#!/usr/bin/env python
# tren.py
# Copyright (c) 2010 TundraWare Inc.
# For Updates See: http://www.tundraware.com/Software/tren
# Program Information
PROGNAME = "tren.py"
BASENAME = PROGNAME.split(".py")[0]
PROGENV = BASENAME.upper()
RCSID = "$Id: tren.py,v 1.135 2010/02/02 17:53:55 tundra Exp $"
VERSION = RCSID.split()[2]
# Copyright Information
CPRT = "(c)"
DATE = "2010"
OWNER = "TundraWare Inc."
RIGHTS = "All Rights Reserved."
COPYRIGHT = "Copyright %s %s, %s %s" % (CPRT, DATE, OWNER, RIGHTS)
PROGVER = PROGNAME + " " + VERSION + (" - %s" % COPYRIGHT)
HOMEPAGE = "http://www.tundraware.com/Software/%s\n" % BASENAME
#----------------------------------------------------------#
# Variables User Might Change #
#----------------------------------------------------------#
#------------------- Nothing Below Here Should Need Changing ------------------#
#----------------------------------------------------------#
# Imports #
#----------------------------------------------------------#
import copy
import getopt
import os
from stat import *
import sys
#----------------------------------------------------------#
# Aliases & Redefinitions #
#----------------------------------------------------------#
#----------------------------------------------------------#
# Constants & Literals #
#----------------------------------------------------------#
#####
# General Program Constants
#####
MAXINCLUDES = 50 # Maximum number of includes allowed
TGTSEQFLG = "" # Indicates non-stat() in SeqTypes table
#####
# Message Formatting Constants
#####
# Make sure these make sense: MAXLINELEN > PADWIDTH + WRAPINDENT
# because of the way line conditioning/wrap works.
MAXLINELEN = 75 # Default max printed line length
PADCHAR = " " # Padding character
PADWIDTH = 30 # Column width
LSTPAD = 13 # Padding to use when dumping lists
WRAPINDENT = 8 # Extra indent on wrapped lines
MINLEN = PADWIDTH + WRAPINDENT + 1 # Minimum line length
#####
# Literals
#####
ALL = "All" # Rename target is whole filename
COMMENT = "#" # Comment character in include files
ESC = "\\" # Escape character
EXT = "Ext" # Rename target is extension
EXTDELIM = "." # Extension delimeter
INCL = "-I" # Include file command line option
NAM = "Nam" # Rename target is name
RENSEP = "=" # Rename command separator: oldRENSEPnew
#----------------------------------------------------------#
# Prompts, & Application Strings #
#----------------------------------------------------------#
#####
# Debug Messages
#####
DEBUGFLAG = "-d"
dCMDLINE = "Command Line"
dCURSTATE = "Current State Of Program Options"
dDEBUG = "DEBUG"
dDUMPOBJ = "Dumping Object %s"
dINCLUDING = "Including file '%s'"
dPROGENV = "$" + PROGENV
dRENREQ = "Renaming Request:"
dRESOLVEDOPTS = "Resolved Command Line"
dSEPCHAR = "-" # Used for debug separator lines
dSEQATIME = "Access Time Sequence:"
dSEQCMD = "Command Line Sequence:"
dSEQCTIME = "Creation Time Sequence:"
dSEQDEV = "Device Sequence:"
dSEQGID = "GID Sequence:"
dSEQINO = "Inode Sequence:"
dSEQMODE = "Mode Sequence:"
dSEQMTIME = "Modification Time Sequence:"
dSEQNLINK = "Nlinks Sequence"
dSEQSIZE = "Size Sequence:"
dSEQTARGS = "Rename Targets:"
dSEQUID = "UID Sequence:"
#####
# Error Messages
#####
eBADARG = "Invalid command line: %s!"
eBADINCL = "option %s requires argument" % INCL
eBADNEWOLD = "Bad -r argument '%s'! Requires exactly one new, old string separator (Default: " + RENSEP + ")"
eBADLEN = "Bad line length '%s'!"
eERROR = "ERROR"
eFILEOPEN = "Cannot open file '%s': %s!"
eLINELEN = "Specified line length too short! Must be at least %s" % MINLEN
eTOOMANYINC = "Too many includes! (Max is %d) Possible circular reference?" % MAXINCLUDES
#####
# Warning Messages
#####
#####
# Usage Prompts
#####
uTable = [PROGVER,
HOMEPAGE,
"usage: " + PROGNAME + " [-1abCcdEefghqtvwXx] [-I file] [-l string] [-r old=new]... file|dir file|dir ...",
" where,",
" -1 Rename only the first instance of the specified string (Default)",
" -a Rename within the entire file or directory name (Default)",
" -C Do case-sensitive renaming (Default)",
" -c Collapse case when doing string substitution.",
" -d Dump debugging information",
" -e Only perform renaming within extension portion of or directory name.",
" -E Continue renaming even after an error is encountered",
" -f Force renaming even if target file or directory name already exists.",
" -g Replace all instances (global rename) of the old string with the new.",
" -h Print help information.",
" -I file Include command line arguments from file",
" -l string File extension delimiter string. (Default: .)",
" -q Quiet mode, do not show progress.",
" -R rensep Separator string for -r rename arguments. (Default: =)",
" -r old=new Replace old with new in file or directory names.",
" -t Test mode, don't rename, just show what the program *would* do",
" -v Print detailed program version information and exit.",
" -w Line length of diagnostic and error output (Default: 75)",
" -X Treat the renaming strings literally (Default)",
" -x Treat the old replacement string as a Python regular expression",
]
#----------------------------------------------------------#
# Global Variables & Data Structures #
#----------------------------------------------------------#
# Program toggle and option defaults
ProgramOptions = {
"DEBUG" : False, # Debugging off
"CASESENSITIVE" : True, # Search is case-sensitive
"ERRORCONTINUE" : False, # Do not continue after error
"FORCERENAM" : False, # Do not rename if target already exists
"GLOBAL" : False, # Only rename first instance of old string
"QUIET" : False, # Display progress
"REGEX" : False, # Do not treat old string as a regex
"TARGET" : ALL, # Can be "All", "Name", or "Ext"
"TESTMODE" : False # Global data structures
}
#--------------------------- Code Begins Here ---------------------------------#
#----------------------------------------------------------#
# Object Base Class Definitions #
#----------------------------------------------------------#
#####
# Container For Holding Rename Targets
#####
class RenameTargets:
"""
This class is used to keep track of all the files and/or
directories we're renaming. When __init__ finishes,
RenNames dictionary will be populated as follows:
fully-qualified name : [ basename,
stat information for the entry,
position in command line args list (0-based)
ascending alpha order of rename targets (O-based)
descending alpha order of rename targets (0-based)
ascending order of appearance by-mode (O-based)
descending order of appearance by-mode (0-based)
ascending order of appearance by-inode (O-based)
descending order of appearance by-inode (0-based)
ascending order of appearance by-devno (O-based)
descending order of appearance by-devno (0-based)
ascending order of appearance by-nlinks (O-based)
descending order of appearance by-nlinks (0-based)
ascending order of appearance by-uid (O-based)
descending order of appearance by-uid (0-based)
ascending order of appearance by-gid (O-based)
descending order of appearance by-gid (0-based)
ascending order of appearance by-atime (0-based)
descending order of appearance by-atime (0-based)
ascending order of appearance by-ctime (0-based)
descending order of appearance by-ctime (0-based)
ascending order of appearance by-mtime (0-based)
descending order of appearance by-mtime (0-based)
ascending order of appearance by-size (0-based)
descending order of appearance by-size (0-based)
]
"""
def __init__(self, targs):
# Dictionary of all rename targets and their stat info
self.RenNames = {}
# Ordered lists used by sequence renaming tokens
args = {} # Keys = 0, Values = Rename targets from command line
modes = {} # Keys = modes, Values = List of corresponding files
inodes = {} # Keys = inodes, Values = List of corresponding files
devs = {} # Keys = devs, Values = List of corresponding files
nlinks = {} # Keys = nlinks, Values = List of corresponding files
uids = {} # Keys = uids, Values = List of corresponding files
gids = {} # Keys = gids, Values = List of corresponding files
atimes = {} # Keys = atimes, Values = List of corresponding files
ctimes = {} # Keys = ctimes, Values = List of corresponding files
mtimes = {} # Keys = mtimes, Values = List of corresponding files
sizes = {} # Keys = sizes, Values = List of corresponding files
# Populate the data structures
cmdorder = 0
for t in targs:
try:
fullname = os.path.abspath(t)
basename = os.path.basename(t)
stats = os.stat(fullname)
except (IOError, OSError) as e:
ErrorMsg(eFILEOPEN % (t, e.args[1]))
sys.exit(1)
# This data structure is used to keep track of everything
# we need to build the sequence renaming token support.
# This makes it easy to add more types later on.
SeqTypes = [ [TGTSEQFLG, args, dSEQTARGS],
[ST_MODE, modes, dSEQMODE],
[ST_INO, inodes, dSEQINO],
[ST_DEV, devs, dSEQDEV],
[ST_NLINK, nlinks, dSEQNLINK],
[ST_UID, uids, dSEQUID],
[ST_GID, gids, dSEQGID],
[ST_ATIME, atimes, dSEQATIME],
[ST_CTIME, ctimes, dSEQCTIME],
[ST_CTIME, mtimes, dSEQMTIME],
[ST_SIZE, sizes, dSEQSIZE],
]
# Incrementally build lists of keys that will later be
# used to create sequence renaming tokens
for seqtype in SeqTypes:
# Handle os.stat() values
if seqtype[0] != TGTSEQFLG:
statval = stats[seqtype[0]]
# Handle non os.stat() stuff
else:
statval = TGTSEQFLG
# Where to put the results
vals = seqtype[1]
if statval in vals:
vals[statval].append(fullname)
else:
vals[statval] = [fullname]
self.RenNames[fullname] = [basename, stats, cmdorder]
cmdorder += 1
# Create the various sorted views we may need
# for sequence renaming tokens
for seqtype in SeqTypes:
view = seqtype[1]
debugmsg = seqtype[2]
vieworder = view.keys()
vieworder.sort()
# Sort alphabetically when multiple filenames
# map to the same key, creating overall
# ordering as we go.
t = []
for i in vieworder:
view[i].sort()
for j in view[i]:
t.append(j)
# Now store the ascending- and descending order it
# the master dictionary
tblz = len(t)
for name in t:
self.RenNames[name].append(t.index(name)) # Ascending index
self.RenNames[name].append(tblz - t.index(name) - 1) # Descending Index
if ProgramOptions["DEBUG"]:
l=[]
for item in vieworder:
DumpList(DebugMsg, debugmsg, item, view[item])
if ProgramOptions["DEBUG"]:
SEPARATOR = dSEPCHAR * MAXLINELEN
DebugMsg(SEPARATOR)
DebugMsg(dDUMPOBJ % str(self))
DebugMsg(SEPARATOR)
# Dump abspath, basename, & stat information
for name in self.RenNames:
DumpList(DebugMsg, name, "", self.RenNames[name])
DebugMsg(SEPARATOR)
# Now get rid of the working dictionaries to free up their memory
del args, modes, inodes, devs, nlinks, uids, gids, atimes, ctimes, mtimes, sizes
# End of class 'RenameTargets'
#----------------------------------------------------------#
# Supporting Function Definitions #
#----------------------------------------------------------#
#####
# Turn A List Into Columns With Space Padding
#####
def ColumnPad(list, padchar=PADCHAR, padwidth=PADWIDTH):
retval = ""
for l in list:
l = str(l)
retval += l + ((padwidth - len(l)) * padchar)
return retval
# End of 'ColumnPad()'
#####
# Condition Line Length With Fancy Wrap And Formatting
#####
def ConditionLine(msg,
padchar=PADCHAR, \
padwidth=PADWIDTH, \
wrapindent=WRAPINDENT ):
retval = []
retval.append(msg[:MAXLINELEN])
msg = msg[MAXLINELEN:]
while msg:
msg = padchar * (padwidth + wrapindent) + msg
retval.append(msg[:MAXLINELEN])
msg = msg[MAXLINELEN:]
return retval
# End of 'ConditionLine()'
#####
# Print A Debug Message
#####
def DebugMsg(msg):
l = ConditionLine(msg)
for msg in l:
PrintStderr(PROGNAME + " " + dDEBUG + ": " + msg)
# End of 'DebugMsg()'
#####
# Debug Dump Of A List
#####
def DumpList(handler, msg, listname, content):
handler(msg)
itemarrow = ColumnPad([listname, " "], padwidth=LSTPAD)
handler(ColumnPad([" ", " %s %s" % (itemarrow, content)]))
# End of 'DumpList()'
#####
# Dump The State Of The Program
#####
def DumpState():
SEPARATOR = dSEPCHAR * MAXLINELEN
DebugMsg(SEPARATOR)
DebugMsg(dCURSTATE)
DebugMsg(SEPARATOR)
opts = ProgramOptions.keys()
opts.sort()
for o in opts:
DebugMsg(ColumnPad([o, ProgramOptions[o]]))
DebugMsg(SEPARATOR)
# End of 'DumpState()'
#####
# Print An Error Message
#####
def ErrorMsg(emsg):
l = ConditionLine(emsg)
for emsg in l:
PrintStderr(PROGNAME + " " + eERROR + ": " + emsg)
# End of 'ErrorMsg()'
#####
# Split -r Argument Into Separate Old And New Strings
#####
def GetOldNew(arg):
escaping = False
numseps = 0
sepindex = 0
i = 0
while i < len(arg):
# Scan string ignoring escaped separators
if arg[i:].startswith(RENSEP):
if (i > 0 and (arg[i-1] != ESC)) or i == 0:
sepindex = i
numseps += 1
i += len(RENSEP)
else:
i += 1
if numseps != 1:
ErrorMsg(eBADNEWOLD % arg)
sys.exit(1)
else:
old, new = arg[:sepindex], arg[sepindex + len(RENSEP):]
old = old.replace(ESC + RENSEP, RENSEP)
new = new.replace(ESC + RENSEP, RENSEP)
return [old, new]
# End of 'GetOldNew()'
#####
# Print To stderr
#####
def PrintStderr(msg, trailing="\n"):
sys.stderr.write(msg + trailing)
# End of 'PrintStderr()'
#####
# Print To stdout
#####
def PrintStdout(msg, trailing="\n"):
sys.stdout.write(msg + trailing)
# End of 'PrintStdout'
#####
# Process Include Files On The Command Line
#####
def ProcessIncludes(OPTIONS):
# Make sure the include argument has trailing space or the
# code below will break.
OPTIONS = " ".join(OPTIONS).replace(INCL, INCL+" ").split()
# Resolve include file references allowing for nested includes.
# This has to be done here separate from the command line options so
# that getopt() processing below will "see" the included statements.
NUMINCLUDES = 0
while " ". join(OPTIONS).find(INCL) > -1:
# Get the index of the next include to process.
# It cannot be the last item because this means the filename
# to include is missing.
i = OPTIONS.index(INCL)
if i == len(OPTIONS)-1:
ErrorMsg(eBADARG % eBADINCL)
sys.exit(1)
file = OPTIONS[i+1] ; lhs = OPTIONS[:i] ; rhs = OPTIONS[i+2:]
# Keep track of- and limit the number of includes allowed
# This is an easy way to stop circular (infinite) includes.
NUMINCLUDES += 1
if NUMINCLUDES >= MAXINCLUDES:
ErrorMsg(eTOOMANYINC)
sys.exit(1)
# Replace insert option on the command line with that file's contents.
# Handle comments within lines.
try:
n = []
f = open(file)
for l in f.readlines():
l = l.split(COMMENT)[0]
n += l.split()
f.close()
if ProgramOptions["DEBUG"]:
DebugMsg(dINCLUDING % file)
OPTIONS = lhs + n + rhs
except IOError as e:
ErrorMsg(eFILEOPEN % (file, e.args[1]))
sys.exit(1)
return OPTIONS
# End of 'ProcessIncludes()'
#####
# Print Usage Information
#####
def Usage():
for line in uTable:
PrintStdout(line)
# End of 'Usage()'
#----------------------------------------------------------#
# Program Entry Point #
#----------------------------------------------------------#
#####
# Command Line Preprocessing
#
# Some things have to be done *before* the command line
# options can actually be processed. This includes:
#
# 1) Prepending any options specified in the environment variable.
#
# 2) Resolving any include file references
#
# 3) Separating the command line into [options ... filenames ..]
# groupings so that user can interweave multiple options
# and names on the command line.
#
# 4) Building the data structures that depend on the file/dir names
# specified for renaming. We have to do this first, because
# -r renaming operations specified on the command line will
# need this information if they make use of renaming tokens.
#
#####
# Process any options set in the environment first, and then those
# given on the command line
OPTIONS = sys.argv[1:]
envopt = os.getenv(PROGENV)
if envopt:
OPTIONS = envopt.split() + OPTIONS
# Check for debug manually to see if we want
# debug info about includes
if DEBUGFLAG in OPTIONS:
ProgramOptions["DEBUG"] = True
# Deal with include files
OPTIONS = ProcessIncludes(OPTIONS)
# Check for debug manually again before we process the options
# to get debug info on command line expansion
if DEBUGFLAG in OPTIONS:
ProgramOptions["DEBUG"] = True
RenRequests = []
try:
opts, args = getopt.getopt(OPTIONS, '1abbCcdEefghl:qR:r:tvw:Xx]')
except getopt.GetoptError as e:
ErrorMsg(eBADARG % e.args[0])
sys.exit(1)
# Now process the options
for opt, val in opts:
if opt == "-1":
ProgramOptions["GLOBAL"] = False
if opt == "-a":
ProgramOptions["TARGET"] = ALL
if opt == "-b":
ProgramOptions["TARGET"] = NAM
if opt == "-C":
ProgramOptions["CASESENSITIVE"] = True
if opt == "-c":
ProgramOptions["CASESENSITIVE"] = False
if opt == "-d":
DumpState()
if opt == "-E":
ProgramOptions["ERRORCONTINUE"] = True
if opt == "-e":
ProgramOptions["TARGET"] = EXT
if opt == "-f":
ProgramOptions["FORCERENAM"] = True
if opt == "-g":
ProgramOptions["GLOBAL"] = True
if opt == "-h":
Usage()
sys.exit(0)
if opt == "-l":
EXTDELIM = val
if opt == "-q":
ProgramOptions["QUIET"] = True
if opt == '-R':
RENSEP = val
if opt == "-r":
old, new = GetOldNew(val)
RenRequests.append([old,new, copy.deepcopy(ProgramOptions)])
if opt == "-t":
ProgramOptions["TESTMODE"] = True
if opt == "-v":
PrintStdout(RCSID)
sys.exit(0)
if opt == "-w":
try:
l = int(val)
except:
ErrorMsg(eBADLEN % val)
sys.exit(1)
if l < MINLEN:
ErrorMsg(eLINELEN)
sys.exit(1)
MAXLINELEN = l
if opt == "-X":
ProgramOptions["REGEX"] = False
if opt == "-x":
ProgramOptions["REGEX"] = True
if ProgramOptions["DEBUG"]:
# Dump what we know about the command line
DumpList(DebugMsg, dCMDLINE, "", sys.argv)
DumpList(DebugMsg, dPROGENV, "", envopt)
DumpList(DebugMsg, dRESOLVEDOPTS, "", OPTIONS)
# Create and populate an object with rename targets. We have to
# do *before* we process any renaming requests because they may make
# reference to renaming tokens that only can be resolved with the
# contents of the 'targs' data structure.
targs = None
if args:
targs = RenameTargets(args)
# Display outstanding renaming requests if we're debugging
if ProgramOptions["DEBUG"]:
for i in RenRequests:
DumpList(DebugMsg, dRENREQ , "", i)
# Release the target container if we created one
if targs:
del targs
tundra