Mercurial > moin > 1.9
changeset 4146:3b36f0cfc656
(Re)moved normalizePagename from Request into wikiutil library
| author | Florian Krupicka <florian.krupicka@googlemail.com> |
|---|---|
| date | Sat, 07 Jun 2008 19:55:27 +0200 |
| parents | 5bd2da7053a3 |
| children | 95e9e24da409 |
| files | MoinMoin/_tests/test_wikiutil.py MoinMoin/action/CopyPage.py MoinMoin/action/PackagePages.py MoinMoin/action/RenamePage.py MoinMoin/formatter/__init__.py MoinMoin/mail/mailimport.py MoinMoin/request/__init__.py MoinMoin/request/_tests/test_request.py MoinMoin/theme/__init__.py MoinMoin/user.py MoinMoin/wikiutil.py |
| diffstat | 11 files changed, 133 insertions(+), 134 deletions(-) [+] |
line wrap: on
line diff
--- a/MoinMoin/_tests/test_wikiutil.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/_tests/test_wikiutil.py Sat Jun 07 19:55:27 2008 +0200 @@ -954,4 +954,86 @@ assert relative_page == wikiutil.RelPageName(current_page, absolute_page) +class TestNormalizePagename(object): + + def testPageInvalidChars(self): + """ request: normalize pagename: remove invalid unicode chars + + Assume the default setting + """ + test = u'\u0000\u202a\u202b\u202c\u202d\u202e' + expected = u'' + result = wikiutil.normalize_pagename(test) + assert result == expected + + def testNormalizeSlashes(self): + """ request: normalize pagename: normalize slashes """ + cases = ( + (u'/////', u''), + (u'/a', u'a'), + (u'a/', u'a'), + (u'a/////b/////c', u'a/b/c'), + (u'a b/////c d/////e f', u'a b/c d/e f'), + ) + for test, expected in cases: + result = wikiutil.normalize_pagename(test) + assert result == expected + + def testNormalizeWhitespace(self): + """ request: normalize pagename: normalize whitespace """ + cases = ( + (u' ', u''), + (u' a', u'a'), + (u'a ', u'a'), + (u'a b c', u'a b c'), + (u'a b / c d / e f', u'a b/c d/e f'), + # All 30 unicode spaces + (config.chars_spaces, u''), + ) + for test, expected in cases: + result = wikiutil.normalize_pagename(test) + assert result == expected + + def testUnderscoreTestCase(self): + """ request: normalize pagename: underscore convert to spaces and normalized + + Underscores should convert to spaces, then spaces should be + normalized, order is important! + """ + cases = ( + (u' ', u''), + (u' a', u'a'), + (u'a ', u'a'), + (u'a b c', u'a b c'), + (u'a b / c d / e f', u'a b/c d/e f'), + ) + for test, expected in cases: + result = wikiutil.normalize_pagename(test) + assert result == expected + +class TestGroupPages(object): + + def setup_method(self, method): + self.config = self.TestConfig(page_group_regex=r'.+Group') + + def teardown_method(self, method): + del self.config + + def testNormalizeGroupName(self): + """ request: normalize pagename: restrict groups to alpha numeric Unicode + + Spaces should normalize after invalid chars removed! + """ + cases = ( + # current acl chars + (u'Name,:Group', u'NameGroup'), + # remove than normalize spaces + (u'Name ! @ # $ % ^ & * ( ) + Group', u'Name Group'), + ) + for test, expected in cases: + # validate we are testing valid group names + if wikiutil.isGroupPage(test, self.request.cfg): + result = wikiutil.normalize_pagename(test) + assert result == expected + coverage_modules = ['MoinMoin.wikiutil']
--- a/MoinMoin/action/CopyPage.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/action/CopyPage.py Sat Jun 07 19:55:27 2008 +0200 @@ -47,7 +47,7 @@ _ = self._ form = self.form newpagename = form.get('newpagename', [u''])[0] - newpagename = self.request.normalizePagename(newpagename) + newpagename = wikiutil.normalize_pagename(newpagename, self.cfg) comment = form.get('comment', [u''])[0] comment = wikiutil.clean_input(comment)
--- a/MoinMoin/action/PackagePages.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/action/PackagePages.py Sat Jun 07 19:55:27 2008 +0200 @@ -187,7 +187,7 @@ pages = [] for pagename in pagelist: - pagename = self.request.normalizePagename(pagename) + pagename = wikiutil.normalize_pagename(pagename, self.request.cfg) if pagename: page = Page(self.request, pagename) if page.exists() and self.request.user.may.read(pagename):
--- a/MoinMoin/action/RenamePage.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/action/RenamePage.py Sat Jun 07 19:55:27 2008 +0200 @@ -46,7 +46,7 @@ _ = self._ form = self.form newpagename = form.get('newpagename', [u''])[0] - newpagename = self.request.normalizePagename(newpagename) + newpagename = wikiutil.normalize_pagename(newpagename, self.cfg) comment = form.get('comment', [u''])[0] comment = wikiutil.clean_input(comment)
--- a/MoinMoin/formatter/__init__.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/formatter/__init__.py Sat Jun 07 19:55:27 2008 +0200 @@ -94,7 +94,7 @@ return '' if not pagename and page: pagename = page.page_name - pagename = self.request.normalizePagename(pagename) + pagename = wikiutil.normalize_pagename(pagename, self.request.cfg) if pagename and pagename not in self.pagelinks: self.pagelinks.append(pagename)
--- a/MoinMoin/mail/mailimport.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/mail/mailimport.py Sat Jun 07 19:55:27 2008 +0200 @@ -184,7 +184,7 @@ generate_summary = True pagename = pagename[1:].lstrip() - pagename = request.normalizePagename(pagename) + pagename = wikiutil.normalize_pagename(pagename, request.cfg) if choose_html and msg['html']: content = "{{{#!html\n%s\n}}}" % msg['html'].replace("}}}", "} } }")
--- a/MoinMoin/request/__init__.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/request/__init__.py Sat Jun 07 19:55:27 2008 +0200 @@ -910,47 +910,6 @@ name = u'/'.join(decoded) return name - def normalizePagename(self, name): - """ Normalize page name - - Prevent creating page names with invisible characters or funny - whitespace that might confuse the users or abuse the wiki, or - just does not make sense. - - Restrict even more group pages, so they can be used inside acl lines. - - @param name: page name, unicode - @rtype: unicode - @return: decoded and sanitized page name - """ - # Strip invalid characters - name = config.page_invalid_chars_regex.sub(u'', name) - - # Split to pages and normalize each one - pages = name.split(u'/') - normalized = [] - for page in pages: - # Ignore empty or whitespace only pages - if not page or page.isspace(): - continue - - # Cleanup group pages. - # Strip non alpha numeric characters, keep white space - if wikiutil.isGroupPage(self, page): - page = u''.join([c for c in page - if c.isalnum() or c.isspace()]) - - # Normalize white space. Each name can contain multiple - # words separated with only one space. Split handle all - # 30 unicode spaces (isspace() == True) - page = u' '.join(page.split()) - - normalized.append(page) - - # Assemble components into full pagename - name = u'/'.join(normalized) - return name - def read(self, n=None): """ Read n bytes from input stream. """ raise NotImplementedError @@ -1173,7 +1132,7 @@ path = '/' + path if path.startswith('/'): - pagename = self.normalizePagename(path) + pagename = wikiutil.normalize_pagename(path, self.cfg) else: pagename = None
--- a/MoinMoin/request/_tests/test_request.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/request/_tests/test_request.py Sat Jun 07 19:55:27 2008 +0200 @@ -13,89 +13,6 @@ from MoinMoin.request import HeadersAlreadySentException -class TestNormalizePagename(object): - - def testPageInvalidChars(self): - """ request: normalize pagename: remove invalid unicode chars - - Assume the default setting - """ - test = u'\u0000\u202a\u202b\u202c\u202d\u202e' - expected = u'' - result = self.request.normalizePagename(test) - assert result == expected - - def testNormalizeSlashes(self): - """ request: normalize pagename: normalize slashes """ - cases = ( - (u'/////', u''), - (u'/a', u'a'), - (u'a/', u'a'), - (u'a/////b/////c', u'a/b/c'), - (u'a b/////c d/////e f', u'a b/c d/e f'), - ) - for test, expected in cases: - result = self.request.normalizePagename(test) - assert result == expected - - def testNormalizeWhitespace(self): - """ request: normalize pagename: normalize whitespace """ - cases = ( - (u' ', u''), - (u' a', u'a'), - (u'a ', u'a'), - (u'a b c', u'a b c'), - (u'a b / c d / e f', u'a b/c d/e f'), - # All 30 unicode spaces - (config.chars_spaces, u''), - ) - for test, expected in cases: - result = self.request.normalizePagename(test) - assert result == expected - - def testUnderscoreTestCase(self): - """ request: normalize pagename: underscore convert to spaces and normalized - - Underscores should convert to spaces, then spaces should be - normalized, order is important! - """ - cases = ( - (u' ', u''), - (u' a', u'a'), - (u'a ', u'a'), - (u'a b c', u'a b c'), - (u'a b / c d / e f', u'a b/c d/e f'), - ) - for test, expected in cases: - result = self.request.normalizePagename(test) - assert result == expected - - -class TestGroupPages(object): - - def setup_method(self, method): - self.config = self.TestConfig(page_group_regex=r'.+Group') - - def teardown_method(self, method): - del self.config - - def testNormalizeGroupName(self): - """ request: normalize pagename: restrict groups to alpha numeric Unicode - - Spaces should normalize after invalid chars removed! - """ - cases = ( - # current acl chars - (u'Name,:Group', u'NameGroup'), - # remove than normalize spaces - (u'Name ! @ # $ % ^ & * ( ) + Group', u'Name Group'), - ) - for test, expected in cases: - # validate we are testing valid group names - if wikiutil.isGroupPage(self.request, test): - result = self.request.normalizePagename(test) - assert result == expected - class TestHTTPHeaders(object): std_headers = ['Status: 200 OK', 'Content-type: text/html; charset=%s' % config.charset]
--- a/MoinMoin/theme/__init__.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/theme/__init__.py Sat Jun 07 19:55:27 2008 +0200 @@ -361,7 +361,7 @@ pass # Handle regular pagename like "FrontPage" - pagename = request.normalizePagename(pagename) + pagename = wikiutil.normalize_pagename(pagename, request.cfg) # Use localized pages for the current user if localize:
--- a/MoinMoin/user.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/user.py Sat Jun 07 19:55:27 2008 +0200 @@ -201,7 +201,7 @@ @param name: user name, unicode """ normalized = normalizeName(name) - return (name == normalized) and not wikiutil.isGroupPage(request, name) + return (name == normalized) and not wikiutil.isGroupPage(name, request.cfg) def encodeList(items):
--- a/MoinMoin/wikiutil.py Fri Jun 06 14:26:55 2008 +0200 +++ b/MoinMoin/wikiutil.py Sat Jun 07 19:55:27 2008 +0200 @@ -709,14 +709,14 @@ return request.cfg.cache.page_template_regexact.search(pagename) is not None -def isGroupPage(request, pagename): +def isGroupPage(pagename, cfg): """ Is this a name of group page? @param pagename: the page name @rtype: bool @return: true if page is a form page """ - return request.cfg.cache.page_group_regexact.search(pagename) is not None + return cfg.cache.page_group_regexact.search(pagename) is not None def filterCategoryPages(request, pagelist): @@ -2222,6 +2222,47 @@ ############################################################################# ### Misc ############################################################################# +def normalize_pagename(name, cfg): + """ Normalize page name + + Prevent creating page names with invisible characters or funny + whitespace that might confuse the users or abuse the wiki, or + just does not make sense. + + Restrict even more group pages, so they can be used inside acl lines. + + @param name: page name, unicode + @rtype: unicode + @return: decoded and sanitized page name + """ + # Strip invalid characters + name = config.page_invalid_chars_regex.sub(u'', name) + + # Split to pages and normalize each one + pages = name.split(u'/') + normalized = [] + for page in pages: + # Ignore empty or whitespace only pages + if not page or page.isspace(): + continue + + # Cleanup group pages. + # Strip non alpha numeric characters, keep white space + if isGroupPage(page, cfg): + page = u''.join([c for c in page + if c.isalnum() or c.isspace()]) + + # Normalize white space. Each name can contain multiple + # words separated with only one space. Split handle all + # 30 unicode spaces (isspace() == True) + page = u' '.join(page.split()) + + normalized.append(page) + + # Assemble components into full pagename + name = u'/'.join(normalized) + return name + def taintfilename(basename): """ Make a filename that is supposed to be a plain name secure, i.e.