@metaprogrammer when do you guys expect a migrations to be in beta? Is it a platform for continuous migration?

I only ask because I'm presently rolling one myself for my own work.

What do you mean by platform?

I mean is it going to integrate with the diagram editor?

will there be api hooks?

Not in the first version

Ok I'm currently doing some work to do that with the diagram editor - should we collab?

bearing in mind I use Pony to save time because I'm terrible at database design without it

is there a curl request you have somewhere that authenticates diagram editor?

Please explain your last sentense

sorry - Pony has an editor so I'd like to authenticate to pull the code from it with cURL or python requests

I have an account - I'd like to know if that breaks terms of service

if not I'd like to know if you have some basic code to authenticate (since I don't know how you guys wrote it) so that I can pull the code - thats all 🙂

Ah, I don't think we have such functionality now.

I will write the code and send you a pull request

Currently our migrations are not tied to editor

I was thinking it might be a fun exercise

Ok, cool, let's try

yah I'm keen to whip up something for continuious migration

let me analyze post request for auth

Keep in mind that editor is not open source software

Yes so I'm wondering how I might be breaking your terms by doing this

XHRPOSThttps://m.stripe.com/4 [HTTP/2 200 OK 367ms] Response Cookies m expires 2030-02-04T19:03:15.000Z httpOnly true samesite Unset secure true value dospkcodpkocdpskocpskosdpkospkcs Request Cookies m dospkcodpkocdpskocpskosdpkospkcs

I think you don't break our terms by doing this

ok

this is where the POST goes to --> https://m.stripe.com/4

what is this?

so they handle your auth and billing etc

ok I will write some code now - brb

We use stripe to bill accounts, I don't think you need to access stripe to do anything with migrations

It seems to be where the auth response comes from

when you login

We handle auth at backend, stripe is not used for it

Hey @minskmaz I’m away this weekend Let’s catch up on Monday and see how we can collaborate on this

ok!

What do you develop @minskmaz ? Can you share the link?

Oh there's nothing to show yet. I started this morning on continuous migration concept. Haven't put into version control. I use twisted / klein / treq - is that ok or should I do in python requests?

@minskmaz twisted 😱

Hi, what do you mean by "the code stops right there"? The code hangs is some lock?

What is the last log record that you can see?

When you create an object, like x = MyObject(foo=10, bar=20) the object doesn't become saved to database right now, it will be saved later when it become necessary (like, before exit from db_session or before executing a query) It is possible that your application creates object in memory and than hangs for some reason unrelated to Pony (like, waiting for some queue or lock) After object creation you can force insert by performing explicit flush(). It is not necessary, but maybe it will make for you easier to debug the problem

Are you sure that pony-related code is executed at all?

I mean, in db session

Yes, I mean the following code, that wrapped in db_session

Try to use db_session with sql_debug=True option Maybe the debug turned off for some reason?

Or maybe the log messages are suppressed?

Do you have explicit flush() after object creation?

And how the line where you create an object looks like?

I suspect some typo somewhere

Do print after flush executed correctly?

What is the reason for rollback? Do you see some exception?

Can you show me the whole method? You can send it privately if you wish

Normal execution should not lead to rollback, and if you have rollback that there should be a reason for this (explicit rollback() or catching an exception)

So the reason was an unhandled exception during object creation which does not appear in output, because the function was executed in a separate thread

Hello. I have an issue with show() method. Full code: # Person class is just a Person from official docs mary = Person.get(name='Mary') orm.show(mary) Then I get an error: --------------------------------------------- AttributeError Traceback (most recent call last) <ipython-input-22-86c678a1794a> in <module> ----> 1 orm.show(mary) ... Large Error ... :\users\user\appdata\local\programs\python\python37\lib\site-packages\pony\orm\core.py in __init__(self, query, limit, offset, lazy) 6292 __slots__ = '_query', '_limit', '_offset', '_items', '_expr_type', '_col_names' 6293 def __init__(self, query, limit, offset, lazy): -> 6294 translator = query._translator 6295 self._query = query 6296 self._limit = limit AttributeError: 'list' object has no attribute '_translator' You can see full error log in file below

How can I solve it? All of the code I used is code from official "Get Started". I translated this page to Jupyter Notebook (it's how I learn new libraries) and got this error.

If you want to look at this notebook with error

I think the problem is that you are trying to show a pony object rather than a query. It is normally used like Person.select(lambda p: p.name == ‘Mary’).show()

You could do mary.to_dict() to show the object’s data

You could do mary.to_dict() to show the object’s data

Yes, maybe, but I just run code from official docs

Where do the official docs show that use of “show”?

Where do the official docs show that use of “show”?

https://docs.ponyorm.org/firststeps.html#getting-objects

Ah you are right

Is it bug?

I am using Python 3.7

Hi kop! I think this is a bug, I'll try to fix it a bit later today

Hi kop! I think this is a bug, I'll try to fix it a bit later today

Thanks. I love Pony so much

can I see any big projects written using Pony?

Hi there, I'm currently in the design phase for a new web application providing league results and ranking tables. Due to the fact that this app should serve different leagues, I'd like to break things up in smaller portions, means that I try to use a schema in PostgreSQL like <league>.<league_abbr>_<season>_results and <league>.<league_abbr>_<season>_ranking. The _ranking tables are calculated through Pandas, which only can either append or recreate when using their direct to_sql-function on dataframes. The main problem now is that I'd like to use a common model with various table names, which may also change in the frontend when the visitor views results of more than just one league, maybe also from multiple seasons. And it should be able to read from existing tables not created through the ORM layer itself. I'm now somehow stuck on how to solve this. Maybe one of you has some idea or a solution in the bottom drawer. 🙂 Many thansk in advance! Regards, Thomas

can I see any big projects written using Pony?

I think most big projects are commercial and therefore not available to share.

Thomas, maybe it would be easier to not have many tables and work out how to take the results from pandas and use pony to insert them into the database?

I think most big projects are commercial and therefore not available to share.

I just want to see any real cases of Pony usage

E.g how to create separate threading class of Database and put entities into it for make ability to use it in other threads

For example, we used Pony in https://fineartbiblio.com/ (it is a commercial project, sources are not available)

For example, we used Pony in https://fineartbiblio.com/ (it is a commercial project, sources are not available)

Sorry, it doesn't suit me

For example, we used Pony in https://fineartbiblio.com/ (it is a commercial project, sources are not available)

But this projects is great! I was using it.

E.g how to create separate threading class of Database and put entities into it for make ability to use it in other threads

Hi, is this related to me? If yes, is there an example out there how to do this?