Generate from Unicode database instead of manualy coding.

[andersch]

BPO	1498930
Nosy	@malemburg, @loewis, @florentx
Superseder	bpo-1571184: Generate numeric/space/linebreak from Unicode database.
Files	Unicodedata.patch: Generate functions instead of manualy update them. isspace.patch: Update unicode.isspace().

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = 'https://github.com/loewis'
closed_at = <Date 2010-04-01.11:58:25.421>
created_at = <Date 2006-06-01.15:21:35.000>
labels = ['interpreter-core', 'type-feature']
title = 'Generate from Unicode database instead of manualy coding.'
updated_at = <Date 2010-04-01.11:58:25.419>
user = 'https://bugs.python.org/andersch'

bugs.python.org fields:

activity = <Date 2010-04-01.11:58:25.419>
actor = 'flox'
assignee = 'loewis'
closed = True
closed_date = <Date 2010-04-01.11:58:25.421>
closer = 'flox'
components = ['Interpreter Core']
creation = <Date 2006-06-01.15:21:35.000>
creator = 'andersch'
dependencies = []
files = ['7303', '7304']
hgrepos = []
issue_num = 1498930
keywords = ['patch']
message_count = 4.0
messages = ['50405', '50406', '83925', '102086']
nosy_count = 4.0
nosy_names = ['lemburg', 'loewis', 'andersch', 'flox']
pr_nums = []
priority = 'normal'
resolution = 'duplicate'
stage = 'resolved'
status = 'closed'
superseder = '1571184'
type = 'enhancement'
url = 'https://bugs.python.org/issue1498930'
versions = ['Python 3.1', 'Python 2.7']

[andersch]

This patch makes _PyUnicode_IsLinebreak,
_PyUnicode_ToNumeric and _PyUnicode_IsWhitespace
generated from the Unicode database instead of being
subject to manual updates. This ought to ease the
transition to Unicode version 5.0.0.

Allso, the definitions of whitespaces has changed
between v3.2 and v4.1 so the test had to be updated.

[andersch]

Logged In: YES
user_id=621306

With the beta1 out the door I thought We might rather want a
minimal fix for _PyUnicode_IsWhitespace. So here one is.

[malemburg]

You may not know it, but these functions are generated from the Unicode
database.

However, because these functions need to be fast and are small enough,
they were not converted to the unicodetype_db approach and instead left
as they were originally implemented: as switch statements for the
compilers to optimize.

Is there any reason why this would need to change for Unicode 5.0 ?

[florentx]

Fixed with bpo-1571184.