diff options
author | Barry Warsaw <barry@python.org> | 2002-06-28 23:41:42 (GMT) |
---|---|---|
committer | Barry Warsaw <barry@python.org> | 2002-06-28 23:41:42 (GMT) |
commit | 062749ac577be62c2f14fe61c2167c9c5b909ea3 (patch) | |
tree | 9c97c5c0052cdf237b51b7caa2f16fabb771d489 | |
parent | 7c75c99a10691998b7e1506ede4f98eec28bd226 (diff) | |
download | cpython-062749ac577be62c2f14fe61c2167c9c5b909ea3.zip cpython-062749ac577be62c2f14fe61c2167c9c5b909ea3.tar.gz cpython-062749ac577be62c2f14fe61c2167c9c5b909ea3.tar.bz2 |
_split_header(): The code here was terminally broken because it didn't
know anything about RFC 2047 encoded headers. Fortunately we have a
perfectly good header splitter in Header.encode(). So we just call
that to give us a properly formatted and split header.
Header.encode() didn't know about "highest-level syntactic breaks" but
that's been fixed now too.
-rw-r--r-- | Lib/email/Generator.py | 64 |
1 files changed, 11 insertions, 53 deletions
diff --git a/Lib/email/Generator.py b/Lib/email/Generator.py index 64db084..0f78861 100644 --- a/Lib/email/Generator.py +++ b/Lib/email/Generator.py @@ -11,6 +11,8 @@ import random from types import ListType, StringType from cStringIO import StringIO +from email.Header import Header + EMPTYSTRING = '' SEMISPACE = '; ' BAR = '|' @@ -149,17 +151,17 @@ class Generator: # headers. text = '%s: %s' % (h, v) if self.__maxheaderlen > 0 and len(text) > self.__maxheaderlen: - text = self._split_header(text) + text = self._split_header(h, text) print >> self._fp, text # A blank line always separates headers from body print >> self._fp - def _split_header(self, text): + def _split_header(self, name, text): maxheaderlen = self.__maxheaderlen # Find out whether any lines in the header are really longer than # maxheaderlen characters wide. There could be continuation lines # that actually shorten it. Also, replace hard tabs with 8 spaces. - lines = [s.replace('\t', SPACE8) for s in text.split('\n')] + lines = [s.replace('\t', SPACE8) for s in text.splitlines()] for line in lines: if len(line) > maxheaderlen: break @@ -167,56 +169,12 @@ class Generator: # No line was actually longer than maxheaderlen characters, so # just return the original unchanged. return text - rtn = [] - for line in text.split('\n'): - splitline = [] - # Short lines can remain unchanged - if len(line.replace('\t', SPACE8)) <= maxheaderlen: - splitline.append(line) - rtn.append(SEMINLTAB.join(splitline)) - else: - oldlen = len(line) - # Try to break the line on semicolons, but if that doesn't - # work, try to split on folding whitespace. - while len(line) > maxheaderlen: - i = line.rfind(';', 0, maxheaderlen) - if i < 0: - break - splitline.append(line[:i]) - line = line[i+1:].lstrip() - if len(line) <> oldlen: - # Splitting on semis worked - splitline.append(line) - rtn.append(SEMINLTAB.join(splitline)) - continue - # Splitting on semis didn't help, so try to split on - # whitespace. - parts = re.split(r'(\s+)', line) - # Watch out though for "Header: longnonsplittableline" - if parts[0].endswith(':') and len(parts) == 3: - rtn.append(line) - continue - first = parts.pop(0) - sublines = [first] - acc = len(first) - while parts: - len0 = len(parts[0]) - len1 = len(parts[1]) - if acc + len0 + len1 < maxheaderlen: - sublines.append(parts.pop(0)) - sublines.append(parts.pop(0)) - acc += len0 + len1 - else: - # Split it here, but don't forget to ignore the - # next whitespace-only part - splitline.append(EMPTYSTRING.join(sublines)) - del parts[0] - first = parts.pop(0) - sublines = [first] - acc = len(first) - splitline.append(EMPTYSTRING.join(sublines)) - rtn.append(NLTAB.join(splitline)) - return NL.join(rtn) + # The `text' argument already has the field name prepended, so don't + # provide it here or the first line will get folded too short. + h = Header(text, maxlinelen=maxheaderlen, + # For backwards compatibility, we use a hard tab here + continuation_ws='\t') + return h.encode() # # Handlers for writing types and subtypes |