-
-
Notifications
You must be signed in to change notification settings - Fork 248
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
[14.0][Crédito de impostos]: Crédito de impostos pt1 - adiciona stock price e impostos creditáveis por Tax/CST #3158
base: 14.0
Are you sure you want to change the base?
Conversation
Hi @renatonlima, |
Creio que a lógica pode ficar no fiscal, até por que o crédito precisa ser revisado por vários atores:
|
08e4cd1
to
0a1121e
Compare
0a1121e
to
e3de308
Compare
e3de308
to
c6272b3
Compare
Movido para o l10n_br_fiscal! |
c6272b3
to
61a74a9
Compare
a gente ta mais eh tentando fatiar o modulo l10n_br_fiscal que ja ta muito acima do peso dos padroes que fazem o sucesso da OCA. Nao seria melhor extrair um novo modulo para isso? |
Para emitir o SPED fiscal e importar o XML dos documentos vamos precisar dessa estrutura. Então pode ser melhor deixar esse código no fiscal mesmo... Se não vamos ter muitos módulos de cola pra levar isso até o estoque e o compras. Fazer um esforço pra fechar essa parte e a importação de nota, pois libera o SPED. E aí entendo melhor a arquitetura final da pra remover oq não precisa do fiscal sem problemas, ou até dividir ele em peqienos módulos. |
@rvalyi acredito que o stock price seja na verdade uma correção do valor de estoque que já está sendo calculado no fiscal e não está correto por ter diferenças com os lançamentos contábeis. Por outro lado entendo que a modificação do valor do custo do produto pode ser uma alteração drástica no funcionamento do sistema. Busquei suavizar essa adaptação inserindo uma booleano capaz de manter o funcionamento atual do sistema na configuração da empresa, veja na linha:
|
@DiegoParadeda pode ser melhor vc colocar esses campos na visão do mixin, pois ai vc remove a necessidade de alterar as views dos módulos purchase, stock e account: l10n_br_fiscal/views/document_fiscal_line_mixin_view.xml |
I don't think this is a valid argument to bloat the l10n_br_fiscal module further. With such reasoning one could say: "by law it's mandatory to post stock valuation moves in the accounting, so the stock_account module and even the stock module should be merged into the account module." Again, just like the vast majority of Odoo account module users use the account module just for invoicing, the vast majority of the l10n_br_fiscal module users will also use it just for their invoicing and not full blow accounting. Full blown accouting is nice, but certainly not a valid reason to bloat to death the code that 90% of the module users will not use. I think we should double check with @renatonlima but you are also forcing me to insist that this is a very serious issue with have with the Kmee company contributions in the repo... You quite never do the clean up work, even you are almost not maintaining the modules you add in the repo, often staying months without giving feedback about modules you added, sometimes staying like 2 years without contributing much. Yet, from time to time @mileo comes here and just like a spoilt child insists that feature X or Y should be included in a timely manner to module he didn't do nor maintain to match his own very agenda all the other contributors should promptly adapt too. Sorry but it sucks and this has been 10 years like this already. I don't know about anybody behaving like that across all the OCA repos. Sometimes features like the DFe were added by @mileo to the l10n_br_fiscal modules in a hurry and stayed like 4 years adding 1000 lines of dead code inside the l10n_br_fiscal module which already weights more than 20k lines of code. So no sorry, the time has come to be smarter about how to make things more modular. Look how hard it is later to extract just like 1200 lines from the l10n_br_fiscal module #3012 Again, if Kmee were doing the cleanup work, if the signal to noise rate of the contributions were as good as the ones from Escodoo or Engenere for instance, that would be a quite different story to take time to add feature X or Y to some module. But given this is a very recurring issue from even days before you @DiegoParadeda were contributing in the repo, I have to tell it stright again, and yes in English because eventually this is an issue to escalate with the OCA. |
@rvalyi why do you think my argument isn't valid? Please, let's stick to the point |
sorry, this pressure to bloat already extremely bloated modules (that uou guys largely bloated) without never doing the cleanup part is a very recurring issue always coming from the same company Kmee. I don't know if you guys have no clue why I complain about this, but in either cases you are forcing me to repeat it straight. @mileo it looks you think you were born with an unlimited credit for pressuring for bad contributions in the repo. I don't think that can run like that for ever so I prefer making it clear again sorry. Please guys when contributing to some open source project ask yourself always: what am I bringing to the project, and what energy will I be sucking from the project maintainers im exchange. That balance cannot run negative forever. |
I understand your point, and I'm not looking to add any pressure. However, I wanted to bring to your attention an issue regarding the compute methods for product.product.standard_price in the module. It seems these methods do not take taxes into consideration. In practice, this oversight results in the standard_price field being populated incorrectly when creating a purchase and stock receipt for a product with the AVCO option turned on. I believe the Pull Request I've submitted addresses this issue, or aims to do so. However, I could use your input on the next steps. Would you prefer we relocate these functions to another module like l10n_br_fiscal_extended? Alternatively, should we delay this feature until after a complete refactor of the module for improved cleanliness? I'm seeking clarification because I'm unsure about the best course of action. Your repeated reference to @mileo has also contributed to my confusion. |
generic names like l10n_br_fiscal_extended shoukd be avoided and if you look carefully in the OCA, you won't find such module names. Again contributing advanced stuff to the OCA is all about experience and there is a very famous quote: There are only two hard things in Computer Science: cache invalidation and naming things. -- Phil Karlton So instead try to be feature specific. Good modules size in in the OCA are from 100 lines to some 4000. Anything larger than that should prevent you from sleeping untill you find a solution to improve it. Here l10n_br_fiscal is more than 20k lines, period. Another very basic and important thing to bare in mind: fixing a design after you created it poorly, deployed it in production in many companies, let other modules depend on it in several branches will at the bare minimum cost an order of magnitude to fix it after compared to fixing it BEFORE. So in doubt always abstain from adding features, use things like git-aggretate to get the job done without introducing technical debt for the others. So no sorry, making the situation worse and promise you'll fix it later doesn't work for me. Sorry but Kmee has already burnt any credibility here about this during these 10 years. The case with l10n_br_dfe was just an example but if there is any doubt I can easily come with a list with a dozen of such things, including payment gateway module logging credit card numbers for more than one year, easy peasy. May be you Diego aren't so aware about this, but that doesn't change facts and at the end of the day if your employer will never pay you to fix things later, well it's reasonable we expect things won't before different this time. Remember reputation takes years and years to build and can be destroyed in a split second. sorry I didn't think too much about how to do it properly. Recently I already had to save the XML NFe importation or the SPED from being screwed the same way. You guys cannot expect we will save the project every week for free to accommodate your very own needs. What I can tell so far is that this should not be a problem to have some SPED module depend on some extra tax credit module though I gave example to simply test if a module is present or not when serializating the SPED registers so no hard dependency might even be needed. As for importing XML NFe's, that should require a phD either to come with a small hook that would trigger an override in your tax credit module. Again, you guys should learn more from the OCA ecosystem before tryjng to contribute complex parts of the project. Regards. |
@DiegoParadeda BTW I just checked and the payment gateway that has been logging credit card numbers for more than one year in the repo while you guys just walked way presenting yourselves as key people for the projects was authored... So you should probably be able to understand when I say I cannot trust people from Kmee when they ask "can we do this little extra shit and we promiss we will fix it later?" |
@rvalyi |
I have a different interpretation: forcing bad stuff into the project is the real disrespect of the maintainer time in 1st place, not complaining about it for the tenth time. Really if you think that's me being wrong about this I think you should come first with a few dozens of modules endorsed by other repos of the OCA, 10 years is far enough time to prove you are the good guys you claim to be. There is no way you pile poor code into the project, including major security issues and behave like everyday is a new chance and we should always forgive it all and keep loosing our time with your PRs over and over. |
This is a work in progress, I'm not forcing anything. Feel free to add your code review and I will surely consider it. Bonus points if you can do it respectfully |
I have my doubts about this.
So no sorry, it's way too much to silent it. If you look carefully you will see that in my entire OCA contributions, I never had to talk to people the way I had to do we a few persons from Kmee. Never. Is it because of the project? Look if I had to talk like this with people from Escodoo or Engenere. Answer: never. THE PROBLEM IS WITH YOU. |
I was in a rush a few hours earlier, so may be my message was a bit confused: so basically, my problem is not so much with you @DiegoParadeda but it's much more with the track records from the company you are working for, that is Kmee and @mileo. Though that doesn't exonerate you entirely because the payment gateway logging credit card numbers for a full year was your work and you also left your website_* modules with broken tests annoying all other real contributors for months 2 or 3 years ago. So this should be no secret that with part of the bad track record from Kmee in the project I mentioned earlier, me and Renato are pissed off to say the least. Now I was also stressed such feature would be merged in a hurry by @mileo just like he merged #3165 in just 5 hours earlier this week which wasn't a bad PR, but he also has track of having merged in a hurry extremely bad contributions such as #928 or #820 just to mention a few ones. So basically I started complaining more explicitly about bloating l10n_br_fiscal further without Kmee never tackling any clean up job (and even just inside the modules or libs you guys are leading there is plenty of room to express all your talent for making clean things) because I was concerned this PR marked as ready for review would be yet another bloat @mileo would merge and walk away leaving us with the extra burden in the module which is now more than 20k lines of code already largely because of this lack of programming skills when introducing new features. Again, just like for the IAS2 thing, nobody here is complaining you are attempting to implement legal feature X or Y. It's just really about the way guys are trying to implement these things. It does matter a lot. the OCA modules are done as a volunteer work with little resources, OCA/l10n-brazil is already one of the very most active project out of the 250 OCA repo, l10n_br_fiscal is already the largest of the 3000 OCA modules, so there isn't much room left for bloating things in suboptimal ways as this is already sucking way too much energy to migrate and maintain the way it is. |
Parte 1 de #2635