-#!/usr/bin/python
+#! /usr/bin/python
#
# 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>
#
# 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
+
+# minimum line size, we add a zero-sized breakable space every
+# LINESIZE characters
+linesize = 20
+tabsize = 8
+inputfile = sys.stdin
+outputfile = sys.stdout
+exclude_headers = False
+show_CR = False
+show_hunk_infos = False
-import sys, re, htmlentitydefs
html_hdr="""<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html><head>
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 {}
+ 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>
- <table>
"""
html_footer="""
-</table></body></html>
+</body></html>
+"""
+
+table_hdr="""
+ <table class="diff">
+"""
+
+table_footer="""
+</table>
"""
DIFFON="\x01"
DIFFOFF="\x02"
-buffer=[]
+buf=[]
add_cpt, del_cpt = 0,0
line1, line2 = 0,0
hunk_off1, hunk_size1, hunk_off2, hunk_size2 = 0,0,0,0
-# minimum line size, we add a zero-sized breakable space every
-# LINESIZE characters
-LINESIZE=20
-TAB=8
+
+# Characters we're willing to word wrap on
+WORDBREAK=" \t;.,/):"
def sane(x):
r=""
for c in s:
# used by diffs
if c==DIFFON:
- t += '<spanclass="changed2">'
+ t += '<span class="diffchanged2">'
elif c==DIFFOFF:
t += "</span>"
# special highlighted chars
elif c=="\t" and ponct==1:
- n = TAB-(i%TAB)
+ n = tabsize-(i%tabsize)
if n==0:
- n=TAB
- t += ('<spanclass="ponct">»</span>'+' '*(n-1))
- i += n
+ n=tabsize
+ t += ('<span class="diffponct">»</span>'+' '*(n-1))
+ elif c==" " and ponct==1:
+ t += '<span class="diffponct">·</span>'
elif c=="\n" and ponct==1:
- t += '<spanclass="ponct">\</span>'
+ if show_CR:
+ t += '<span class="diffponct">\</span>'
else:
t += c
i += 1
+
+ if linesize and (WORDBREAK.count(c)==1):
+ t += '​'
+ i=0
if linesize and i>linesize:
i=0
t += "​"
-
- if ponct==1:
- t = t.replace(' ', '<spanclass="ponct">·</span>')
- t = t.replace("spanclass", "span class")
return t
def add_comment(s):
- sys.stdout.write('<tr class="misc"><td colspan="4">%s</td></tr>\n'%convert(s))
+ outputfile.write('<tr class="diffmisc"><td colspan="4">%s</td></tr>\n'%convert(s))
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))
+ outputfile.write("<tr><th colspan='2'>%s</th>"%convert(f1, linesize=linesize))
+ outputfile.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))
+ global hunk_off1, hunk_size1, hunk_off2, hunk_size2
+ global show_hunk_infos
+ if show_hunk_infos:
+ outputfile.write('<tr class="diffhunk"><td colspan="2">Offset %d, %d lines modified</td>'%(hunk_off1, hunk_size1))
+ outputfile.write('<td colspan="2">Offset %d, %d lines modified</td></tr>\n'%(hunk_off2, hunk_size2))
+ else:
+ # ⋮ - vertical ellipsis
+ outputfile.write('<tr class="diffhunk"><td colspan="2">⋮</td><td colspan="2">⋮</td></tr>');
+
def add_line(s1, s2):
global line1
global line2
if s1==None and s2==None:
- type="unmodified"
- elif s1==None:
- type="added"
- elif s2==None:
- type="deleted"
+ type_name="unmodified"
+ elif s1==None or s1=="":
+ type_name="added"
+ elif s2==None or s1=="":
+ type_name="deleted"
elif s1==s2:
- type="unmodified"
+ type_name="unmodified"
else:
- type="changed"
+ type_name="changed"
s1,s2 = linediff(s1, s2)
- sys.stdout.write('<tr class="%s">'%type)
+ outputfile.write('<tr class="diff%s">'%type_name)
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>')
+ outputfile.write('<td class="diffline">%d </td>'%line1)
+ outputfile.write('<td class="diffpresent">')
+ outputfile.write(convert(s1, linesize=linesize, ponct=1))
+ outputfile.write('</td>')
else:
s1=""
- sys.stdout.write('<td colspan="2"> </td>')
+ outputfile.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>')
+ outputfile.write('<td class="diffline">%d </td>'%line2)
+ outputfile.write('<td class="diffpresent">')
+ outputfile.write(convert(s2, linesize=linesize, ponct=1))
+ outputfile.write('</td>')
else:
s2=""
- sys.stdout.write('<td colspan="2"></td>')
+ outputfile.write('<td colspan="2"></td>')
- sys.stdout.write('</tr>\n')
+ outputfile.write('</tr>\n')
if s1!="":
line1 += 1
def empty_buffer():
- global buffer
+ global buf
global add_cpt
global del_cpt
if del_cpt == 0 or add_cpt == 0:
- for l in buffer:
+ for l in buf:
add_line(l[0], l[1])
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:
add_line(s0, s1)
add_cpt, del_cpt = 0,0
- buffer = []
+ buf = []
+def parse_input():
+ global buf, add_cpt, del_cpt
+ global line1, line2
+ global hunk_off1, hunk_size1, hunk_off2, hunk_size2
-sys.stdout.write(html_hdr)
+ if not exclude_headers:
+ outputfile.write(html_hdr)
+ outputfile.write(table_hdr)
+
+ while True:
+ l=inputfile.readline()
+ if l=="":
+ break
-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)
+ m=re.match('^--- ([^\s]*)', l)
+ if m:
+ empty_buffer()
+ file1=m.groups()[0]
+ l=inputfile.readline()
+ m=re.match('^\+\+\+ ([^\s]*)', 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:
- file2=m.groups()[0]
- add_filename(file1, file2)
- hunk_off1, hunk_size1, hunk_off2, hunk_size2 = 0,0,0,0
- continue
+ 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
+ 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()
+ hunk_size1 -= 1
+ hunk_size2 -= 1
+ buf.append((l[1:], l[1:]))
+ 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)
-
+ outputfile.write(table_footer)
+ if not exclude_headers:
+ outputfile.write(html_footer)
+
+
+def usage():
+ print '''
+diff2html.py [-i file] [-o file] [-x]
+diff2html.py -h
+
+Transform a unified diff from stdin to a colored side-by-side HTML
+page on stdout.
+
+ -i file set input file, else use stdin
+ -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 inputfile, outputfile
+ global exclude_headers, show_CR, show_hunk_infos
+
+ try:
+ opts, args = getopt.getopt(sys.argv[1:], "hi:o:xt:l:rk",
+ ["help", "input=", "output=",
+ "exclude-html-headers", "tabsize=",
+ "linesize=", "show-cr", "show-hunk-infos"])
+ except getopt.GetoptError, err:
+ print str(err) # will print something like "option -a not recognized"
+ usage()
+ sys.exit(2)
+ output = None
+ verbose = False
+ for o, a in opts:
+ if o in ("-h", "--help"):
+ usage()
+ sys.exit()
+ elif o in ("-i", "--input"):
+ inputfile = open(a, "r")
+ elif o in ("-o", "--output"):
+ outputfile = open(a, "w")
+ 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"
+ parse_input()
-empty_buffer()
-sys.stdout.write(html_footer)
+if __name__ == "__main__":
+ main()
git.droids-corp.org / diff2html.git / blobdiff