On Jun 23, 2010, at 1:48 AM, Thierry Volpiatto wrote:
To reconcile,
i convert the csv file of my bank to an org table, then i split window
in three, one with org table, one with ledger file, and one with the
reconcile buffer.
Just so you know, Ledger 3.0 has a new command: convert.
Give
On Jun 21, 2010, at 3:02 PM, Leandro Henrique Oliveira Fernandes wrote:
When I run:
-M -p from last month reg Supermercado
Error: Unexpected date period token 'last'
I will get this fixed.
-M -b 2009/10/01 -r -d /Assets:Leandro/ reg Expenses:Supermercado
Error: Cannot
On Jun 19, 2010, at 4:32 PM, ed wrote:
Don't think so, nm ledger shows e.g. mpfr and boost symbols as
unresolved.
Ok, then my packaging script isn't working properly for 10.5.
John
On Jun 21, 2010, at 3:02 PM, Leandro Henrique Oliveira Fernandes wrote:
-M -p from last month reg Supermercado
Error: Unexpected date period token 'last'
This is working now in the 'next' branch.
John
On Jun 19, 2010, at 7:24 AM, ed wrote:
Erm, a thought, won't you still need all the shared libraries to run
this binary?
It should have put them into the .dmg. It didn't?
John
On Jun 18, 2010, at 7:01 AM, ed wrote:
File ./acprep, line 640, in current_version
assert(match)
AssertionError
This is the master branch. Where would I send the dmg?
What's in your version.m4 file, at the top of the source tree?
You can upload .dmg's here:
On Jun 18, 2010, at 12:51 PM, TR wrote:
Just to be clear, the metadata field Payee can be below each
posting?
Yes, that is the intended usage actually.
John
On Jun 17, 2010, at 7:32 AM, Brent Pinkney wrote:
Also, is there anyway ledger can auto-magically apply interest to an account
periodically. I have a gawk script to do it, but was just wondering.
No. Since the aim of Ledger is not content creation, but rather reporting,
it has no capacity of
On Jun 16, 2010, at 7:14 PM, Leandro Henrique Oliveira Fernandes wrote:
Here in Brazil I can divide my creditcard purchases in installments.
So, maybe in current month I can have installments of some purchases
that I did in previous months (That is why I used --effective). So,
I have to list
On Jun 16, 2010, at 9:06 AM, ed wrote:
FYI:
boost 1.42
gettext 0.18.1.1
gmp 5.01
mpfr 2.4.2
pcre 8.01
==
All 5 tests passed
==
Awesome! Can you try running this now:
./acprep bindmg
To see if it can successfully bundle up the binary into a .dmg?
Here are the coverage numbers for 'make check':
http://ftp.newartisans.com/pub/ledger/lcov/ledger-proof/src/index.html
Have a little ways to go yet. 61% coverage overall.
John
On Jun 15, 2010, at 7:13 AM, John Wiegley wrote:
Have a little ways to go yet. 61% coverage overall.
Hmmm... I've found that my use of boost::optional pretty much assures that
I'll never get too high on branch coverage. Consider the following:
int foo(const boost::optionalbar_t x
Since we are inching ever closer to 3.0, I am going to start versioning pushes
to master. Today's version is 3.0.0-20100615, and constitutes BETA1. I'll be
pushing to master less frequently now, as I try to see how long each of these
releases can stand on its own. Writing the manual will
On Jun 13, 2010, at 4:52 AM, Martin Michlmayr wrote:
However, I'm not sure I like the syntax with the two colons. Given
that the meta-data syntax has never been part of an official release,
what do you think of changing the existing Key: Value syntax to
Key: Value and then you could drop the
On Jun 13, 2010, at 5:34 AM, Martin Michlmayr wrote:
This isn't really ledger related but I hope someone can help. I
thought I understood actual vs effective dates but it seems I'm more
confused than I thought.
Actual and effective have no internal meaning to Ledger. They mean whatever
you
This is what you need, and it's not simple:
ledger -F '%(scrub(display_total) 0 ? ansify_if(justify(( +
to_string(abs(scrub(display_total))) + ), 20, 20 + prepend_width, true,
false), red) : justify(scrub(display_total), 20, 20 + prepend_width, true,
color)) %(!options.flat ? depth_spacer
I realized last night that Ledger was *this* close to supporting typed metadata,
so I was able to add it in less than an hour.
Regular metadata looks like this:
; Key: Value
This is now just a special case. The general form for metadata is:
; Key:: Value Expression
So, these two are
Consider the following transaction:
2010-06-12 Sample
Assets:Brokerage 10 APPL
Assets:Brokerage $-100
In this case, what actually gets added to Brokerage is not APPL, but:
10 APPL {$10} [2010/06/12]
But what if you wanted that $10 price to be fixated, as in {=$10}?
These can occur in many places:
; Within an automated transaction, the assert is evaluated every time
; a posting is matched, with the expression context set to the
; matched posting.
= /Food/
assert account(Expenses:Food).total = $100
2010-06-12
On Jun 10, 2010, at 12:11 PM, dtromero wrote:
Any help, resources, or pushes in the right direction would be
appreciated. Thanks!
What operating system and architecture are you on? Are you willing to build
source code, or do you want a pre-packaged binary?
John
On Jun 10, 2010, at 3:50 PM, Roel Vanhout wrote:
So that's where I'm at, at the moment. At this moment I'm thinking
that boost::ptr_deque will be the cleanest/easiest solution but I
haven't tried it yet and I have to go in 15 minutes. I'll let you know
when I have news, but if you have any
On Jun 10, 2010, at 3:50 PM, Roel Vanhout wrote:
At this moment I'm thinking
that boost::ptr_deque will be the cleanest/easiest solution but I
haven't tried it yet
Ok, I'm about to push this to the 'next' branch in 10 mins. ptr_deque was
extremely easy to use.
John
On Jun 10, 2010, at 8:49 PM, David Brown wrote:
Perhaps it would be worth a release just for this patch? I've had to
apply something similar on every machine I've build ledger on, and
most distros include it in their build instructions for 2.6.2.
I don't even remember the process I used to
One thing standing between now and Ledger 3.0 is my desire to ensure that
everything marked fixed really is fixed. To that end, I'd like to turn
every closed bug into a regression test, or else mark the bug as CLOSED to
indicate it can't be tested.
If you are interested in helping out, turn one
On Jun 9, 2010, at 10:29 AM, Roel Vanhout wrote:
I'm trying out ledger for a project I'm starting shortly. I'm mostly
making plans and forecasts at this moment, with a small startup cost
left and right, with the plan to transition smoothly into the project
once money really starts to move
On Jun 8, 2010, at 7:51 AM, ed wrote:
I would be content with 3 levels of settings:
I think this is likely an ideal. After much thinking on this, I realized that
I can't do amount-specific valuation; and that account-specific valuation
cannot occur in all cases (such as when you use reg
On Jun 8, 2010, at 8:47 AM, Martin Michlmayr wrote:
However, I'd like to use them together. Say I own shares in EUR and
GBP. I can use -B and -V to see the basis vs market cost. But maybe
I'd like to see the market cost of all shares (including those I
bought with GBP) in EUR, i.e. -V -X
On Nov 30, 2009, at 10:45 AM, jleija wrote:
Note that in the July 15th transaction the converted amount is -29.41,
when it should had been -30.00 because its enclosing fixed mxn
region sets mxn to $0.1. The -29.41 comes out of multiplying -300.00
mxn by 0.0980392156862745, which is the rate
On Jun 5, 2010, at 8:21 AM, ed wrote:
That didn't work
Please try editing the 'acprep' file, and removing the addition of -fast to the
configure args. It should be fairly obvious which lines to comment out.
John
On May 23, 2010, at 7:56 PM, Zack Williams wrote:
I also found something else regarding calculations on commodity prices
- given this input file:
Ledger doesn't calculate the EMED value if you use a small whole
number for the commodity that does the conversion from hours to
rate/hour:
On Jun 4, 2010, at 7:16 PM, Felipe Magno de Almeida wrote:
I would like to translate to Brazilian Portuguese as well.
Thank you so much, Felipe. I've added you to the list and will contact you
once things are finalized.
John
On Feb 27, 2010, at 5:37 PM, thierry wrote:
On below tests, ledger v3 is 21 and 46 times slower that ledger v2.
After several days of intensive examination, I think now there is little help
for what you're seeing. The slowness comes from (a) the size of your data file
and how many
On Jun 4, 2010, at 6:02 AM, John Wiegley wrote:
After optimization work (current 'next' branch):
reg, no flags 5.00s
reg, -V13.39s
reg, -X12.91s
Note too that with --no-rounding, I can get -V and -X numbers within only 2x
slower. That is, if you
On May 23, 2010, at 6:22 PM, Peter Keen wrote:
Most of us are pretty familiar with ledger by now but I don't think
the greater community knows much about it. I wrote up a little blog
post explaining why I use it and a few small examples on how it works.
This was great, Peter! It brought many
On Jun 4, 2010, at 11:06 AM, ed wrote:
Smiles feebly ... next branch? could I have an acprep incantation
please? You can't ssh into my laptop but I'd be happy to email you the
config.log if that would help
Type:
git checkout next ; ./acprep opt update
That should do it for you. :)
John
On Jun 4, 2010, at 11:06 AM, ed wrote:
Smiles feebly ... next branch? could I have an acprep incantation
please? You can't ssh into my laptop but I'd be happy to email you the
config.log if that would help
Type:
git checkout next ; ./acprep opt update
That should do it for you. :)
John
On Jun 4, 2010, at 8:24 AM, Martin Michlmayr wrote:
I believe this is a bug in ledger. I independently ran into this the
other day: http://newartisans.com/bugzilla/show_bug.cgi?id=213
It has some explanation of what I believe is going on.
In your case, too, Martin, you need to use the -B
On Jun 2, 2010, at 5:33 AM, ed wrote:
Erm, bump? Any joy for those of us still running perfectly serviceable
antiques?
I need to see config.log. Is there any chance I could ssh into your system?
John
This is just to clarify, and for the record:
P1 Intended for the next push of 'next'
P2 Intended for the next push of 'master'
P3 Intended for the next point release, e.g. 3.0.x
P4 Intended for the next minor release, e.g. 3.x.0
P5 Intended for the next major release, e.g. x.0.0
By the way, to all Ledger users: I consider compiler warnings, on any system,
to be an error. Please enter a bug for me if you see them, and include the
warning, how you built it, and the details of your system (OS version, etc).
Thanks,
John
On Jun 2, 2010, at 7:47 PM, Matthew Palmer wrote:
You can always force the issue by adding -Werror to the compiler flags.
I don't want to be a nuisance, but acprep --warn does just this (plus turning
on a ton of other warning flags).
John
On May 30, 2010, at 12:28 PM, leandroohf wrote:
The first problem is that option -r does not output related accounts
when I use --effective options together. Bellow, It follows a
example of my ledger file
I'm not sure if this worked properly in 2.x or not, but it should work fine in
3.0. I
On May 31, 2010, at 5:27 AM, Martin Michlmayr wrote:
So, is the third solution the right one or did I miss anything?
This sounds right to me. The -B report should always reflect actual
commodities changing hands, and must *always* balance to zero across all
accounts. If you want to see
On May 31, 2010, at 10:02 AM, Felipe Magno de Almeida wrote:
Also, Brazil uses comma and is not in Europe :)
Ok, ok. ;) How about --comma-decimals?
John
If you have a transaction, following by a series of pricing entries, -V and -G
will now inject revaluation postings for each intermediate pricing entry,
rather than simply the difference between actual postings. Thus, for example:
1994/01/01 Achat Rialto
Actif:Fixe:Rialto 2
In an effort to make bug tracking more collaborative, I have created a Bugzilla
for Ledger. Report your bugs here:
http://newartisans.com/bugzilla/enter_bug.cgi
The bugs will also be tracked in Org-mode, since that's what I use for private
scheduling. The integration is actually a snap,
On Feb 27, 2010, at 3:36 PM, thierry wrote:
Hello John, did you find any time to work on this one? Is it on your
todo list?
It's bug #199 now. :) I'll make this top priority now.
http://newartisans.com/bugzilla/show_bug.cgi?id=199
John
On May 27, 2010, at 8:03 AM, Thomas Bikeev wrote:
Is there a way to serialize ledger journal out from interactive (REPL)
mode?
1. ledger -f journal.acc
2. while in *REPL* do stuff (add / alter entries)
3. write back to disk
There are no ledger commands which modify the data file. Is that
On May 25, 2010, at 5:35 AM, Xavier Shay wrote:
I'm on a mac and just downloaded the latest dmg. As far as I can tell,
gnucash support is not currently in the next branch.
gnucash support is missing in 3.0. Though if there is still sufficient
interest, I would consider porting it from 2.x.
On May 24, 2010, at 12:41 PM, thierry wrote:
But may I ask to dig further about --exchange impact?
I absolutely will. :)
John
On Mar 25, 2010, at 5:51 PM, Drew Raines wrote:
I'm getting this error on Debian 5 (Boost 1.35).
src/textual.cc:683: error: ‘is_regular_file’ was not declared in this scope
The latest code in 'next' branch should fix this.
John
On Mar 22, 2010, at 3:06 AM, Martin Michlmayr wrote:
It seems that --sort d is not honoured when -p is also used:
This is now working again.
John
On Feb 27, 2010, at 3:37 PM, thierry wrote:
On below tests, ledger v3 is 21 and 46 times slower that ledger v2.
Some specs:
- CPU is AMD Athlon(tm) Dual Core Processor 4850e, 2500MHz, 64 bits
- RAM is 1790MiB
- OS is Ubuntu 9.10
While I am working on this: how did you build ledger3? If you
Well, Ledger's slow down was not --exchange, as I thought. In fact, disabling
market valuation entirely had no effect on speed whatsoever.
What's killing us is the exact rounding that 3.0 now supports in the market
value report. That is, if the total between two lines in a register report
On Mar 13, 2010, at 1:10 PM, thierry wrote:
I've found that results of 'balance' command are different, depending
on the syntax of the amount.
Actually, I was expecting same result, whatever the way the amount is
spelled.
$ ledger -f test8.ledger bal
$270 Assets:Broker
On Mar 22, 2010, at 5:11 AM, Martin Michlmayr wrote:
With ledger v3 (from git master), print doesn't honour --date-format:
$ ledger -f b.dat --date-format %Y-%m-%d print
This has been fixed (in git next).
John
On Mar 21, 2010, at 8:31 PM, Zack Williams wrote:
I'm getting a few different errors with the git master version - most
specifically, it seems that if I use a different path to the file, or
put the file path in quotes, I either get a different error, or it
actually works:
Do you have any
On Mar 25, 2010, at 6:47 PM, Martin Michlmayr wrote:
ledger v3's print command will put the comments into the next line
when you have two semi-colons. It works correctly with only one
semi-colon. ledger v2 handled this correctly.
It's not the presence of double-semis, but the length of the
On Mar 28, 2010, at 6:26 PM, thierry wrote:
I do not understand why the --flat option removes any output in below
example. Is this a bug?
$ ledger -f test_14.ledger bal ^Expenses --depth=2 --flat
$
This is expected behavior. --depth=2 means that only accounts at depth 2 or 1
will be
Then try git reset --hard 017492ef^ instead.
John
On May 20, 2010, at 8:51 AM, ed wrote:
ledger ed$ git revert 017492ef
fatal: Commit 017492ef5e80003073c5d053252d4a68a44260ae is a merge but
no -m option was given.
Sorry, I'm barely literate in git
On May 19, 11:03 am, John Wiegley
On May 19, 2010, at 4:08 AM, ed wrote:
Yes with yesterday's update, I tend to update ledger on this machine
most days
Assuming you use git, try this:
git revert 017492ef
./acprep opt make -- -prefix=/usr/local
If that works, I know what the problem is.
To undo this change, use:
git
Something further I wanted to note about Ledger 3.0:
1. Any amount or cost expression surrounded by parentheses is a value
expression:
Expenses:Food ($10 + $3)
2. Every value expression has an evaluation context. For a regular
posting, the context is its own self:
On May 18, 2010, at 6:49 PM, John Wiegley wrote:
Something further I wanted to note about Ledger 3.0:
And for even more trickery, consider this:
= /foo/
(Expenses:Food) (any(account =~ /MasterCard/) ? amount * 0.10 : amount *
0.20)
If any posting's account matches the regex foo
On May 18, 2010, at 6:55 PM, Zack Williams wrote:
Ah... tried that. It works, but it doesn't do a straight replace of
$account with the account name. In my case, I'd like to print out a
per-employee tax amount for each of EFICA and EMED, and what seems
obvious to me would be something like
On May 17, 2010, at 9:54 AM, Simon Michael wrote:
Did you make a fix ? I haven't found it in master or next branch..
Yes, I pushed it to master last night. I reverted the commit which added the
use of `rx' in ledger.el.
John
On May 13, 2010, at 7:42 PM, Peter Ross wrote:
can I use tags in value expressions? I'm trying to set an automated
transaction that would select only tagged entries.
Here is how I automatically deduct GST (VAT in the rest of the world)
from any expenses which are tagged :gst:
= tag gst
On May 11, 2010, at 1:39 AM, Thomas wrote:
Is it possible to enter into ledger timestamped entries?
No. There's a lng story behind that, since at one point I did try to make
transactions have time-level granularity. But it becomes an absolute nightmare
when crossing timezones, etc.
So
On May 1, 2010, at 11:01 AM, thierry wrote:
I have found that the leger print command compute what's inside
parenthesis. See below.
This is another bug. The information exists to do what you want, it was simply
an oversight.
John
On Apr 30, 2010, at 12:12 PM, thierry wrote:
I have found that the leger print command filters out the balance
assertions. Actually I am using print as a pretty printer to
reformat my input file after modification by hand. I would have
expected that balance assertions are kept from print
On Apr 29, 2010, at 7:58 AM, Peter Ross wrote:
All I've found is an email which mentions using %(account), but no
other information.
Is there any documentation anywhere?
Not yet. Essentialy the new format is %[-][MAX][.MIN](VALEXPR). But it can
get a little crazy due to its flexibility,
On Apr 28, 2010, at 7:52 AM, Peter Ross wrote:
With the above the balance sheet and income and expense report don't
match. The balance sheet should change by the amount of profit earnt.
Ah, you need a bit more then:
D 1,000.00 AUD
D 1,000.00 EUR
2009/01/01 First sale
Assets:Bank
In Ledger, currencies can co-exist in the same account, so all you need is this:
D 1,000.00 AUD
D 1,000.00 EUR
2009/01/01 First sale
Assets:Bank 1.00 AUD
Income:Sales
2009/02/01 Marketing (1 EUR = 2.00 AUD)
Expenses:Marketing
On Apr 27, 2010, at 8:01 PM, Martin Michlmayr wrote:
I've tried both
ledger -f - bal Assets: '!%type=Dup'
and
ledger -f - bal '!%type=Dup' Assets:
but in both cases it shows all accounts, not just my assets.
My current workaround is:
ledger -f - print '!%type=Dup' tmp.ledger
ledger
On Apr 22, 2010, at 1:25 PM, Felipe Magno de Almeida wrote:
I've noticed a lot of exceptions being catched by pointer. Just out of
curiosity, why is this?
You must be looking at the code for 2.x? I highly recommend using what is
current in the Git project (http://github.com/jwiegley/ledger),
Ledger 2.6 can read OFX directy if you had libofx installed when you
built Ledger.
John
On Apr 18, 2010, at 3:00 PM, Zack Williams zdw...@gmail.com wrote:
What's the best way to convert an OFX file I've obtained from a
financial institution into a ledger file? I'd prefer to do the
On Apr 18, 2010, at 5:35 PM, Zack Williams wrote:
I know it can, I'm just wondering what CLI or file syntax I need to
use to make it happen, which I can't seem to find.
Just: ledger -f file.ofx print
I have ledger 2.6.2 compiled out of macports with the ofx variant
installed, so I'd assume
This is actually already implemented in ldg-post.el, part of the new
ledger-mode. :) Our implementations are pretty similar.
John
On Apr 14, 2010, at 3:57 PM, David Glasser wrote:
(defun hash-keys (hashtable)
Return all keys in hashtable.
(let (allkeys)
(maphash (lambda (kk vv) (setq
On Apr 13, 2010, at 4:10 PM, David Glasser wrote:
(While on the topic of ledger.el: it seems to only have partial
support for multi-file-!include setups: there's a ledger-master-file
but only the report commands use it, not reconcile mode, add-entry, or
completion. I think I see how to make
Could you implement something by simply using metadata?
John
On Apr 11, 2010, at 5:20 PM, David Glasser wrote:
I'm just starting to use ledger to implement finances for an
organization. The organization is a home with a bunch of folks where
we share costs. I'm thinking that the virtual
On Apr 11, 2010, at 5:29 PM, David Glasser wrote:
Hmm, well, other than the fact that I can't use metadata since I don't
know how to get a 3.0 build on Mac 10.5 :) I think that would work
for the simplest bit of this (if I used a BudgetUtil201004 tag and
tagged postings with it) but that
On Apr 10, 2010, at 3:09 PM, David Glasser wrote:
When I'm using the emacs mode and I hit enter after the first line of
the entry or the first posting, I basically always want to indent.
Anyone know the magic to make that happen?
Usually one binds RET to `indent-according-to-mode', but I'm
On Apr 10, 2010, at 6:36 PM, Zack Williams wrote:
Specifically, if you use a single whitespace character between the end
of the account name and the amount in a transaction, it complains
could not balance this transaction (too many missing amounts),
because I think it's including the amount
On Apr 5, 2010, at 2:11 AM, Martin Michlmayr wrote:
Would it be possible to print an error when a bogus date is found?
Believe it or not, the strptime function itself does not validation of dates.
I'd love to do full validation, but it will require writing my own parser at
some point.
John
On Apr 2, 2010, at 12:45 PM, Martin Michlmayr wrote:
ledger bal will also put this commodity in quotation marks. Is this
necessary? IMHO it would be nicer if ledger bal would strip the
quotation mark.
I think that should be doable. It's really only ledger print that must quote
them.
John
On Apr 2, 2010, at 2:02 PM, Martin Michlmayr wrote:
Does ledge have some ifdef like capability that would allow me to
exclude certain transactions if a given file is included?
Usually the way I solve scenarios like this is with a meta file. For
example, you have two files: year2009.ledger
On Apr 2, 2010, at 2:12 PM, Martin Michlmayr wrote:
I agree that this meta file approach would work but it seems a bit
like a hack to me. Don't you think the ifdef logic I suggested
would be useful?
Well, there is another, less hackish approach that Simon's post made me think
of. Just add
Thank you very much for these excellent bug reports, Martin. I will
add all of to the list.
Hello from the It's A Small World ride at Disney World, btw. Gotta
love the iPhone. :-)
John
On Mar 25, 2010, at 5:45 PM, Martin Michlmayr t...@cyrius.com wrote:
Given the file
2008/03/03
A
It is a bug.
John
On Mar 25, 2010, at 6:47 PM, Martin Michlmayr t...@cyrius.com wrote:
Given this input file:
2008/03/03
A 100 @@ 1000.00 EUR ; @ 10.00 ; @ 10.00
B
ledger v3's print command will put the comments into the next line
when you have two semi-colons. It
On Mar 22, 2010, at 5:11 AM, Martin Michlmayr wrote:
With ledger v3 (from git master), print doesn't honour --date-format:
Yes, this is intended. The output of print is supposed to be directly
parseable by Ledger, which means the date is always written out in the same
form in which it
On Mar 16, 2010, at 12:07 AM, Hilco Wijbenga wrote:
I'm not sure this is actually a problem but I thought I should mention
it. When building the master branch, I get broken pipes at the very
beginning (see below). The build does (appear to) finish correctly.
(The test command ./ledger -f
On Mar 16, 2010, at 12:36 AM, Hilco Wijbenga wrote:
$ git checkout -b maint origin/maint
$ ./acprep update
acprep is the maint branch is ancient, and doesn't work this way. You have to
run ./acprep, and then do the configure/make yourself.
John
On Mar 16, 2010, at 12:49 AM, Hilco Wijbenga wrote:
You might want to make that a bit clearer on
http://github.com/jwiegley/ledger/. Especially, since you prefer
people use the BETA. The way I read it, you're supposed to use
./acprep update.
Yes, at this point the master is actually more
On Mar 13, 2010, at 1:10 PM, thierry wrote:
I've found that results of 'balance' command are different, depending
on the syntax of the amount.
Actually, I was expecting same result, whatever the way the amount is
spelled.
That's a great bug, Thierry. Thank you!
John
On Mar 11, 2010, at 12:15 PM, Greg wrote:
Anybody ever use a memo field with ledger? I need to put short
descriptions in for each transaction.
Example:
2004/04/09 Viva Italiano
Expenses:Food $11.00
Expenses:Tips$2.50
Liabilities:MasterCard $-13.50
On Mar 1, 2010, at 10:39 AM, Thierry Daucourt wrote:
What I can not understand is why ledger -V output of test32.ledger is in FRF,
even if I put directive D 1000,00 EUR.
Is there a way to force the output in EUR, my currency ?
Currency selection with -V can be very complex. Use -X EUR.
Thierry,
It looks like you've exposed a somewhat pathological case with regard to
multiple currencies. Ledger is doing far too much make work than it should be.
Is there any chance I can gain access to your data file? Use ledger print
--anon may, in your case, yield a file which still
Hmm... I never expected people to use spaces in unquoted regexps You can
do this:
/Frais financiers/
But I suppose space should be allowable too.
John
On Mar 1, 2010, at 4:08 PM, thierry wrote:
I found an unexpected result while using the reg command, with
regular expression
On Feb 28, 2010, at 5:28 AM, Thierry Daucourt wrote:
How did you build it?
cd ledger ./acprep update
Try using acprep opt update instead, since right now you are building with
full debugging on. That tends to make things many times slower.
John
On Mar 1, 2010, at 12:51 AM, Peter Keen wrote:
Is it normal for the master branch of ledger to take hours to compile?
It's going on two and a half right now in the middle of acprep opt
make. I would attach the actual output but my system is just sitting
there mostly locked up compiling.
On Mar 1, 2010, at 1:16 AM, Peter Keen wrote:
On Sun, Feb 28, 2010 at 10:07 PM, John Wiegley jwieg...@gmail.com wrote:
It all depends on your machine and what else it's doing. On my MacBook Pro,
with 3G of RAM, I can run this:
acprep -j2 out update
But only if I'm not doing anything
1001 - 1100 of 1223 matches
Mail list logo
