Merge pull request #2 from sissaschool/develop

Develop

.github/CODEOWNERS

@@ -1,4 +1,4 @@
 # The following owners will be the default owners for everything
 # in the repo unless a later match takes precedence.
-sportran/*      @lorisercole
-sportran_gui/*  @rikigigi
+sportran/      @lorisercole
+sportran_gui/  @rikigigi

.github/workflows/ci.yml

@@ -39,7 +39,7 @@ jobs:
 
     strategy:
       matrix:
-        python-version: ['3.6', '3.7', '3.8', '3.9', '3.10']
+        python-version: ['3.7', '3.8', '3.9', '3.10', '3.11']
 
     steps:
     - name: Checkout

.github/workflows/publish.yml

@@ -60,7 +60,7 @@ jobs:
 
     strategy:
       matrix:
-        python-version: ['3.6', '3.7', '3.8', '3.9', '3.10']
+        python-version: ['3.7', '3.8', '3.9', '3.10', '3.11']
 
     steps:
     - name: Checkout

MANIFEST.in

@@ -3,9 +3,9 @@ include LICENSE.txt
 include setup.json
 include sportran/README.md
 include sportran/metadata.json
-include sportran/utils/plot_style.mplstyle
-exclude sportran/utils/blocks.py
-exclude sportran/utils/blockanalysis.py
+exclude sportran/utils/obsolete/
 include sportran_gui/README_GUI.md
 include sportran_gui/assets/icon.gif
 include sportran_gui/assets/languages.json
+include sportran/plotter/styles/api_style.mplstyle
+include sportran/plotter/styles/cli_style.mplstyle

README.md

@@ -9,7 +9,7 @@ A code to estimate transport coefficients from the cepstral analysis of a multi-
 https://sportran.readthedocs.io
 
 ### References
- - [Ercole L., Bertossa R., Bisacchi S., and Baroni S., "_SporTran: a code to estimate transport coefficients from the cepstral analysis of (multivariate) current time series_", *arXiv*:2202.11571 (2022)](https://arxiv.org/abs/2202.11571), submitted to *Comput. Phys. Commun.*
+ - [Ercole L., Bertossa R., Bisacchi S., and Baroni S., "_SporTran: a code to estimate transport coefficients from the cepstral analysis of (multivariate) current time series_", *Comput. Phys. Commun.*, 108470](https://doi.org/10.1016/j.cpc.2022.108470), [*arXiv*:2202.11571 (2022)](https://arxiv.org/abs/2202.11571)
  - (cepstral analysis) [Ercole, Marcolongo, Baroni, *Sci. Rep.* **7**, 15835 (2017)](https://doi.org/10.1038/s41598-017-15843-2)
  - (multicomponent systems) [Bertossa, Grasselli, Ercole, Baroni, *Phys. Rev. Lett.* **122**, 255901 (2019)](https://doi.org/10.1103/PhysRevLett.122.255901) ([arXiv](https://arxiv.org/abs/1808.03341))
  - (review) [Baroni, Bertossa, Ercole, Grasselli, Marcolongo, *Handbook of Materials Modeling* (2018)](https://doi.org/10.1007/978-3-319-50257-1_12-1) ([arXiv](https://arxiv.org/abs/1802.08006))

setup.json

@@ -1,21 +1,21 @@
 {
     "name": "sportran",
-    "version": "1.0.0rc2",
+    "version": "1.0.0rc4",
     "author": "Loris Ercole, Riccardo Bertossa, Sebastiano Bisacchi",
     "author_email": "[email protected]",
     "description": "Cepstral Data Analysis of current time series for Green-Kubo transport coefficients",
     "license": "GPL 3",
     "url": "https://github.com/sissaschool/sportran",
     "keywords": "cepstral data analysis thermal conductivity transport coefficients physics green-kubo",
-    "python_requires": ">=3.6.*, <4",
+    "python_requires": ">=3.7, <4",
     "classifiers": [
         "Development Status :: 5 - Production/Stable",
         "Programming Language :: Python",
-        "Programming Language :: Python :: 3.6",
         "Programming Language :: Python :: 3.7",
         "Programming Language :: Python :: 3.8",
         "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
+        "Programming Language :: Python :: 3.11",
         "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
         "Operating System :: OS Independent",
         "Intended Audience :: Science/Research",
@@ -42,7 +42,7 @@
         ],
         "tests": [
             "pytest>=5.1.0",
-            "pytest-regressions==2.2.0",
+            "pytest-regressions>=2.4.2",
             "pandas>=1.1.0",
             "testbook",
             "ipykernel"

sportran/analysis.py

@@ -130,7 +130,7 @@ def main():
     input_file_group.add_argument('--split', type=int, default=1,
             help='Build a time series with n*m independent processes (n is the number of processes of the original timeseries, m is the number provided with --split). The length of the new time series will be [original length]/m. (optional)')
 
-    lammps_group = input_file_group.add_argument_group('LAMMPS format settings')
+    lammps_group = parser.add_argument_group('LAMMPS input file format settings')
     lammps_group.add_argument('--run-keyword', type=str,
             help='Keyword that identifies the run to be read: a specific comment line placed just before the run command (only for "lammps" format)')
     lammps_group.add_argument('--structure', type=str,
@@ -211,6 +211,18 @@ def main():
     return 0
 
 
+def concatenate_if_not_none_with_labels(concat, labels=None):
+    out_arr = []
+    out_label = ''
+    if labels is None:
+        labels = ['' for i in concat]
+    for arr, label in zip(concat, labels):
+        if arr is not None:
+            out_arr.append(arr)
+            out_label += f' {label}'
+    return np.concatenate([out_arr], axis=1).transpose(), f'{out_label}\n'
+
+
 def run_analysis(args):
 
     inputfile = args.inputfile
@@ -472,8 +484,8 @@ def average(data, name, units=''):
 
     if not no_text_out:
         outfile_name = output + '.psd.dat'
-        outarray = np.c_[j.freqs_THz, j.psd, j.fpsd, j.logpsd, j.flogpsd]
-        outfile_header = 'freqs_THz  psd  fpsd  logpsd  flogpsd\n'
+        outarray, outfile_header = concatenate_if_not_none_with_labels(
+            [j.freqs_THz, j.psd, j.fpsd, j.logpsd, j.flogpsd], ['freqs_THz', 'psd', 'fpsd', 'logpsd', 'flogpsd'])
         np.savetxt(outfile_name, outarray, header=outfile_header, fmt=fmt)
         if j.MANY_CURRENTS:
             outfile_name = output + '.cospectrum.dat'
@@ -483,15 +495,17 @@ def average(data, name, units=''):
             np.savetxt(outfile_name, np.column_stack([outarray.real, outarray.imag]), fmt=fmt)
 
             outfile_name = output + '.cospectrum.filt.dat'
-            outarray = np.c_[j.freqs_THz,
-                             j.fcospectrum.reshape(
-                                 (j.fcospectrum.shape[0] * j.fcospectrum.shape[1], j.fcospectrum.shape[2])).transpose()]
-            np.savetxt(outfile_name, np.column_stack([outarray.real, outarray.imag]), fmt=fmt)
+            if j.fcospectrum is not None:
+                outarray = np.c_[j.freqs_THz,
+                                 j.fcospectrum.reshape((j.fcospectrum.shape[0] * j.fcospectrum.shape[1],
+                                                        j.fcospectrum.shape[2])).transpose()]
+                np.savetxt(outfile_name, np.column_stack([outarray.real, outarray.imag]), fmt=fmt)
 
         if resample:
             outfile_name = output + '.resampled_psd.dat'
-            outarray = np.c_[jf.freqs_THz, jf.psd, jf.fpsd, jf.logpsd, jf.flogpsd]
-            outfile_header = 'freqs_THz  psd  fpsd  logpsd  flogpsd\n'
+            outarray, outfile_header = concatenate_if_not_none_with_labels(
+                [jf.freqs_THz, jf.psd, jf.fpsd, jf.logpsd, jf.flogpsd],
+                ['freqs_THz', 'psd', 'fpsd', 'logpsd', 'flogpsd'])
             np.savetxt(outfile_name, outarray, header=outfile_header, fmt=fmt)
 
         outfile_name = output + '.cepstral.dat'
@@ -623,7 +637,8 @@ def write_old_binary(self, output):
         """Write old binary format."""
         opts = {'allow_pickle': False}
         optsa = {'axis': 1}
-        outarray = np.c_[self.j_freqs_THz, self.j_fpsd, self.j_flogpsd, self.j_psd, self.j_logpsd]
+        outarray, _ = concatenate_if_not_none_with_labels(
+            [self.j_freqs_THz, self.j_fpsd, self.j_flogpsd, self.j_psd, self.j_logpsd])
         np.save(output + '.psd.npy', outarray, **opts)
 
         if self.j_cospectrum is not None:
@@ -634,7 +649,8 @@ def write_old_binary(self, output):
             outarray = np.c_[self.j_freqs_THz, self.j_fcospectrum.reshape(-1, self.j_fcospectrum.shape[-1]).transpose()]
             np.save(output + '.cospectrum.filt.npy', outarray, **opts)
 
-        outarray = np.c_[self.jf_freqs_THz, self.jf_psd, self.jf_fpsd, self.jf_logpsd, self.jf_flogpsd]
+        outarray, _ = concatenate_if_not_none_with_labels(
+            [self.jf_freqs_THz, self.jf_psd, self.jf_fpsd, self.jf_logpsd, self.jf_flogpsd])
         np.save(output + '.resampled_psd.npy', outarray, **opts)
 
         outarray = np.c_[self.jf_cepf_logpsdK, self.jf_cepf_logpsdK_THEORY_std, self.jf_cepf_logtau,

sportran/i_o/read_lammps_log.py

@@ -50,6 +50,7 @@
 import numpy as np
 from time import time
 from sportran.utils import log
+from io import BytesIO, StringIO
 
 
 def is_string(string):
@@ -67,21 +68,51 @@ def is_vector_variable(string):
     return bracket
 
 
-def file_length(filename):
+def _get_file_length(f):
     i = -1
-    with open(filename) as f:
-        for i, l in enumerate(f, 1):
-            pass
+    for i, l in enumerate(f, 1):
+        pass
     return i
 
 
-def data_length(filename):
+def file_length(file):
+    i = -1
+    if isinstance(file, bytes):
+        f = BytesIO(file)
+        i = _get_file_length(f)
+        f.close()
+    elif isinstance(file, str):
+        with open(file) as f:
+            i = _get_file_length(f)
+    else:
+        raise ValueError('Unsupported data type for file: {}'.format(type(file)))
+
+    return i
+
+
+def _get_data_length(f):
+    i = 0
+    while is_string(f.readline().split()[0]):   # skip text lines
+        pass
+    for i, l in enumerate(f, 2):
+        pass
+
+    return i
+
+
+def data_length(file):
     i = 0
-    with open(filename) as f:
-        while is_string(f.readline().split()[0]):   # skip text lines
-            pass
-        for i, l in enumerate(f, 2):
-            pass
+
+    if isinstance(file, bytes):
+        f = BytesIO(file)
+        i = _get_data_length(f)
+        f.close()
+    elif isinstance(file, str):
+        with open(file) as f:
+            i = _get_data_length(f)
+    else:
+        raise ValueError('Unsupported data type for file: {}'.format(type(file)))
+
     return i
 
 
@@ -119,7 +150,7 @@ class LAMMPSLogFile(object):
 
 #############################################################################
   Example script:
-     jfile = LAMMPSLogFile(filename, run_keyword='PRODUCTION RUN')
+     jfile = LAMMPSLogFile(data_file, run_keyword='PRODUCTION RUN')
      jfile.read_datalines(NSTEPS=100, start_step=0, select_ckeys=['Step', 'Temp', 'flux'])
      print(jfile.data)
 
@@ -129,16 +160,20 @@ class LAMMPSLogFile(object):
 #############################################################################
     """
 
-    def __init__(self, filename, run_keyword=None, select_ckeys=None, **kwargs):
+    def __init__(self, data_file, run_keyword=None, select_ckeys=None, **kwargs):
         """
-        LAMMPSLogFile(filename, run_keyword, select_ckeys, **kwargs)
+        LAMMPSLogFile(data_file, run_keyword, select_ckeys, **kwargs)
 
         **kwargs:
             endrun_keyword  [default: 'Loop time']
             group_vectores  [default: True]
             GUI             [default: False]
         """
-        self.filename = filename
+
+        if not isinstance(data_file, (bytes, str)):
+            raise ValueError('Unsupported dat type for file: {}'.format(type(data_file)))
+
+        self.data_file = data_file
         self.run_keyword = run_keyword
         self.select_ckeys = select_ckeys
         self.endrun_keyword = kwargs.get('endrun_keyword', 'Loop time')
@@ -152,14 +187,14 @@ def __init__(self, filename, run_keyword=None, select_ckeys=None, **kwargs):
             global FloatProgress, display
 
         self._open_file()
-        self.MAX_NSTEPS = file_length(self.filename)
+        self.MAX_NSTEPS = file_length(self.data_file)
         self._read_ckeys(self.run_keyword, group_vectors)
         self.ckey = None
         return
 
     def __repr__(self):
         msg = 'TableFile:\n' + \
-              '  filename:      {}\n'.format(self.filename) + \
+              '  data_file:      {}\n'.format(self.data_file) + \
               '  all_ckeys:     {}\n'.format(self.all_ckeys) + \
               '  select_ckeys:  {}\n'.format(self.select_ckeys) + \
               '  used ckey:     {}\n'.format(self.ckey) + \
@@ -169,10 +204,13 @@ def __repr__(self):
 
     def _open_file(self):
         """Open the file."""
-        try:
-            self.file = open(self.filename, 'r')
-        except:
-            raise ValueError('File does not exist.')
+        if isinstance(self.data_file, bytes):
+            self.file = StringIO(BytesIO(self.data_file).read().decode('utf-8'))
+        elif isinstance(self.data_file, str):
+            try:
+                self.file = open(self.data_file, 'r')
+            except:
+                raise ValueError('File does not exist.')
         return
 
     def _read_ckeys(self, run_keyword, group_vectors=True):