Skip to content

Get it running correctly, more information is needed #133

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
kleinias opened this issue Jul 9, 2018 · 5 comments
Closed

Get it running correctly, more information is needed #133

kleinias opened this issue Jul 9, 2018 · 5 comments
Assignees
Labels

Comments

@kleinias
Copy link

kleinias commented Jul 9, 2018

I installed flat by the commands from the manual:

virtualenv --python=python3 env 
. env/bin/activate
pip install FoLiA-Linguistic-Annotation-Tool
export PYTHONPATH=/home/bingobongo/code/flat/
export DJANGO_SETTINGS_MODULE=settings
django-admin migrate --run-syncdb
export PYTHONPATH=/home/bingobongo/code/flat/
export DJANGO_SETTINGS_MODULE=settings
django-admin migrate --run-syncdb
django-admin createsuperuser

with the default settings, without the raise error line. and then started the servers:

django-admin runserver &
foliadocserve -d "/home/bingobongo/code/flat/" -p 8080 --git 

If I go to "localhost:8000", I can log in to the created account and upload a file I converted to folia-xml. But then I get some error, before some submission.

auswahl_013

Looks similar to issue #128, but selecting twice is not required for me.
Is there something wrong with my configuration?

If there is some help, let me know!

PS: And I have a problem, how to write a new set definition. I read the documentation, but its a little bit sparse, if one hasn't any idea. Have you anywhere an example, where I could get some idea from, how it could look in the end?

@proycon proycon self-assigned this Jul 10, 2018
@proycon proycon added the to do label Jul 10, 2018
@proycon
Copy link
Owner

proycon commented Jul 19, 2018

Sorry for the delay! Something seems wrong indeed, that __pycache__ part in the URL should not be there at all. I assume it's not in the URL you explicitly access either? Did you set a BASE_PREFIX variable in your FLAT configuration? (I wonder if the pycache somehow got injected through that variable, it's still a bit of mystery to me what could have caused it)

Is this the installation as provided by LaMachine or installed and configured from scratch?

@kleinias
Copy link
Author

kleinias commented Jul 24, 2018

Ok... so the "pycache" comes into that because I took that as namespace and that was, what I saw now, that I could not choose something else at the site, when I should upload the file. The menu looked like this:

menu_021

Then I created a directory, and then I could choose all other directories in the working directory as namespace. I get the same error with another namespace instead of the pycache:

auswahl_022

All my installation procedure is here, its just the lines from the installation guide: install it with the install_script, the settings file modified and the lines to start in another script. you just would have to change the path. thats my setup.

installsettingsstart.zip

@proycon
Copy link
Owner

proycon commented Aug 7, 2018

Sorry it took a while again because of the holiday period. It seems you set the WORKDIR aka folia document root wrong, you have it pointing to the code, but it should be an (initially empty) directory where the FoLiA documents will be stored. That explains why you saw all those directories as possible namespaces.

Make sure to start foliadocserve with that same directory specified as document root.

I hope this gets you a bit further?

@proycon proycon closed this as completed Aug 7, 2018
@proycon proycon reopened this Aug 7, 2018
@kleinias
Copy link
Author

I changed these paths.
There was later another process listening on the port and doing some first step, but now I got it working :) thx.

I have a few questions:

Can I somehow import annotations from parsers?

Do you have some kind of documentation, how to build customized annotation schemes?

Is there a possibility to see my annotations as arcs?

And some note:
I get different error messages, if I don't do correct annotations, like: "TypeError: Cannot read property 'trim' of null". It would seem better to me, if that would point me to the error, that I did, than some error in the program.

@proycon
Copy link
Owner

proycon commented Aug 17, 2018

Can I somehow import annotations from parsers?

FLAT uses the FoLiA format. Some parsers like Frog output this natively, for some others there are converters (e.g. Alpino), for others you may need to implement a converter yourself. I'm planning on working on a converter from the CONNL-U format if that helps.

Do you have some kind of documentation, how to build customized annotation schemes?

Yes, see the documentation on FoLiA Set Definitions, chapter 3 of the FoLiA Documentation. Also see the examples in https://github.com/proycon/folia/tree/master/setdefinitions

Is there a possibility to see my annotations as arcs?

Not really at nicely visualised arcs no, the best we have is the visualisations you can toggle on/off in the global annotations menu.

And some note:
I get different error messages, if I don't do correct annotations, like: "TypeError: Cannot read property 'trim' of null". It would seem better to me, if that would point me to the error, that I did, than some error in the program.

Hmm yes, that sounds pretty uninformative, I'd consider it a bug, so feel free to open an issue on that one.

Also be aware that the default Full configuration of FLAT exposes a lot of functionality you don't need and which will confuse the annotator, make sure to customize your configuration to enable only those features you need (see the administration guide), I might have to ship some simpler configurations with FLAT out of the box.

@proycon proycon closed this as completed Feb 7, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants