cpython/Lib/fileinput.py

414 lines
14 KiB
Python
Raw Normal View History

"""Helper class to quickly write a loop over all standard input files.
Typical use is:
import fileinput
for line in fileinput.input():
process(line)
This iterates over the lines of all files listed in sys.argv[1:],
defaulting to sys.stdin if the list is empty. If a filename is '-' it
is also replaced by sys.stdin. To specify an alternative list of
filenames, pass it as the argument to input(). A single file name is
also allowed.
Functions filename(), lineno() return the filename and cumulative line
number of the line that has just been read; filelineno() returns its
line number in the current file; isfirstline() returns true iff the
line just read is the first line of its file; isstdin() returns true
iff the line was read from sys.stdin. Function nextfile() closes the
current file so that the next iteration will read the first line from
the next file (if any); lines not read from the file will not count
towards the cumulative line count; the filename is not changed until
after the first line of the next file has been read. Function close()
closes the sequence.
Before any lines have been read, filename() returns None and both line
numbers are zero; nextfile() has no effect. After all lines have been
read, filename() and the line number functions return the values
pertaining to the last line read; nextfile() has no effect.
All files are opened in text mode by default, you can override this by
setting the mode parameter to input() or FileInput.__init__().
If an I/O error occurs during opening or reading a file, the IOError
exception is raised.
If sys.stdin is used more than once, the second and further use will
return no lines, except perhaps for interactive use, or if it has been
explicitly reset (e.g. using sys.stdin.seek(0)).
Empty files are opened and immediately closed; the only time their
presence in the list of filenames is noticeable at all is when the
last file opened is empty.
It is possible that the last line of a file doesn't end in a newline
character; otherwise lines are returned including the trailing
newline.
Class FileInput is the implementation; its methods filename(),
lineno(), fileline(), isfirstline(), isstdin(), nextfile() and close()
correspond to the functions in the module. In addition it has a
readline() method which returns the next input line, and a
__getitem__() method which implements the sequence behavior. The
sequence must be accessed in strictly sequential order; sequence
access and readline() cannot be mixed.
Optional in-place filtering: if the keyword argument inplace=1 is
passed to input() or to the FileInput constructor, the file is moved
to a backup file and standard output is directed to the input file.
This makes it possible to write a filter that rewrites its input file
in place. If the keyword argument backup=".<some extension>" is also
given, it specifies the extension for the backup file, and the backup
file remains around; by default, the extension is ".bak" and it is
deleted when the output file is closed. In-place filtering is
disabled when standard input is read. XXX The current implementation
does not work for MS-DOS 8+3 filesystems.
Performance: this module is unfortunately one of the slower ways of
processing large numbers of input lines. Nevertheless, a significant
speed-up has been obtained by using readlines(bufsize) instead of
readline(). A new keyword argument, bufsize=N, is present on the
input() function and the FileInput() class to override the default
buffer size.
XXX Possible additions:
- optional getopt argument processing
- isatty()
- read(), read(size), even readlines()
"""
import sys, os
2001-01-20 19:34:12 -04:00
__all__ = ["input","close","nextfile","filename","lineno","filelineno",
"isfirstline","isstdin","FileInput"]
_state = None
DEFAULT_BUFSIZE = 8*1024
2006-02-19 17:26:07 -04:00
def input(files=None, inplace=0, backup="", bufsize=0,
mode="r", openhook=None):
"""input([files[, inplace[, backup[, mode[, openhook]]]]])
Create an instance of the FileInput class. The instance will be used
as global state for the functions of this module, and is also returned
to use during iteration. The parameters to this function will be passed
2002-05-23 12:15:30 -03:00
along to the constructor of the FileInput class.
"""
global _state
if _state and _state._file:
raise RuntimeError, "input() already active"
_state = FileInput(files, inplace, backup, bufsize, mode, openhook)
return _state
def close():
"""Close the sequence."""
global _state
state = _state
_state = None
if state:
state.close()
def nextfile():
"""
Close the current file so that the next iteration will read the first
line from the next file (if any); lines not read from the file will
not count towards the cumulative line count. The filename is not
changed until after the first line of the next file has been read.
Before the first line has been read, this function has no effect;
it cannot be used to skip the first file. After the last line of the
2002-05-23 12:15:30 -03:00
last file has been read, this function has no effect.
"""
if not _state:
raise RuntimeError, "no active input()"
return _state.nextfile()
def filename():
"""
Return the name of the file currently being read.
2002-05-23 12:15:30 -03:00
Before the first line has been read, returns None.
"""
if not _state:
raise RuntimeError, "no active input()"
return _state.filename()
def lineno():
"""
Return the cumulative line number of the line that has just been read.
Before the first line has been read, returns 0. After the last line
2002-05-23 12:15:30 -03:00
of the last file has been read, returns the line number of that line.
"""
if not _state:
raise RuntimeError, "no active input()"
return _state.lineno()
def filelineno():
"""
Return the line number in the current file. Before the first line
has been read, returns 0. After the last line of the last file has
2002-05-23 12:15:30 -03:00
been read, returns the line number of that line within the file.
"""
if not _state:
raise RuntimeError, "no active input()"
return _state.filelineno()
def fileno():
"""
Return the file number of the current file. When no file is currently
opened, returns -1.
"""
if not _state:
raise RuntimeError, "no active input()"
return _state.fileno()
def isfirstline():
"""
Returns true the line just read is the first line of its file,
2002-05-23 12:15:30 -03:00
otherwise returns false.
"""
if not _state:
raise RuntimeError, "no active input()"
return _state.isfirstline()
def isstdin():
"""
Returns true if the last line was read from sys.stdin,
2002-05-23 12:15:30 -03:00
otherwise returns false.
"""
if not _state:
raise RuntimeError, "no active input()"
return _state.isstdin()
class FileInput:
"""class FileInput([files[, inplace[, backup[, mode[, openhook]]]]])
2002-05-23 12:15:30 -03:00
Class FileInput is the implementation of the module; its methods
filename(), lineno(), fileline(), isfirstline(), isstdin(), fileno(),
nextfile() and close() correspond to the functions of the same name
in the module.
In addition it has a readline() method which returns the next
input line, and a __getitem__() method which implements the
sequence behavior. The sequence must be accessed in strictly
2002-05-23 12:15:30 -03:00
sequential order; random access and readline() cannot be mixed.
"""
2006-02-19 17:26:07 -04:00
def __init__(self, files=None, inplace=0, backup="", bufsize=0,
mode="r", openhook=None):
if isinstance(files, basestring):
files = (files,)
else:
if files is None:
files = sys.argv[1:]
if not files:
files = ('-',)
else:
files = tuple(files)
self._files = files
self._inplace = inplace
self._backup = backup
self._bufsize = bufsize or DEFAULT_BUFSIZE
self._savestdout = None
self._output = None
self._filename = None
self._lineno = 0
self._filelineno = 0
self._file = None
self._isstdin = False
self._backupfilename = None
self._buffer = []
self._bufindex = 0
# restrict mode argument to reading modes
if mode not in ('r', 'rU', 'U', 'rb'):
raise ValueError("FileInput opening mode must be one of "
"'r', 'rU', 'U' and 'rb'")
self._mode = mode
if inplace and openhook:
raise ValueError("FileInput cannot use an opening hook in inplace mode")
Merged revisions 55407-55513 via svnmerge from svn+ssh://pythondev@svn.python.org/python/branches/p3yk ................ r55413 | fred.drake | 2007-05-17 12:30:10 -0700 (Thu, 17 May 2007) | 1 line fix argument name in documentation; match the implementation ................ r55430 | jack.diederich | 2007-05-18 06:39:59 -0700 (Fri, 18 May 2007) | 1 line Implements class decorators, PEP 3129. ................ r55432 | guido.van.rossum | 2007-05-18 08:09:41 -0700 (Fri, 18 May 2007) | 2 lines obsubmit. ................ r55434 | guido.van.rossum | 2007-05-18 09:39:10 -0700 (Fri, 18 May 2007) | 3 lines Fix bug in test_inspect. (I presume this is how it should be fixed; Jack Diedrich, please verify.) ................ r55460 | brett.cannon | 2007-05-20 00:31:57 -0700 (Sun, 20 May 2007) | 4 lines Remove the imageop module. With imgfile already removed in Python 3.0 and rgbimg gone in Python 2.6 the unit tests themselves were made worthless. Plus third-party libraries perform the same function much better. ................ r55469 | neal.norwitz | 2007-05-20 11:28:20 -0700 (Sun, 20 May 2007) | 118 lines Merged revisions 55324-55467 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r55348 | georg.brandl | 2007-05-15 13:19:34 -0700 (Tue, 15 May 2007) | 4 lines HTML-escape the plain traceback in cgitb's HTML output, to prevent the traceback inadvertently or maliciously closing the comment and injecting HTML into the error page. ........ r55372 | neal.norwitz | 2007-05-15 21:33:50 -0700 (Tue, 15 May 2007) | 6 lines Port rev 55353 from Guido: Add what looks like a necessary call to PyErr_NoMemory() when PyMem_MALLOC() fails. Will backport. ........ r55377 | neal.norwitz | 2007-05-15 22:06:33 -0700 (Tue, 15 May 2007) | 1 line Mention removal of some directories for obsolete platforms ........ r55380 | brett.cannon | 2007-05-15 22:50:03 -0700 (Tue, 15 May 2007) | 2 lines Change the maintainer of the BeOS port. ........ r55383 | georg.brandl | 2007-05-16 06:44:18 -0700 (Wed, 16 May 2007) | 2 lines Bug #1719995: don't use deprecated method in sets example. ........ r55386 | neal.norwitz | 2007-05-16 13:05:11 -0700 (Wed, 16 May 2007) | 5 lines Fix bug in marshal where bad data would cause a segfault due to lack of an infinite recursion check. Contributed by Damien Miller at Google. ........ r55389 | brett.cannon | 2007-05-16 15:42:29 -0700 (Wed, 16 May 2007) | 6 lines Remove the gopherlib module. It has been raising a DeprecationWarning since Python 2.5. Also remove gopher support from urllib/urllib2. As both imported gopherlib the usage of the support would have raised a DeprecationWarning. ........ r55394 | raymond.hettinger | 2007-05-16 18:08:04 -0700 (Wed, 16 May 2007) | 1 line calendar.py gets no benefit from xrange() instead of range() ........ r55395 | brett.cannon | 2007-05-16 19:02:56 -0700 (Wed, 16 May 2007) | 3 lines Complete deprecation of BaseException.message. Some subclasses were directly accessing the message attribute instead of using the descriptor. ........ r55396 | neal.norwitz | 2007-05-16 23:11:36 -0700 (Wed, 16 May 2007) | 4 lines Reduce the max stack depth to see if this fixes the segfaults on Windows and some other boxes. If this is successful, this rev should be backported. I'm not sure how close to the limit we should push this. ........ r55397 | neal.norwitz | 2007-05-16 23:23:50 -0700 (Wed, 16 May 2007) | 4 lines Set the depth to something very small to try to determine if the crashes on Windows are really due to the stack size or possibly some other problem. ........ r55398 | neal.norwitz | 2007-05-17 00:04:46 -0700 (Thu, 17 May 2007) | 4 lines Last try for tweaking the max stack depth. 5000 was the original value, 4000 didn't work either. 1000 does work on Windows. If 2000 works, that will hopefully be a reasonable balance. ........ r55412 | fred.drake | 2007-05-17 12:29:58 -0700 (Thu, 17 May 2007) | 1 line fix argument name in documentation; match the implementation ........ r55427 | neal.norwitz | 2007-05-17 22:47:16 -0700 (Thu, 17 May 2007) | 1 line Verify neither dumps or loads overflow the stack and segfault. ........ r55446 | collin.winter | 2007-05-18 16:11:24 -0700 (Fri, 18 May 2007) | 1 line Backport PEP 3110's new 'except' syntax to 2.6. ........ r55448 | raymond.hettinger | 2007-05-18 18:11:16 -0700 (Fri, 18 May 2007) | 1 line Improvements to NamedTuple's implementation, tests, and documentation ........ r55449 | raymond.hettinger | 2007-05-18 18:50:11 -0700 (Fri, 18 May 2007) | 1 line Fix beginner mistake -- don't mix spaces and tabs. ........ r55450 | neal.norwitz | 2007-05-18 20:48:47 -0700 (Fri, 18 May 2007) | 1 line Clear data so random memory does not get freed. Will backport. ........ r55452 | neal.norwitz | 2007-05-18 21:34:55 -0700 (Fri, 18 May 2007) | 3 lines Whoops, need to pay attention to those test failures. Move the clear to *before* the first use, not after. ........ r55453 | neal.norwitz | 2007-05-18 21:35:52 -0700 (Fri, 18 May 2007) | 1 line Give some clue as to what happened if the test fails. ........ r55455 | georg.brandl | 2007-05-19 11:09:26 -0700 (Sat, 19 May 2007) | 2 lines Fix docstring for add_package in site.py. ........ r55458 | brett.cannon | 2007-05-20 00:09:50 -0700 (Sun, 20 May 2007) | 2 lines Remove the rgbimg module. It has been deprecated since Python 2.5. ........ r55465 | nick.coghlan | 2007-05-20 04:12:49 -0700 (Sun, 20 May 2007) | 1 line Fix typo in example (should be backported, but my maintenance branch is woefully out of date) ........ ................ r55472 | brett.cannon | 2007-05-20 12:06:18 -0700 (Sun, 20 May 2007) | 2 lines Remove imageop from the Windows build process. ................ r55486 | neal.norwitz | 2007-05-20 23:59:52 -0700 (Sun, 20 May 2007) | 1 line Remove callable() builtin ................ r55506 | neal.norwitz | 2007-05-22 00:43:29 -0700 (Tue, 22 May 2007) | 78 lines Merged revisions 55468-55505 via svnmerge from svn+ssh://pythondev@svn.python.org/python/trunk ........ r55468 | neal.norwitz | 2007-05-20 11:06:27 -0700 (Sun, 20 May 2007) | 1 line rotor is long gone. ........ r55470 | neal.norwitz | 2007-05-20 11:43:00 -0700 (Sun, 20 May 2007) | 1 line Update directories/files at the top-level. ........ r55471 | brett.cannon | 2007-05-20 12:05:06 -0700 (Sun, 20 May 2007) | 2 lines Try to remove rgbimg from Windows builds. ........ r55474 | brett.cannon | 2007-05-20 16:17:38 -0700 (Sun, 20 May 2007) | 4 lines Remove the macfs module. This led to the deprecation of macostools.touched(); it completely relied on macfs and is a no-op on OS X according to code comments. ........ r55476 | brett.cannon | 2007-05-20 16:56:18 -0700 (Sun, 20 May 2007) | 3 lines Move imgfile import to the global namespace to trigger an import error ASAP to prevent creation of a test file. ........ r55477 | brett.cannon | 2007-05-20 16:57:38 -0700 (Sun, 20 May 2007) | 3 lines Cause posixfile to raise a DeprecationWarning. Documented as deprecated since Ptyhon 1.5. ........ r55479 | andrew.kuchling | 2007-05-20 17:03:15 -0700 (Sun, 20 May 2007) | 1 line Note removed modules ........ r55481 | martin.v.loewis | 2007-05-20 21:35:47 -0700 (Sun, 20 May 2007) | 2 lines Add Alexandre Vassalotti. ........ r55482 | george.yoshida | 2007-05-20 21:41:21 -0700 (Sun, 20 May 2007) | 4 lines fix against r55474 [Remove the macfs module] Remove "libmacfs.tex" from Makefile.deps and mac/mac.tex. ........ r55487 | raymond.hettinger | 2007-05-21 01:13:35 -0700 (Mon, 21 May 2007) | 1 line Replace assertion with straight error-checking. ........ r55489 | raymond.hettinger | 2007-05-21 09:40:10 -0700 (Mon, 21 May 2007) | 1 line Allow all alphanumeric and underscores in type and field names. ........ r55490 | facundo.batista | 2007-05-21 10:32:32 -0700 (Mon, 21 May 2007) | 5 lines Added timeout support to HTTPSConnection, through the socket.create_connection function. Also added a small test for this, and updated NEWS file. ........ r55495 | georg.brandl | 2007-05-21 13:34:16 -0700 (Mon, 21 May 2007) | 2 lines Patch #1686487: you can now pass any mapping after '**' in function calls. ........ r55502 | neal.norwitz | 2007-05-21 23:03:36 -0700 (Mon, 21 May 2007) | 1 line Document new params to HTTPSConnection ........ r55504 | neal.norwitz | 2007-05-22 00:16:10 -0700 (Tue, 22 May 2007) | 1 line Stop using METH_OLDARGS ........ r55505 | neal.norwitz | 2007-05-22 00:16:44 -0700 (Tue, 22 May 2007) | 1 line Stop using METH_OLDARGS implicitly ........ ................
2007-05-22 15:11:13 -03:00
elif openhook and not hasattr(openhook, '__call__'):
raise ValueError("FileInput openhook must be callable")
self._openhook = openhook
def __del__(self):
self.close()
def close(self):
self.nextfile()
self._files = ()
2002-03-26 16:28:40 -04:00
def __iter__(self):
return self
def __next__(self):
try:
line = self._buffer[self._bufindex]
except IndexError:
pass
else:
self._bufindex += 1
self._lineno += 1
self._filelineno += 1
return line
line = self.readline()
if not line:
2002-03-26 16:28:40 -04:00
raise StopIteration
return line
2002-04-15 22:38:40 -03:00
2002-03-26 16:28:40 -04:00
def __getitem__(self, i):
if i != self._lineno:
raise RuntimeError, "accessing lines out of order"
try:
return self.__next__()
2002-03-26 16:28:40 -04:00
except StopIteration:
raise IndexError, "end of input reached"
def nextfile(self):
savestdout = self._savestdout
self._savestdout = 0
if savestdout:
sys.stdout = savestdout
output = self._output
self._output = 0
if output:
output.close()
file = self._file
self._file = 0
if file and not self._isstdin:
file.close()
backupfilename = self._backupfilename
self._backupfilename = 0
if backupfilename and not self._backup:
try: os.unlink(backupfilename)
except OSError: pass
self._isstdin = False
self._buffer = []
self._bufindex = 0
def readline(self):
try:
line = self._buffer[self._bufindex]
except IndexError:
pass
else:
self._bufindex += 1
self._lineno += 1
self._filelineno += 1
return line
if not self._file:
if not self._files:
return ""
self._filename = self._files[0]
self._files = self._files[1:]
self._filelineno = 0
self._file = None
self._isstdin = False
self._backupfilename = 0
if self._filename == '-':
self._filename = '<stdin>'
self._file = sys.stdin
self._isstdin = True
else:
if self._inplace:
self._backupfilename = (
2007-08-16 11:35:24 -03:00
self._filename + (self._backup or ".bak"))
try: os.unlink(self._backupfilename)
except os.error: pass
# The next few lines may raise IOError
os.rename(self._filename, self._backupfilename)
self._file = open(self._backupfilename, self._mode)
try:
perm = os.fstat(self._file.fileno()).st_mode
except OSError:
self._output = open(self._filename, "w")
else:
fd = os.open(self._filename,
os.O_CREAT | os.O_WRONLY | os.O_TRUNC,
perm)
self._output = os.fdopen(fd, "w")
try:
if hasattr(os, 'chmod'):
os.chmod(self._filename, perm)
except OSError:
pass
self._savestdout = sys.stdout
sys.stdout = self._output
else:
# This may raise IOError
if self._openhook:
self._file = self._openhook(self._filename, self._mode)
else:
self._file = open(self._filename, self._mode)
self._buffer = self._file.readlines(self._bufsize)
self._bufindex = 0
if not self._buffer:
self.nextfile()
# Recursive call
return self.readline()
def filename(self):
return self._filename
def lineno(self):
return self._lineno
def filelineno(self):
return self._filelineno
def fileno(self):
if self._file:
try:
return self._file.fileno()
except ValueError:
return -1
else:
return -1
def isfirstline(self):
return self._filelineno == 1
def isstdin(self):
return self._isstdin
def hook_compressed(filename, mode):
ext = os.path.splitext(filename)[1]
if ext == '.gz':
import gzip
return gzip.open(filename, mode)
elif ext == '.bz2':
import bz2
return bz2.BZ2File(filename, mode)
else:
return open(filename, mode)
def hook_encoded(encoding):
import codecs
def openhook(filename, mode):
return codecs.open(filename, mode, encoding)
return openhook
def _test():
import getopt
inplace = 0
backup = 0
opts, args = getopt.getopt(sys.argv[1:], "ib:")
for o, a in opts:
if o == '-i': inplace = 1
if o == '-b': backup = a
for line in input(args, inplace=inplace, backup=backup):
if line[-1:] == '\n': line = line[:-1]
if line[-1:] == '\r': line = line[:-1]
print("%d: %s[%d]%s %s" % (lineno(), filename(), filelineno(),
isfirstline() and "*" or "", line))
print("%d: %s[%d]" % (lineno(), filename(), filelineno()))
if __name__ == '__main__':
_test()