-#!/usr/bin/python
+#! /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
# Transform a unified diff from stdin to a colored
# side-by-side HTML page on stdout.
#
-# Author: Olivier MATZ <zer0@droids-corp.org>
+# Authors: Olivier MATZ <zer0@droids-corp.org>
+# Alan De Smet <adesmet@cs.wisc.edu>
+# Sergey Satskiy <sergey.satskiy@gmail.com>
+# scito <info at scito.ch>
#
# Inspired by diff2html.rb from Dave Burt <dave (at) burt.id.au>
# (mainly for html theme)
+#
+# TODO:
+# - The sane function currently mashes non-ASCII characters to "."
+# Instead be clever and convert to something like "xF0"
+# (the hex value), and mark with a <span>. Even more clever:
+# Detect if the character is "printable" for whatever definition,
+# and display those directly.
+
+import sys, re, htmlentitydefs, getopt, StringIO, codecs, datetime
-import sys, re, htmlentitydefs
-
-html_hdr="""<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
- <html><head>
- <meta name="generator" content="diff2html.rb" />
- <title>HTML Diff</title>
- <style>
- table { border:0px; border-collapse:collapse; width: 100%; font-size:0.75em; font-family: Lucida Console, monospace }
- td.line { color:#8080a0 }
- th { background: black; color: white }
- tr.unmodified td { background: #D0D0E0 }
- tr.hunk td { background: #A0A0A0 }
- tr.added td { background: #CCFFCC }
- tr.deleted td { background: #FFCCCC }
- tr.changed td { background: #FFFFA0 }
- span.changed2 { background: #E0C880 }
- span.ponct { color: #B08080 }
- tr.misc td {}
- tr.separator td {}
- </style>
- </head>
- <body>
- <table>
+# minimum line size, we add a zero-sized breakable space every
+# LINESIZE characters
+linesize = 20
+tabsize = 8
+show_CR = False
+encoding = "utf-8"
+lang = "en"
+
+desc = "File comparison"
+dtnow = datetime.datetime.now()
+modified_date = "%s+01:00"%dtnow.isoformat()
+
+html_hdr = """<!DOCTYPE html>
+<html lang="{5}" dir="ltr"
+ xmlns:dc="http://purl.org/dc/terms/">
+<head>
+ <meta charset="{1}" />
+ <meta name="generator" content="diff2html.py (http://git.droids-corp.org/gitweb/?p=diff2html)" />
+ <!--meta name="author" content="Fill in" /-->
+ <title>HTML Diff{0}</title>
+ <link rel="shortcut icon" href="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQAgMAAABinRfyAAAACVBMVEXAAAAAgAD///+K/HwIAAAAJUlEQVQI12NYBQQM2IgGBQ4mCIEQW7oyK4phampkGIQAc1G1AQCRxCNbyW92oQAAAABJRU5ErkJggg==" type="image/png" />
+ <meta property="dc:language" content="{5}" />
+ <!--meta property="dc:date" content="{3}" /-->
+ <meta property="dc:modified" content="{4}" />
+ <meta name="description" content="{2}" />
+ <meta property="dc:abstract" content="{2}" />
+ <style>
+ table {{ border:0px; border-collapse:collapse; width: 100%; font-size:0.75em; font-family: Lucida Console, monospace }}
+ td.line {{ color:#8080a0 }}
+ th {{ background: black; color: white }}
+ tr.diffunmodified td {{ background: #D0D0E0 }}
+ tr.diffhunk td {{ background: #A0A0A0 }}
+ tr.diffadded td {{ background: #CCFFCC }}
+ tr.diffdeleted td {{ background: #FFCCCC }}
+ tr.diffchanged td {{ background: #FFFFA0 }}
+ span.diffchanged2 {{ background: #E0C880 }}
+ span.diffponct {{ color: #B08080 }}
+ tr.diffmisc td {{}}
+ tr.diffseparator td {{}}
+ </style>
+</head>
+<body>
"""
-html_footer="""
-</table></body></html>
+html_footer = """
+<footer>
+ <p>Modified at {1}. HTML formatting created by <a href="http://git.droids-corp.org/gitweb/?p=diff2html;a=summary">diff2html</a>. </p>
+</footer>
+</body></html>
"""
-DIFFON="\x01"
-DIFFOFF="\x02"
+table_hdr = """
+ <table class="diff">
+"""
-buffer=[]
-add_cpt, del_cpt = 0,0
-line1, line2 = 0,0
-hunk_off1, hunk_size1, hunk_off2, hunk_size2 = 0,0,0,0
+table_footer = """
+</table>
+"""
-# minimum line size, we add a zero-sized breakable space every
-# LINESIZE characters
-LINESIZE=20
-TAB=8
+DIFFON = "\x01"
+DIFFOFF = "\x02"
+
+buf = []
+add_cpt, del_cpt = 0, 0
+line1, line2 = 0, 0
+hunk_off1, hunk_size1, hunk_off2, hunk_size2 = 0, 0, 0, 0
+
+
+# Characters we're willing to word wrap on
+WORDBREAK = " \t;.,/):"
def sane(x):
- r=""
+ r = ""
for i in x:
j = ord(i)
- if i not in ['\t', '\n'] and ((j < 32) or (j >= 127)):
- r=r+"."
+ if i not in ['\t', '\n'] and (j < 32):
+ r = r + "."
else:
- r=r+i
+ r = r + i
return r
def linediff(s, t):
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 ]))
-
- 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))
+ 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)]
+
+
+ d[0][0] = (0, (0, 0))
for i in range(m+1)[1:]:
- d[i][0] = (i,(i-1,0))
+ d[i][0] = (i,(i-1, 0))
for j in range(n+1)[1:]:
- d[0][j] = (j,(0,j-1))
+ d[0][j] = (j,(0, j-1))
for i in range(m+1)[1:]:
for j in range(n+1)[1:]:
cost = 0
else:
cost = 1
- d[i][j] = min((d[i-1][j][0] + 1, (i-1,j)),
- (d[i][j-1][0] + 1, (i,j-1)),
- (d[i-1][j-1][0] + cost, (i-1,j-1)))
-
- l=[]
- coord = (m,n)
- while coord != (0,0):
+ d[i][j] = min((d[i-1][j][0] + 1, (i-1, j)),
+ (d[i][j-1][0] + 1, (i, j-1)),
+ (d[i-1][j-1][0] + cost, (i-1, j-1)))
+
+ l = []
+ coord = (m, n)
+ while coord != (0, 0):
l.insert(0, coord)
- x,y = coord
+ x, y = coord
coord = d[x][y][1]
l1 = []
l2 = []
for coord in l:
- cx,cy = coord
+ cx, cy = coord
child_val = d[cx][cy][0]
-
+
father_coord = d[cx][cy][1]
- fx,fy = father_coord
+ fx, fy = father_coord
father_val = d[fx][fy][0]
diff = (cx-fx, cy-fy)
- if diff == (0,1):
+ if diff == (0, 1):
l1.append("")
l2.append(DIFFON + t[fy] + DIFFOFF)
- elif diff == (1,0):
+ elif diff == (1, 0):
l1.append(DIFFON + s[fx] + DIFFOFF)
l2.append("")
elif child_val-father_val == 1:
l1.append(s[fx])
l2.append(t[fy])
- r1,r2 = (reduce(lambda x,y:x+y, l1), reduce(lambda x,y:x+y, l2))
- return r1,r2
+ r1, r2 = (reduce(lambda x, y:x+y, l1), reduce(lambda x, y:x+y, l2))
+ return r1, r2
def convert(s, linesize=0, ponct=0):
- i=0
- t=""
- l=[]
+ i = 0
+ t = u""
for c in s:
# used by diffs
- if c==DIFFON:
- t += '<spanclass="changed2">'
- elif c==DIFFOFF:
- t += "</span>"
+ if c == DIFFON:
+ t += u'<span class="diffchanged2">'
+ elif c == DIFFOFF:
+ t += u"</span>"
# 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
- elif c=="\t" and ponct==1:
- n = TAB-(i%TAB)
- if n==0:
- n=TAB
- t += ('<spanclass="ponct">»</span>'+' '*(n-1))
- i += n
- elif c=="\n" and ponct==1:
- t += '<spanclass="ponct">\</span>'
+ elif c == "\t" and ponct == 1:
+ n = tabsize-(i%tabsize)
+ if n == 0:
+ n = tabsize
+ t += (u'<span class="diffponct">»</span>'+' '*(n-1))
+ elif c == " " and ponct == 1:
+ t += u'<span class="diffponct">·</span>'
+ elif c == "\n" and ponct == 1:
+ if show_CR:
+ t += u'<span class="diffponct">\</span>'
else:
t += c
i += 1
- if linesize and i>linesize:
- i=0
- t += "​"
-
- if ponct==1:
- t = t.replace(' ', '<spanclass="ponct">·</span>')
- t = t.replace("spanclass", "span class")
-
+
+ if linesize and (WORDBREAK.count(c) == 1):
+ t += u'​'
+ i = 0
+ if linesize and i > linesize:
+ i = 0
+ t += u"​"
+
return t
-def add_comment(s):
- sys.stdout.write('<tr class="misc"><td colspan="4">%s</td></tr>\n'%convert(s))
+def add_comment(s, output_file):
+ output_file.write(('<tr class="diffmisc"><td colspan="4">%s</td></tr>\n'%convert(s)).encode(encoding))
-def add_filename(f1, f2):
- sys.stdout.write("<tr><th colspan='2'>%s</th>"%convert(f1, linesize=LINESIZE))
- sys.stdout.write("<th colspan='2'>%s</th></tr>\n"%convert(f2, linesize=LINESIZE))
-def add_hunk():
- global hunk_off1
- global hunk_size1
- global hunk_off2
- global hunk_size2
- sys.stdout.write('<tr class="hunk"><td colspan="2">Offset %d, %d lines modified</td>'%(hunk_off1, hunk_size1))
- sys.stdout.write('<td colspan="2">Offset %d, %d lines modified</td></tr>\n'%(hunk_off2, hunk_size2))
+def add_filename(f1, f2, output_file):
+ output_file.write(("<tr><th colspan='2'>%s</th>"%convert(f1, linesize=linesize)).encode(encoding))
+ output_file.write(("<th colspan='2'>%s</th></tr>\n"%convert(f2, linesize=linesize)).encode(encoding))
-def add_line(s1, s2):
+
+def add_hunk(output_file, show_hunk_infos):
+ if show_hunk_infos:
+ output_file.write('<tr class="diffhunk"><td colspan="2">Offset %d, %d lines modified</td>'%(hunk_off1, hunk_size1))
+ output_file.write('<td colspan="2">Offset %d, %d lines modified</td></tr>\n'%(hunk_off2, hunk_size2))
+ else:
+ # ⋮ - vertical ellipsis
+ output_file.write('<tr class="diffhunk"><td colspan="2">⋮</td><td colspan="2">⋮</td></tr>')
+
+
+def add_line(s1, s2, output_file):
global line1
global line2
- if s1==None and s2==None:
- type="unmodified"
- elif s1==None:
- type="added"
- elif s2==None:
- type="deleted"
- elif s1==s2:
- type="unmodified"
+ if s1 == None and s2 == None:
+ type_name = "unmodified"
+ elif s1 == None or s1 == "":
+ type_name = "added"
+ elif s2 == None or s1 == "":
+ type_name = "deleted"
+ elif s1 == s2:
+ type_name = "unmodified"
else:
- type="changed"
- s1,s2 = linediff(s1, s2)
-
- sys.stdout.write('<tr class="%s">'%type)
- if s1!=None and s1!="":
- sys.stdout.write('<td class="line">%d </td>'%line1)
- sys.stdout.write('<td>')
- sys.stdout.write(convert(s1, linesize=LINESIZE, ponct=1))
- sys.stdout.write('</td>')
+ type_name = "changed"
+ s1, s2 = linediff(s1, s2)
+
+ output_file.write(('<tr class="diff%s">' % type_name).encode(encoding))
+ if s1 != None and s1 != "":
+ output_file.write(('<td class="diffline">%d </td>' % line1).encode(encoding))
+ output_file.write('<td class="diffpresent">'.encode(encoding))
+ output_file.write(convert(s1, linesize=linesize, ponct=1).encode(encoding))
+ output_file.write('</td>')
else:
- s1=""
- sys.stdout.write('<td colspan="2"> </td>')
-
- if s2!=None and s2!="":
- sys.stdout.write('<td class="line">%d </td>'%line2)
- sys.stdout.write('<td>')
- sys.stdout.write(convert(s2, linesize=LINESIZE, ponct=1))
- sys.stdout.write('</td>')
+ s1 = ""
+ output_file.write('<td colspan="2"> </td>')
+
+ if s2 != None and s2 != "":
+ output_file.write(('<td class="diffline">%d </td>'%line2).encode(encoding))
+ output_file.write('<td class="diffpresent">')
+ output_file.write(convert(s2, linesize=linesize, ponct=1).encode(encoding))
+ output_file.write('</td>')
else:
- s2=""
- sys.stdout.write('<td colspan="2"></td>')
+ s2 = ""
+ output_file.write('<td colspan="2"></td>')
- sys.stdout.write('</tr>\n')
+ output_file.write('</tr>\n')
- if s1!="":
+ if s1 != "":
line1 += 1
- if s2!="":
+ if s2 != "":
line2 += 1
-def empty_buffer():
- global buffer
+def empty_buffer(output_file):
+ global buf
global add_cpt
global del_cpt
if del_cpt == 0 or add_cpt == 0:
- for l in buffer:
- add_line(l[0], l[1])
+ for l in buf:
+ add_line(l[0], l[1], output_file)
elif del_cpt != 0 and add_cpt != 0:
l0, l1 = [], []
- for l in buffer:
+ for l in buf:
if l[0] != None:
l0.append(l[0])
if l[1] != None:
l1.append(l[1])
- max = (len(l0) > len(l1)) and len(l0) or len(l1)
- for i in range(max):
+ max_len = (len(l0) > len(l1)) and len(l0) or len(l1)
+ for i in range(max_len):
s0, s1 = "", ""
- if i<len(l0):
+ if i < len(l0):
s0 = l0[i]
- if i<len(l1):
+ if i < len(l1):
s1 = l1[i]
- add_line(s0, s1)
-
- add_cpt, del_cpt = 0,0
- buffer = []
-
-
-
-sys.stdout.write(html_hdr)
-
-while True:
- l=sys.stdin.readline()
- if l=="":
- break
-
- m=re.match('^--- ([^\s]*)', l)
- if m:
- empty_buffer()
- file1=m.groups()[0]
- l=sys.stdin.readline()
- m=re.match('^\+\+\+ ([^\s]*)', l)
+ add_line(s0, s1, output_file)
+
+ add_cpt, del_cpt = 0, 0
+ buf = []
+
+
+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:
+ 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()
+ if l == "":
+ break
+
+ m = re.match('^--- ([^\s]*)', l)
+ if m:
+ empty_buffer(output_file)
+ file1 = 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
+
+ m = re.match("@@ -(\d+),?(\d*) \+(\d+),?(\d*)", l)
if m:
- file2=m.groups()[0]
- add_filename(file1, file2)
- hunk_off1, hunk_size1, hunk_off2, hunk_size2 = 0,0,0,0
- continue
-
- m=re.match("@@ -(\d+),?(\d*) \+(\d+),?(\d*)", l)
- if m:
- empty_buffer()
- hunk_data = map(lambda x:x=="" and 1 or int(x), m.groups())
- hunk_off1, hunk_size1, hunk_off2, hunk_size2 = hunk_data
- line1, line2 = hunk_off1, hunk_off2
- add_hunk()
- continue
-
- if hunk_size1 == 0 and hunk_size2 == 0:
- empty_buffer()
- add_comment(l)
- continue
-
- if re.match("^\+", l):
- add_cpt += 1
- hunk_size2 -= 1
- buffer.append((None, l[1:]))
- continue
-
- if re.match("^\-", l):
- del_cpt += 1
- hunk_size1 -= 1
- buffer.append((l[1:], None))
- continue
-
- if re.match("^\ ", l) and hunk_size1 and hunk_size2:
- empty_buffer()
- hunk_size1 -= 1
- hunk_size2 -= 1
- buffer.append((l[1:], l[1:]))
- continue
-
- empty_buffer()
- add_comment(l)
-
-
-empty_buffer()
-sys.stdout.write(html_footer)
+ empty_buffer(output_file)
+ hunk_data = map(lambda x:x=="" and 1 or int(x), m.groups())
+ hunk_off1, hunk_size1, hunk_off2, hunk_size2 = hunk_data
+ line1, line2 = hunk_off1, hunk_off2
+ add_hunk(output_file, show_hunk_infos)
+ continue
+
+ if hunk_size1 == 0 and hunk_size2 == 0:
+ empty_buffer(output_file)
+ add_comment(l, output_file)
+ continue
+
+ if re.match("^\+", l):
+ add_cpt += 1
+ hunk_size2 -= 1
+ buf.append((None, l[1:]))
+ continue
+
+ if re.match("^\-", l):
+ del_cpt += 1
+ hunk_size1 -= 1
+ buf.append((l[1:], None))
+ continue
+
+ if re.match("^\ ", l) and hunk_size1 and hunk_size2:
+ empty_buffer(output_file)
+ hunk_size1 -= 1
+ hunk_size2 -= 1
+ buf.append((l[1:], l[1:]))
+ continue
+
+ empty_buffer(output_file)
+ add_comment(l, output_file)
+
+ empty_buffer(output_file)
+ output_file.write(table_footer.encode(encoding))
+ if not exclude_headers:
+ output_file.write(html_footer.format("", dtnow.strftime("%d.%m.%Y")).encode(encoding))
+
+
+def usage():
+ print '''
+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
+ -h show help and exit
+'''
+
+def main():
+ global linesize, tabsize
+ global show_CR
+ global encoding
+
+ input_file_name = ''
+ output_file_name = ''
+
+ exclude_headers = False
+ show_hunk_infos = False
+
+ try:
+ opts, args = getopt.getopt(sys.argv[1:], "he:i:o:xt:l:rk",
+ ["help", "encoding=", "input=", "output=",
+ "exclude-html-headers", "tabsize=",
+ "linesize=", "show-cr", "show-hunk-infos"])
+ except getopt.GetoptError, err:
+ print unicode(err) # will print something like "option -a not recognized"
+ usage()
+ sys.exit(2)
+ verbose = False
+ for o, a in opts:
+ if o in ("-h", "--help"):
+ usage()
+ sys.exit()
+ elif o in ("-e", "--encoding"):
+ encoding = a
+ elif o in ("-i", "--input"):
+ input_file = codecs.open(a, "r", encoding)
+ input_file_name = a
+ elif o in ("-o", "--output"):
+ 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"):
+ tabsize = int(a)
+ elif o in ("-l", "--linesize"):
+ linesize = int(a)
+ elif o in ("-r", "--show-cr"):
+ show_CR = True
+ elif o in ("-k", "--show-hunk-infos"):
+ show_hunk_infos = True
+ else:
+ assert False, "unhandled option"
+
+ # 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()
git.droids-corp.org / diff2html.git / blobdiff