mirror of
https://github.com/nodejs/node.git
synced 2024-12-01 16:10:02 +01:00
a30839576c
Make "--with-intl=none" the default and add "intl-none" option to vcbuild.bat. If icu data is missing print a warning unless either --download=all or --download=icu is set. If set then automatically download, verify (MD5) and unpack the ICU data if not already available. There's a "list" of URLs being used, but right now only the first is picked up. The logic works something like this: * If there is no directory deps/icu, * If no zip file (currently icu4c-54_1-src.zip), * Download zip file (icu-project.org -> sf.net) * Verify the MD5 sum of the zipfile * If bad, print error and exit * Unpack the zipfile into deps/icu * If deps/icu now exists, use it, else fail with help text Add the configuration option "--with-icu-source=..." Usage: * --with-icu-source=/path/to/my/other/icu * --with-icu-source=/path/to/icu54.zip * --with-icu-source=/path/to/icu54.tgz * --with-icu-source=http://example.com/icu54.tar.bz2 Add the configuration option "--with-icu-locals=...". Allows choosing which locales are used in the "small-icu" case. Example: configure --with-intl=small-icu --with-icu-locales=tlh,grc,nl (Also note that as of this writing, neither Klingon nor Ancient Greek are in upstream CLDR data. Serving suggestion only.) Don't use hard coded ../../out paths on windows. This was suggested by @misterdjules as it causes test failures. With this fix, "out" is no longer created on windows and the following can run properly: python tools/test.py simple Reduce space by about 1MB with ICU 54 (over without this patch). Also trims a few other source files, but only conditional on the exact ICU version used. This is to future-proof - a file that is unneeded now may be needed in future ICUs. Also: * Update distclean to remove icu related files * Refactor some code into tools/configure.d/nodedownload.py * Update docs * Add test PR-URL: https://github.com/joyent/node/pull/8719 Fixes: https://github.com/joyent/node/issues/7676#issuecomment-64704230 [trev.norris@gmail.com small change to test's whitespace and logic] Signed-off-by: Trevor Norris <trev.norris@gmail.com>
357 lines
12 KiB
Python
Executable File
357 lines
12 KiB
Python
Executable File
#!/usr/bin/python
|
|
#
|
|
# Copyright (C) 2014 IBM Corporation and Others. All Rights Reserved.
|
|
#
|
|
# @author Steven R. Loomis <srl@icu-project.org>
|
|
#
|
|
# This tool slims down an ICU data (.dat) file according to a config file.
|
|
#
|
|
# See: http://bugs.icu-project.org/trac/ticket/10922
|
|
#
|
|
# Usage:
|
|
# Use "-h" to get help options.
|
|
|
|
import sys
|
|
import shutil
|
|
# for utf-8
|
|
reload(sys)
|
|
sys.setdefaultencoding("utf-8")
|
|
|
|
import optparse
|
|
import os
|
|
import json
|
|
import re
|
|
|
|
endian=sys.byteorder
|
|
|
|
parser = optparse.OptionParser(usage="usage: mkdir tmp ; %prog -D ~/Downloads/icudt53l.dat -T tmp -F trim_en.json -O icudt53l.dat" )
|
|
|
|
parser.add_option("-P","--tool-path",
|
|
action="store",
|
|
dest="toolpath",
|
|
help="set the prefix directory for ICU tools")
|
|
|
|
parser.add_option("-D","--input-file",
|
|
action="store",
|
|
dest="datfile",
|
|
help="input data file (icudt__.dat)",
|
|
) # required
|
|
|
|
parser.add_option("-F","--filter-file",
|
|
action="store",
|
|
dest="filterfile",
|
|
help="filter file (JSON format)",
|
|
) # required
|
|
|
|
parser.add_option("-T","--tmp-dir",
|
|
action="store",
|
|
dest="tmpdir",
|
|
help="working directory.",
|
|
) # required
|
|
|
|
parser.add_option("--delete-tmp",
|
|
action="count",
|
|
dest="deltmpdir",
|
|
help="delete working directory.",
|
|
default=0)
|
|
|
|
parser.add_option("-O","--outfile",
|
|
action="store",
|
|
dest="outfile",
|
|
help="outfile (NOT a full path)",
|
|
) # required
|
|
|
|
parser.add_option("-v","--verbose",
|
|
action="count",
|
|
default=0)
|
|
|
|
parser.add_option('-L',"--locales",
|
|
action="store",
|
|
dest="locales",
|
|
help="sets the 'locales.only' variable",
|
|
default=None)
|
|
|
|
parser.add_option('-e', '--endian', action='store', dest='endian', help='endian, big, little or host, your default is "%s".' % endian, default=endian, metavar='endianness')
|
|
|
|
(options, args) = parser.parse_args()
|
|
|
|
optVars = vars(options)
|
|
|
|
for opt in [ "datfile", "filterfile", "tmpdir", "outfile" ]:
|
|
if optVars[opt] is None:
|
|
print "Missing required option: %s" % opt
|
|
sys.exit(1)
|
|
|
|
if options.verbose>0:
|
|
print "Options: "+str(options)
|
|
|
|
if (os.path.isdir(options.tmpdir) and options.deltmpdir):
|
|
if options.verbose>1:
|
|
print "Deleting tmp dir %s.." % (options.tmpdir)
|
|
shutil.rmtree(options.tmpdir)
|
|
|
|
if not (os.path.isdir(options.tmpdir)):
|
|
os.mkdir(options.tmpdir)
|
|
else:
|
|
print "Please delete tmpdir %s before beginning." % options.tmpdir
|
|
sys.exit(1)
|
|
|
|
if options.endian not in ("big","little","host"):
|
|
print "Unknown endianness: %s" % options.endian
|
|
sys.exit(1)
|
|
|
|
if options.endian is "host":
|
|
options.endian = endian
|
|
|
|
if not os.path.isdir(options.tmpdir):
|
|
print "Error, tmpdir not a directory: %s" % (options.tmpdir)
|
|
sys.exit(1)
|
|
|
|
if not os.path.isfile(options.filterfile):
|
|
print "Filterfile doesn't exist: %s" % (options.filterfile)
|
|
sys.exit(1)
|
|
|
|
if not os.path.isfile(options.datfile):
|
|
print "Datfile doesn't exist: %s" % (options.datfile)
|
|
sys.exit(1)
|
|
|
|
if not options.datfile.endswith(".dat"):
|
|
print "Datfile doesn't end with .dat: %s" % (options.datfile)
|
|
sys.exit(1)
|
|
|
|
outfile = os.path.join(options.tmpdir, options.outfile)
|
|
|
|
if os.path.isfile(outfile):
|
|
print "Error, output file does exist: %s" % (outfile)
|
|
sys.exit(1)
|
|
|
|
if not options.outfile.endswith(".dat"):
|
|
print "Outfile doesn't end with .dat: %s" % (options.outfile)
|
|
sys.exit(1)
|
|
|
|
dataname=options.outfile[0:-4]
|
|
|
|
|
|
## TODO: need to improve this. Quotes, etc.
|
|
def runcmd(tool, cmd, doContinue=False):
|
|
if(options.toolpath):
|
|
cmd = os.path.join(options.toolpath, tool) + " " + cmd
|
|
else:
|
|
cmd = tool + " " + cmd
|
|
|
|
if(options.verbose>4):
|
|
print "# " + cmd
|
|
|
|
rc = os.system(cmd)
|
|
if rc is not 0 and not doContinue:
|
|
print "FAILED: %s" % cmd
|
|
sys.exit(1)
|
|
return rc
|
|
|
|
## STEP 0 - read in json config
|
|
fi= open(options.filterfile, "rb")
|
|
config=json.load(fi)
|
|
fi.close()
|
|
|
|
if (options.locales):
|
|
if not config.has_key("variables"):
|
|
config["variables"] = {}
|
|
if not config["variables"].has_key("locales"):
|
|
config["variables"]["locales"] = {}
|
|
config["variables"]["locales"]["only"] = options.locales.split(',')
|
|
|
|
if (options.verbose > 6):
|
|
print config
|
|
|
|
if(config.has_key("comment")):
|
|
print "%s: %s" % (options.filterfile, config["comment"])
|
|
|
|
## STEP 1 - copy the data file, swapping endianness
|
|
## The first letter of endian_letter will be 'b' or 'l' for big or little
|
|
endian_letter = options.endian[0]
|
|
|
|
runcmd("icupkg", "-t%s %s %s""" % (endian_letter, options.datfile, outfile))
|
|
|
|
## STEP 2 - get listing
|
|
listfile = os.path.join(options.tmpdir,"icudata.lst")
|
|
runcmd("icupkg", "-l %s > %s""" % (outfile, listfile))
|
|
|
|
fi = open(listfile, 'rb')
|
|
items = fi.readlines()
|
|
items = [items[i].strip() for i in range(len(items))]
|
|
fi.close()
|
|
|
|
itemset = set(items)
|
|
|
|
if (options.verbose>1):
|
|
print "input file: %d items" % (len(items))
|
|
|
|
# list of all trees
|
|
trees = {}
|
|
RES_INDX = "res_index.res"
|
|
remove = None
|
|
# remove - always remove these
|
|
if config.has_key("remove"):
|
|
remove = set(config["remove"])
|
|
else:
|
|
remove = set()
|
|
|
|
# keep - always keep these
|
|
if config.has_key("keep"):
|
|
keep = set(config["keep"])
|
|
else:
|
|
keep = set()
|
|
|
|
def queueForRemoval(tree):
|
|
global remove
|
|
if not config.has_key("trees"):
|
|
# no config
|
|
return
|
|
if not config["trees"].has_key(tree):
|
|
return
|
|
mytree = trees[tree]
|
|
if(options.verbose>0):
|
|
print "* %s: %d items" % (tree, len(mytree["locs"]))
|
|
# do varible substitution for this tree here
|
|
if type(config["trees"][tree]) == str or type(config["trees"][tree]) == unicode:
|
|
treeStr = config["trees"][tree]
|
|
if(options.verbose>5):
|
|
print " Substituting $%s for tree %s" % (treeStr, tree)
|
|
if(not config.has_key("variables") or not config["variables"].has_key(treeStr)):
|
|
print " ERROR: no variable: variables.%s for tree %s" % (treeStr, tree)
|
|
sys.exit(1)
|
|
config["trees"][tree] = config["variables"][treeStr]
|
|
myconfig = config["trees"][tree]
|
|
if(options.verbose>4):
|
|
print " Config: %s" % (myconfig)
|
|
# Process this tree
|
|
if(len(myconfig)==0 or len(mytree["locs"])==0):
|
|
if(options.verbose>2):
|
|
print " No processing for %s - skipping" % (tree)
|
|
else:
|
|
only = None
|
|
if myconfig.has_key("only"):
|
|
only = set(myconfig["only"])
|
|
if (len(only)==0) and (mytree["treeprefix"] != ""):
|
|
thePool = "%spool.res" % (mytree["treeprefix"])
|
|
if (thePool in itemset):
|
|
if(options.verbose>0):
|
|
print "Removing %s because tree %s is empty." % (thePool, tree)
|
|
remove.add(thePool)
|
|
else:
|
|
print "tree %s - no ONLY"
|
|
for l in range(len(mytree["locs"])):
|
|
loc = mytree["locs"][l]
|
|
if (only is not None) and not loc in only:
|
|
# REMOVE loc
|
|
toRemove = "%s%s%s" % (mytree["treeprefix"], loc, mytree["extension"])
|
|
if(options.verbose>6):
|
|
print "Queueing for removal: %s" % toRemove
|
|
remove.add(toRemove)
|
|
|
|
def addTreeByType(tree, mytree):
|
|
if(options.verbose>1):
|
|
print "(considering %s): %s" % (tree, mytree)
|
|
trees[tree] = mytree
|
|
mytree["locs"]=[]
|
|
for i in range(len(items)):
|
|
item = items[i]
|
|
if item.startswith(mytree["treeprefix"]) and item.endswith(mytree["extension"]):
|
|
mytree["locs"].append(item[len(mytree["treeprefix"]):-4])
|
|
# now, process
|
|
queueForRemoval(tree)
|
|
|
|
addTreeByType("converters",{"treeprefix":"", "extension":".cnv"})
|
|
addTreeByType("stringprep",{"treeprefix":"", "extension":".spp"})
|
|
addTreeByType("translit",{"treeprefix":"translit/", "extension":".res"})
|
|
addTreeByType("brkfiles",{"treeprefix":"brkitr/", "extension":".brk"})
|
|
addTreeByType("brkdict",{"treeprefix":"brkitr/", "extension":"dict"})
|
|
addTreeByType("confusables",{"treeprefix":"", "extension":".cfu"})
|
|
|
|
for i in range(len(items)):
|
|
item = items[i]
|
|
if item.endswith(RES_INDX):
|
|
treeprefix = item[0:item.rindex(RES_INDX)]
|
|
tree = None
|
|
if treeprefix == "":
|
|
tree = "ROOT"
|
|
else:
|
|
tree = treeprefix[0:-1]
|
|
if(options.verbose>6):
|
|
print "procesing %s" % (tree)
|
|
trees[tree] = { "extension": ".res", "treeprefix": treeprefix, "hasIndex": True }
|
|
# read in the resource list for the tree
|
|
treelistfile = os.path.join(options.tmpdir,"%s.lst" % tree)
|
|
runcmd("iculslocs", "-i %s -N %s -T %s -l > %s" % (outfile, dataname, tree, treelistfile))
|
|
fi = open(treelistfile, 'rb')
|
|
treeitems = fi.readlines()
|
|
trees[tree]["locs"] = [treeitems[i].strip() for i in range(len(treeitems))]
|
|
fi.close()
|
|
if(not config.has_key("trees") or not config["trees"].has_key(tree)):
|
|
print " Warning: filter file %s does not mention trees.%s - will be kept as-is" % (options.filterfile, tree)
|
|
else:
|
|
queueForRemoval(tree)
|
|
|
|
def removeList(count=0):
|
|
# don't allow "keep" items to creep in here.
|
|
global remove
|
|
remove = remove - keep
|
|
if(count > 10):
|
|
print "Giving up - %dth attempt at removal." % count
|
|
sys.exit(1)
|
|
if(options.verbose>1):
|
|
print "%d items to remove - try #%d" % (len(remove),count)
|
|
if(len(remove)>0):
|
|
oldcount = len(remove)
|
|
hackerrfile=os.path.join(options.tmpdir, "REMOVE.err")
|
|
removefile = os.path.join(options.tmpdir, "REMOVE.lst")
|
|
fi = open(removefile, 'wb')
|
|
for i in remove:
|
|
print >>fi, i
|
|
fi.close()
|
|
rc = runcmd("icupkg","-r %s %s 2> %s" % (removefile,outfile,hackerrfile),True)
|
|
if rc is not 0:
|
|
if(options.verbose>5):
|
|
print "## Damage control, trying to parse stderr from icupkg.."
|
|
fi = open(hackerrfile, 'rb')
|
|
erritems = fi.readlines()
|
|
fi.close()
|
|
#Item zone/zh_Hant_TW.res depends on missing item zone/zh_Hant.res
|
|
pat = re.compile("""^Item ([^ ]+) depends on missing item ([^ ]+).*""")
|
|
for i in range(len(erritems)):
|
|
line = erritems[i].strip()
|
|
m = pat.match(line)
|
|
if m:
|
|
toDelete = m.group(1)
|
|
if(options.verbose > 5):
|
|
print "<< %s added to delete" % toDelete
|
|
remove.add(toDelete)
|
|
else:
|
|
print "ERROR: could not match errline: %s" % line
|
|
sys.exit(1)
|
|
if(options.verbose > 5):
|
|
print " now %d items to remove" % len(remove)
|
|
if(oldcount == len(remove)):
|
|
print " ERROR: could not add any mor eitems to remove. Fail."
|
|
sys.exit(1)
|
|
removeList(count+1)
|
|
|
|
# fire it up
|
|
removeList(1)
|
|
|
|
# now, fixup res_index, one at a time
|
|
for tree in trees:
|
|
# skip trees that don't have res_index
|
|
if not trees[tree].has_key("hasIndex"):
|
|
continue
|
|
treebunddir = options.tmpdir
|
|
if(trees[tree]["treeprefix"]):
|
|
treebunddir = os.path.join(treebunddir, trees[tree]["treeprefix"])
|
|
if not (os.path.isdir(treebunddir)):
|
|
os.mkdir(treebunddir)
|
|
treebundres = os.path.join(treebunddir,RES_INDX)
|
|
treebundtxt = "%s.txt" % (treebundres[0:-4])
|
|
runcmd("iculslocs", "-i %s -N %s -T %s -b %s" % (outfile, dataname, tree, treebundtxt))
|
|
runcmd("genrb","-d %s -s %s res_index.txt" % (treebunddir, treebunddir))
|
|
runcmd("icupkg","-s %s -a %s%s %s" % (options.tmpdir, trees[tree]["treeprefix"], RES_INDX, outfile))
|