kallithea Changeset - 4e565c5d7b7d

Changeset - 4e565c5d7b7d

Parent rev.

Child rev.

[Not reviewed]

default

0 2 0

Mads Kiilerich - 6 years ago 2019-12-15 20:00:38
mads@kiilerich.com

Grafted from: 78a342d95b59

lib: establish py3 compatible strategy for string handling: introducing safe_bytes and deprecating safe_str

The meaning of safe_str will change when moving to py3. All use of safe_str is
thus tech debt that we have to chop off, mostly by moving to either
safe_unicode or safe_bytes ... or dropping because we know what we are doing
and rely on the improved type safety in py3.

2 files changed with 9 insertions and 6 deletions:

kallithea/lib/utils2.py

kallithea/lib/vcs/utils/__init__.py

0 comments (0 inline, 0 general)

kallithea/lib/utils2.py

➞

Show inline comments

@@ @@ -34,25 +34,25 @@ import datetime @@
 import os
 import pwd
 import re
 import time
 import urllib
 import urlobject
 from tg.i18n import ugettext as _
 from tg.i18n import ungettext
 from webhelpers2.text import collapse, remove_formatting, strip_tags
 from kallithea.lib.compat import json
 from kallithea.lib.vcs.utils import safe_str, safe_unicode  # re-export
+from kallithea.lib.vcs.utils import safe_bytes, safe_str, safe_unicode  # re-export
 from kallithea.lib.vcs.utils.lazy import LazyProperty
 def str2bool(_str):
     """
     returns True/False value from given string, it tries to translate the
     string into boolean
     :param _str: string value to translate into boolean
     :rtype: boolean
     :returns: boolean from given string
     """

kallithea/lib/vcs/utils/__init__.py

➞

Show inline comments

@@ @@ -67,66 +67,69 @@ def safe_int(val, default=None): @@
 def safe_unicode(s):
     """
     Safe unicode function. Use a few tricks to turn s into unicode string:
     In case of UnicodeDecodeError with configured default encodings, try to
     detect encoding with chardet library, then fall back to first encoding with
     errors replaced.
     """
     if isinstance(s, unicode):
         return s
-    if not isinstance(s, str):  # use __str__ / __unicode__ and don't expect UnicodeDecodeError
+    if not isinstance(s, bytes):  # use __str__ / __unicode__ and don't expect UnicodeDecodeError
         return unicode(s)
     from kallithea.lib.vcs.conf import settings
     for enc in settings.DEFAULT_ENCODINGS:
         try:
             return unicode(s, enc)
         except UnicodeDecodeError:
             pass
     try:
         import chardet
         encoding = chardet.detect(s)['encoding']
         if encoding is not None:
             return s.decode(encoding)
     except (ImportError, UnicodeDecodeError):
         pass
     return unicode(s, settings.DEFAULT_ENCODINGS[0], 'replace')
-def safe_str(s):
+def safe_bytes(s):
     """
-    Safe str function. Use a few tricks to turn s into bytes string:
+    Safe bytes function. Use a few tricks to turn s into bytes string:
     In case of UnicodeEncodeError with configured default encodings, fall back
     to first configured encoding with errors replaced.
     """
-    if isinstance(s, str):
+    if isinstance(s, bytes):
         return s
-    assert isinstance(s, unicode), s  # don't use safe_str to coerce non-strings
+    assert isinstance(s, unicode), repr(s)  # bytes cannot coerse with __str__ or handle None or int
     from kallithea.lib.vcs.conf import settings
     for enc in settings.DEFAULT_ENCODINGS:
         try:
             return s.encode(enc)
         except UnicodeEncodeError:
             pass
     return s.encode(settings.DEFAULT_ENCODINGS[0], 'replace')
 safe_str = safe_bytes  # safe_str is deprecated - it will be redefined when changing to py3
 # Regex taken from http://www.regular-expressions.info/email.html
 email_re = re.compile(
     r"""[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@"""
     r"""(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?""",
     re.IGNORECASE)
 def author_email(author):
     """
     Returns email address of given author string.
     If author contains <> brackets, only look inside that.
     If any RFC valid email address is found, return that.

0 comments (0 inline, 0 general)