#!/usr/bin/python
"""
========================================================================================
headcode.py - July 2016, Python 3.X + 2.X
Synopsis: a simple HTML file code insertion/deletion script
Author: M. Lutz, learning-python.com
License: provided freely but with no warranties of any kind
UPDATE: for a more-recent tool which achieves some similar goals, see
also https://learning-python.com/thumbspage/build/insert-analytics.py
Adds (if no args) or removes (if any arg) analytics or other -section code
block *IN-PLACE* in all HTML files in a directory tree. Change your tree root path
(HTMLDIR) and insertion code (INSERT) below as desired, and run with 0 args to add
or 1 to remove. In the target use case, each run of this script saves editing 70
files manually - a prime example of the sort of tactical tasks Python excels at.
For an example of this script in action, see:
http://learning-python.com/books/headcode-LOG.txt
Assumes your HTML files have either a "" or "" to use as a replacement
target. Also assumes your bytes INSERT is encoded per a Unicode scheme that is
compatible with the content of all your HTML files -- it's ASCII by default if
you use a bytes literal and the os.linesep encoding here, and ASCII is a subset of
most encodings (e.g., latin1, utf8); but manually encode a str otherwise if needed.
The INSERT encoding constraint derives from this script's use of binary files and
bytes to sidestep Unicode encodings - it processes still-encoded text to avoid
having to know each file's (and hence the INSERT's) encoding. Alternatives:
1) Guess file encodings by trying to open and decode with each of a set of options
2) Look for an encoding type in a tag (if present) by parsing HTML
3) Attempt to guess each file's encoding by inspecting its contents' bytes
Each of these seems overkill for this simple utility (but see genhtml for #1).
========================================================================================
"""
from __future__ import print_function # 2.X compatibility
import sys, os, glob
HTMLDIR = r'C:\MY-STUFF\Training\cd\training-cd-2.7-3.5--oct015' # live
HTMLDIR = r'C:\Users\me\Desktop\training-cd-2.7-3.5--oct015' # test
mode = 'insert' if len(sys.argv) == 1 else 'restore'
print('MODE:', mode)
# assume ASCII compatible with content of files, else encode a str manually
INSERT = b"""
"""
INSERT = INSERT.replace(b'\n', os.linesep.encode('ascii')) # for binary-mode file
cnvcnt = 0
for (dirpath, dirshere, fileshere) in os.walk(HTMLDIR):
for filehere in fileshere:
if filehere.endswith(('.htm', '.html')):
htmlfile = os.path.join(dirpath, filehere)
print(htmlfile)
oldfile = open(htmlfile, 'rb') # ignore Unicode encoding of file content, use bytes
oldtext = oldfile.read() # but assume compatible with INSERT encoding (ASCII?)
oldfile.close()
newfile = open(htmlfile, 'wb')
if mode == 'insert':
newtext = oldtext.replace(b'', (INSERT + b''))
newtext = newtext.replace(b'', (INSERT + b''))
elif mode == 'restore':
newtext = oldtext.replace(INSERT, b'')
newfile.write(newtext)
newfile.close()
cnvcnt += 1
print('DONE: %d files converted.' % cnvcnt)