Python java.io.File() Examples

def create_index():

    lucene.initVM()
    if os.path.exists(prm.index_folder):
        shutil.rmtree(prm.index_folder)

    indexDir = SimpleFSDirectory(File(prm.index_folder))
    writerConfig = IndexWriterConfig(Version.LUCENE_4_10_1, StandardAnalyzer())
    writer = IndexWriter(indexDir, writerConfig)
    wk = wiki.Wiki(prm.pages_path)

    print "%d docs in index" % writer.numDocs()
    print "Reading files from wikipedia..."
    n = 0
    for l in wk.get_text_iter():
        doc = Document()
        doc.add(Field("text", l, Field.Store.YES, Field.Index.ANALYZED))
        doc.add(Field("id", str(n), Field.Store.YES, Field.Index.ANALYZED))
        writer.addDocument(doc)
        n += 1
        if n % 100000 == 0:
            print 'indexing article', n
    print "Indexed %d docs from wikipedia (%d docs in index)" % (n, writer.numDocs())
    print "Closing index of %d docs..." % writer.numDocs()
    writer.close() 

def execute( self, cmd ):
        """Execute cmd in a shell, and return the java.lang.Process instance.
        Accepts either a string command to be executed in a shell,
        or a sequence of [executable, args...].
        """
        shellCmd = self._formatCmd( cmd )

        env = self._formatEnvironment( self.environment )
        try:
            p = Runtime.getRuntime().exec( shellCmd, env, File(os.getcwdu()) )
            return p
        except IOException, ex:
            raise OSError(
                0,
                "Failed to execute command (%s): %s" % ( shellCmd, ex )
                )

    ########## utility methods 

def __init__(self, dir=None, label=None):
        if not dir: dir = os.getcwd()
        if not label: label = "FileTree"
        dir = File(dir)
        self._dir = dir
        self.this = JPanel()
        self.this.setLayout(BorderLayout())

        # Add a label
        self.this.add(BorderLayout.PAGE_START, JLabel(label))

        # Make a tree list with all the nodes, and make it a JTree
        tree = JTree(self._add_nodes(None, dir))
        tree.setRootVisible(False)
        self._tree = tree

        # Lastly, put the JTree into a JScrollPane.
        scrollpane = JScrollPane()
        scrollpane.getViewport().add(tree)
        self.this.add(BorderLayout.CENTER, scrollpane) 

def getatime(path):
    # We can't detect access time so we return modification time. This
    # matches the behaviour in os.stat().
    path = _tostr(path, "getatime")
    f = File(sys.getPath(path))
    if not f.exists():
        raise OSError(0, 'No such file or directory', path)
    return f.lastModified() / 1000.0


# expandvars is stolen from CPython-2.1.1's Lib/ntpath.py:

# Expand paths containing shell variable substitutions.
# The following rules apply:
#       - no expansion within single quotes
#       - no escape character, except for '$$' which is translated into '$'
#       - ${varname} is accepted.
#       - varnames can be made out of letters, digits and the character '_'
# XXX With COMMAND.COM you can use any characters in a variable name,
# XXX except '^|<>='. 

def execute( self, cmd ):
        """Execute cmd in a shell, and return the java.lang.Process instance.
        Accepts either a string command to be executed in a shell,
        or a sequence of [executable, args...].
        """
        shellCmd = self._formatCmd( cmd )

        env = self._formatEnvironment( self.environment )
        try:
            p = Runtime.getRuntime().exec( shellCmd, env, File(os.getcwdu()) )
            return p
        except IOException, ex:
            raise OSError(
                0,
                "Failed to execute command (%s): %s" % ( shellCmd, ex )
                )

    ########## utility methods 

def getatime(path):
    # We can't detect access time so we return modification time. This
    # matches the behaviour in os.stat().
    path = _tostr(path, "getatime")
    f = File(sys.getPath(path))
    if not f.exists():
        raise OSError(0, 'No such file or directory', path)
    return f.lastModified() / 1000.0


# expandvars is stolen from CPython-2.1.1's Lib/ntpath.py:

# Expand paths containing shell variable substitutions.
# The following rules apply:
#       - no expansion within single quotes
#       - no escape character, except for '$$' which is translated into '$'
#       - ${varname} is accepted.
#       - varnames can be made out of letters, digits and the character '_'
# XXX With COMMAND.COM you can use any characters in a variable name,
# XXX except '^|<>='. 

def execute( self, cmd ):
        """Execute cmd in a shell, and return the java.lang.Process instance.
        Accepts either a string command to be executed in a shell,
        or a sequence of [executable, args...].
        """
        shellCmd = self._formatCmd( cmd )

        env = self._formatEnvironment( self.environment )
        try:
            p = Runtime.getRuntime().exec( shellCmd, env, File(os.getcwdu()) )
            return p
        except IOException, ex:
            raise OSError(
                0,
                "Failed to execute command (%s): %s" % ( shellCmd, ex )
                )

    ########## utility methods 

def make_jar_classloader(jar):
        import os
        from java.net import URL, URLClassLoader
        from java.io import File

        if isinstance(jar, bytes): # Java will expect a unicode file name
            jar = jar.decode(sys.getfilesystemencoding())
        jar_url = File(jar).toURI().toURL().toString()
        url = URL(u'jar:%s!/' % jar_url)

        if is_jython_nt:
            # URLJarFiles keep a cached open file handle to the jar even
            # after this ClassLoader is GC'ed, disallowing Windows tests
            # from removing the jar file from disk when finished with it
            conn = url.openConnection()
            if conn.getDefaultUseCaches():
                # XXX: Globally turn off jar caching: this stupid
                # instance method actually toggles a static flag. Need a
                # better fix
                conn.setDefaultUseCaches(False)

        return URLClassLoader([url])

# Filename used for testing 

def getatime(path):
    # We can't detect access time so we return modification time. This
    # matches the behaviour in os.stat().
    path = _tostr(path, "getatime")
    f = File(sys.getPath(path))
    if not f.exists():
        raise OSError(0, 'No such file or directory', path)
    return f.lastModified() / 1000.0


# expandvars is stolen from CPython-2.1.1's Lib/ntpath.py:

# Expand paths containing shell variable substitutions.
# The following rules apply:
#       - no expansion within single quotes
#       - no escape character, except for '$$' which is translated into '$'
#       - ${varname} is accepted.
#       - varnames can be made out of letters, digits and the character '_'
# XXX With COMMAND.COM you can use any characters in a variable name,
# XXX except '^|<>='. 

def __init__(self, index_dir, use_ram=False, jvm_ram=None):
        global lucene_vm_init
        if not lucene_vm_init:
            if jvm_ram:
                # e.g. jvm_ram = "8g"
                print "Increased JVM ram"
                lucene.initVM(vmargs=['-Djava.awt.headless=true'], maxheap=jvm_ram)
            else:
                lucene.initVM(vmargs=['-Djava.awt.headless=true'])
            lucene_vm_init = True
        self.dir = SimpleFSDirectory(File(index_dir))

        self.use_ram = use_ram
        if use_ram:
            print "Using ram directory..."
            self.ram_dir = RAMDirectory(self.dir, IOContext.DEFAULT)
        self.analyzer = None
        self.reader = None
        self.searcher = None
        self.writer = None
        self.ldf = None
        print "Connected to index " + index_dir 

def _init_index(self):

        if not os.path.exists(self.corpus.path):
            os.mkdir(self.corpus.path)
        try:
            searcher = IndexSearcher(SimpleFSDirectory(File(self.corpus.path)), True)
        #except lucene.JavaError:
        except:
            analyzer = self.corpus.analyzer
            writer = IndexWriter(SimpleFSDirectory(File(self.corpus.path)), analyzer, True, IndexWriter.MaxFieldLength.LIMITED)
            writer.setMaxFieldLength(1048576)
            writer.optimize()
            writer.close()

        self.lucene_index = SimpleFSDirectory(File(self.corpus.path))
        self.searcher = IndexSearcher(self.lucene_index, True)
        self.reader = IndexReader.open(self.lucene_index, True)
        self.analyzer = self.corpus.analyzer 

def import_csv(self, csv_file):

        try:
            writer = IndexWriter(SimpleFSDirectory(File(self.corpus.path)), self.analyzer, False,
                                        IndexWriter.MaxFieldLength.LIMITED)
            changed_rows = addmetadata.add_metadata_from_csv(self.searcher, self.reader, writer, csv_file,self.args_dir,
                                                             new_files=True)
            writer.close()
        except UnicodeDecodeError:
            try:
                writer.close()
            except:
                pass
            self.parent.write({'error': 'CSV import failed: file contained non-unicode characters. Please save the file with UTF-8 encoding and try again!'})
            return
        self.parent.write({'message': "CSV import complete: %s rows added." % (changed_rows,)}) 

def execute( self, cmd ):
        """Execute cmd in a shell, and return the java.lang.Process instance.
        Accepts either a string command to be executed in a shell,
        or a sequence of [executable, args...].
        """
        shellCmd = self._formatCmd( cmd )

        env = self._formatEnvironment( self.environment )
        try:
            p = Runtime.getRuntime().exec( shellCmd, env, File(os.getcwdu()) )
            return p
        except IOException, ex:
            raise OSError(
                0,
                "Failed to execute command (%s): %s" % ( shellCmd, ex )
                )

    ########## utility methods 

def execute( self, cmd ):
        """Execute cmd in a shell, and return the java.lang.Process instance.
        Accepts either a string command to be executed in a shell,
        or a sequence of [executable, args...].
        """
        shellCmd = self._formatCmd( cmd )

        env = self._formatEnvironment( self.environment )
        try:
            p = Runtime.getRuntime().exec( shellCmd, env, File(os.getcwd()) )
            return p
        except IOException, ex:
            raise OSError(
                0,
                "Failed to execute command (%s): %s" % ( shellCmd, ex )
                )

    ########## utility methods 

def isdir(path):
    """Test whether a path is a directory"""
    path = _tostr(path, "isdir")
    return File(sys.getPath(path)).isDirectory() 

def getsize(path):
    path = _tostr(path, "getsize")
    f = File(sys.getPath(path))
    size = f.length()
    # Sadly, if the returned length is zero, we don't really know if the file
    # is zero sized or does not exist.
    if size == 0 and not f.exists():
        raise OSError(0, 'No such file or directory', path)
    return size 

def join(path, *args):
    """Join two or more pathname components, inserting os.sep as needed"""
    path = _tostr(path, "join")
    f = File(path)
    for a in args:
        a = _tostr(a, "join")
        g = File(a)
        if g.isAbsolute() or len(f.getPath()) == 0:
            f = g
        else:
            if a == "":
                a = os.sep
            f = File(f, a)
    return asPyString(f.getPath()) 

def isfile(path):
    """Test whether a path is a regular file"""
    path = _tostr(path, "isfile")
    return File(sys.getPath(path)).isFile() 

def isdir(path):
    """Test whether a path is a directory"""
    path = _tostr(path, "isdir")
    return File(sys.getPath(path)).isDirectory() 

def isabs(path):
    """Test whether a path is absolute"""
    path = _tostr(path, "isabs")
    return File(path).isAbsolute() 

def getmtime(path):
    path = _tostr(path, "getmtime")
    f = File(sys.getPath(path))
    if not f.exists():
        raise OSError(0, 'No such file or directory', path)
    return f.lastModified() / 1000.0 

def exists(path):
    """Test whether a path exists.

    Returns false for broken symbolic links.

    """
    path = _tostr(path, "exists")
    return File(sys.getPath(path)).exists() 

def basename(path):
    """Return the final component of a pathname"""
    path = _tostr(path, "basename")
    return asPyString(File(path).getName()) 

def dirname(path):
    """Return the directory component of a pathname"""
    path = _tostr(path, "dirname")
    result = asPyString(File(path).getParent())
    if not result:
        if isabs(path):
            result = path # Must be root
        else:
            result = ""
    return result 

def _abspath(path):
    # Must use normpath separately because getAbsolutePath doesn't normalize
    # and getCanonicalPath would eliminate symlinks.
    return normpath(asPyString(File(sys.getPath(path)).getAbsolutePath())) 

def __init__(self, index_dir, mode, date_format='%Y-%m-%dT%H:%M:%S'):
        """Constructor of Indexer.

        Parameters
        ----------
        index_dir : string
            The location of lucene index
        mode : string
            The mode when opening lucene index. Available values are:
                'create', open new index and overwriting over index,
                'append', open existed index and append.
                'create_or_append', if `index_dir` exists, 'append',
                else 'create'
        date_format : string
            We save datetime field as string, `date_format` specify how to
            format datetime into string.
        """
        # self.store = FSDirectory.open(File(index_dir))
        self.store = FSDirectory.open(Paths.get(index_dir))
        # self.analyzer = StandardAnalyzer(Version.LUCENE_CURRENT)
        self.analyzer = StandardAnalyzer()
        # self.config = IndexWriterConfig(Version.LUCENE_CURRENT, self.analyzer)
        self.config = IndexWriterConfig(self.analyzer)
        self.mode = mode
        self.date_format = date_format
        if mode == 'create_or_append':
            self.config.setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND)
        elif mode == 'create':
            self.config.setOpenMode(IndexWriterConfig.OpenMode.CREATE)
        elif mode == 'append':
            self.config.setOpenMode(IndexWriterConfig.OpenMode.APPEND)
        else:
            raise ValueError('Invalid mode %s', mode)
        self.writer = IndexWriter(self.store, self.config) 

def join(path, *args):
    """Join two or more pathname components, inserting os.sep as needed"""
    path = _tostr(path, "join")
    f = File(path)
    for a in args:
        a = _tostr(a, "join")
        g = File(a)
        if g.isAbsolute() or len(f.getPath()) == 0:
            f = g
        else:
            if a == "":
                a = os.sep
            f = File(f, a)
    return asPyString(f.getPath()) 

def expanduser(path):
    if path[:1] == "~":
        c = path[1:2]
        if not c:
            return gethome()
        if c == os.sep:
            return asPyString(File(gethome(), path[2:]).getPath())
    return path 

def _realpath(path):
    try:
        return asPyString(File(sys.getPath(path)).getCanonicalPath())
    except java.io.IOException:
        return _abspath(path) 

def getmtime(path):
    path = _tostr(path, "getmtime")
    f = File(sys.getPath(path))
    if not f.exists():
        raise OSError(0, 'No such file or directory', path)
    return f.lastModified() / 1000.0