on scanning on linux (market investigation)

he order was to convert the data on the official letter to computer data. Therefore a tool chain of scan device and ocr, thinkable.

  • scan device: device with paper
  • auto feed in Android camera photos of papers

which are to be:

  • non interactive
  • free software
  • free software interfaceable

This is a matrix of 2*3 to the question: Is there a solution available, but if not: in reach? how thinkable? because of most exactly what and most exactly whom hindered to now.

Out of the hips one could previously shoot, that there have been reported of being the printer industry corrupt in high degree with respect to vendor-lock-in. Crooks all. So it does wonder that in the year 2015 there is afaik no free software solution in the world. And I would suspect as the devices are “multi”-compact their scanner they make the prefer deliver them defunct. Yes, that precisely their role. Enemies in enemy land. Is something wrong on my count then I would not be glader than proving the opposite. So the truth of it all is that they – because of hostile manner – hold those all but the last product arbitrarily back. Their PR and their brain wash is all about that, in the governing everywhere. We will find it each corner you will see… And one could anyway take time to make sure if the support for a future solution is on a decent way, btw and by all, no? Which is not more then our lethal interest…

Some first stakes:



Das Scannen läuft über die standardisierte SANE-Schnittstelle – siehe Scanner. Der Automatische Dokumenteneinzug wird auch unterstützt – unter XSane muss man dafür ein Häkchen bei “batch scan mode” setzen.

Ubuntu users say their where one. Fine then? Who mine? No? I can not imaging it was mine, neither can I imagine hp could do a good thing. To gossip, my hp under windows he make me sneak strict pathes; if not he makes me reboot repeatedly because of not using hp-cardridge, which surely can cost or commit some millions of angry user interaction hours. For a long time you would have few lust to trust them, which leads to the certain: You simply can not trust them. What will run, will run bad and hurts you effectively the more compared to what you should expect fairly: So exactly designed.




On a adf-ocr-scanner solution for ubuntu linux


I am needing to set my bureau the most possible to free software. Therefore I search four the functionality, which is a chain out of the components:

staple of documents->scanner device->adf->ocr->open format-> computer data.

Yes and the next requirement is: most possible non interactive. And: ocr-retrieved text inlayed in original scanned image.

Just in case of question what fore:

I am looking to have it integrated in a self build bureau environment software like tryton with python.I would like to grasp an orientation on the current market situation, in order “make sure if the support for a future solution is on a decent way.” (E.g. by sending an request to the fsf priority issue list, if this is a good idea.)

From this one could suggest that it can not be any more very far from it:



So question to you:

-Have you know of such an entire tool chain already working that way? Could you imagine one? Please share your thoughts and opinions unhinderedly.

-Have you heard of/imagined a free software ocr into that scanned output on linux is actually piped to? Which one?

-Which are in your opinion the weakest points in the chain? Is it availability of scanner or adf-capability for linux? Is it the use of closed formats such as pdf?

-ocr-retrieved text inlayed in original scanned image. is this possible? (see screenshot below)adfocr13

Hint: The black stripes is text selection in black color. Ocr did not change the original appearance.

-how do these commercial linux bureaus around fill the feature gap?


Thanks in advance

GNU General Public License

HPLIP (HPLinux Imaging and Printing) project—initiated and led by Hewlett-Packard (HP)

they have no community…

but I comforted somewhat that true.

As of 2013 the supplied printer-drivers support a total of 2,080 HP printer models; many of these are free and open-source (FOSS), licensed under MIT, BSD, and GPL licenses, but others

A GUI toolbox to access all these functions in a friendly way
have you heard that.
* HPAIO SANE backend (hpaio) for flatbed and Automatic Document Feeder
(ADF) scanning using MFP devices


my device is not in the list. Sending them an email What is about my version? prove it? Bugs? Tutorial?



Xsane. …, select ‘Multipage’ (Ctrl+M).

.$ scanimage -b –batch-count=n –source ADF where ‘n’ is number of pages to scan.

From hp-scan command:.

$ hp-scan –adf

looks clean

give it a try thogh.. apt-get and batch somewhat? not?

my computer is somewhat slow,


on proteusing gnuhealth Hello World (2) (login aim table)

Hi there,

on the login aim table

I believe there is no come through through that. Do they have me lock cracking as first trainings session? If that is a design, what kind of design might that be? Who said it?

I got some tips on an alternative method: config.set_xmlrpc(‘http://admin:gnusolidario@health.gnusolidario.org:8069/health28’). So I have together with the manual at least 3 methods for connecting. I have what? I have various kinds of ip-hosts-tryton-server (online to gnuhelth-url to the debian packaged server, to an from scretch out of sources-server, to any other server for example made by a demo script) together fouŕ, to connect to,

(-where is not clear what this tcp port stuff shall and in how far a firewall is needed. One needed a clear network picture about the 3Tiers-protocol, btw. – Which we can not adopt that easy cause they do not let us (log)-in. btw, too.)

And? Yes for each method, I have a source, example, manual url, where the function is documented – and of cause the error message for each one, which is status quo then.

Ok, than we have a crappy html table 3 columns. The bold n indicate which solution we are especially after.

Here is the table


Installed:proteus? by apt-get ii tryton-proteus 3.0.0-3
and gnuhealth:https://en.wikibooks.org/wiki/GNU_Health/Installation#Downloading_and_Installing_GNU_Health
wget [http://ftp.gnu.org/gnu/health/gnuheal …] Uncompress the file: tar xzf gnuhealth-latest.tar.gz cd gnuhealth-* Run the gnuhealth_install.sh script ./gnuhealth_install.sh 3.4.3 Installed versions: ii tryton-proteus 3.0.0-3; trytond-3.4.3 

host method resultn:defunct
gnuheath (local ,handbook) Ankername1./trytond –logconf /home/gnuhealth/gnuhealth/tryton/server/config/gnuhealth_log.conf -v –all -d gnuhealth #from console 1, “five minute long module -work seen in logs. Then: “Update/Init succeed!“, server stopsexport Iapt-get remove tryton-servePYTHONPATH=”/home/gnuhealth/gnuhealth/tryton/server/trytond-3.4.3/bin” ;ipython # console 2from proteus import config, Modelconfig.set_trytond(database_name=’gnuhealth’,database_type=’postgresql’, password=’444nh4′,config_file=’/home/gnuhealth/gnuhealth/tryton/server/config/trytond.conf’)Last error message: ImportError: cannot import name CONFIGn; Exception: Module health not found; 5 minutes later: KeyError: ‘res.user, Exception: Module health_pediatrics not found, Exception: Module health_lifestyle
mportError: No module named trytond.configresult: or is it more a feature gap (debian standard system only)  as not all necessary paths are can not be informed to proteus arbitrarily?
n! verified to be not possible
manual y
pip install “proteus >=3.4,<=3.5cat /home/gnuhealth/gnuhealth/tryton/server/config/trytond.conf|grep -A1 xmpm|grep -v “xmpm\|–”
listen = localhost:8069
xmlrpc = *:8069
ssl_xmlrpc = Falsessl_xmlrpc = False
[Mon May 18 15:20:21 2015] INFO:trytond.server:starting JSON-RPC protocol on localhost:8000
[Mon May 18 15:20:21 2015] INFO:trytond.server:starting XML-RPC protocol on localhost:8069


pokoli,cedk: You were right, my user mistake have been not to follow the hint in the official config-file.

But one needed the handbook to exactly, in order to come up with the idea of building a own group with brackets [] in the config file. Therefore the judgement on this lock picking run is not closed.


gnuhealth (online) -.
manual y
pip install “proteus >=3.4,<=3.5cat /home/gnuhealth/gnuhealth/tryton/server/config/trytond.conf|grep -A1 xmpm|grep -v “xmpm\|–”
listen = localhost:8069
xmlrpc = *:8069
ssl_xmlrpc = Falseconfig.set_xmlrpc(‘http://admin:gnusolidario@health.gnusolidario.org:8069/health28&#8217;)


3. Connect to the GNU Health Demo Database using the following information:
Server: health.gnusolidario.org:8000
Database: health28
User name: admin
Password: gnusolidario

error: [Errno 111] Connection refused
It does not connect because of:

1. firewall

2.not sure if gnusolidario org goes xmlrpc, if not I cannot connect with this function, cannot I. If this host is jasonrpc only, can proteus connect it, at all?

n, will turn out to be impossible
host method result
debian tryton server from proteus import config, Model
config = config.set_trytond(database_name=’aaa’,database_type=’postgresql’, password=’admin’)Source?
manual y
config.set_xmlrpc n, will do
gnuheath (demoscript built) n
manual y
config.set_xmlrpc( n?

on proteusing gnuhealth Hello World


This is on what I stumbled when I search around and asked why there is no obvious answer to a problem such as this:

13:09 wsirc_763 hi,
13:11 wsirc_763 Want to connect to a locally freshly installed gnuhealth with proteus.
13:11 wsirc_763 config = config.set_trytond(database_name=’gnuhealth’,database_type=’postgresql’, password=’344erf4nh4′, config_file = ‘/home/gnuhealth/gnuhealth/tryton/server/config/trytond.conf’)
13:11 wsirc_763 ImportError: cannot import name CONFIG
13:13 wsirc_763 can enter that server manually but still can not enter it with console. Any idea? Thank you.


 #The running hello world:
from proteus import config, Model
config = config.set_trytond(database_name=’aaa’,database_type=’postgresql’, password=’admin’)
# connect to running
#config.set_trytond(database_name=’gnuhealth’,database_type=’postgresql’, password=’444nh4′,config_file=/home/gnuhealth/gnuhealth/tryton/server/config/trytond.conf)
#feedback at data base, check
Module = Model.get(‘ir.module.module’)
sale_module, = Module.find([(‘name’, ‘=’, ‘sale’)])
sale_module.state #Out[33]: u’uninstalled’


Actually this is a way I had prepared to login on a sample server that comes with debian installation server – we got a server out of the source pocket.  So uptonow nobody is speaking with us about proteus. And then? You could have a closer look on the tool previously? And then if you find bugs and or feature gaps, repair the whole stuff in order to use it? Hopefully not.

We are concentrating now on the function:

proteus.config,config.set_trytond() this is the function to be expected the login-into-gnuhealthorwhatever function.

Who had problems with it?

https://www.google.de/search?q=config.set_trytond%28%29&biw=1000&bih=600&ei=lqtYVZPXPIagNsHmgYgL&start=10&sa=N#q=%22config.set_trytond%28%29%22+postgres – almost nobody.

How often applicated?

How is the packaging on proteus structured?




Ok, this could be the most prominent example of having our method been used for.  I do not know what speaks against proteus?

export PYTHONPATH=”/home/gnuhealth/gnuhealth/tryton/server/trytond-3.4.3/bin”
config.set_trytond(database_name=’gnuhealth’,database_type=’postgresql’, password=’444nh4′,config_file=/home/gnuhealth/gnuhealth/tryton/server/config/trytond.conf)

Exception: Module health_pediatrics not found

Ok, wei passed, we are still in this very function and it reclaims us to do what? I have not ordered him to search one I guess. Just to do the login? Where is the point of this message in this very situation?

gnuhealth@:~/gnuhealth/tryton/server/trytond-3.4.3/bin $ find /home/gnuhealth/gnuhealth/ -type d|wc
333 333 23730
gnuhealth@:~/gnuhealth/tryton/server/trytond-3.4.3/bin $ find /home/gnuhealth/gnuhealth/ -type d|grep demo

The demo is out: No it is not?

gnuhealth@:~/gnuhealth/tryton/server/trytond-3.4.3/bin $ find /home//gnuhealth-2.8.1 -name “serverpass*”
Where is the old one?


config = set_config(database, password)


def set_config(database, password):
return pconfig.set_trytond(database, password=password)

here he jumps to proteus sources. How was it invoked?

gnuhealth@:~/gnuhealth/tryton/server/trytond-3.4.3/bin $ ls
trytond wc
gnuhealth@:~/gnuhealth/tryton/server/trytond-3.4.3/bin $ python /home/gnuhealth/gnuhealth-2.8.1/scripts/demo/health_demo_server.py

I used that script and ok, it baked me the binary there.User name: adminPassword: admin hm

 Next order from head quarter: Login to https://en.wikibooks.org/wiki/GNU_Health/The_Demo_database#Online_Demo_Database !


installation of gnuhealth Ubuntu 14.04 (2)

Hi there, again,

On the installation routine’s circumstances

If one’s mind comes on the installation of gnuhealth project. Is not it such that in each circumstance of today’s life one asks for the design behind it? How it was meant once when it was set into the world. What about circumstances? Hey, this project is really kind of politicon, is not it firstly to mention? They run the health hospital system of a whole state, Jamaika. WOW. <—This is the NGO I am member of…

The second thing that needs to be acknowledged, the development seems to be active and by a considerable small team. And therefore new bugs and changes that might make in the long consequence the installation routine – busy or/and broken.

Third thing: These development team is be seen tackling however with new kind of “better-handbook-only” documentation delivery strategy to the end costumers.

So that is the thread to follow: Having_ a closer look to the handbook a wikipedia hosted wikibook then.

Lets recap what I as a defined as such: Testuser, experienced on my way through the installation process. My first impression is: supoptimal.

First: One ever asks oneself: Do they not want to let know how a anybody could install there software intriguedly? That keeps us away half from lust getting through this installation process. Why had not they talked about the warning that could be enduring weeks?

Second: On Wikibooks where the contents is to study from kind of html-generation way, noway in a form pdf-book, so there is yes bread crumb but there is no centralized contents index on the top of the wikibook and I dislike that- I say this just for the difference in the handling which stakes out. To this second point is to add:  Ah, if we have a straight look on the centralized contents index on the top of the wikibook. We found the point sabotage, – no joke trouble shooting. – It is empty. They say something of one could write one oneself an article over the trouble shooting. In the end all bores down to truth that we all should weapon ourselves with logging and tool-analyzing tools that all we could bore down to whichever assembler level of world history. Right? So what? Where are these tool for this way on what prepared, would one indeed ask constructively?

Fourth: This comes effectively so:

One is to ask: If not a more bread crump wise installation could be designable?

Lets open the wikibook Gnuhealth page on installation then:

On the red thread of the gnuhealth installation process

The point in this current gnuhealth installation routine is that it is noway bread crumb navigation, but maybe the opposite of it. The handbook reader is not be able to prepare the necessary for this from the beginning on:

-there is first a data base company hack to fulfill. Which has two steps. The second step waits as the most difficult point in the installation process:


Here is where user graphically-only is urged and where with considerable time risk he can proof most early, if all hours enduring work before were successful. This is what I mean that is neither expectable nor bread crumb.

Then -for the handbooks defense one could could point the newbie from the beginning on to this passage out of this chapter:

Check in the server log (or console) if any error occurs. If so, try to fix the problem (ie. any Python module missing or some unmet dependencies), drop the database just created and repeat the procedure.

And keep the compile script output on stdin and stdout logged next time one might add. Update: Yes and there will be opened a list of useful tools in this situation.

And this is the solution in short words to be prepared on this bottle neck: There are in 8 eight textbox datas to be entered by a user. This is more shining a test for IQ or aberridical behavior . So the best bet is to make a table on a paper with a column 8 interface variables with their textbox names on. Then read carefully which whole information you get out the handbook. Developing a best bet 8 column variation? Yes, this is what it is actually about.

You have to denominate a database and it’s database-user and it’s password. Please build a rational sentence of database, database user and database user password might one ask? Hm. It comes grosser. Yes and then stick the word “PostgreSQL” in it with out stumbling? He?

OK further on there is a server more to who we connect. Tryton server, User, password and two options on tcp/ip on a so call rpc-machine-never-heard-of. The latter seems ok.(nmap localhost) So then we have a user and a password. And here the thriller begins. After some discriminating jury “admin, admin” would be the choice but: No. In the end functioned by trial and error a considerably long time: that admin indeed is the indicated and existing user in Tryton Server, but his password were not admin. By heuristic googling I found some posts of same error message. And I got to the handbook page of resetting the password.  – I had to reset my bash console meanwhile and unsuccessfully. Hm –  So you got two passwords and completed all column.

Here should open up the graphical application of gnuhealth.

on second run: installation of gnuhealth

here are some command lines I found worth to be mentioned. Have fun with it:

apt-get install libxml2-dev libxslt1-dev
apt-get install python-cracklib

# I found that I needed to install them, after two times unsuccessfully had run the compile script. But if one compares here https://en.wikibooks.org/wiki/GNU_Health/Operating_System-Specific_Notes#Install_dependencies_4, they already should had been in..

rm -rf ../gnuhealth
rm -rf /tmp/gnuhealth_installer # for cleaning up in the mean while.

~/gnuhealth-2.8.1 $ ./gnuhealth_install.sh > ./mylog.txt 2>&1 &
#[1] 22678
#gnuhealth@:~/gnuhealth-2.8.1 $
#[1]+ Exit 1 ./gnuhealth_install.sh > ./mylog.txt 2>&1

ps aux|grep trytond
#gnuheal+ 16213 0.0 0.4 364436 15376 pts/3 Sl Mai15 0:07 python ./trytond
#gnuheal+ 21954 0.0 0.0 18932 968 pts/3 S+ 13:55 0:00 grep –color=auto trytond
kill 9 16213

#107 in another Console: tail -f gnuhealth-2.8.1/mylog.txt

./tryton -v –log-level=DEBUG

#INFO:tryton.rpc:common.db.db_exist(None, None, (‘gnuhealth’,))
#INFO:tryton.rpc:common.db.create(None, None, (‘gnuhealth’, ‘6678werhu8’, ‘en_US’, ‘444nteh4’)

./trytond –logconf /home/gnuhealth/gnuhealth/tryton/server/config/gnuhealth_log.conf -v

# Error Message:

#Access denied!
#Sorry, wrong password for the Tryton server. Please try again.

printf ’44tth4\n44ttnh4’|su – postgres -c “createuser –createdb –no-createrole –no-superuser -P gnuhealth”

find /home/gnuhealth/ -name “serverpass*”

printf ‘467677h4\n467677h4’|python serverpass.py # have not checked the printf though. I think finding this ridiculous is serious.

psql -l

su postgres -c ” dropuser gnuhealth”
su postgres -c ” dropdb gnuhealth”
su – postgres -c “psql -P format=unaligned -c ‘\l'”|grep gnuhealth;su – postgres -c “psql -P format=unaligned -c ‘\du'”|grep gnuhealth

#http://www.binarytides.com/install-postgresql-phppgadmin-pgadmin-ubuntu/ for those who want to to inspect their data base at least once graphically.

Screenshots of the chase:







I see then several, do not know how much time it needs. This form is the user having to wait why?




nmap localhost
root@:~# cat 1-tog|grep 8000 |wc
108    2580   18924
root@:~# cat 1-tog|grep 5432 |wc
  1429   34227  257520
root@:~# tcpdump -nvvvi lo >1-tog&

root@:~# cat 1-tog|grep 5432 |grep incor|wc
1418   34062  256313
root@:~# cat 1-tog|grep 5432 |grep -v incor > Flags [R], cksum 0xf532 (correct), seq 3782025849, win 0, length 0 > Flags [R], cksum 0x9e84 (correct), seq 1587608562, win 0, length 0 > Flags [R], cksum 0xe6a1 (correct), seq 3386787989, win 0, length 0 > Flags [P.], cksum 0xfe82 (correct), seq 6381:6471, ack 4457, win 429, options [nop,nop,TS val 123825025 ecr 123825025], length 90 > Flags [R], cksum 0xf331 (correct), seq 3349364286, win 0, length 0
12:26:25.454323 IP (tos 0x0, ttl 64, id 31270, offset 0, flags [DF], proto TCP (6), length 52) > Flags [R], cksum 0x87f2 (correct), seq 925907438, win 0, length 0 > Flags [R], cksum 0x35c9 (correct), seq 3254255950, win 0, length 0 > Flags [R], cksum 0x9239 (correct), seq 4100616809, win 0, length 0 > Flags [R], cksum 0xf740 (correct), seq 2401465000, win 0, length 0 > Flags [R], cksum 0xe58e (correct), seq 3049479609, win 0, length 0
root@:~# cat 1-tog|grep 5432 |grep -v incor

I lost some what the patience and wanted to know what they were talking about. Like Mister Tottle, and out of curiosity because I am new to three Tier. Ask tcpdump: here I liked that http://albanianwizard.org/tcpdump-how-to-the-linux-troubleshooter.albanianwizard

All that noise on the cable about some unmet checksums…

Ok, I have been grown minister of the GNU-Hostital now? Anyway I do not find it good what you do with all that poor Jamaican patients there.

That will be the structure for my first tramp company. So what: this was the installation of GNU Health, it took three days. Thank you for reading!

Vista install in 2 minutes, here we go

For comparison how a service design can be concieved as well: https://www.youtube.com/watch?v=FVbf9tOGwno

installation of gnuhealth: bug on Ubuntu 14.04

Question: What can this kind of error in the middle of the installation process mean? (see in the screenshot the suggested method for installing, namely taken from the official wikibooks book “GNU Heatlh”) . Has anybody an idea?


x86_64-linux-gnu-gcc -pthread -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fPIC -I/tmp/pip_build_gnuhealth/lxml/src/lxml/includes -I/usr/include/python2.7 -c src/lxml/lxml.etree.c -o build/temp.linux-x86_64-2.7/src/lxml/lxml.etree.o -w

In file included from src/lxml/lxml.etree.c:239:0:

/tmp/pip_build_gnuhealth/lxml/src/lxml/includes/etree_defs.h:14:31: fatal error: libxml/xmlversion.h: No such file or directory

#include “libxml/xmlversion.h”


compilation terminated.

error: command ‘x86_64-linux-gnu-gcc’ failed with exit status 1

There we walk, till what? One key element of chain is broken again:



How long shall I wait? What happened to them? Who are they? And now what would be the next guess, first repairing the tor-chat wheel and then retürning to search the chat of gnu-health for posting my question? If not that, what else.

Or an other channel, namely the mailing lists?



Yes, I see the mailing list is quite frequent. So once more passwort username, uffuffuff would be the best option. I guess.