Production Docs: FX data: Initialise the spot FX data in MongoDB from .csv files; error in repocsv_spotfx_prices.py #1346
Comments
joshwigginton
changed the title
|
Well that is a bit pointless as you will just read in csv prices and write to csv prices!!! fixed in last commit (develop branch) |
CEllis40
added a commit
to CEllis40/pysystemtrade
that referenced
this issue
Mar 12, 2024
* final piece of long only DO * final piece of long only DO * final piece of long only DO * merge latest, black * final piece of long only DO * tell Black to ignore one line https://black.readthedocs.io/en/stable/the_black_code_style/current_style.html * fix to issue with reduce only and no trade in DO * fix to issue with reduce only and no trade in DO * can now pass constraint functions to DO * die * Improve log messages when there is a problem emailing a report (cherry picked from commit ca311cc) * Fix stack handler bug (cherry picked from commit 4de5eba) * Fix stack handler bug (cherry picked from commit 8488b99) * fix net position error in roll code * checked into wrong branch * starting to build in parquet * starting to build in parquet * parquet create files * parquet create files * parquet create files * missing extension * missing extension * added backup function * added capital * deal with ts in parquet * replace arctic adjusted prices with parquet - read the discussion! * single point of xfer * single point of xfer * reorder backup order; * use parquet rather than arctic capital * revert to arctic capital * add explicit column name to capital df * revert to parquet capital * revert to parquet capital * contract prices in backup * change ident repr for contracts and freq * per contract prices * much cleaner way to switch arctic/parquet * bug * bug * bug * bug * bug * removed redundant pointers.py * removed all direct arctic references * fixed backup * fixed backup * fix for class adding * accident indent * accident indent * added multiple price backup for parquet * parquet backup now skips if already done * parquet backup now skips if already done * default is now parquet for multiple prices * default is now parquet for multiple prices * back up to parquet now interactive * added fx to backup * added fx to backup * parquet now default for fx * weird error fixed * weird error fixed * confusion over spread data * added historic spreads to backup * switched default of spread to parquet * added optimal position data * switched optimal positions to parquet * added position data to backup * added position data to backup * added position data to backup * added position data to backup * added position data to backup * switched default for position storage to parquet * added parquet backup changed control config * accidentally wrote over setup and requirements * accidentally wrote over setup and requirements * black and started tweaking codde to run * Update backup_parquet_data_to_remote.py * Update backup_parquet_data_to_remote.py * fix weird IB date string * fix future warning * fix future warning * fixed .append issue * supress warning fix bfill error * fixed append in margn * now correct warning syntax * fixed backup issue * slippage reporting append error * update spread costs * removed annoying warning from volumes * removed annoying warning from contract positions * fix append error in trade reporting * fix append error in pandl reporting * fix append error in trade reporting * removed future warning from pandl calculator * removed warning from capital data * more pandas fixes * remove warning in price changes * updated version * bug reading spread data * arctic was still in setup and requirements * blacked * update Black version to 23.11.0 * quick test runs python 3.10 * HANG-TECH --> HANGTECH in spread costs (cherry picked from commit a372c5c) * slow test runs python 3.10 * python 3.10.13 * add a default parquet store to config * fix date format in algo tests * fix date format in roll config tests * EDOLLAR -> SOFR * undo monkeypatch work * skip weird 'ufunc' failing test * filter out nans with pd.isnull(), and convert to numeric with pd.to_numeric() * unskip test * fix robcarver17#1296 * couple more new pandas issues * Update strategy_functions.py * Update vol.py can now forward and backfill in chained method * remove future warning * remove future warning properly speed up percentage calculation * fix project name * black * black * refactor away uses of log_with_attributes() in the algo code * refactoring away specific_log() * additional logger test for method 'clear' * refactor away uses of log_with_attributes() in the order and stack handler code * refactor away uses of log.setup() in fx code * Update vol.py * Formula fixes in risk report header * fix * black * docs: update ARM64 MacOS installation instructions * Update installation.md Fix suspected typo in command line for apple silicon. * Update setup.py Versions of 'matplotlib' beyond 3.8.0 do not offer support for Python 3.8. * repointing data refs to rawdata refs to make it easier to override * repointing data refs to rawdata refs to make it easier to override * repointing data refs to rawdata refs to make it easier to override * black * remove references to log.setup() from comments and messages * remove references to log.setup() from POC * removing log.setup() tests * removing setup() from adapter * removing reference to pst_logger * logging docs updated * link to python 3.10 logging docs * remove old POC logging doc * temp override needs to handle missing data * black * spelling * grammar * fix shadow cost default * spelling * spelling * spelling * spelling * inherite -> inherit * augemented -> augmented * overriden -> overridden * spelling * grammar * raise Exception, not str * Can only see emails now, not logs or errors * fix menu ID for broker orders * ago -> algo * missing ) * type not constructor * black * print branch name in job * remove unused method _resolve_names_and_add() * attempt to fix log error * replace futures_contract_object.log() with temp log_attributes() * Correct IBEX spread cost entry * initial * new docs doc outlining some recent changes * installation doc updated for new dependencies * link to new recent_changes doc * link to new recent_changes doc, TOC rebuilt * fix slow tests in master * black * latest action versions * Update instrumentconfig.csv * Update instrumentconfig.csv * Update spreadcosts.csv * Revert "Updating commissions and spreadcosts based on my observations" * setup not properly merged * fixes split frequency price file support - no longer using '/' character * adds support for import of split frequency price files * link to bc-utils usage guide * allow csv_data_path to be passed to __init__() * fix version so it matches CHANGELOG * also generating merged prices when importing split frequency CSV prices * Bump pyyaml from 5.3.1 to 5.4 Bumps [pyyaml](https://github.com/yaml/pyyaml) from 5.3.1 to 5.4. - [Changelog](https://github.com/yaml/pyyaml/blob/main/CHANGES) - [Commits](yaml/pyyaml@5.3.1...5.4) --- updated-dependencies: - dependency-name: pyyaml dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com> * fix robcarver17#1346 * updated some commissions on new instruments * commit --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: rob <robcarver17@googlemail.com> Co-authored-by: Andy Geach <andy@bugorfeature.net> Co-authored-by: Robert Carver <robcarver17@users.noreply.github.com> Co-authored-by: todd <3578666+tgibson11@users.noreply.github.com> Co-authored-by: Rob Carver <robcarver17@gmail.com> Co-authored-by: Mendel Friedman <77807315+meldinman@users.noreply.github.com> Co-authored-by: yuntai <yuntai@users.noreply.github.com> Co-authored-by: Hari Limaye <harilimaye@gmail.com> Co-authored-by: Abhishek Kumar <abhishek@corewars.org> Co-authored-by: fafuja <46684536+fafuja@users.noreply.github.com> Co-authored-by: Matthijs Vákár <m.i.l.vakar@uu.nl> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I am setting up a production instance and following the guide and ran into an issue. The guide mentions:
FX data:
Initialise the spot FX data in MongoDB from .csv files (this will be out of date, but you will update it in a moment)
Running:
pysystemtrade/blob/master/sysinit/futures/repocsv_spotfx_prices.py
I get the following error:
Traceback (most recent call last):
File "/home/josh/pysystemtrade/sysinit/futures/repocsv_spotfx_prices.py", line 26, in
db_fx_price_data.add_fx_prices(
File "/home/josh/pysystemtrade/sysdata/fx/spotfx.py", line 176, in add_fx_prices
if self.is_code_in_data(code):
File "/home/josh/pysystemtrade/sysdata/fx/spotfx.py", line 167, in is_code_in_data
if code in self.get_list_of_fxcodes():
File "/home/josh/pysystemtrade/sysdata/fx/spotfx.py", line 234, in get_list_of_fxcodes
raise NotImplementedError(USE_CHILD_CLASS_ERROR)
NotImplementedError: You need to use a child class of fxPricesData
My workaround:
I was able to make this change to fix it. I am just learning the system/code, not sure if this is the best fix or will cause other issues.
The text was updated successfully, but these errors were encountered: