Hi all, On Friday, 12 February 2021 at 02:20:11 UTC+1 [email protected] wrote:
> On Thu, Feb 11, 2021 at 6:23 PM [email protected] <[email protected]> > wrote: > >> Thank you both, >> >> Martin for the beangulp documentation and Dan for the clear instructions. >> >> As a result I've managed to convert one of my scripts quite easily. >> >> However on running the extract command, I am immediately hit with an >> issue with the rounding of the amount. >> >> What used to be in v2 >> 2020-12-29 * "INTEREST PAID AFTER TAX 0.00 DEDUCTED, 7.69" >> Assets:Current:S 7.69 GBP >> >> is now >> 2020-12-29 * "INTEREST PAID AFTER TAX 0.00 DEDUCTED, 7.69" >> Assets:Current:S 7.69000000000000039079850466805510222911834716796875 >> GBP >> > > This is likely related to the fact that the display context from your > ledger file isn't loaded. > (In v3 I'm completely reviewing all the tolerances and precision and how > this is done.) > > So I have had a success although I don't understand why. Last night I ended up converting my old import config with all my importers in it into beangulp and since then the display context issue is not visible. It's strange that it now works given my importers are the same but perhaps the problem depends on the order of loading of importers or the fact that my test script only had one importer? > > I think beangulp is quite neat but I have some concerns around its day to >> day use but perhaps that is because I have not been using it properly yet. >> > > To be clear, beangulp isn't really a new thing. It's literally the > beancount.ingest code I've copied over so we can make it better and evolve > it a bit faster. > > > I have all my importers in their own git repos, most of which are not >> production ready and so are not on pypi. I guess in fact most peoples >> importers may be so specific to their needs that they would not be put on >> pypi either. >> >> So in world where we all have multiple importers in development, would >> the recommendation be a) to install the importers into the venv of beangulp >> or b) to create a venv for each importer and install beangulp into each or >> c) some other option? >> >> a) Is easier to use because one would not need to switch venvs while >> importing. >> b) Protects against package issues between importers. >> c) Perhaps it's enough to add my importers to the PYTHON_PATH and hope >> for the best. >> > > I don't know. > We have stable: v2 (beancount v2 + beancount.ingest) and development: > beancount/master + beangulp. > Choose one. > If things occasionally breaking due to changes bothers you, stick with v2. > If you want the goodness of new things, and are willing to make changes > when necessary, use the master branch and development repos. > > After converting my old import config i'm finding the process easy enough. My workflow is entirely in Vim so I've had to learn how to use a virtual environment from inside Vim but that was easy enough. Moving forward as I would like to make changes to the CSV importer so I will use beangulp for importing and beancount v3 as much as possible, but if something breaks, presumably I can continue using beangulp for importing and v2 for the rest? Kind regards, Jakub Jamro. > > >> >> Kind regards, >> Kuba. >> >> On Thursday, 11 February 2021 at 23:26:51 UTC+1 [email protected] wrote: >> >>> On 11/02/2021 22:53, [email protected] wrote: >>> > Regarding importer development on v2, if I was to make changes to >>> > csv.py, will the changes be migrated from the v2 branch into beangulp? >>> > >>> > I'm still unsure of how to use beangulp and I would like to make >>> changes >>> > to the importers so will I need to port the changes over myself once I >>> > familiarise myself with the new world? >>> >>> Hello Kuba, >>> >>> currently the v2 codebase is in maintenance mode: only bug fixes and >>> changes that can be trivially merged into v3 are likely to be accepted. >>> This disqualifies most changes to the ingestion framework that are not >>> fixes for bugs affecting users. >>> >>> Also, if anything, I expect the CSV importer in beangulp to become >>> simpler, and not to gain new features. It was intended as an example to >>> demonstrate how to write an importer and grow into something with a know >>> for every aspect, and it deviated from its original scope. >>> >>> However, if you have some improvements in mind, we can discuss them case >>> by case. Please feel free to send PRs for beangulp. >>> >>> So far, the only thing that changed in bengulp is the command line >>> interface, thus all existing importers can be easily migrated to it by >>> simply fixing the imports. >>> >>> Installing beangulp should be easy: >>> >>> $ python -m venv gulp >>> $ cd gulp >>> $ ./bin/pip install git+https://github.com/beancount/beancount/ >>> $ ./bin/pip install git+https://github.com/beancount/beangulp/ >>> >>> or >>> >>> $ git checkout https://github.com/beancount/beangulp/ >>> $ cd beangulp >>> $ python -m venv . >>> $ ./bin/pip install git+https://github.com/beancount/beancount/ >>> $ ./bin/python setup.py develop >>> >>> if you want to hack on the code, or similar incantations. >>> >>> Cheers, >>> Dan >>> >> -- >> > You received this message because you are subscribed to the Google Groups >> "Beancount" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> > To view this discussion on the web visit >> https://groups.google.com/d/msgid/beancount/b8f0474c-ee18-46b9-8952-ff56b9e08ddcn%40googlegroups.com >> >> <https://groups.google.com/d/msgid/beancount/b8f0474c-ee18-46b9-8952-ff56b9e08ddcn%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> > -- You received this message because you are subscribed to the Google Groups "Beancount" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/beancount/198987ad-03c8-48bc-b848-2ce253e680c5n%40googlegroups.com.
