tools/parsers: add new parsing functions

.gitignore

@@ -213,3 +213,5 @@ dev/
 # Jupytext sync is a tool for writing notebooks in `.ipynb` format, and syncing them to a `.md` file
 # https://marketplace.visualstudio.com/items?itemName=caenrigen.jupytext-sync
 .jupytext-sync-ipynb
+
+examples

src/aiida_epw/tools/parsers.py

@@ -1,16 +1,87 @@
-"""Manual parsing functions for post-processing."""
+"""Manual parsing functions for post-processing. These functions can either be used independently or as helper functions of the `EpwParser` class.
+It would be good to always have a `parsed_data` dictionary as an output."""
 
 import numpy
+import re
+import io
 
 Ry2eV = 13.605662285137
 
 
+### This parse function is used for parsing the `band.eig` and `phband.freq` files.
+def parse_epw_bands(file_content):
+    """Parse the contents of a band structure file generated by EPW software."""
+    parsed_data = {}
+
+    nbnd, nks = (
+        int(v)
+        for v in re.search(r"&plot nbnd=\s+(\d+), nks=\s+(\d+)", file_content).groups()
+    )
+    kpt_pattern = re.compile(r"\s([\s-][\d\.]+)" * 3)
+    band_pattern = re.compile(r"\s+([-\d\.]+)" * nbnd)
+
+    kpoints = []
+    bands = []
+
+    for number, line in enumerate(file_content.splitlines()):
+        match_kpt = re.search(kpt_pattern, line)
+        if match_kpt and number % 2 == 1:
+            kpoints.append(list(match_kpt.groups()))
+
+        match_band = re.search(band_pattern, line)
+        if match_band and number % 2 == 0:
+            bands.append(list(match_band.groups()))
+
+    parsed_data["kpoints"] = numpy.array(kpoints, dtype=float)
+    parsed_data["bands"] = numpy.array(bands, dtype=float)
+
+    return parsed_data
+
+
+def parse_epw_eldos(file_content):
+    """Parse the contents of the `.dos` file.
+    parameters:
+        file_content: the content of the `.dos` file as a string.
+    returns:
+        parsed_data: a dictionary containing the energy, the electronic density of states and the integrated density of states.
+    """
+    parsed_data = {}
+
+    dos = numpy.loadtxt(io.StringIO(file_content), dtype=float, comments="#")
+
+    parsed_data["energy"] = dos[:, 0]
+    parsed_data["edos"] = dos[:, 1]
+    parsed_data["integrated_dos"] = dos[:, 2]
+    return parsed_data
+
+
+def parse_epw_phdos(file_content):
+    """Parse the contents of the `.phdos` file.
+    parameters:
+        file_content: the content of the `.phdos` file as a string.
+    returns:
+        parsed_data: a dictionary containing the frequency, the phonon density of states. Note that there are multiple phonon density of states for different smearing.
+    """
+    parsed_data = {}
+
+    phdos = numpy.loadtxt(io.StringIO(file_content), dtype=float, skiprows=1)
+    parsed_data["frequency"] = phdos[:, 0]
+    parsed_data["phdos"] = phdos[:, 1:]
+    return parsed_data
+
+
 def parse_epw_a2f(file_content):
+    """Parse the contents of the `.a2f` file.
+    parameters:
+        file_content: the content of the `.a2f` file as a string.
+    returns:
+        parsed_data: a dictionary containing the frequency, the spectral function for different smearing, and metadata such as electron smearing, Fermi window and summed el-ph coupling strength.
+    """
     parsed_data = {}
 
     a2f, footer = file_content.split("\n Integrated el-ph coupling")
 
-    a2f_array = numpy.array([line.split() for line in a2f.split("\n")], dtype=float)
+    a2f_array = numpy.array([line.split() for line in a2f.split("\n")[1:]], dtype=float)
     parsed_data["frequency"] = a2f_array[:, 0]
     parsed_data["a2f"] = a2f_array[:, 1:]
 
@@ -31,3 +102,62 @@ def parse_epw_a2f(file_content):
                 parsed_data[property] = float(line.split()[-1])
 
     return parsed_data
+
+
+def parse_epw_max_eigenvalue(file_content):
+    """Parse the max_eigenvalue part of the `stdout` file when solving the linearized Eliashberg equation."""
+    parsed_data = {}
+    re_pattern = re.compile(r"\s+([\d\.]+)\s+([\d\.-]+)\s+\d+\s+[\d\.]+\s+\d+\n")
+    parsing_block = file_content.split(
+        "Finish: Solving (isotropic) linearized Eliashberg"
+    )[0]
+
+    parsed_data["max_eigenvalue"] = numpy.array(
+        re_pattern.findall(parsing_block), dtype=float
+    )
+    return parsed_data
+
+
+def parse_epw_imag_iso(file_contents, prefix="aiida"):
+    """Parse the isotropic gap functions from EPW isotropic Eliashberg equation calculation.
+    parameters:
+        folder: the folder containing the `imag_iso` files. When serving as a helper function, it can take a `Retrieved` folder from aiida .
+        When used independently, it can take a local folder.
+        prefix: the prefix of the `imag_iso` files.
+    returns:
+        parsed_data: a dictionary containing the isotropic gap functions of numpy array type and the corresponding temperatures as keys.
+    """
+    parsed_data = {}
+    pattern_iso = re.compile(rf"^{prefix}\.imag_iso_(\d{{3}}\.\d{{2}})$")
+
+    for filename, file_content in file_contents.items():
+        match = pattern_iso.match(filename)
+        if match:
+            T = float(match.group(1))
+            gap_function = numpy.loadtxt(
+                io.StringIO(file_content), dtype=float, comments="#", skiprows=1
+            )
+            parsed_data[T] = gap_function
+    return parsed_data
+
+
+def parse_epw_imag_aniso_gap0(file_contents, prefix="aiida"):
+    """Parse the anisotropic gap functions from EPW anisotropic Eliashberg equation calculation.
+    parameters:
+        file_contents: a dictionary containing the file contents with filename as keys.
+        prefix: the prefix of the `imag_aniso_gap0` files.
+    returns:
+        parsed_data: a sorted dictionary containing the anisotropic gap functions of numpy array type and the corresponding temperatures as keys.
+    """
+    parsed_data = {}
+    pattern_aniso_gap0 = re.compile(rf"^{prefix}\.imag_aniso_gap0_(\d{{3}}\.\d{{2}})$")
+
+    for filename, file_content in file_contents.items():
+        match = pattern_aniso_gap0.match(filename)
+        if match:
+            T = float(match.group(1))
+            gap_function = numpy.loadtxt(
+                io.StringIO(file_content), dtype=float, comments="#", skiprows=1
+            )
+            parsed_data[T] = gap_function
+    return parsed_data