Add support for idendification of function definitions since it's needed

in some cases in order to prevent some odd function contents from being
mistakenly identified as invalid variable assignments. For example, this
line from _gcc-specs-directive_raw() is commonly found in environment.bz2
files:

$1=="*"directive":"  { pspec=spec; spec=""; outside=0; next }

svn path=/main/trunk/; revision=9431

diff --git a/bin/filter-bash-environment.py b/bin/filter-bash-environment.py
index 691d406da7f06e8f32e4e948baaafd0e63ffcfa2..0c1e67911ea0e10cdb3eb0f2bc37c5aad03ab8c7 100755 (executable)
--- a/bin/filter-bash-environment.py
+++ b/bin/filter-bash-environment.py
@@ -12,6 +12,8 @@ egrep_compat_map = {
 }
 
 here_doc_re = re.compile(r'.*\s<<[-]?(\w+)$')
+func_start_re = re.compile(r'^\s*[-\w]*\s*\(\)\s*$')
+func_end_re = re.compile(r'^\}$')
 
 def compile_egrep_pattern(s):
        for k, v in egrep_compat_map.iteritems():
@@ -20,6 +22,7 @@ def compile_egrep_pattern(s):
 
 def filter_bash_environment(pattern, file_in, file_out):
        here_doc_delim = None
+       in_func = None
        for line in file_in:
                if here_doc_delim is not None:
                        if here_doc_delim.match(line):
@@ -31,13 +34,26 @@ def filter_bash_environment(pattern, file_in, file_out):
                        here_doc_delim = re.compile("^%s$" % here_doc.group(1))
                        file_out.write(line)
                        continue
+               # Note: here-documents are handled before fuctions since otherwise
+               # it would be possible for the content of a here-document to be
+               # mistaken as the end of a function.
+               if in_func:
+                       if func_end_re.match(line) is not None:
+                               in_func = None
+                       file_out.write(line)
+                       continue
+               in_func = func_start_re.match(line)
+               if in_func is not None:
+                       file_out.write(line)
+                       continue
                if pattern.match(line) is None:
                        file_out.write(line)
 
 if __name__ == "__main__":
        description = "Filter out any lines that match a given PATTERN " + \
-               "while leaving bash here-documents intact. The PATTERN should " + \
-               "use python regular expression syntax but [:space:] and " + \
+               "while leaving bash function definitions and here-documents " + \
+               "intact. The PATTERN should use python regular expression syntax" + \
+               " but [:digit:], [:space:] and " + \
                "[:alnum:] character classes will be automatically translated " + \
                "for compatibility with egrep syntax."
        usage = "usage: %s PATTERN" % os.path.basename(sys.argv[0])