GUIed'' interface to bring most of the
remainder of CBBSH up to date.
Ravinder Bhumbla (rbhumbla@atol.ucsd.edu) has contributed an
Emacs forms-mode for browsing/editing .cbb files from within Emacs.
Emacs version 19 has a forms mode which can be used to browse
regularly ordered data. It is a convenient tool when you are not at
an X terminal.
Just read the file .../contrib/emacs-forms into an Emacs
buffer. Edit the path to your .cbb file where it says ``EDIT HERE''
and then type M-x forms-mode. Emacs will ask if you want to evaluate
the current buffer contents. Type ``yes''. (Ravi)
CBB stores it's data file in a simple tab-delimited format. If you
are careful about maintaining the correct tab-delimited format, it is
quite possible to edit your files with just about any text editor.
You will definitely want to make a backup copy of your .cbb file
before attempting this. It is very easy to make a mistake and corrupt
a transaction record. However, this can be a useful technique for
many tasks.
For instance, lets say my wife just informed me that comestibles are
spelled ``Food'' not ``Fude''. I could use a text editor to do a
global substition through my entire .cbb file to fix this mistake.
Another use could be for deleting or duplicating blocks of
transactions.
This simple format gives you last-resort options when you need to do
something that the graphical interface was not designed to do.
Most spreadsheets and database applications can load tab-delimited
files. This is another alternative for manipulating or viewing .cbb
files in ways that the graphical front end cannot.
CBB stores the values of a few standard variables in the .cbbrc.tcl file. Whenever CBB starts, it first sources this
file. This file is regenerated based on the current value of the
relevant variables whenever an account is created, loaded, or saved.
Any additions/deletions will be lost. However, changes to the values
of the variables will be maintained. Some of the preferences that can
be set in your .cbbrc.tcl file are: debugging on or off, balloon
help on or off, USA or international date format, autosave interval,
main list box dimensions, account list box height, maximum number of
splits, font selection, color selection, and your favorite web
browser. Please look in your .cbbrc.tcl for a complete and
current list of preferences that can be set.
The first field (everything before the <tab>) is the name to
appear on the External menu. The second field (everything after
the <tab> is the command line of the program to execute. Before
launching the external program, CBB will save the current contents of
the buffer to a temporary file. This allows the external program to
have access to any changes that have been made, but not yet saved.
The %t is replaced with the full path name of this temporary
file. The %a is replaced with the full name of the current
account file.
CBB has a ``simple'' well defined interface for installing new
reports. Each report is a stand alone ``program'' which understands a
predefined set of command line options and displays its output to stdout.
When the Configure Reports program is launched, it looks for the
file, .../lib/cbb/reports/reports.conf. This file consists of
entries like the following:
Transaction List rep-txn-list.pl
Txn List by Category rep-by-cat.pl
Short List by Category rep-by-cat-shrt.pl
Each line contains two fields. The report name which will appear in
the dialog box, and the executable name of the report. CBB scans this
file when it starts up in order to create the Reports menu.
CBB assumes that all report executables will also be located in the
.../lib/cbb/reports/ directory. Each report must accept the
following options where date if of the form mm/dd/[yy[yy]] and
account-list is a list of a CBB (ASCII) format account files:
Usage: report [ -from date ] [ -to date] account-list
The report executable will read the entries in the specified account
files, ignore any entries outside the specified date range, and print
its output to stdout. The last thing the report should print is a
single line containing the text none.
When a report is selected from the Reports menu, CBB displays the
Report Configuration dialog box. When the user clicks on Generate
Report, CBB saves the current contents of its ``buffer'' to a
temporary file. It then looks up the corresponding report executable,
and launches it with the above options. CBB reads the output of the
executable (stopping when it receives a line containing only the text
none) and routes it according what the user specified in the
Report Configuration dialog box.
The procedure for creating and installing new graphs is analogous to
the reports procedure. One thing to note, the graph executable should
wait for a carriage return from stdin before exiting. This is CBB's
way of letting the graph executable know that the user is done looking
at the graph.
When CBB is invoked with the -devel option, a Devel menu
is added to the menu bar. This menu will allow you to re-``source''
the various Tk pieces of CBB. With this option enabled you can reload
various pieces of the code after modifying them without quiting and
restarting CBB. This can greatly speed up the development process in
some situtations.
This works by executing the Tk source command on the selected
file. To use this feature effectively, you need to be aware of the
contents of a particular Tk file. Effectively, you are re-running
that code at the time you are sourcing it. Therefore, old procedure
definitions will be replaced if they were changed. The thing to watch
out for is code that is executed outside of procedure calls. This
code will be executed when you re-source the file. This may or may
not cause a problem if this code is intended to be run only once on
startup.
Activating and using this menu is something you only need to be
concerned about if you are actively developing code.
The FAQ tends to resist emacs TeX mode. This is most annoying.
Because of this I have placed the FAQ in its own separate file called
...Ta da!!! ...``FAQ''. The FAQ is also available on the CBB
web page at: http://www.menet.umn.edu/~clolson/cbb/