n97fit 1.0.1__tar.gz

n97fit-1.0.1/AUTHORS

@@ -0,0 +1,9 @@
+The following people contributed to the code:
+
+Takakiyo Nakazawa - Center for Atmospheric and Oceanic Studies (CAOS), Tohoku University
+
+Misa Ishizawa - Environment and Climate Change Canada
+
+Taku Umezawa - National Institute for Environmental Studies (NIES)
+
+Hisashi Yashiro - National Institute for Environmental Studies (NIES)

n97fit-1.0.1/LICENSE

@@ -0,0 +1,25 @@
+Copyright (c) 1991-2024, n97fit developers
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+* Redistributions of source code must retain the above copyright notice,
+  this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+* Neither the name of the copyright holder nor the names of its contributors
+  may be used to endorse or promote products derived from this software
+  without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE 
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+

n97fit-1.0.1/MANIFEST.in

@@ -0,0 +1 @@
+include AUTHORS

n97fit-1.0.1/PKG-INFO

@@ -0,0 +1,181 @@
+Metadata-Version: 2.1
+Name: n97fit
+Version: 1.0.1
+Summary:  This is a program of data selection and curve fitting by a digital filtering technique including Reinsch-type cubic spline, Fourier harmonics and linear interpolation. It is to derive the fitted curve to the observed data ( CO2, d13C,..), as well as to separate the seasonal, short- and long-term variations from the data. 
+Author: Takakiyo Nakazawa
+Maintainer-email: Hisashi Yashio <yashiro.hisashi@nies.go.jp>
+License: Copyright (c) 1991-2024, n97fit developers
+        All rights reserved.
+        
+        Redistribution and use in source and binary forms, with or without
+        modification, are permitted provided that the following conditions are met:
+        * Redistributions of source code must retain the above copyright notice,
+          this list of conditions and the following disclaimer.
+        * Redistributions in binary form must reproduce the above copyright notice,
+          this list of conditions and the following disclaimer in the documentation
+          and/or other materials provided with the distribution.
+        * Neither the name of the copyright holder nor the names of its contributors
+          may be used to endorse or promote products derived from this software
+          without specific prior written permission.
+        
+        THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+        ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+        WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+        DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE 
+        FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+        (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+        LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+        ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+        (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+        SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+        
+        
+Classifier: Development Status :: 4 - Beta
+Classifier: License :: OSI Approved :: BSD License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Atmospheric Science
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+License-File: AUTHORS
+Requires-Dist: numpy
+Requires-Dist: scipy>=1.10
+Requires-Dist: pandas<3,>=2.0
+Provides-Extra: dev
+Requires-Dist: build; extra == "dev"
+
+n97fit: Curve fitting tool for trace gas observations
+======
+
+As known as "Nakazawa-fitting"
+
+## Reference Paper
+
+Nakazawa, T., Ishizawa, M., Higuchi, K. And Trivett, N.B.A. (1997), Two Curve Fitting Methods Applied to CO2 Flask Data. Environmetrics, 8: 197-218. [URL]
+
+[URL]: https://doi.org/10.1002/(SICI)1099-095X(199705)8:3<197::AID-ENV248>3.0.CO;2-C
+
+## Authors
+
+- Takakiyo Nakazawa - Original work
+
+See the AUTHORS file for details
+
+## License
+
+This project is licensed under the BSD 3-clause License - see the LICENSE file for details
+
+## Acknowledgments
+
+We would like to thank Prof. Shinji Morimoto and the staff and students of the Center for Atmospheric and Oceanic Studies in Tohoku University, who contributed to the improvement of this tool.
+
+# How to use
+
+## Prerequisites
+
+- python >= 3.8
+- numpy
+- pandas
+
+## Running as a module
+
+    $> python -m n97fit [input_file] [config_file] [output_dir]
+
+## Input data format
+
+    YYYYMMDD obsID   Lat   Lon   Hgt    Value  Anl.Date QF
+    19891011  2135    13   144     2   131.17  19891023 -1
+    19891012  2137    15   152     2   167.82  19891023  0
+    ...
+
+    YYYY     : Observation year
+    MM       : Observation month
+    DD       : Observation day
+    obsID    : Observation ID (e.g. Flask number)
+    Lat      : Latitude
+    Log      : Longtitude
+    Hgt      : Altitude
+    Value    : Time series of concentration, isotope ratio, etc.
+    Anl.Date : Analyzing/extracting day
+    QF       : Quality flag
+
+## Outputs
+
+- data\_all.txt       : All of the input data
+- data\_eff\_ave.txt  : Input data, averaged good quality data (flag=0) by "Range of day"
+- data\_eq\_ave.txt   : Input data, averaged and aligned to the output interval of fitting results
+- data\_rej.txt       : Input data, outliers
+- fit\_all.txt        : Fitting results
+- fit\_annual.txt     : Fitting results, annual mean
+- fit\_monthly.txt    : Fitting results, monthly mean
+- fit\_multimonth.txt : Fitting results, multi-monthly mean
+- fit\_onemonth.txt   : Fitting results, one harmonics
+- fit\_period.txt     : Fitting results, mean of the whole period
+
+# Release Notes
+
+## 2024-04-01
+
+- Version 1.0.1
+- Initial release of python version
+
+## Original Release Notes (FORTRAN version)
+
+
+    Program : lsfbf4-9-2.f
+
+          This is a program of data selection and curve fitting
+       by a digital filtering technique including Reinsch-type cubic
+       spline, Fourier harmonics and linear interpolation. It is to
+       derive the fitted curve to the observed data ( CO2, d13C,..),
+       as well as to separate the seasonal, short- and long-term
+       variations from the data.
+
+          The average seasonal cycle is represented by harmonics
+       (NH=0-6), but if you give NH, the average cycle with NH+1 is
+       first calculated and then smoothed by a Butterworth filter
+       which attenuates signals with a period of NH+1 to 50 %.
+          Linear interpolation is used to calculate equally spaced
+       data from residuals which are obtained by extracting the
+       average seasonal cycle from actual data.
+          The long-term trend and irregular parts of the seasonal
+       cycle are driven by smoothing equally spaced data ,obtained
+       above, with the Butterworth filter.  The extrapolated data
+       for the trend and overall fit are given by the fit to the
+       actual data with a Reinsch spline (T5=1825days), which are
+       altered by an iterative procedure.
+          Prior to use of this program, the number of harmonics has
+       to be determined by the program 'PERI'.  The digital filter
+      form of Reinsch spline is shown explicitly by the programs
+       'COX1' or 'COX' in which the transfer function is plotted.
+
+          KD must be sufficiently larger than the value of ((910
+       +2920)/(mean data interval expressed in day)+data number of
+       the record). KO must be sufficiently larger than the value
+       of (910+2920+length of the data record expressed in day).
+       If you want to change KD, you must change KD of subroutine
+       or function Fourier, Frc1, Reinsch, Setup, Chol1d, Smooth,
+       Extrad1 and Judge1,too.  If you want to change KO, you must
+       change N or KO in subroutines Bwf, Scycle, Yac and Judge1,
+       too.
+
+                                 By T. Nakazawa (Mar. 8, 1991)
+
+    * Before digital filtering, data values with more than 3 standard
+     errors of estimated fit are identified as outliers and rejected
+     from further analysis. (dafault setting)
+
+    * The order of Butterworth filter is set to be 26th for long-term
+     and to be 16th for seasonal cycle. (dafault setting)
+
+                                Last modified by M. Ishizawa
+                                                 (Feb. 4, 2002)
+
+    Fortran 90 version refactoring
+                        Developed by Hisashi Yashio (2008)

n97fit-1.0.1/README.md

@@ -0,0 +1,129 @@
+n97fit: Curve fitting tool for trace gas observations
+======
+
+As known as "Nakazawa-fitting"
+
+## Reference Paper
+
+Nakazawa, T., Ishizawa, M., Higuchi, K. And Trivett, N.B.A. (1997), Two Curve Fitting Methods Applied to CO2 Flask Data. Environmetrics, 8: 197-218. [URL]
+
+[URL]: https://doi.org/10.1002/(SICI)1099-095X(199705)8:3<197::AID-ENV248>3.0.CO;2-C
+
+## Authors
+
+- Takakiyo Nakazawa - Original work
+
+See the AUTHORS file for details
+
+## License
+
+This project is licensed under the BSD 3-clause License - see the LICENSE file for details
+
+## Acknowledgments
+
+We would like to thank Prof. Shinji Morimoto and the staff and students of the Center for Atmospheric and Oceanic Studies in Tohoku University, who contributed to the improvement of this tool.
+
+# How to use
+
+## Prerequisites
+
+- python >= 3.8
+- numpy
+- pandas
+
+## Running as a module
+
+    $> python -m n97fit [input_file] [config_file] [output_dir]
+
+## Input data format
+
+    YYYYMMDD obsID   Lat   Lon   Hgt    Value  Anl.Date QF
+    19891011  2135    13   144     2   131.17  19891023 -1
+    19891012  2137    15   152     2   167.82  19891023  0
+    ...
+
+    YYYY     : Observation year
+    MM       : Observation month
+    DD       : Observation day
+    obsID    : Observation ID (e.g. Flask number)
+    Lat      : Latitude
+    Log      : Longtitude
+    Hgt      : Altitude
+    Value    : Time series of concentration, isotope ratio, etc.
+    Anl.Date : Analyzing/extracting day
+    QF       : Quality flag
+
+## Outputs
+
+- data\_all.txt       : All of the input data
+- data\_eff\_ave.txt  : Input data, averaged good quality data (flag=0) by "Range of day"
+- data\_eq\_ave.txt   : Input data, averaged and aligned to the output interval of fitting results
+- data\_rej.txt       : Input data, outliers
+- fit\_all.txt        : Fitting results
+- fit\_annual.txt     : Fitting results, annual mean
+- fit\_monthly.txt    : Fitting results, monthly mean
+- fit\_multimonth.txt : Fitting results, multi-monthly mean
+- fit\_onemonth.txt   : Fitting results, one harmonics
+- fit\_period.txt     : Fitting results, mean of the whole period
+
+# Release Notes
+
+## 2024-04-01
+
+- Version 1.0.1
+- Initial release of python version
+
+## Original Release Notes (FORTRAN version)
+
+
+    Program : lsfbf4-9-2.f
+
+          This is a program of data selection and curve fitting
+       by a digital filtering technique including Reinsch-type cubic
+       spline, Fourier harmonics and linear interpolation. It is to
+       derive the fitted curve to the observed data ( CO2, d13C,..),
+       as well as to separate the seasonal, short- and long-term
+       variations from the data.
+
+          The average seasonal cycle is represented by harmonics
+       (NH=0-6), but if you give NH, the average cycle with NH+1 is
+       first calculated and then smoothed by a Butterworth filter
+       which attenuates signals with a period of NH+1 to 50 %.
+          Linear interpolation is used to calculate equally spaced
+       data from residuals which are obtained by extracting the
+       average seasonal cycle from actual data.
+          The long-term trend and irregular parts of the seasonal
+       cycle are driven by smoothing equally spaced data ,obtained
+       above, with the Butterworth filter.  The extrapolated data
+       for the trend and overall fit are given by the fit to the
+       actual data with a Reinsch spline (T5=1825days), which are
+       altered by an iterative procedure.
+          Prior to use of this program, the number of harmonics has
+       to be determined by the program 'PERI'.  The digital filter
+      form of Reinsch spline is shown explicitly by the programs
+       'COX1' or 'COX' in which the transfer function is plotted.
+
+          KD must be sufficiently larger than the value of ((910
+       +2920)/(mean data interval expressed in day)+data number of
+       the record). KO must be sufficiently larger than the value
+       of (910+2920+length of the data record expressed in day).
+       If you want to change KD, you must change KD of subroutine
+       or function Fourier, Frc1, Reinsch, Setup, Chol1d, Smooth,
+       Extrad1 and Judge1,too.  If you want to change KO, you must
+       change N or KO in subroutines Bwf, Scycle, Yac and Judge1,
+       too.
+
+                                 By T. Nakazawa (Mar. 8, 1991)
+
+    * Before digital filtering, data values with more than 3 standard
+     errors of estimated fit are identified as outliers and rejected
+     from further analysis. (dafault setting)
+
+    * The order of Butterworth filter is set to be 26th for long-term
+     and to be 16th for seasonal cycle. (dafault setting)
+
+                                Last modified by M. Ishizawa
+                                                 (Feb. 4, 2002)
+
+    Fortran 90 version refactoring
+                        Developed by Hisashi Yashio (2008)

n97fit-1.0.1/pyproject.toml

@@ -0,0 +1,66 @@
+[build-system]
+requires = ["setuptools>=64"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "n97fit"
+version = "1.0.1"
+license = {file = "LICENSE"}
+
+authors = [
+  {name = "Takakiyo Nakazawa"},
+]
+
+maintainers = [
+  {name = "Hisashi Yashio", email = "yashiro.hisashi@nies.go.jp"}
+]
+
+description = """
+ This is a program of data selection and curve fitting \
+ by a digital filtering technique including Reinsch-type cubic \
+ spline, Fourier harmonics and linear interpolation. It is to \
+ derive the fitted curve to the observed data ( CO2, d13C,..), \
+ as well as to separate the seasonal, short- and long-term \
+ variations from the data. \
+"""
+readme = "README.md"
+
+dependencies = [
+    "numpy",
+    "scipy>=1.10",
+    "pandas>=2.0,<3"
+]
+
+requires-python = ">=3.8"
+
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "License :: OSI Approved :: BSD License",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Topic :: Scientific/Engineering :: Atmospheric Science",
+]
+
+keywords = [""]
+
+[project.urls]
+#Homepage = ""
+#Documentation = ""
+#Repository = ""
+
+[project.scripts]
+n97fit = "n97fit:cli_main"
+
+[project.optional-dependencies]
+dev = [
+    "build"
+]
+
+[tool.setuptools]
+package-dir = {"" = "src"}
+include-package-data = true

n97fit-1.0.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

n97fit-1.0.1/src/n97fit/__init__.py

@@ -0,0 +1,6 @@
+from .n97fit import N97Fit, logger
+from .__main__ import cli_main
+
+from .io import read_n97format, write_fixedformat
+
+__all__ = ["N97Fit", "logger", "read_n97format", "write_fixedformat", "cli_main"]

n97fit-1.0.1/src/n97fit/__main__.py

@@ -0,0 +1,194 @@
+from typing import List
+from pathlib import Path
+
+from pandas import read_fwf
+
+from .io import read_n97format, write_fixedformat, logger
+from . import N97Fit
+
+
+def readcfg(file, nrows: int = None) -> List:
+    """
+    Read configuration file and return a numpy array.
+
+    Parameters:
+      n (int): Number of rows to read from the file.
+      filename (os.PathLike): Path to the configuration file.
+
+    Returns:
+      np.ndarray: Numpy array containing the configuration data.
+    """
+
+    logger.info("")
+    logger.info(" Config file         : %s", file)
+    logger.info(" +Number of list     : %d", nrows)
+
+    dtypes = {
+        "label": str,
+        "value": int,
+    }
+
+    conf = read_fwf(
+        file,
+        widths=[34, 4],
+        dtype=dtypes,
+        names=dtypes.keys(),
+        nrows=nrows,
+        skiprows=0,
+        header=None,
+    )
+
+    for i, row in conf.iterrows():
+        logger.info(" +%-34s %3d", row["label"], row["value"])
+
+    return conf["value"].values.tolist()
+
+
+def getargs():
+    import argparse
+
+    parser = argparse.ArgumentParser(
+        prog="n97fit",
+        description="A program of data selection and curve fitting.",
+    )
+    parser.add_argument("input", type=Path, help="input data file")
+    parser.add_argument("config", type=Path, help="config file")
+    parser.add_argument("outdir", type=Path, help="output directory")
+    args, _ = parser.parse_known_args()
+
+    return args.input, args.config, args.outdir
+
+
+def cli_main():
+    input, config, outdir = getargs()
+
+    [avgday, sigma, outday, nhm, cutlo, orderlo, orderhi, sw_iteration] = readcfg(
+        file=config, nrows=8
+    )
+
+    # Read input data
+    data = read_n97format(file=input)
+
+    # create N97Fit instance
+    n97fit = N97Fit(
+        avgday=avgday,
+        sigma=sigma,
+        nhm=nhm,
+        cutlo=cutlo,
+        orderlo=orderlo,
+        orderhi=orderhi,
+        sw_iteration=sw_iteration,
+        loglevel="INFO",
+    )
+
+    # exec fitting
+    n97fit.fit(data)
+
+    # fetch results
+    all_data = n97fit.data_all()
+    rej_data = n97fit.data_rej()
+    eff_ave = n97fit.data_eff_ave()
+    eq_ave = n97fit.data_eq_ave()
+    fit_all = n97fit.all(interval=outday)
+    fit_annual = n97fit.annual()
+    fit_monthly = n97fit.monthly()
+    fit_multimonth = n97fit.multimonth()
+    fit_period = n97fit.period()
+    fit_onemonth = n97fit.onemonth()
+
+    # write output
+    logger.info("")
+    logger.info(" * * * * * START Output * * * * *")
+
+    outdir = Path(outdir)
+    outdir.mkdir(parents=True, exist_ok=True)
+    logger.info(" Output directory    : %s", outdir)
+
+    write_fixedformat(
+        all_data,
+        Path(outdir, "data_all.txt"),
+        format={"flg": "%3d"},
+        float_format="%12.5f",
+    )
+
+    write_fixedformat(
+        rej_data,
+        Path(outdir, "data_rej.txt"),
+        format={"flg": "%3d"},
+        float_format="%12.5f",
+    )
+
+    write_fixedformat(
+        eff_ave,
+        Path(outdir, "data_eff_ave.txt"),
+        float_format="%12.5f",
+        date_format="%(excel)11d",
+    )
+
+    write_fixedformat(
+        eq_ave,
+        Path(outdir, "data_eq_ave.txt"),
+        float_format="%12.5f",
+        date_format="%(excel)11d",
+    )
+
+    write_fixedformat(
+        fit_all,
+        Path(outdir, "fit_all.txt"),
+        format={"year": "%9.5f"},
+        float_format="%13.7f",
+        date_format="%(excel)7d",
+    )
+
+    write_fixedformat(
+        fit_annual,
+        Path(outdir, "fit_annual.txt"),
+        format={"year": "%4d"},
+        float_format="%13.7f",
+        fillna={
+            "ave_fit": -999.9999999,
+            "stdev_fit": -99.9999999,
+            "ave_trend": -999.9999999,
+            "stdev_trend": -99.9999999,
+        },
+    )
+
+    write_fixedformat(
+        fit_monthly,
+        Path(outdir, "fit_monthly.txt"),
+        format={"year": "%4d", "mn": "%2d"},
+        float_format="%13.7f",
+        fillna={
+            "ave_fit": -999.9999999,
+            "stdev_fit": -99.9999999,
+            "ave_trend": -999.9999999,
+            "stdev_trend": -99.9999999,
+        },
+    )
+
+    write_fixedformat(
+        fit_onemonth,
+        Path(outdir, "fit_onemonth.txt"),
+        float_format="%13.7f",
+    )
+
+    write_fixedformat(
+        fit_multimonth,
+        Path(outdir, "fit_multimonth.txt"),
+        format={"period": "%10s", "mn": "%2d"},
+        float_format="%13.7f",
+    )
+
+    write_fixedformat(
+        fit_period,
+        Path(outdir, "fit_period.txt"),
+        format={"period": "%10s"},
+        float_format="%13.7f",
+    )
+
+    logger.info("")
+    logger.info(" ************** E N D ***************")
+
+
+if __name__ == "__main__":
+    cli_main()

n97fit-1.0.1/src/n97fit/date/__init__.py

@@ -0,0 +1,8 @@
+from .cday import date2cday, cday2date
+from .ijmapping import ijmapping
+
+__all__ = [
+    "date2cday",
+    "cday2date",
+    "ijmapping",
+]

n97fit-1.0.1/src/n97fit/date/cday.py

@@ -0,0 +1,54 @@
+from typing import Union
+from datetime import date, timedelta
+
+import pandas as pd
+
+# start date
+ys = 1904
+sdate = date(ys, 1, 1)
+
+# cache dictionaries
+cday_dict = dict()
+date_dict = dict()
+
+
+def date2cday(dt: Union[date, pd.Timestamp]) -> int:
+    """Converts a date to a day count from 1904-01-01
+
+    Args:
+        year (int): year
+        month (int): month
+        day (int): day
+
+    Returns:
+        int: day count
+    """
+    if isinstance(dt, pd.Timestamp):
+        dt = dt.date()
+
+    if dt in cday_dict:
+        return cday_dict[dt]
+
+    cday = dt.toordinal() - sdate.toordinal()
+    cday_dict[dt] = cday
+
+    return cday
+
+
+def cday2date(cday: int) -> date:
+    """Converts a day count from 1904-01-01 to a date
+
+    Args:
+        cday (int): day count
+
+    Returns:
+        date: date
+    """
+
+    if cday in date_dict:
+        return date_dict[cday]
+
+    cdate = sdate + timedelta(days=int(cday))
+    date_dict[cday] = cdate
+
+    return cdate