#!/usr/bin/env python3
import re
import sys
import os
import codecs
canonical_url = 'https://slurm.schedmd.com/'
include_pat = r'()'
include_regex = re.compile(include_pat)
canonical_pat = r'()'
canonical_regex = re.compile(canonical_pat)
page_title_pat = r'()'
page_title_regex = re.compile(page_title_pat)
url_pat = r'(\s+href\s*=\s*")([^"#]+)(#[^"]+)?(")'
url_regex = re.compile(url_pat)
first_header_pat = r'(<[h|H]1>\s*([a-zA-Z0-9_ ()\'/-\\.]+)\s*)'
first_header_regex = re.compile(first_header_pat)
version_pat = r'(@SLURM_VERSION@)'
version_regex = re.compile(version_pat)
ids = {}
title = ''
dirname = ''
# Insert tags for options
# Two styles are processed.
# pppppp
# has tag --pppppp or -P , --pppppp
# has tag hhhh also has tag has tag ":
header_num = lineIn[2]
# Don't make links to h1 tags - no need, since already at top of page
if header_num == "1":
return lineIn;
posEnd = lineIn.find("" % header_num)
if posEnd == -1:
return lineIn;
contents = lineIn[4:posEnd]
id_name = contents.replace(' ','-')
if id_name in ids:
ids[id_name] += 1
id_name += "_" + str(ids[id_name])
else:
ids[id_name] = 0
id_name = "SECTION_%s" % (id_name)
lineOt = '%s ' % (header_num, contents, id_name, id_name, header_num)
return lineOt
if lineIn[0:4] != "":
return lineIn
if lineIn[4:7] == "":
bold_tag = True
elif lineIn[4:7] == "":
bold_tag = False
else:
return lineIn
contentsBgn = 4
contentsEnd = lineIn.find("")
contents = lineIn[contentsBgn:contentsEnd]
posBgn = lineIn.find("--")
if posBgn == -1:
# 1st form
posBgn = 7
else:
posBgn += 2
if bold_tag:
posEnd = lineIn.find(" ",posBgn)
else:
posEnd = lineIn.find("",posBgn)
if posEnd == -1:
# poorly constructed
return lineIn
id_name = lineIn[posBgn:posEnd]
id_name = id_name.replace(' ','-')
if id_name in ids:
ids[id_name] += 1
id_name += "_" + str(ids[id_name])
else:
ids[id_name] = 0
id_name = "OPT_%s" % (id_name)
link_and_contents = '%s" + link_and_contents + " "
return lineOt
def llnl_references(line):
manStr = "Refer to mc_support.html"
htmlStr = 'Refer to mc_support '
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'https://slurm.schedmd.com/mc_support.html '
htmlStr = 'the mc_support document'
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'https://slurm.schedmd.com/dist_plane.html. '
htmlStr = 'the dist_plane document'
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = '<https://slurm.schedmd.com/mpi_guide.html >'
htmlStr = 'mpi_guide '
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = '(https://slurm.schedmd.com/power_save.html). '
htmlStr = 'power_save '
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'https://slurm.schedmd.com/cons_tres.html '
htmlStr = 'cons_tres '
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'https://slurm.schedmd.com/cons_tres_share.html '
htmlStr = 'cons_tres_share '
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'https://slurm.schedmd.com/gang_scheduling.html '
htmlStr = 'gang_scheduling '
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'https://slurm.schedmd.com/preempt.html '
htmlStr = 'preempt '
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'logrotate '
htmlStr = 'logrotate'
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'logrotate '
htmlStr = 'logrotate'
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'path_resolution '
htmlStr = 'path_resolution'
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
manStr = 'path_resolution '
htmlStr = 'path_resolution'
lineFix = line.replace(manStr,htmlStr)
if lineFix != line:
return lineFix
return line
def relative_reference(lineIn):
lineOt = ""
refAnchor="man2html "
while posHREF != -1:
posRefAnchor = lineIn.find(refAnchor,cursor)
lineOt = lineOt + lineIn[cursor:posRefAnchor+lenRefAnchor]
cursor = posHREF + lenRef + 3
lineOt = lineOt + '"'
posQuote = lineIn.find('"',cursor)
lineOt = lineOt + lineIn[cursor:posQuote] + ".html"
cursor = posQuote
posHREF = lineIn.find(fullRef,cursor)
lineOt = lineOt + lineIn[cursor:]
if lineOt != lineIn:
break;
if lineOt == "":
return lineIn
return lineOt
def include_virtual(matchobj):
global dirname
if dirname:
filename = dirname + '/' + matchobj.group(2)
else:
filename = matchobj.group(2)
if os.access(filename, os.F_OK):
lines = open(filename, 'r').read()
return lines
else:
return matchobj.group(0)
def canonical_rewrite(matchobj):
global newfilename
return 'Slurm Workload Manager - ' + title + ' '
def url_rewrite(matchobj):
global dirname
if dirname:
localpath = dirname + '/' + matchobj.group(2)
else:
localpath = matchobj.group(2)
if matchobj.group(2)[-6:] == '.shtml' and os.access(localpath, os.F_OK):
location = matchobj.group(2)
if matchobj.group(3) is None:
newname = location[:-6] + '.html'
else:
newname = location[:-6] + '.html' + matchobj.group(3)
return matchobj.group(1) + newname + matchobj.group(4)
else:
return matchobj.group(0)
def version_rewrite(matchobj):
global version
return version
files = []
version = sys.argv[1]
for f in sys.argv[4:]:
dirname, basefilename = os.path.split(f)
posLastDot = basefilename.rfind(".")
mhtmlname = basefilename[:posLastDot] + ".mhtml"
cmd = "man2html < " + f + "> " + mhtmlname
os.system(cmd)
print(">>>>>>> " + mhtmlname)
files.append(mhtmlname)
for filename in files:
newfilename = filename[:-6] + '.html'
print('Converting', filename, '->', newfilename)
shtml = codecs.open(filename, 'r', encoding='utf-8')
html = codecs.open(newfilename, 'w', encoding='utf-8')
for line in shtml.readlines():
result = first_header_regex.match(line)
if result:
title = result.group(2)
break
shtml.seek(0)
lines = open(sys.argv[2], 'r').read()
lines = lines.replace(".shtml",".html")
lines = version_regex.sub(version_rewrite, lines)
lines = page_title_regex.sub(page_title_rewrite, lines)
lines = canonical_regex.sub(canonical_rewrite, lines)
html.write(lines)
# html.write()
for line in shtml.readlines():
# Remove html header/footer created by man2html
if line == "Content-type: text/html\n":
continue
if line == "Content-type: text/html; charset=UTF-8\n":
continue
if line[:6] == "":
continue
if line[:7] == "":
continue
if line[:7] == "