Merge pull request #1311 from pyiron/output_files

Add output files class

.ci_support/environment-docs.yml

@@ -11,6 +11,7 @@ dependencies:
 - h5io_browser =0.0.7
 - h5py =3.10.0
 - jinja2 =3.1.3
+- monty =2024.2.2
 - numpy =1.26.3
 - pandas =2.2.0
 - pint =0.23

.ci_support/environment-old.yml

@@ -6,6 +6,7 @@ dependencies:
 - h5io_browser =0.0.6
 - h5py =3.6.0
 - jinja2 =2.11.3
+- monty =2024.2.2
 - numpy =1.23.5
 - pandas =2.0.0
 - pint =0.18

.ci_support/environment.yml

@@ -8,6 +8,7 @@ dependencies:
 - h5io_browser =0.0.7
 - h5py =3.10.0
 - jinja2 =3.1.3
+- monty =2024.2.2
 - numpy =1.26.3
 - pandas =2.2.0
 - pint =0.23

pyiron_base/jobs/job/core.py

@@ -10,6 +10,7 @@
 import os
 import posixpath
 import shutil
+from typing import List
 import warnings
 
 from pyiron_base.interfaces.has_groups import HasGroups
@@ -26,11 +27,15 @@
     _job_is_compressed,
     _job_compress,
     _job_decompress,
+    _job_list_files,
+    _job_read_file,
     _job_delete_files,
     _job_delete_hdf,
     _job_remove_folder,
 )
 from pyiron_base.state import state
+from pyiron_base.utils.deprecate import deprecate
+from pyiron_base.jobs.job.extension.files import FileBrowser
 
 __author__ = "Jan Janssen"
 __copyright__ = (
@@ -134,6 +139,12 @@ def __init__(self, project, job_name):
     def content(self):
         return self._hdf5_content
 
+    @property
+    def files(self):
+        return FileBrowser(working_directory=self.working_directory)
+
+    files.__doc__ = FileBrowser.__doc__
+
     @property
     def job_name(self):
         """
@@ -595,6 +606,7 @@ def get_job_id(self, job_specifier=None):
             return response[-1]["id"]
         return None
 
+    @deprecate("use job.files.list()")
     def list_files(self):
         """
         List files inside the working directory
@@ -605,9 +617,7 @@ def list_files(self):
         Returns:
             list: list of file names
         """
-        if os.path.isdir(self.working_directory):
-            return os.listdir(self.working_directory)
-        return []
+        return _job_list_files(self)
 
     def list_childs(self):
         """
@@ -898,9 +908,11 @@ def __getitem__(self, item):
         """
 
         if item in self.list_files():
-            file_name = posixpath.join(self.working_directory, "{}".format(item))
-            with open(file_name) as f:
-                return f.readlines()
+            warnings.warn(
+                "Using __getitem__ on a job to access files in deprecated: use job.files instead!",
+                category=DeprecationWarning,
+            )
+            return _job_read_file(self, item)
 
         # first try to access HDF5 directly to make the common case fast
         try:

pyiron_base/jobs/job/extension/files.py

@@ -0,0 +1,118 @@
+import os
+from typing import List
+from pyiron_base.jobs.job.util import (
+    _working_directory_list_files,
+    _working_directory_read_file,
+)
+
+
+class FileBrowser:
+    """
+    Allows to browse the files in a job directory.
+
+    By default this object prints itself as a listing of the job directory and
+    the files inside.
+
+    >>> job.files
+    /path/to/my/job:
+    \tpyiron.log
+    \terror.out
+
+    Access to the names of files is provided with :meth:`.list`
+
+    >>> job.files.list()
+    ['pyiron.log', 'error.out', 'INCAR']
+
+    Access to the contents of files is provided by indexing into this object,
+    which returns a list of lines in the file
+
+    >>> job.files['error.out']
+    ["Oh no\n", "Something went wrong!\n"]
+
+    The :meth:`.tail` method prints the last lines of a file to stdout
+
+    >>> job.files.tail('error.out', lines=1)
+    Something went wrong!
+
+    For files that have valid python variable names can also be accessed by
+    attribute notation
+
+    >>> job.files.INCAR # doctest: +SKIP
+    File('INCAR')
+    """
+
+    __slots__ = ("_working_directory",)
+
+    def __init__(self, working_directory):
+        self._working_directory = working_directory
+
+    def _get_file_dict(self):
+        return {
+            f.replace(".", "_"): f
+            for f in _working_directory_list_files(
+                working_directory=self._working_directory
+            )
+        }
+
+    def __dir__(self):
+        return list(self._get_file_dict().keys()) + super().__dir__()
+
+    def list(self) -> List[str]:
+        """
+        List all files in the working directory of the job.
+        """
+        return _working_directory_list_files(working_directory=self._working_directory)
+
+    def _ipython_display_(self):
+        path = self._job.working_directory + ":"
+        files = [
+            "\t" + f
+            for f in _working_directory_list_files(
+                working_directory=self._working_directory
+            )
+        ]
+        print(os.linesep.join([path, *files]))
+
+    def tail(self, file: str, lines: int = 100):
+        """
+        Print the last lines of a file.
+
+        Args:
+            file (str): filename
+            lines (int): number of lines to print
+
+        Raises:
+            FileNotFoundError: if the given file does not exist
+        """
+        print(
+            *_working_directory_read_file(
+                working_directory=self._working_directory, file_name=file, tail=lines
+            ),
+            sep="",
+        )
+
+    def __getitem__(self, item):
+        if item not in _working_directory_list_files(
+            working_directory=self._working_directory
+        ):
+            raise KeyError(item)
+
+        return File(os.path.join(self._working_directory, item))
+
+    def __getattr__(self, item):
+        try:
+            return self[self._get_file_dict()[item]]
+        except KeyError:
+            raise AttributeError(item) from None
+
+
+class File(str):
+    def tail(self, lines: int = 100):
+        print(
+            *_working_directory_read_file(
+                working_directory=os.path.dirname(self),
+                file_name=os.path.basename(self),
+                tail=lines,
+            ),
+            sep="",
+        )