rlib-users — RLIB Users Mailing List

Re: [Rlib-users] problem with r.detailcnt ( RLIB 1.3.4 )]

[Bob D. <bd...@si...>]

Configure it with --disable-doc

- bob



On Thu, 2005-07-07 at 17:02 -0400, Bob Doan wrote:
>         When running make (1.3.4) on Centos 4.0:
>         
>         Package Fancyhdr Warning: fancyhdr's E option without twoside
>         option is 
>         useless
>          on input line 461.
>         LaTeX Warning: Reference `INTRO' on page iii undefined on
>         input line 518.
>         LaTeX Warning: Reference `27' on page iii undefined on input
>         line 542.
>         (tons of these)
>         .....
>         Error: pdfjadetex: libpng: internal error
>          ==> Fatal error occurred, the output PDF file is not
>         finished!
>         ......
>         
>         
>         Any ideas?

-- 
Bob Doan <bd...@si...>

Re: [Rlib-users] problem with r.detailcnt ( RLIB 1.3.4 )]

[Bob D. <bd...@si...>]

Hi,

Is this happening in PDF or HTML or both?

Also can you add a column and just print out r.detailcnt

And see what comes out (does the number continue to increase?)

- bob

On Thu, 2005-07-07 at 17:02 -0400, Bob Doan wrote:
>         Hi,
>         I am using r.detailcnt to colorize my rows, but
>         at page 4 of my report it don't works anymore.
>         I am using:
>         <Line bgcolor="iif(r.detailcnt%2,'white','0xebebeb')">
>         
>         Please how I can solve this?
>         
>         -- 
>         Everton Luis Berz
>         Nucleo de Sistemas :: FACCAT - Faculdades de Taquara
>         +55 51 541 6600 - R.647
>         ICQ 7807919
-- 
Bob Doan <bd...@si...>

Re: [Rlib-users] problem with r.detailcnt ( RLIB 1.3.4 )

[Everton L. B. <ev...@fa...>]

I am using rlib 1.3.4. At 1.2.1 the bug not happens.
-- 
Everton

Everton Luis Berz escreveu:
> Hi,
> I am using r.detailcnt to colorize my rows, but
> at page 4 of my report it don't works anymore.
> I am using:
> <Line bgcolor="iif(r.detailcnt%2,'white','0xebebeb')">
> 
> Please how I can solve this?
> 

[Rlib-users] Make Error: pdfjadetex

[Stephen J. L. Jr. <sla...@o1...>]

When running make (1.3.4) on Centos 4.0:

Package Fancyhdr Warning: fancyhdr's E option without twoside option is 
useless
 on input line 461.
LaTeX Warning: Reference `INTRO' on page iii undefined on input line 518.
LaTeX Warning: Reference `27' on page iii undefined on input line 542.
(tons of these)
.....
Error: pdfjadetex: libpng: internal error
 ==> Fatal error occurred, the output PDF file is not finished!
......


Any ideas?

[Rlib-users] problem with r.detailcnt

[Everton L. B. <ev...@fa...>]

Hi,
I am using r.detailcnt to colorize my rows, but
at page 4 of my report it don't works anymore.
I am using:
<Line bgcolor="iif(r.detailcnt%2,'white','0xebebeb')">

Please how I can solve this?

-- 
Everton Luis Berz
Nucleo de Sistemas :: FACCAT - Faculdades de Taquara
+55 51 541 6600 - R.647
ICQ 7807919

[Rlib-users] RLIB 1.3.4 Pre Release

[Bob D. <bd...@si...>]

All,

If anyone has time can you try:
http://www.sicom.com/~bdoan/rlib-1-3.4.tar.gz

This release includes numerous bug fixes along with pretty serious
double free bug in RPDF which was cauing RLIB to crash randomly and/or
you could not run RLIB in a loop generating reports.

Let me know if it works for you

- Bob

[Rlib-users] GCC4 warning fixes part4

[Zoltan B. <zb...@du...>]

Hi,

here's the next part. Changelog:

***********************************************************
Silence warnings in datasource.c. The original code gave these
kinds of warnings on GCC4:
warning: ISO C forbids conversion of function pointer to object pointer t=
ype
warning: ISO C forbids initialization between function pointer and 'void =
*'
warning: ISO C forbids assignment between function pointer and 'void *'
The above warnings sound serious but on the other hand, explicitly
casting the address of a function pointer to a (gpointer *) gave this
warning on older GCCs:
warning: dereferencing type-punned pointer will break strict-aliasing rul=
es
Aliasing function pointers to gpointer using unions solves this, using
   union {
     gpointer ptr;
     void (*function)(void);
   }
allows the assignment to the gpointer and calling function() without
warnings.
***********************************************************

Best regards,
Zolt=E1n B=F6sz=F6rm=E9nyi

[Rlib-users] Need Help in Report

[Nasir B. <na...@ge...>]

Hi There,

I am trying to develop a PHP script which is generating pdf report using =
RLIB. I require to make a report for YTD  total data for multiple =
accounts. In addition to YTD total  I am also required to display =
transactions for each account for a particular day. I have two queries. =
One calculates YTD total  for each account and the other gives previous =
day's transactions for each account. It is possible that accounts may =
not have any transaction for this day.

I want to display the output in following format :


Account No   1                                                           =
               YTD Total                      Query:1
			=09
			Transaction 1
			Trassaction 2							Query:2
			Trassaction 3
-------------------------------------------------------------------------=
-------------------------------------------
Account No   2                                                           =
               YTD Total                      Query:1
			=09
			Transaction 1
			Trassaction 2							Query:2
			Trassaction 3
-------------------------------------------------------------------------=
-------------------------------------------
Account No   3                                                           =
               YTD Total		Query:1
-------------------------------------------------------------------------=
--------------------------------------------
Account No   4                                                           =
               YTD Total		Query:1
-------------------------------------------------------------------------=
--------------------------------------------
Account No   5                                                           =
               YTD Total                      Query:1
			=09
			Transaction 1
			Trassaction 2							Query:2
			Trassaction 3
-------------------------------------------------------------------------=
-------------------------------------------
Account No   6                                                           =
               YTD Total                      Query:1
			=09
			Transaction 1
			Trassaction 2							Query:2
			Trassaction 3
-------------------------------------------------------------------------=
-------------------------------------------
Account No   7                                                           =
               YTD Total		Query:1
-------------------------------------------------------------------------=
------------------------------------------=20

As you can see some of the accounts do not have any trasaction listed. I =
tried to use the break feature in XML, but I am unable to display the =
accounts
that do not have any transactions.

Please suggest me how I can solve this issue.

Thank you,

Nasir Baig

Gelber Group, L.L.C.

Re: [Rlib-users] Rlib 1.3.3 make error

[Bob D. <bd...@si...>]

Try http://www.sicom.com/~bdoan/rlib-1.3.4.tar.gz

I think I fixed them all. (I sometimes compile on a RH 6.1 machine)

Also for those of you with outstanding patches I'll review them this
weekend.  I'm really busy @ work and @ home.  sorry guys

- bob

On Thu, 2005-06-23 at 12:38 -0400, Shannon Weyrick wrote:
> The problem is that function (and a few other's in rlib) has a variable 
> declaration after code. Newer versions of gcc don't care but older 
> versions do. If you reorder the variables a bit it compiles. Sorry I had 
> a patch for this but I've misplaced it - luckily it's a pretty quick fix.
> 
> Shannon
> 
> Michael McInally wrote:
> > ./configure --prefix=/usr --disable-gd --disable-python
> > --disable-postgre
> > 
> > make
> > 
> > graphing.c: In function `rlib_graph':
> > graphing.c:244: parse error before `minor_tick'
> > graphing.c:246: `minor_tick' undeclared (first use in this function)
> > graphing.c:246: (Each undeclared identifier is reported only once
> > graphing.c:246: for each function it appears in.)
> > make[2]: *** [graphing.lo] Error 1
> > make[2]: Leaving directory `/home/mcinallym/rlib-1.3.3/libsrc'
> > make[1]: *** [all-recursive] Error 1
> > make[1]: Leaving directory `/home/mcinallym/rlib-1.3.3'
> > make: *** [all] Error 2
> > 
> > 
> > 
> > Can't I disable graphing somehow?  Or is this a problem somewhere else?
> > 
> > 
> > 
> 
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
> from IBM. Find simple to follow Roadmaps, straightforward articles,
> informative Webcasts and more! Get everything you need to get up to
> speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
> _______________________________________________
> Rlib-users mailing list
> Rli...@li...
> https://lists.sourceforge.net/lists/listinfo/rlib-users

Re: [Rlib-users] Rlib 1.3.3 make error

[Shannon W. <we...@ro...>]

The problem is that function (and a few other's in rlib) has a variable 
declaration after code. Newer versions of gcc don't care but older 
versions do. If you reorder the variables a bit it compiles. Sorry I had 
a patch for this but I've misplaced it - luckily it's a pretty quick fix.

Shannon

Michael McInally wrote:
> ./configure --prefix=/usr --disable-gd --disable-python
> --disable-postgre
> 
> make
> 
> graphing.c: In function `rlib_graph':
> graphing.c:244: parse error before `minor_tick'
> graphing.c:246: `minor_tick' undeclared (first use in this function)
> graphing.c:246: (Each undeclared identifier is reported only once
> graphing.c:246: for each function it appears in.)
> make[2]: *** [graphing.lo] Error 1
> make[2]: Leaving directory `/home/mcinallym/rlib-1.3.3/libsrc'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/home/mcinallym/rlib-1.3.3'
> make: *** [all] Error 2
> 
> 
> 
> Can't I disable graphing somehow?  Or is this a problem somewhere else?
> 
> 
> 

[Rlib-users] Rlib 1.3.3 make error

[Michael M. <mci...@pi...>]

./configure --prefix=/usr --disable-gd --disable-python
--disable-postgre

make

graphing.c: In function `rlib_graph':
graphing.c:244: parse error before `minor_tick'
graphing.c:246: `minor_tick' undeclared (first use in this function)
graphing.c:246: (Each undeclared identifier is reported only once
graphing.c:246: for each function it appears in.)
make[2]: *** [graphing.lo] Error 1
make[2]: Leaving directory `/home/mcinallym/rlib-1.3.3/libsrc'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/mcinallym/rlib-1.3.3'
make: *** [all] Error 2



Can't I disable graphing somehow?  Or is this a problem somewhere else?



-- 

Michael McInally
Software Developer
Picasso Fish
mci...@pi...
(519)-426-8981

Re: [Rlib-users] Forcing the Euro symbol

[Bob D. <bd...@si...>]

The main problem is w/ PDF fonts.  Adobe doesn't have a courier font
that has the Euro in it.  So you would have to use a custom font (Which
RPDF does not support yet)

- bob

On Wed, 2005-06-15 at 15:55 -0500, Derek Giromini wrote:
>=20
> How can I force the Euro symbol to appear without changing
> localization? US-ASCII should have the symbol, but I can't find a
> reliable way to force it to appear.
>=20
> In short, I'd like format string like: format=3D"'=E2=82=AC%10.2f'"
>=20
> --=20
> Derek Giromini=20
> Gelber Group, L.L.C.=20
> desk: +1.312.692.2843=20
> pager: +1.312.689.0378
>=20

[Rlib-users] Forcing the Euro symbol

[Derek G. <dgi...@ge...>]

How can I force the Euro symbol to appear without changing localization? =
US-ASCII should have the symbol, but I can't find a reliable way to =
force it to appear.

In short, I'd like format string like: format=3D"'EUR%10.2f'"

--
Derek Giromini
Gelber Group, L.L.C.
desk: +1.312.692.2843
pager: +1.312.689.0378

Re: [Rlib-users] undefined symbol: rlib_add_datasource_postgre

[Bob D. <bd...@si...>]

Recompile RLIB w/ --disable-postgre

FYI RLIB 1.3.0 is really old

Also if you don't mind me asking what are you using RLIB for?

- bob

[Rlib-users] undefined symbol: rlib_add_datasource_postgre

[Michael M. <mci...@pi...>]

OK, I have no idea what happened.

My system configuration has not changed in months.

I have been using the RLIB reporting module for reports in a development
system.  RLIB is also installed on the production server, and is still
working.


Just today, I attempted to view my reports on the dev system, and I
receive this message:


Warning: dl(): Unable to load dynamic library '/usr/lib/php4/librlib.so'
- /usr/lib/php4/librlib.so: undefined symbol:
rlib_add_datasource_postgre in /...dir.../report_volunteers.php on line
4


I tried to re-compile RLIB 1.3.0 with:

./configure --prefix=/usr --disable-utf8
make clean
make
make install

Restarted mySQL
Restarted apache


It still reports the error.  The script does not use postgresql, so I do
not know why RLIB is trying to find it...



-- 

Michael McInally
Software Developer
Picasso Fish
mci...@pi...
(519)-426-8981

[Rlib-users] Re: accent problems

[Bob D. <bd...@si...>]

Carol,

Download RLIB 1.3.3  And just configure it with --prefix=/usr

It will fix all of your problems.

Also

**DRUM ROLL**

We now are working on a WIKI manual 

http://newrlib.sicom.com


Here's what I have started on the function reference so far

http://newrlib.sicom.com/index.php?title=Documentation_Function



Please help edit the pages and improve the documentation.  I will be out
of town until Friday but Mike @ SICOM Can help you out

Mike will be spending time daily working on the documentation.  Other
people from the community will help also.  Together we can make a
rocking manual :)
> 
> I have tried on several occasions to figure out the downloads and
> installs.  I think I have the docbook-dtds, but the
> docbook-style-dsssl is not clear to me on how to install it.  The
> install directions that are there (actually on a website link) look
> like they are for a windows app.  I feel too pressed to get things
> done to spend a lot of time with trying to get all this.  Is your
> expectation that anyone who downloads all the stuff to install rlib is
> also going to download all this other stuff for the documentation?
> 
Not any more

Cheers!

- bob

[Rlib-users] New C# Bindings and Such

[Bob D. <bd...@si...>]

All,

I resynced all the bindings to the latest API.  This time I wrote a
script to do it so doing this in the future will be much easier and less
error prone

I also added c# bindings which work! :)

So At this moment we have

API: c, php, perl, python, java, c#

INPUTS: mysql, postgresql, xml, array data sources, odbc (which brings
in tons of other databases)

OUTPUTS: PDF, HTML, TXT, CSV

40+ functions inside the RLIB engine

Parts, graphs, custom functions, split details (mailing labels) call
backs, iterations, spec files, utf8, locales, and a working win32 port.

All wrapped up in a nice GPL package

I never could have imagined how much RLIB was going to improve when we
GPL'd it 2 years ago, and I think the best is yet to come.

Here are my plans:

For the next 2-3 months I want to make the documentation perfect and
provide nice RPMS/ Windows installers to get RLIB

I worked more on the WIKI some more over the weekend:

http://newrlib.sicom.com/

Anyone can add documentation.  Please document things that you know of.
If you don't know what something does just stub it out w/ a TODO.

I plan to have a lot of examples in the documentation.  You can see that
from what I have so far.  Please please please help document.  I suck at
it.

Thanks!

- bob

-- 
Bob Doan <bd...@si...>

[Rlib-users] HTML problem reproduction

[Shannon W. <we...@ro...>]

Greets,

I worked on tracking the HTML problems down more. I found where the 
problem is, but couldn't muster the skill to eradicate it.

The first problem was the one I mentioned in the last email - if 
trim_links is turned on, it's not closing <span> tags for bgcolor in 
some cases. The fix I had for that is attached.

The next problem is related and again specific to trim_links and 
bgcolor. The problem appears to be in 
rlib_layout_find_common_properties_in_a_line(), which also gets my 
nomination for Longest Function name.

The second part of the function loops through extra_data looking for 
where to start and stop link flags. But if trim links is on there is a 
bgcolor for that Line then in some situations it gets out of sync and 
again ends up with mismatched <span> tags. This all happens around lines 
637-662 in layout.c

I've attached a sample php script and xml report file that demonstrates 
the problem in various ways (including a work around). Also it 
demonstrates the fact that trim_links only works if the link text is 
left aligned and doesn't work if it's center or right aligned.

Bob, regarding the new Doc Wiki - I think it's a great idea. I'd love to 
contribute but unfortunately I'm one of the one's in the dark about most 
of the functionality that's not documented in the current docs :)

Thanks,
Shannon

[Rlib-users] Calling all documentation volunteers

[Bob D. <bd...@si...>]

In order to improve the RLIB Documentation I'm going to move it to a
wiki where selected people can edit it/improve it.

Right now I have set up a wiki (Media Wiki) where the documentation will
be.  It's very easy to edit it.

http://newrlib.sicom.com

Who wants to help????

- bob

-- 
Bob Doan <bd...@si...>

[Rlib-users] Re: [Rlib-devel] RFE Real national encodings, etc

[Bob D. <bd...@si...>]

On Sat, 2005-05-14 at 08:51 +0200, Zoltan Boszormenyi wrote:
> Hi,
> 
> I came across this site: http://www.fpdf.org/
> This stuff can also create PDFs but can also embed
> fonts in given charsets in them. The whole stuff
> is written in PHP and the license is freeware
> for any (business or home) use so we could rewrite
> it in C for RPDF.

I was going to use that as my model for the font loading.  Problem is I
have been so busy w/ other things I have not had the chance.  My guess
is that you should be able to port it ;)

If you make the font loading work in RPDF I can handle the plumbing in
RLIB so it could use the different "fixed" fonts

> 
> Also, how about modifying RLIB and RPDF in such a way
> that field width could be given in some metric units
> and also be able to use proportional fonts?
> I showed RLIB to some of my collegues and they said
> the reports would be nicer if the font wouldn't be courier.
> PDF provides 14 default fonts and also the embedded ones.

Two problems here.  Having different field heights on the fixed font in
RLIB would not be too hard at this point because a few months ago I made
RLIB inspect the whole "line" for the max height because of images on a
line.

Non Fixed fonts are going to be harder.  RLIB currently uses a FIXED
layout engine and everything really boils down char counts of things.
Like the "first_name" field is 20 chars long.  RLIB knows say the width
of a char of a 12 point font so it knows it will take up say
20(chars)*12(point)*(600 pix per point)/72.00(DPI)

Rlib won't be able to do this because width_of("W") != width_of("i").

There are two ways to solve the problem.  One is allow the user to
specify the width of a field in percent so say first_name width is "20%"

This is not exactly ideal because someone's first name might be
"WWWWWWWWWWWWWWWWWWWWWWWWWWWW" which would be really wide and might not
fit.

The proper thing to do would then be to change RLIB from a 1 pass payout
engine to a 2 pass layout engine for non fixed with fonts.  It would lay
things out much like HTML does.  By first going though and looking at
the widths of all the fields in a table and then determine the final
width of the columns that way.

But before any of this can happen we need to have font loading in
RPDF ;)

> 
> Also, it would be nice if the manual was more comprehensive.

Agree.

> I am just started using RLIB and the manual doesn't help too much.
> I haven't used any reporting software before, so I am a real newbie
> in this field. E.g. the manual doesn't document all internal variables,
> I know totpages  exists because you discussed it on the mailing list.
> It required a fair amount of experimenting to find out that
> a sum of a field across all records needs to be in <ReportFooter>
> to be displayed at the end and doesn't need any breaks.
> It's logical but not documented. A dummies' guide for RLIB with many
> examples, please! ;-)

We're working on it.  I think I'm going to move the RLIB documentation
to a wiki for a little while so everyone in the community can edit it.

*************************************************************

Who out there in user land would be willing to help out????????

**************************************************************

For examples see: src/examples/php

But we really need to beef things up.

FYI this weekend I plan to poke @ c# support in RLIB and a script to
automatically generate all the bindings and what not
-- 
Bob Doan <bd...@si...>

Re: [Rlib-users] patch, problems, question

[Bob D. <bd...@si...>]

> Patch works great, thanks.

Anyone out there in User Land know if any other report writer can allow
you to do this?????  It seems like a nice features to have...

> 
> I researched the HTML problem I was having and it turns out to be a bug 
> related to trim_links. Circa line 279 in layout.c, if it does the trim 
> links a call to rlib_layout_output_extras_start is made which (in html) 
> starts a <span> tag if there is a background color. However, there  is 
> never a call to _end after that so that span tag is never closed.
> 
> I tried to hack this by adding a call to end_draw_cell_background() and 
> it fixes it somewhat but I'm pretty sure this isn't correct, or at least 
> all that needs to be done.
> 

Send a patch and let me have a look

Re: [Rlib-users] patch, problems, question

[Shannon W. <we...@ro...>]

Bob Doan wrote:
> Attached is an untested patch which should fix the problem
> 
> let me know
> 
> - bob

Hey Bob,

Patch works great, thanks.

I researched the HTML problem I was having and it turns out to be a bug 
related to trim_links. Circa line 279 in layout.c, if it does the trim 
links a call to rlib_layout_output_extras_start is made which (in html) 
starts a <span> tag if there is a background color. However, there  is 
never a call to _end after that so that span tag is never closed.

I tried to hack this by adding a call to end_draw_cell_background() and 
it fixes it somewhat but I'm pretty sure this isn't correct, or at least 
all that needs to be done.

Please advise :)

Shannon

Re: [Rlib-users] patch, problems, question

[Bob D. <bd...@si...>]

Attached is an untested patch which should fix the problem

let me know

- bob

On Fri, 2005-05-13 at 17:30 -0400, Shannon Weyrick wrote:
> > 
> > And then in the XML do:
> > 
> > <field value="bobdoan(mikeroth('bacon', 'eggs'))" width="50"/>
> > 
> > http://www.sicom.com/~bdoan/rlib-1.3.4.tar.gz
> > 
> > see src/examples/php/array[xml|php] as an example
> > 
> > - Bob
> 
> Hey Bob,
> 
> Quick (hopefully) simple bug: the arguments are showing up to the PHP 
> function in the reverse order. In your example, one would expect it to 
> show up as "Bacon and Eggs" in the output but in fact it shows up "Eggs 
> and Bacon".
> 
> Other than that it looks great, thanks a lot.
> 
> Shannon
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click
> _______________________________________________
> Rlib-users mailing list
> Rli...@li...
> https://lists.sourceforge.net/lists/listinfo/rlib-users
-- 
Bob Doan <bd...@si...>

Re: [Rlib-users] patch, problems, question

[Shannon W. <we...@ro...>]

> 
> And then in the XML do:
> 
> <field value="bobdoan(mikeroth('bacon', 'eggs'))" width="50"/>
> 
> http://www.sicom.com/~bdoan/rlib-1.3.4.tar.gz
> 
> see src/examples/php/array[xml|php] as an example
> 
> - Bob

Hey Bob,

Quick (hopefully) simple bug: the arguments are showing up to the PHP 
function in the reverse order. In your example, one would expect it to 
show up as "Bacon and Eggs" in the output but in fact it shows up "Eggs 
and Bacon".

Other than that it looks great, thanks a lot.

Shannon

Re: [Rlib-users] patch, problems, question

[Shannon W. <we...@ro...>]

Bob Doan wrote:
> 
> And then in the XML do:
> 
> <field value="bobdoan(mikeroth('bacon', 'eggs'))" width="50"/>
> 
> http://www.sicom.com/~bdoan/rlib-1.3.4.tar.gz
> 
> see src/examples/php/array[xml|php] as an example
> 
> You owe me a crash fix now ;)
> 
> - Bob
> 

Woo hoo :)

Looks great, I'll start playing with it shortly. That's going to be 
really powerful. For example I know I can use that right away with 
urlencode() because some of my "link" attributes need it.

I'll definitely look in the crash bug.

You might also be getting another patch for the html renderer. I'm 
actually having a lot of problems with bad generated HTML breaking the 
pages and it's kind of critical for me to fix that now so I'm about to 
dive in.

Thanks,
Shannon