Hi Stefano, I agree. Many things should be more straightforward and
better documented.
I think we could pick out a few common tasks to focus our
tool-building/documenting efforts on. Eg:
1. importing bank data and CSV generally. All of the tools and basic
generic workflows for this should be described on one page. Focus on
CSV, but we should mention OFX too (ledger-autosync is arguably best at
this with its download feature).
2. exporting all data and reports as CSV
3. moving data between the ledger-likes (ledger, hledger, beancount...).
Again, all tools and techniques gathered on one page. All existing
formats should be listed. The output of "ledger print" is a sort of
lowest common denominator, I propose we give it a name and decree that
every tool should import this as a basic interchange format. And/or a
standardised CSV representation of it, such as "hledger print -O csv"
4. moving data from and to other accounting tools (gnucash, moneydance,
excel, quick{en,books}, mobile account apps)
5. manual data entry. Editors and their modes, ledger entry, hledger add
and other prompting tools, hledger-web, recurring entry scripts, etc.
6. a catalog of journal entries covering all common transactions
On 2/4/16 12:47 AM, Stefano Zacchiroli wrote:
> I've the feeling that, right now, the lack of a generic framework ---
> generic both on the front of data source (CSV, OFX, weird bank formats,
> etc.) and on that of output formats (ledger, beancount, etc.) --- for
> semi-automatically importing transactions is perhaps the most
> significant limiting factor for the adoption of CLI accounting.
>
> I went myself through the ad-hoc automation of my work-flow for
> importing transactions from my bank, scripting together the web outside
> of browsers suite [1] and icsv2ledger. It works decently enough for me,
> but the CLI accounting community cannot really expect every newcomer to
> go through that hacking process if it wants the community to flourish.
>
> [1]: http://weboob.org/
>
> Martin is right that the most front-end part of the import chain (web
> scraping in most cases) will always remain a case-by-case business. But
> there, communities like weboob can feel that niche quite nicely, if only
> they will manage to grow and be diverse enough. (Right now that
> community is very much skewed toward supporting French banks, with very
> sparse support for other international banks.)
>
> But the rest of the toolchain, from CSV down to your favorite CLI
> accounting tool can really do better in terms of reference tools and
> automation. I'm sorry I haven't had time/energy to contribute myself to
> ledgerhub, because the design looked pretty solid; I'm looking forward
> to the new rewrite :-)
>
> </ramble>
>
--
---
You received this message because you are subscribed to the Google Groups
"Ledger" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
