wesnoth/data/tools/wmlscope

#!/usr/bin/env python
#
# wmlscope -- generate reports on WML macro and resource usage
#
# By Eric S. Raymond, April 2007.
#
# This tool cross-references macro definitions with macro calls, and
# resource (sound or image) files with uses of the resources in WML.
# and generates various useful reports from such cross-references.
# It also checks actual macro arguments against types implied by the formals
#
# (Most of the work is done by a cross-referencer class that is also
# used elsewhere, e.g. by wmlmove.)
#
# It takes a list of directories as arguments; if none is given, it
# behaves as though the current directory had been specified as a
# single argument.  Each directory is treated as a separate domain for
# macro and resource visibility purposes.
#
# There are two kinds of namespace, exporting and non-exporting.
# Exporting namespaces make all their resources and macro names
# globally visible.  You can make a namespace exporting by embedding
# a comment like this in it:
#
#     # wmlscope: export=yes
#
# Wesnoth core data is an exporting namespace.  Campaigns are non-exporting;
# they should contain the declaration
#
#     # wmlscope: export=no
#
# somewhere.  wmlscope will complain when it sees a namespace with no export
# property, then treat it as non-exporting.
#
# You can tell wmlscope to ignore stretches of config files
# with the following magic comments:
#
#     # wmlscope: start ignoring
#     # wmlscope: stop ignoring
#
# Similarly, you can tell wmlscope to ignore multiple or duplicate macro
# definitions in a range of lines with the following magic comments:
#
#     # wmlscope: start conditionals
#     # wmlscope: stop conditionals
#
# The following magic comment:
#
#     # prune FOOBAR
#
# will cause wmlscope to forget about all but one of the definitions of FOOBAR
# it has seen.  This will be useful mainly for symbols that have different
# definitions enabled by an #ifdef.
#
#
# This tool does catch one kind of implicit reference: if an attack name
# is specified but no icon is given, the attack icon will default to
# a name generated from the attack name.  This behavior can be suppressed
# by adding a magic comment containing the string "no-icon" to the name= line.
#
# The checking done by this tool has a couple of flaws:
#
# (1) It doesn't actually evaluate file inclusions.  Instead, any
# macro definition satisfies any macro call made under the same
# directory.  Exception: when an #undef is detected, the macro is
# tagged local and not visible outside the span of lines where it's
# defined.
#
# (2) It doesn't read [binary_path] tags, as this would require
# implementing a WML parser.  Instead, it assumes that a resource-file
# reference can be satisfied by any matching image file from anywhere
# in the same directory it came from.  The resources under the *first*
# directory argument (only) are visible everywhere.
#
# (3) A reference with embedded {}s in a macro will have the macro's
# formal args substituted in at WML evaluation time.  Instead, this
# tool treats each {} as a .* wildcard and considers the reference to
# match *every* resource filename that matches that pattern.  Under
# appropriate circumstances this might report a resource filename
# statically matching the pattern as having been referenced even
# though none of the actual macro calls would actually generate it.
#
# Problems (1) and (2) imply that this tool might conceivably report
# that a reference has been satisfied when under actual
# WML-interpreter rules it has not.
#
# The reporting format is compatible with GNU Emacs compile mode.
#
# For debugging purposes, an in-line comment of the form
#
#     # wmlscope: warnlevel NNN
#
# sets the warning level.

import sys, os, time, re, getopt, hashlib
from wesnoth.wmltools import *

def interpret(lines, css):
    "Interpret the ! convention for .cfg comments."
    inlisting = False
    outstr = '<p class="%s">' % css
    for line in lines:
        line = line.strip()
        if inlisting:
            if line and line[0] != '!':
                outstr += "</pre>\n<p>"
                inlisting = False
        else:
            if not line:
                outstr += "</p><p>"
                continue
            if line[0] == '!':
                outstr += "</p>\n<pre class='listing'>"
                inlisting = True
                bracketdepth = curlydepth = 0
        line = line.replace("<", "&lt;").replace(">", "&gt;").replace("&", "&amp;")
        if inlisting:
            outstr += line[1:] + "\n"
        else:
            outstr += line + "\n"
    if not inlisting:
        outstr += "</p>\n"
    else:
        outstr += "</pre>\n"
    outstr = outstr.replace("<p></p>", "")
    outstr = outstr.replace("\n\n</pre>", "\n</pre>")
    return outstr

class CrossRefLister(CrossRef):
    "Cross-reference generator with reporting functions"
    def xrefdump(self, pred=None):
        "Report resolved macro references."
        sorted = self.xref.keys()
        sorted.sort()
        for name in sorted:
            for defn in self.xref[name]:
                if pred and not pred(name, defn):
                    continue
                if defn.undef:
                    type = "local"
                else:
                    type = "global"
                nrefs = len(defn.references)
                if nrefs == 0:
                    print "%s: %s macro %s is unused" % (defn, type, name)
                else:
                    print "%s: %s macro %s is used in %d files:" % (defn, type, name, nrefs)
                defn.dump_references()
        sorted = self.fileref.keys()
        sorted.sort()
        for name in sorted:
            defloc = self.fileref[name]
            if pred and not pred(name, defloc):
                continue
            nrefs = len(defloc.references)
            if nrefs == 0:
                print "Resource %s is unused" % defloc
            else:
                print "Resource %s is used in %d files:" % (defloc, nrefs)
            defloc.dump_references()

    def unresdump(self):
        "Report unresolved references, arity mismatches, duplicate unit IDs."
        # First the unresolved references
        if len(self.unresolved) == 0 and len(self.missing) == 0:
            print "# No unresolved references"
        else:
            #print self.fileref.keys()
            for (name, reference) in self.unresolved + self.missing:
                print "%s: Unresolved reference -> %s" % (reference, name)
        mismatched = []
        sorted = self.xref.keys()
        sorted.sort()
        for name in sorted:
            for defn in self.xref[name]:
                m = defn.mismatches()
                if m.references:
                    mismatched.append((name, m))
        # Then the type mismatches
        if mismatched:
            print "# Mismatched references:"
            for (n, m) in mismatched:
                print "%s: macro %s(%s) has mismatches:" % (m, n, ", ".join(["{}={}".format(x, formaltype(x)) for x in m.args]))
                for (file, refs) in m.references.items():
                    for (ln, args) in refs:
                        print '"%s", line %d: %s(%s) with signature (%s)' % (file, ln, n, ", ".join(args), ", ".join(["{}={}".format(f, actualtype(a)) for f,a in zip(m.args, args)]))

    def undersized(self):
        "Report undersized images that cannot be safely overlaid on a hex."
        try:
            from PIL import Image
            for (namespace, filename) in xref.filelist.generator():
                if filename.endswith(".png"):
                    try:
                        im = Image.open(filename)
                        (x, y) = im.size
                        if x <= 60 or y <= 60:
                            print "%s: %d by %d" % (filename, x, y)
                    except IOError:
                        sys.stderr.write("%s: PIL internal error\n" % filename)
        except ImportError:
            sys.stderr.write("Install Python Imaging Library to enable size check.\n")
    def duplicates(self, exportonly):
        "Dump duplicate unit IDs."
        duplicate_latch = False
        for (key, value) in self.unit_ids.items():
            if len(value) > 1:
                if exportonly and not [x for x in value if self.exports(x.namespace)]:
                    continue
                if not duplicate_latch:
                    print "# Duplicate IDs"
                    duplicate_latch = True
                print "%s: occurs %d times as unit ID" % (key, len(value))
                for ref in value:
                    print "%s: exported=%s" % (ref, self.exports(ref.namespace))

    def typelist(self, branch):
        "Dump actual and formal arguments for macros in specified file"
        already_seen = []
        sorted = self.xref.keys()
        sorted.sort()
        for name in sorted:
            for defn in self.xref[name]:
                for (filename, refs) in defn.references.items():
                    if filename.endswith(branch):
                        if name not in already_seen:
                            already_seen.append(name)
                            print "%s: macro %s(%s):" % (defn, name, ", ".join(["{}={}".format(x, formaltype(x)) for x in defn.args]))
                        for (ln, args) in refs:
                            print '"%s", line %d: %s(%s) with signature (%s)' % (filename, ln, name, ", ".join(args), ", ".join(["{}={}".format(f, actualtype(a)) for f,a in zip(defn.args, args)]))
    def deflist(self, pred=None):
        "List all resource definitions."
        sorted = self.xref.keys()
        sorted.sort()
        for name in sorted:
            for defn in self.xref[name]:
                if not pred or pred(name, defn):
                    print "macro", name, " ".join(["{}={}".format(x, formaltype(x)) for x in defn.args])
        sorted = self.fileref.keys()
        sorted.sort()
        for name in sorted:
            defloc = self.fileref[name]
            if not pred or pred(name, defloc):
                print "resource", name
        sorted = self.unit_ids.keys()
        sorted.sort()
        for uid in sorted:
            print "unit", uid

    def unchecked(self, fp):
        "List all macro definitions with untyped formals."
        unchecked = []
        defcount = 0
        callcount = 0
        unresolvedcount = 0
        for name in self.xref.keys():
            for defn in self.xref[name]:
                defcount += 1
                callcount += len(defn.references)
                if None in map(formaltype, defn.args):
                    for (i, d) in enumerate(defn.args):
                        if formaltype(d) is None:
                            defn.args[i] += "?"
                    unchecked.append((name, defn))
                    unresolvedcount += len(defn.references)
        if unchecked:
            print "# %d of %d (%d%%) macro definitions and %d of %d calls (%d%%) have untyped formals:" \
                  % (len(unchecked),
                     defcount,
                     int((100 * len(unchecked)) / defcount),
                     unresolvedcount,
                     callcount,
                     int((100 * unresolvedcount) / callcount))
            unchecked.sort(lambda a, b: cmp(a[1], b[1]))
            for (name, defn) in unchecked:
                print "%s: %s(%s)" % (defn, name, ", ".join(defn.args))

    def extracthelp(self, pref, fp):
        "Deliver all macro help comments in HTML form."
        # Bug: finds only the first definition of each macro in scope.
        doclist = self.xref.keys()
        doclist = [x for x in doclist if self.xref[x][0].docstring.count("\n") > 1]
        doclist.sort(lambda x, y: cmp(self.xref[x][0], self.xref[y][0]))
        outstr = ""
        filename = None
        filenamelist = []
        counted = 0
        for name in doclist:
            entry = self.xref[name][0]
            if entry.filename != filename:
                if counted:
                    outstr += "</dl>\n"
                counted += 1
                filename = entry.filename
                if filename.startswith(pref):
                    displayname = filename[len(pref):]
                else:
                    displayname = filename
                outstr += "<p class='toplink'>(<a href='#global'>top</a>)</p>\n"
                outstr += "<h1 id='file:" + displayname + "' class='file_header'>From file: " + displayname + "</h1>\n"
                filenamelist.append(displayname)
                hdr = []
                dfp = open(filename)
                for line in dfp:
                    line = line.lstrip()
                    if line and line.startswith("#textdomain"):
                        continue
                    if line and line[0] == '#':
                        hdr.append(line[1:])
                    else:
                        break
                dfp.close()
                if hdr:
                    outstr += interpret(hdr, "file_explanation")
                outstr += "<dl>\n"
            if entry.docstring:
                lines = entry.docstring.split("\n")
                header = lines.pop(0).split()
                if lines and not lines[-1]:	# Ignore trailing blank lines
                    lines.pop()
                if not lines:	# Ignore definitions without a docstring
                    continue
                outstr += "\n<dt id='" + header[0] + "'>\n"
                outstr += "<em class='macro_name'>" + header[0] + "</em>"
                if header[1:]:
                    outstr += " <em class='macro_formals'>"+" ".join(header[1:])+"</em>"
                outstr += "\n</dt>\n"
                outstr += "<dd>\n"
                outstr += interpret(lines, "macro_explanation")
                outstr += "</dd>\n"
        outstr += "</dl>\n"
        outstr += "<p class='toplink'>(<a href='#global'>top</a>)</p>\n"
        linkheaderstr = "<p class='filelist'>Covered macro files:"
        for filename in filenamelist:
            linkheaderstr += " <a href='#file:" + filename + "'>"
            linkheaderstr += filename + "</a>"
        linkheaderstr += "</p>\n"
        fp.write(linkheaderstr)
        fp.write(outstr)

if __name__ == "__main__":
    def help():
        sys.stderr.write("""\
Usage: macroscope [options] dirpath
    Options may be any of these:
    -h, --help                   Emit this help message and quit
    -c, --crossreference         Report resolved macro references (implies -w 1)
    -C, --collisions             Report duplicate resource files
    -d, --deflist                Make definition list
    -e regexp, --exclude regexp  Ignore files matching the specified regular expression
    -f regexp, --from regexp     Report only on things defined in files matching regexp
    -l, --listfiles              List files that will be processed
    -r ddd, --refcount=ddd       Report only on macros w/references in ddd files
    -t fname, --typecheck fname  List actual & formal argtypes for calls in fname
    -u, --unresolved             Report unresolved macro references
    -w, --warnlevel              Set to 1 to warn of duplicate macro definitions
    --forced-used regexp         Ignore refcount 0 on names matching regexp
    --extracthelp                Extract help from macro definition comments.
    --unchecked                  Report all macros with untyped formals.
   Options may be followed by any number of directiories to check.  If no
   directories are given, all files under the current directory are checked.
""")

    try:
        # Process options
        (options, arguments) = getopt.getopt(sys.argv[1:], "cCdhe:f:lpr:t:uw:",
                                             [
                                              'crossreference',
                                              'collisions',
                                              'definitions',
                                              'exclude=',
                                              'extracthelp',
                                              'force-used=',
                                              'from=',
                                              'help',
                                              'listfiles',
                                              'progress',
                                              'refcount=',
                                              'typelist=',
                                              'unchecked',
                                              'unresolved',
                                              'warnlevel=',
                                              ])
        crossreference = definitions = listfiles = unresolved = extracthelp = False
        from_restrict = None
        refcount_restrict = None
        forceused = []
        exclude = []
        warnlevel = 0
        collisions = False
        typelist = None
        unchecked = False
        progress = False
        for (switch, val) in options:
            if switch in ('-h', '--help'):
                help()
                sys.exit(0)
            if switch in ('-f', '--from'):
                from_restrict = val
            elif switch in ('-c', '--crossreference'):
                crossreference = True
                warnlevel = 1
            elif switch in ('-C', '--collisions'):
                collisions = True
            elif switch in ('-d', '--definitions'):
                definitions = True
            elif switch in ('-e', '--exclude'):
                exclude.append(val)
            elif switch == '--extracthelp':
                extracthelp = True
            elif switch == '--force-used':
                forceused.append(val)
            elif switch in ('-l', '--listfiles'):
                listfiles = True
            elif switch in ('-d', '--progress'):
                progress = True
            elif switch in ('-r', '--refcount'):
                refcount_restrict = int(val)
            elif switch == '--unchecked':
                unchecked = True
            elif switch in ('-t', '--typelist'):
                typelist = val
            elif switch in ('-u', '--unresolved'):
                unresolved = True
            elif switch in ('-w', '--warnlevel'):
                warnlevel = int(val)

        # in certain situations, Windows' command prompt appends a double quote
        # to the command line parameters. This block takes care of this issue.
        for i,arg in enumerate(arguments):
            if arg.endswith('"'):
                arguments[i] = arg[:-1]

        forceused = "|".join(forceused)
        if len(arguments):
            dirpath = arguments
        else:
            dirpath = ['.']
        if not extracthelp:
            print "# Wmlscope reporting on %s" % time.ctime()
            print "# Invocation: %s" % " ".join(sys.argv)
            print "# Working directory: %s" % os.getcwd()
            starttime = time.time()
        xref = CrossRefLister(dirpath, "|".join(exclude), warnlevel, progress)
        if not extracthelp:
            print "#Cross-reference time: %d seconds" % (time.time()-starttime)
        if extracthelp:
            xref.extracthelp(dirpath[0], sys.stdout)
        elif unchecked:
            xref.unchecked(sys.stdout)
        elif listfiles:
            for (namespace, filename) in xref.filelist.generator():
                print filename
        if collisions:
            collisions = []
            for (namespace, filename) in xref.filelist.generator():
                ifp = open(filename)
                collisions.append(hashlib.md5(ifp.read()).digest())
                ifp.close()
            collisions = zip(xref.filelist.flatten(), collisions)
            hashcounts = {}
            for (n, h) in collisions:
                hashcounts[h] = hashcounts.get(h, 0) + 1
            collisions = [(n,h) for (n,h) in collisions if hashcounts[h] > 1]
            collisions.sort(lambda (n1, h1), (n2, h2): cmp(h1, h2))
            lasthash = None
            for (n, h) in collisions:
                if h != lasthash:
                    print "%%"
                    lasthash = h
                print n
            xref.duplicates(exportonly=False)
        elif typelist:
            xref.typelist(typelist)
        elif crossreference or definitions or listfiles or unresolved:
            def predicate(name, defloc):
                if from_restrict and not re.search(from_restrict, defloc.filename):
                    return False
                if refcount_restrict!=None \
                       and len(defloc.references) != refcount_restrict \
                       or (refcount_restrict == 0 and forceused and re.search(forceused, name)):
                    return False
                return True
            if crossreference:
                if xref.noxref:
                    print >>sys.stderr, "macroscope: can't make cross-reference, input included a definitions file."
                else:
                    xref.xrefdump(predicate)
            if definitions:
                xref.deflist(predicate)
            if unresolved:
                #xref.undersized()
                xref.unresdump()
                xref.duplicates(exportonly=True)
    except KeyboardInterrupt:
        print >>sys.stderr, "wmlscope: aborted."

# wmlscope ends here