Skip to content

Latest commit

 

History

History
188 lines (138 loc) · 7.87 KB

CHANGELOG.md

File metadata and controls

188 lines (138 loc) · 7.87 KB

BICO Changelog

v1.6.7 | 30 Sep 2024

  • Fixed bug where the script would crash when there were inconsistencies or errors in the raw binary stream during the conversion of GA_DIAG_CODE. The crash was found in file 2021111013.X00 when extracting SIGNAL_STRENGTH from GA_DIAG_CODE and occurred most likely because of an erroneous raw binary file. In case an error (ValueError) occurs when trying to read SIGNAL_STRENGTH, the value is now set to -9999. Among many files over many years and sites, 2021111013.X00 was so far the only file where this error occurred.

v1.6.5 | 8 Apr 2024

  • Added: new datablock QCL-B, used at site CH-CHA in 2012
  • Added: new datablock QCL-D, used at site CH-CHA in 2009

v1.6.4 | 7 Apr 2024

  • Updated env: PyQt5 needed specific declaration to work in poetry env
  • Now showing all members when clicking on drop-down menus

v1.6.3 | 21 Mar 2024

  • Fixed version number

v1.6.2 | 21 Mar 2024

  • Added info about missing AGC value from IRGA to datablock info IRGA75-R1.md
  • Added new site CH-FOR, CH-HON and CH-TAN to site selection

v1.6.1 | 7 Feb 2024

Addition of new datablock info files

Added info about datablocks recorded by the rECord data logging script. These files are already logged in ASCII format and thus they do not need to be converted. Info about these files is stored in bico, to keep them close to where info about other datablocks are stored, all in one place. Thus, these files are informational only and are not used for binary conversions.

  • Added: new datablock info file R350-R1.md
  • Added: new datablock info file IRGA75-R1.md

v1.6.0 | 14 Nov 2023

  • Added: new datablock R350-B. As it turns out, this datablock was already relevant for R3-50 data logged from 2017 onwards. Up to now we used the R350-A datablock for these data. The difference is that R350-B also records the status address and status data, similar to HS50-A. The size of the datablock is 12 Byte, old data that were converted using the R350-A datablock instead of R350-B still produced correct data for wind and sonic temperature.
  • Added pdf r3-50-manual_200502.pdf to folder /docs. This is the manufacturer manual for R3-50 sonic anemometers from Feb 2005.
  • Updated links to GitHub repo.

v1.5.1 | 21 Oct 2023

v1.5.0 - 26 Jun 2023

  • Added: new datablock QCL-L, used at site CH-LAE in 2005
  • Added: new datablock QCL-ISO, used at site CH-LAE in 2008 (uncertain format)
  • Added: new datablock QCL-L2, used at site CH-LAE in 2005 (uncertain format)
  • Added pdf sonicread_20190503.pdf to folder /docs. This document describes the logging software Sonic Anemometer Data Aquisition Program (sonicread) used for eddy covariance raw data since 2004 (ongoing). Contains some info about data columns and their conversion to proper units.
  • Added pdf qcldoc_20200515.pdf to folder /docs. This document describes the logging software Quantum Cascade Laser (QCL) Data Aquisition Program (qclread) used for Aerodyne QCL instruments and contains some info about column order.
  • Added pdf lgrdoc_20180601.pdf to folder /docs. This document describes the logging software Los Gatos Research (LGR) Data Aquisition Program (lgrread) used for Los Gatos LGR instruments and contains some info about column order.
  • Changed: more padding/width for comboboxes to avoid text cutoffs

v1.4.0 - 15 May 2023

  • Now using poetry for dependency management (instead of conda)
    • The package versions for pandas, numpy, matplotlib and others remain the same as before
  • Removed: start_bico.py (is no longer used)
  • Removed: environment.yml (is no longer used)

v1.3.0 - 11 Aug 2022

  • Added: new datablock QCL-C3
  • Added: More descriptions in GUI about what numbers mean, e.g. using 0 for Row Limit Per File
  • Improved: Better (clearer) plotting of high-res data
  • Changed: Hi-res and aggregated plots do not use scientific notation (e.g., 1e9) or an offset (e.g., +1e9)
  • Updated links in sidebar
  • CRITICAL FIX: In bin.py, the return of line self.check_if_dblock_size_zero(dblock_true_size=dblock_true_size) was not stored in any variable, therefore this check seems to not have worked. Line was changed to end_of_data_reached = self.check_if_dblock_size_zero(dblock_true_size=dblock_true_size). Then indentation was removed for the following break command if end_of_data_reached: break # Stop for loop to really stop the for-loop. I assume this does not have major implications on the conversion but only acts as an additional check if data are still available.

v1.2.3 - 7 Dec 2021

  • Fixed: Script folder was wrong when running from CLI

v1.2.2 - 7 Dec 2021

  • Added: CLI flag -a to avoid duplicates when converting files. Is only considered if script is started via CLI.
  • Added: CLI flag -d to convert only the most recent number of days, e.g. -d 3 to convert files from the last 3 days. Is only considered if script is started via CLI.
  • Changed: Settings file is now called BICO.settings

v1.2.1 - 5 Dec 2021

  • Added automatic detection of working directory

v1.2 - 5 Dec 2021

  • Included CLI support for automatic script execution without GUI
  • Code refactoring: separated the conversion functions (BicoEngine) from GUI (BicoGUI)
  • Added: new datablock QCL-C2 (used at CH-DAV)
  • Changed: switching back to conda for dependency management

v1.1 - 14 Nov 2021

  • BICO is now using poetry for dependency management and packaging
  • The virtual env created by poetry is specific to BICO and stored alongside the source code in folder .venv
  • Updated Python version to 3.9.7, packages were updated to their newest versions
  • Removed conda env
  • Renamed source code folder to src
  • Adjusted folder structure
  • Added support for CLI execution
  • Solved some minor warnings

v1.0 - 20 Sep 2021

  • Updated version number to 1.
  • bug: removed small naming bug in HS100-A.md (SA_DIAG_TYPE, SA_DIAG_VAL)
  • bug: QCL-A3 was incorrectly labelled QCL-A in converted raw data files

v0.5.2 - 17 Mar 2021

  • added: datablock IRGA72-A-GN1.dblock

v0.5.1 - 4 Mar 2021

  • added: datablock IRGA72-B-GN1.dblock
  • added: link to datablock descriptions
  • the BICO run id is now shown in the "BICO finished." text message
  • some minor text adjustments, clarifications in datablock descriptions

v0.5.0 - 23 Feb 2021

  • bug: wrong number of bits for SIGNAL_STRENGTH in IRGA72-B.dblock
  • bug: reading order for extracted variables was wrong in IRGA72-B.dblock
  • bug: wrong number of bits for AGC in IRGA75-A.dblock and IRGA75-A-GN1.dblock
  • bug: reading order for extracted variables was wrong in IRGA75-A.dblock and IRGA75-A-GN1.dblock
  • changed: the 'apply_gain' option for SIGNAL_STRENGTH in IRGA72 datablocks was changed from 6.6666666666666666 to 6.67 to be more in line with their manual. This means that the max signal strength that can be reached is 100.05:
    • IRGA72-A.dblock
    • IRGA72-B.dblock
  • added: median is now also shown in hires plots
  • added: link for CHANGELOG

v0.4.0 - 15 Feb 2021

  • changed: For QCL and LGR instruments, PRESS_CELL in Torr is now converted to hPa.
    For the conversion, the parameter
    'apply_gain': 1.33322368 was used (was previously set to 1).
    The following .dblock files are affected:
    • LGR-A.dblock
    • QCL-A.dblock
    • QCL-A2.dblock
    • QCL-A4.dblock
    • QCL-C.dblock
    • not changed: QCL-A3.dblock because it does not record pressure
  • changed: CHANGELOG is now a markup file (.md)

v0.3.2 - 12 Feb 2021

  • fixed: GUI now shows correct date for version

v0.3.1 - 12 Feb 2021

  • small bug: IRGA75-A-GN1.dblock: datablock was incorrectly named IRGA75-A
  • added: CHANGELOG.txt

Ideas

  • TODO LATER parallelize, multiprocessing? import multiprocessing as mp print("Number of processors: ", mp.cpu_count())