ipyvasp 0.9.84__py2.py3-none-any.whl → 0.9.86__py2.py3-none-any.whl

ipyvasp/__init__.py

@@ -18,6 +18,7 @@ __all__ = [  # For documentation purpose
     "webshow",
     "load_results",
     "parse_text",
+    "get_lines",
     "summarize",
 ]
 

ipyvasp/_version.py

@@ -1 +1 @@
-__version__ = "0.9.84"
+__version__ = "0.9.86"

ipyvasp/core/parser.py

@@ -1,4 +1,4 @@
-__all__ = ["Vasprun", "Vaspout", "minify_vasprun", "xml2dict"]
+__all__ = ["Vasprun", "Vaspout", "minify_vasprun", "xml2dict","read"]
 
 import re
 from io import StringIO
@@ -114,6 +114,35 @@ class Vaspout(DataSource):
     def __init__(self, path):
         raise NotImplementedError("Vaspout is not implemented yet.")
 
+def read(file, start_match, stop_match=r'\n', nth_match=1, skip_last=False,apply=None):
+    """Reads a part of the file between start_match and stop_match and returns a generator. It is lazy and fast.
+    `start_match` and `stop_match`(default is end of same line) are regular expressions. `nth_match` is the number of occurence of start_match to start reading.
+    `skip_last` is used to determine whether to keep or skip last line.
+    `apply` should be None or `func` to transform each captured line.
+    """
+    if "|" in start_match:
+        raise ValueError(
+            "start_match should be a single match, so '|' character is not allowed."
+        )
+    with Path(file).open("r") as f:
+        lines = islice(f, None)  # this is fast
+        matched = False
+        n_start = 1
+        for line in lines:
+            if re.search(start_match, line, flags=re.DOTALL):
+                if nth_match != n_start:
+                    n_start += 1
+                else:
+                    matched = True
+            if matched and re.search(
+                stop_match, line, flags=re.DOTALL
+            ):  # avoid stop before start
+                matched = False
+                if not skip_last:
+                    yield apply(line) if callable(apply) else line
+                break  # stop reading
+            if matched:  # should be after break to handle last line above
+                yield apply(line) if callable(apply) else line
 
 class Vasprun(DataSource):
     "Reads vasprun.xml file lazily. It reads only the required data from the file when a plot or data access is requested."
@@ -129,36 +158,14 @@ class Vasprun(DataSource):
             skipk if isinstance(skipk, (int, np.integer)) else self.get_skipk()
         )
 
-    def read(self, start_match, stop_match, nth_match=1, skip_last=False):
+    def read(self, start_match, stop_match=r'\n', nth_match=1, skip_last=False,apply=None):
         """Reads a part of the file between start_match and stop_match and returns a generator. It is lazy and fast.
-        `start_match` and `stop_match` are regular expressions. `nth_match` is the number of occurence of start_match to start reading.
+        `start_match` and `stop_match`(default is end of same line) are regular expressions. `nth_match` is the number of occurence of start_match to start reading.
         `skip_last` is used to determine whether to keep or skip last line.
+        `apply` should be None or `func` to transform each captured line.
         """
-        if "|" in start_match:
-            raise ValueError(
-                "start_match should be a single match, so '|' character is not allowed."
-            )
-
-        with self.path.open("r") as f:
-            lines = islice(f, None)  # this is fast
-            matched = False
-            n_start = 1
-            for line in lines:
-                if re.search(start_match, line, flags=re.DOTALL):
-                    if nth_match != n_start:
-                        n_start += 1
-                    else:
-                        matched = True
-                if matched and re.search(
-                    stop_match, line, flags=re.DOTALL
-                ):  # avoid stop before start
-                    matched = False
-                    if not skip_last:
-                        yield line
-
-                    break  # stop reading
-                if matched:  # should be after break to handle last line above
-                    yield line
+        kws = {k:v for k,v in locals().items() if k !='self'}
+        return read(self.path,**kws)
 
     def get_skipk(self):
         "Returns the number of k-points to skip in band structure plot in case of HSE calculation."

ipyvasp/misc.py

@@ -70,10 +70,8 @@ class OUTCAR:
     def path(self):
         return self._path
 
-    @_sub_doc(vp.Vasprun.read)
-    @_sig_kwargs(vp.Vasprun.read, ("self",))
-    def read(self, start_match, stop_match, **kwargs):
-        return vp.Vasprun.read(
-            self, start_match, stop_match, **kwargs
-        )  # Pass all the arguments to the function
+    @_sub_doc(vp.read)
+    def read(self, start_match, stop_match=r'\n', nth_match=1, skip_last=False,apply=None):
+        kws = {k:v for k,v in locals().items() if k !='self'}
+        return vp.read(self.path, **kws)  # Pass all the arguments to the function
         

ipyvasp/utils.py

@@ -1,5 +1,6 @@
 __all__ = [
     "get_file_size",
+    "get_lines",
     "set_dir",
     "interpolate_data",
     "rolling_mean",
@@ -10,6 +11,7 @@ __all__ = [
 
 import re
 import os
+import io
 from contextlib import contextmanager
 from pathlib import Path
 from inspect import signature, getdoc
@@ -26,13 +28,45 @@ import matplotlib.pyplot as plt
 def get_file_size(path: str):
     """Return file size"""
     if (p := Path(path)).is_file():
-        size = p.stat.st_size
-        for unit in ["Bytes", "KB", "MB", "GB", "TB"]:
+        size = p.stat().st_size
+        for unit in ["B", "KB", "MB", "GB", "TB"]:
             if size < 1024.0:
                 return "%3.2f %s" % (size, unit)
             size /= 1024.0
     else:
         return ""
+    
+def get_lines(f, indices):
+    """Read lines by indexing from an opened file pointer `f`. Negative indexing is supported to read lines from end.
+    Returns a single str of line if one integer given, otherwise a list of lines.
+    This consumes a lot less memory then indexing over `f.readlines()[index]`.
+
+    >>> with open('some_file','r') as f:
+    >>>     get_lines(f, -1) # last line
+    >>>     get_lines(f, range(5)) # first 5 lines
+    >>>     get_lines(f, range(-5,0)) # last 5 lines
+    """
+    if not isinstance(f, io.TextIOWrapper):
+        raise TypeError(f"f should be file-like object. got {type(f)}")
+    
+    return_line = False
+    if isinstance(indices, int):
+        indices = [indices]
+        return_line = True
+
+    if not isinstance(indices, (tuple,list, range)):
+        raise TypeError(f"indices should int/list/tuple/range, got {type(indices)}")
+    
+    f.seek(0)
+    if min(indices) < 0:
+        if not hasattr(f, '_nlines'): # do this once, assuming file is not changed while reading
+            f._nlines = sum(1 for _ in enumerate(f))
+            f.seek(0)
+
+        indices = [i + (f._nlines if i < 0 else 0) for i in indices] # make all positive
+    
+    lines = [l for i, l in enumerate(f) if i in indices]
+    return lines[0] if return_line else lines
 
 
 def _sig_kwargs(from_func, skip_params=()):

ipyvasp/widgets.py

@@ -40,7 +40,7 @@ import plotly.graph_objects as go
 from . import utils as gu
 from . import lattice as lat
 from .core import serializer, parser as vp, plot_toolkit as ptk
-from .utils import _sig_kwargs, _sub_doc
+from .utils import _sig_kwargs, _sub_doc, get_file_size
 from ._enplots import _fmt_labels
 
 
@@ -68,7 +68,7 @@ def summarize(files, func, **kwargs):
     for name, path in files.items():
         output = func(path, **kwargs)
         if not isinstance(output, dict):
-            raise TypeError("Function must return a dictionary.")
+            raise TypeError("Function must return a dictionary to create DataFrame.")
 
         if "FILE" in output:
             raise KeyError(
@@ -115,6 +115,8 @@ class Files:
     Use methods on return such as `summarize`, `with_name`, `filtered`, `interact` and others.
 
     >>> Files(root_1, glob_1,...).add(root_2, glob_2,...) # Fully flexible to chain
+
+    WARNING: Don't use write operations on paths in files in batch mode, it can cause unrecoverable data loss.
     """
     def __init__(self, path_or_files = '.', glob = '*', exclude = None,files_only = False, dirs_only=False):
         if isinstance(path_or_files, Files):
@@ -168,10 +170,6 @@ class Files:
     def __add__(self, other):
         raise NotImplementedError("Use self.add method instead!")
 
-    def map(self,func):
-        "Map files to a function!"
-        return map(func, self._files)
-
     def with_name(self, name):
         "Change name of all files. Only keeps existing files."
         return self.__class__([f.with_name(name) for f in self._files])
@@ -420,6 +418,54 @@ class Files:
     def bands_widget(self, height='450px'):
         "Get BandsWidget instance with these files."
         return BandsWidget(files=self._files, height=height)
+    
+    def map(self,func, to_df=False):
+        """Map files to a function that takes path as argument. 
+        If `to_df=True`, func may return a dict to create named columns, or just two columns will be created.
+        Otherwise returns generator of elemnets `(path, func(path))`.
+        If you need to operate on opened file pointer, use `.mapf` instead.
+        
+        >>> import ipyvasp as ipv
+        >>> files = ipv.Files(...)
+        >>> files.map(lambda path: ipv.read(path, '<pattern>',apply = lambda line: float(line.split()[0])))
+        >>> files.map(lambda path: ipv.load(path), to_df=True) 
+        """
+        if to_df:
+            return self._try_return_df(func)
+        return ((path, func(path)) for path in self._files) # generator must
+    
+    def _try_return_df(self, func):
+        try: return summarize(self._files,func)
+        except: return pd.DataFrame(((path, func(path)) for path in self._files))
+    
+    def mapf(self, func, to_df=False,mode='r', encoding=None):
+        """Map files to a function that takes opened file pointer as argument. Opened files are automatically closed and should be in readonly mode.
+        Load files content into a generator sequence of  tuples like `(path, func(open(path)))` or DataFrame if `to_df=True`.
+        If `to_df=True`, func may return a dict to create named columns, or just two columns will be created.
+        If you need to operate on just path, use `.map` instead.
+        
+        >>> import json
+        >>> import ipyvasp as ipv
+        >>> files = ipv.Files(...)
+        >>> files.mapf(lambda fp: json.load(fp,cls=ipv.DecodeToNumpy),to_df=True) # or use ipv.load(path) in map
+        >>> files.mapf(lambda fp: ipv.get_lines(fp, range(5)) # read first five lines
+        >>> files.mapf(lambda fp: ipv.get_lines(fp, range(-5,0)) # read last five lines
+        >>> files.mapf(lambda fp: ipv.get_lines(fp, -1) # read last line
+        """
+        if not mode in 'rb':
+            raise ValueError("Only 'r'/'rb' mode is allowed in this context!")
+        
+        def loader(path):
+            with open(path, mode=mode,encoding=encoding) as f:
+                return func(f)
+
+        if to_df:
+            return self._try_return_df(loader)
+        return ((path, loader(path)) for path in self._files) # generator must
+        
+    def stat(self):
+        "Get files stat as DataFrame. Currently only size is supported."
+        return self.summarize(lambda path: {"size": get_file_size(path)})
 
 
 @fix_signature

{ipyvasp-0.9.84.dist-info → ipyvasp-0.9.86.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ipyvasp
-Version: 0.9.84
+Version: 0.9.86
 Summary: A processing tool for VASP DFT input/output processing in Jupyter Notebook.
 Home-page: https://github.com/massgh/ipyvasp
 Author: Abdul Saboor

{ipyvasp-0.9.84.dist-info → ipyvasp-0.9.86.dist-info}/RECORD

@@ -1,25 +1,25 @@
-ipyvasp/__init__.py,sha256=rlorju9arMtHw1QRYPljday-PyZWJdSCxg4lw3g6t0Q,1409
+ipyvasp/__init__.py,sha256=a9rUl8DdqIiZsj03LMJFnJKIuxmtmjmN2Dqj6hmjUls,1427
 ipyvasp/__main__.py,sha256=eJV1TZSiT8mC_VqAeksNnBI2I8mKMiPkEIlwikbtOjI,216
 ipyvasp/_enplots.py,sha256=D38paN8zqZgluNAwmCwcocd7-_h_T0HTGolI1eBkDes,37484
 ipyvasp/_lattice.py,sha256=kOseNCIWt-VCnkhFQZEcsXhyNYobjfqNfAl3seXHiVU,105584
-ipyvasp/_version.py,sha256=3xnjEM8AMDNSU-FRZtXZ39MIuWcHwZXBmMyzZckPxEs,24
+ipyvasp/_version.py,sha256=Ni3S6NaV45wRiYfJLhXykakqV8br57IhjNd9paj9-MY,24
 ipyvasp/bsdos.py,sha256=JvYvHLqMp3eVaJ0amD-9kxp7FehQIFq3WFUxsO5dj0Q,31794
 ipyvasp/cli.py,sha256=aWFEVhNmnW8eSOp5uh95JaDwLQ9K9nlCQcbnOSuhWgw,6844
 ipyvasp/evals_dataframe.py,sha256=-sqxK7LPV6sYDO_XXmZ80FznOaXTkVdbqJKKvTUtMak,20637
 ipyvasp/lattice.py,sha256=VfyhmbpRHA3nePWUmaoMmCiAehby_VvDuDcp34OK3rA,33685
-ipyvasp/misc.py,sha256=SZJ_ePUR2-HEKYTEpDHVRVE7zpIQVTCjiuw0BCC9UTU,2349
+ipyvasp/misc.py,sha256=gPA-71V_fuzyvinHR4P1kYMm90lJYdt24733pA5_4Kk,2340
 ipyvasp/potential.py,sha256=tzA73c5lkp6ahLSJchMrU043-QWaOV0nIOUA7VMmfKQ,11408
 ipyvasp/surface.py,sha256=MjE5oB0wW6Pca_C-xu8rN6OMH7lUEeNPNyM7Kz_Im-8,23766
-ipyvasp/utils.py,sha256=nw_oglztXfl0CZmo63PBstI4didTkuXhThYqhMP8DVI,14247
-ipyvasp/widgets.py,sha256=X4gPkTy--aFZpXGCQJVg4HByt5-mpVB6gNiFmTTIyRg,46628
+ipyvasp/utils.py,sha256=_AD1LMXpy-wNHJpTx4uEahQyDWv_KY70FGxoZCEjVLY,15598
+ipyvasp/widgets.py,sha256=dMMRD5mfTEPnG5j04wuQviGkKP8tcI5oqInny4qDd7I,49209
 ipyvasp/core/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-ipyvasp/core/parser.py,sha256=TmRoIhyIGJ4nr32tb1MAAy5clUZh0CQh-2hP8mnTaVc,38221
+ipyvasp/core/parser.py,sha256=if5DsBpDS7qCIlWsl6JD0ovRBk01h0CdKYa4vE9SSzw,38945
 ipyvasp/core/plot_toolkit.py,sha256=V-IQo7MrOhmNCGpWHIMtV04TmmubgFvGgH_yd4YmoX4,36158
 ipyvasp/core/serializer.py,sha256=v0ma4htirybtQo_wFhIEjkRoZMQk9ETDz85iloaq3YY,38427
 ipyvasp/core/spatial_toolkit.py,sha256=dXowREhiFzBvvr5f_bApzFhf8IzjH2E2Ix90oCBUetY,14885
-ipyvasp-0.9.84.dist-info/LICENSE,sha256=F3SO5RiAZOMfmMGf1KOuk2g_c4ObvuBJhd9iBLDgXoQ,1263
-ipyvasp-0.9.84.dist-info/METADATA,sha256=hX-hNmmx-Yh8dpfR3ZGAigi5cKt7yLQhI6vIWDiPqnI,2421
-ipyvasp-0.9.84.dist-info/WHEEL,sha256=iYlv5fX357PQyRT2o6tw1bN-YcKFFHKqB_LwHO5wP-g,110
-ipyvasp-0.9.84.dist-info/entry_points.txt,sha256=C7m0Sjmr14wFjflCkWXLzr5N6-cQj8uJC9n82mUtzt8,44
-ipyvasp-0.9.84.dist-info/top_level.txt,sha256=ftziWlMWu_1VpDP1sRTFrkfBnWxAi393HYDVu4wRhUk,8
-ipyvasp-0.9.84.dist-info/RECORD,,
+ipyvasp-0.9.86.dist-info/LICENSE,sha256=F3SO5RiAZOMfmMGf1KOuk2g_c4ObvuBJhd9iBLDgXoQ,1263
+ipyvasp-0.9.86.dist-info/METADATA,sha256=62X139t3R61h2ug7ZYsTQwTKSewuGwfyfS_kS7QpJhc,2421
+ipyvasp-0.9.86.dist-info/WHEEL,sha256=iYlv5fX357PQyRT2o6tw1bN-YcKFFHKqB_LwHO5wP-g,110
+ipyvasp-0.9.86.dist-info/entry_points.txt,sha256=C7m0Sjmr14wFjflCkWXLzr5N6-cQj8uJC9n82mUtzt8,44
+ipyvasp-0.9.86.dist-info/top_level.txt,sha256=ftziWlMWu_1VpDP1sRTFrkfBnWxAi393HYDVu4wRhUk,8
+ipyvasp-0.9.86.dist-info/RECORD,,