HTML Diff

X-Git-Url: http://git.droids-corp.org/?p=diff2html.git;a=blobdiff_plain;f=diff2html.py;h=71ac19a3343b450c0d07a1a2683531caa9167129;hp=6e2751e4d7a585f73de28abd6de0fc2ba2ef96d3;hb=659c1a60bdf4fc73098bc97b48db70b68d82ad23;hpb=ad3d66724c62f87c720185bad6f6be71627339de diff --git a/diff2html.py b/diff2html.py index 6e2751e..71ac19a 100644 --- a/diff2html.py +++ b/diff2html.py @@ -1,4 +1,5 @@ #! /usr/bin/python +# coding=utf-8 # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by @@ -18,52 +19,77 @@ # Transform a unified diff from stdin to a colored # side-by-side HTML page on stdout. # -# Authors: Olivier MATZ +# Authors: Olivier Matz # Alan De Smet +# Sergey Satskiy +# scito # # Inspired by diff2html.rb from Dave Burt # (mainly for html theme) # # TODO: # - The sane function currently mashes non-ASCII characters to "." -# Instead be clever and convert to something like "xF0" +# Instead be clever and convert to something like "xF0" # (the hex value), and mark with a . Even more clever: # Detect if the character is "printable" for whatever definition, # and display those directly. - -import sys, re, htmlentitydefs, getopt +import sys, re, htmlentitydefs, getopt, StringIO, codecs, datetime +try: + from simplediff import diff, string_diff +except ImportError: + sys.stderr.write("info: simplediff module not found, only linediff is available\n") + sys.stderr.write("info: it can be downloaded at https://github.com/paulgb/simplediff\n") # minimum line size, we add a zero-sized breakable space every # LINESIZE characters linesize = 20 tabsize = 8 show_CR = False - - -html_hdr = """ - - - HTML Diff - - - +encoding = "utf-8" +lang = "en" +algorithm = 0 + +desc = "File comparison" +dtnow = datetime.datetime.now() +modified_date = "%s+01:00"%dtnow.isoformat() + +html_hdr = """ + + + + + + HTML Diff{0} + + + + + + + + + """ html_footer = """ + """ @@ -85,27 +111,29 @@ hunk_off1, hunk_size1, hunk_off2, hunk_size2 = 0, 0, 0, 0 # Characters we're willing to word wrap on -WORDBREAK = " \t;.,/):" +WORDBREAK = " \t;.,/):-" def sane(x): r = "" for i in x: j = ord(i) - if i not in ['\t', '\n'] and ((j < 32) or (j >= 127)): + if i not in ['\t', '\n'] and (j < 32): r = r + "." else: r = r + i return r def linediff(s, t): + ''' + Original line diff algorithm of diff2html. It's character based. + ''' if len(s): - s = str(reduce(lambda x, y:x+y, [ sane(c) for c in s ])) + s = unicode(reduce(lambda x, y:x+y, [ sane(c) for c in s ])) if len(t): - t = str(reduce(lambda x, y:x+y, [ sane(c) for c in t ])) + t = unicode(reduce(lambda x, y:x+y, [ sane(c) for c in t ])) m, n = len(s), len(t) d = [[(0, 0) for i in range(n+1)] for i in range(m+1)] - x = [[(0, 0) for i in range(n+1)] for i in range(m+1)] d[0][0] = (0, (0, 0)) @@ -161,19 +189,75 @@ def linediff(s, t): return r1, r2 +def diff_changed(old, new): + ''' + Returns the differences basend on characters between two strings + wrapped with DIFFON and DIFFOFF using `diff`. + ''' + con = {'=': (lambda x: x), + '+': (lambda x: DIFFON + x + DIFFOFF), + '-': (lambda x: '')} + return "".join([(con[a])("".join(b)) for a, b in diff(old, new)]) + + +def diff_changed_ts(old, new): + ''' + Returns a tuple for a two sided comparison based on characters, see `diff_changed`. + ''' + return (diff_changed(new, old), diff_changed(old, new)) + + +def word_diff(old, new): + ''' + Returns the difference between the old and new strings based on words. Punctuation is not part of the word. + + Params: + old the old string + new the new string + + Returns: + the output of `diff` on the two strings after splitting them + on whitespace (a list of change instructions; see the docstring + of `diff`) + ''' + separator_pattern = '(\W+)'; + return diff(re.split(separator_pattern, old, flags=re.UNICODE), re.split(separator_pattern, new, flags=re.UNICODE)) + + +def diff_changed_words(old, new): + ''' + Returns the difference between two strings based on words (see `word_diff`) + wrapped with DIFFON and DIFFOFF. + + Returns: + the output of the diff expressed delimited with DIFFON and DIFFOFF. + ''' + con = {'=': (lambda x: x), + '+': (lambda x: DIFFON + x + DIFFOFF), + '-': (lambda x: '')} + return "".join([(con[a])("".join(b)) for a, b in word_diff(old, new)]) + + +def diff_changed_words_ts(old, new): + ''' + Returns a tuple for a two sided comparison based on words, see `diff_changed_words`. + ''' + return (diff_changed_words(new, old), diff_changed_words(old, new)) + + def convert(s, linesize=0, ponct=0): i = 0 - t = "" + t = u"" for c in s: # used by diffs if c == DIFFON: - t += '' + t += u'' elif c == DIFFOFF: - t += "" + t += u"" # special html chars elif htmlentitydefs.codepoint2name.has_key(ord(c)): - t += "&%s;" % (htmlentitydefs.codepoint2name[ord(c)]) + t += u"&%s;" % (htmlentitydefs.codepoint2name[ord(c)]) i += 1 # special highlighted chars @@ -181,32 +265,34 @@ def convert(s, linesize=0, ponct=0): n = tabsize-(i%tabsize) if n == 0: n = tabsize - t += ('»'+' '*(n-1)) + t += (u'»'+' '*(n-1)) elif c == " " and ponct == 1: - t += '·' + t += u'·' elif c == "\n" and ponct == 1: if show_CR: - t += '\' + t += u'\' else: t += c i += 1 if linesize and (WORDBREAK.count(c) == 1): - t += '' + t += u'' i = 0 if linesize and i > linesize: i = 0 - t += "" + t += u"" return t def add_comment(s, output_file): - output_file.write('%s\n'%convert(s)) + output_file.write(('%s\n'%convert(s)).encode(encoding)) + def add_filename(f1, f2, output_file): - output_file.write("%s"%convert(f1, linesize=linesize)) - output_file.write("%s\n"%convert(f2, linesize=linesize)) + output_file.write(("%s"%convert(f1, linesize=linesize)).encode(encoding)) + output_file.write(("%s\n"%convert(f2, linesize=linesize)).encode(encoding)) + def add_hunk(output_file, show_hunk_infos): if show_hunk_infos: @@ -221,6 +307,9 @@ def add_line(s1, s2, output_file): global line1 global line2 + orig1 = s1 + orig2 = s2 + if s1 == None and s2 == None: type_name = "unmodified" elif s1 == None or s1 == "": @@ -231,22 +320,27 @@ def add_line(s1, s2, output_file): type_name = "unmodified" else: type_name = "changed" - s1, s2 = linediff(s1, s2) - - output_file.write('' % type_name) + if algorithm == 1: + s1, s2 = diff_changed_words_ts(orig1, orig2) + elif algorithm == 2: + s1, s2 = diff_changed_ts(orig1, orig2) + else: # default + s1, s2 = linediff(orig1, orig2) + + output_file.write(('' % type_name).encode(encoding)) if s1 != None and s1 != "": - output_file.write('%d ' % line1) - output_file.write('') - output_file.write(convert(s1, linesize=linesize, ponct=1)) + output_file.write(('%d ' % line1).encode(encoding)) + output_file.write(''.encode(encoding)) + output_file.write(convert(s1, linesize=linesize, ponct=1).encode(encoding)) output_file.write('') else: s1 = "" output_file.write(' ') if s2 != None and s2 != "": - output_file.write('%d '%line2) + output_file.write(('%d '%line2).encode(encoding)) output_file.write('') - output_file.write(convert(s2, linesize=linesize, ponct=1)) + output_file.write(convert(s2, linesize=linesize, ponct=1).encode(encoding)) output_file.write('') else: s2 = "" @@ -289,15 +383,16 @@ def empty_buffer(output_file): buf = [] -def parse_input(input_file, output_file, +def parse_input(input_file, output_file, input_file_name, output_file_name, exclude_headers, show_hunk_infos): global add_cpt, del_cpt global line1, line2 global hunk_off1, hunk_size1, hunk_off2, hunk_size2 if not exclude_headers: - output_file.write(html_hdr) - output_file.write(table_hdr) + title_suffix = ' ' + input_file_name + output_file.write(html_hdr.format(title_suffix, encoding, desc, "", modified_date, lang).encode(encoding)) + output_file.write(table_hdr.encode(encoding)) while True: l = input_file.readline() @@ -308,10 +403,12 @@ def parse_input(input_file, output_file, if m: empty_buffer(output_file) file1 = m.groups()[0] - l = input_file.readline() - m = re.match('^\+\+\+ ([^\s]*)', l) - if m: - file2 = m.groups()[0] + while True: + l = input_file.readline() + m = re.match('^\+\+\+ ([^\s]*)', l) + if m: + file2 = m.groups()[0] + break add_filename(file1, file2, output_file) hunk_off1, hunk_size1, hunk_off2, hunk_size2 = 0, 0, 0, 0 continue @@ -353,46 +450,51 @@ def parse_input(input_file, output_file, add_comment(l, output_file) empty_buffer(output_file) - output_file.write(table_footer) + output_file.write(table_footer.encode(encoding)) if not exclude_headers: - output_file.write(html_footer) + output_file.write(html_footer.format("", dtnow.strftime("%d.%m.%Y")).encode(encoding)) def usage(): print ''' -diff2html.py [-i file] [-o file] [-x] +diff2html.py [-e encoding] [-i file] [-o file] [-x] diff2html.py -h Transform a unified diff from stdin to a colored side-by-side HTML page on stdout. +stdout may not work with UTF-8, instead use -o option. -i file set input file, else use stdin + -e encoding set file encoding (default utf-8) -o file set output file, else use stdout -x exclude html header and footer -t tabsize set tab size (default 8) -l linesize set maximum line size is there is no word break (default 20) -r show \\r characters -k show hunk infos + -a algo line diff algorithm (0: linediff characters, 1: word, 2: simplediff characters) (default 0) -h show help and exit ''' def main(): global linesize, tabsize global show_CR + global encoding + global algorithm - input_file = sys.stdin - output_file = sys.stdout + input_file_name = '' + output_file_name = '' exclude_headers = False show_hunk_infos = False try: - opts, args = getopt.getopt(sys.argv[1:], "hi:o:xt:l:rk", - ["help", "input=", "output=", + opts, args = getopt.getopt(sys.argv[1:], "he:i:o:xt:l:rka:", + ["help", "encoding=", "input=", "output=", "exclude-html-headers", "tabsize=", - "linesize=", "show-cr", "show-hunk-infos"]) + "linesize=", "show-cr", "show-hunk-infos", "algorithm="]) except getopt.GetoptError, err: - print str(err) # will print something like "option -a not recognized" + print unicode(err) # will print something like "option -a not recognized" usage() sys.exit(2) verbose = False @@ -400,10 +502,14 @@ def main(): if o in ("-h", "--help"): usage() sys.exit() + elif o in ("-e", "--encoding"): + encoding = a elif o in ("-i", "--input"): - input_file = open(a, "r") + input_file = codecs.open(a, "r", encoding) + input_file_name = a elif o in ("-o", "--output"): - output_file = open(a, "w") + output_file = codecs.open(a, "w") + output_file_name = a elif o in ("-x", "--exclude-html-headers"): exclude_headers = True elif o in ("-t", "--tabsize"): @@ -414,11 +520,29 @@ def main(): show_CR = True elif o in ("-k", "--show-hunk-infos"): show_hunk_infos = True + elif o in ("-a", "--algorithm"): + algorithm = int(a) else: assert False, "unhandled option" - parse_input(input_file, output_file, + + # Use stdin if not input file is set + if not ('input_file' in locals()): + input_file = codecs.getreader(encoding)(sys.stdin) + + # Use stdout if not output file is set + if not ('output_file' in locals()): + output_file = codecs.getwriter(encoding)(sys.stdout) + + parse_input(input_file, output_file, input_file_name, output_file_name, exclude_headers, show_hunk_infos) +def parse_from_memory(txt, exclude_headers, show_hunk_infos): + " Parses diff from memory and returns a string with html " + input_stream = StringIO.StringIO(txt) + output_stream = StringIO.StringIO() + parse_input(input_stream, output_stream, '', '', exclude_headers, show_hunk_infos) + return output_stream.getvalue() + if __name__ == "__main__": main()