sitescripts/extensions/pad/validation.py - Issue 5723465818570752: Issue 520 - Generate PAD files for download portals when updating download links

Unified Diff: sitescripts/extensions/pad/validation.py

Issue 5723465818570752: Issue 520 - Generate PAD files for download portals when updating download links (Closed)

Patch Set: Created May 26, 2014, 10 a.m.

Use n/p to move between diff chunks; N/P to move between comments.

Jump to:

View side-by-side diff with in-line comments

« sitescripts/extensions/pad/__init__.py ('K') | « sitescripts/extensions/pad/language.py ('k') | sitescripts/extensions/template/pad.xml » ('j') | sitescripts/extensions/template/pad.xml » ('J')
Expand Comments ('e') | Collapse Comments ('c') | Hide Comments ('s')

Index: sitescripts/extensions/pad/validation.py

===================================================================

new file mode 100644

--- /dev/null

+++ b/sitescripts/extensions/pad/validation.py

@@ -0,0 +1,117 @@

+# This file is part of the Adblock Plus web scripts,

+# Adblock Plus is free software: you can redistribute it and/or modify

+# it under the terms of the GNU General Public License version 3 as

+# published by the Free Software Foundation.

+# Adblock Plus is distributed in the hope that it will be useful,

+# but WITHOUT ANY WARRANTY; without even the implied warranty of

+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

+# GNU General Public License for more details.

+# You should have received a copy of the GNU General Public License

+# along with Adblock Plus. If not, see <http://www.gnu.org/licenses/>.

+import re, warnings

+try:

+ import libxml2

+except ImportError:

+ libxml2 = None

+ warnings.warn('libxml2 is not installed, can not validate PAD files')

+FIELDS = [

+ ('/XML_DIZ_INFO/MASTER_PAD_VERSION_INFO/MASTER_PAD_VERSION', r'^4\.0\Z'),

+ ('/XML_DIZ_INFO/MASTER_PAD_VERSION_INFO/MASTER_PAD_EDITOR', r'^[^<\x09]{0,100}\Z'),

+ ('/XML_DIZ_INFO/MASTER_PAD_VERSION_INFO/MASTER_PAD_INFO', r'^[^<\x09]{0,1000}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Company_Name', r'^[^<\x09]{2,40}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Address_1', r'^[a-zA-Z0-9\xbc-\xff .\-,#\/\x27]{0,40}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Address_2', r'^[a-zA-Z\xbc-\xff0-9 .\-,#\/\x27]{0,40}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/City_Town', r'^[a-zA-Z\xbc-\xff0-9 .\-,#\/\x27]{2,40}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/State_Province', r'^[a-zA-Z\xbc-\xff0-9 .\-,\/]{0,30}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Zip_Postal_Code', r'^[^<\x09]{0,20}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Country', r'^[a-z A-Z\xbc-\xff\x27-]{2,40}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Company_WebSite_URL', r'^http:\/\/.{2,120}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Contact_Info/Author_First_Name', r'^[^<\x09]{2,30}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Contact_Info/Author_Last_Name', r'^[^<\x09]{2,30}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Contact_Info/Author_Email', r'^.{2,30}\@.{2,63}\..{2,20}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Contact_Info/Contact_First_Name', r'^[^<\x09]{2,30}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Contact_Info/Contact_Last_Name', r'^[^<\x09]{2,30}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Contact_Info/Contact_Email', r'^.{2,30}\@.{2,63}\..{2,20}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Support_Info/Sales_Email', r'^.{2,30}\@.{2,63}\..{2,20}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Support_Info/Support_Email', r'^.{2,30}\@.{2,63}\..{2,20}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Support_Info/General_Email', r'^.{2,30}\@.{2,63}\..{2,20}\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Support_Info/Sales_Phone', r'^\+{0,2}(([0-9#*()-\/_] *){7,40})?\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Support_Info/Support_Phone', r'^\+{0,2}(([0-9#*()-\/_] *){7,40})?\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Support_Info/General_Phone', r'^\+{0,2}(([0-9#*()-\/_] *){7,40})?\Z'),

+ ('/XML_DIZ_INFO/Company_Info/Support_Info/Fax_Phone', r'^\+{0,2}(([0-9#*()-\/_] *){7,40})?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Name', r'^[^<\x09]{1,40}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Version', r'^[a-zA-Z0-9_.\-]{1,15}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Release_Month', r'^(0[1-9]|1[0-2])\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Release_Day', r'^(0[1-9]|[12][0-9]|3[01])\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Release_Year', r'^(19|20|21)[0-9]{2}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Cost_Dollars', r'^([0-9]+(\.[0-9]{2})?)?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Cost_Other_Code', r'^(AED|AFN|ALL|AMD|ANG|AOA|ARS|AUD|AWG|AZM|BAM|BBD|BDT|BGN|BHD|BIF|BMD|BND|BOB|BRL|BSD|BTN|BWP|BYR|BZD|CAD|CDF|CHF|CLP|CNY|COP|COU|CRC|CSD|CZK|CUP|CVE|CYP|DJF|DKK|DOP|DZD|EEK|EGP|ERN|ETB|EUR|FJD|FKP|GBP|GEL|GHC|GIP|GMD|GNF|GTQ|GYD|HKD|HNL|HRK|HTG|HUF|IDR|ILS|INR|IQD|IRR|ISK|JMD|JOD|JPY|KES|KGS|KHR|KMF|KPW|KRW|KWD|KYD|KZT|LAK|LBP|LKR|LRD|LSL|LTL|LVL|LYD|MAD|MDL|MGA|MKD|MMK|MNT|MOP|MRO|MTL|MUR|MVR|MWK|MXN|MYR|MZN|NAD|NGN|NIO|NOK|NPR|NZD|OMR|PAB|PEN|PGK|PHP|PKR|PLN|PYG|QAR|RON|RUB|RWF|SAR|SBD|SCR|SDD|SEK|SGD|SHP|SIT|SKK|SLL|SOS|SRD|STD|SYP|SZL|THB|TJS|TMM|TND|TOP|TRY|TTD|TWD|TZS|UAH|UGX|USD|UYU|UZS|VEB|VND|VUV|WST|XAF|XCD|XOF|XPF|YER|ZAR|ZMK|ZWD)?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Cost_Other', r'^([0-9]+(\.[0-9]{2})?)?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Install_Support', r'^(Install and Uninstall|Install Only|No InstallSupport|Uninstall Only)\Z'),

+ ('/XML_DIZ_INFO/Program_Info/File_Info/File_Size_Bytes', r'^[0-9]{3,16}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/File_Info/File_Size_K', r'^[0-9.]{1,12}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/File_Info/File_Size_MB', r'^[0-9.]{1,8}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Expire_Info/Has_Expire_Info', r'^[YyNn]\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Expire_Info/Expire_Count', r'^[0-9]{0,15}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Expire_Info/Expire_Based_On', r'^(Days|Uses|Either\/Or)?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Expire_Info/Expire_Other_Info', r'^[^<\x09]{0,100}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Expire_Info/Expire_Month', r'^(0[1-9]|1[0-2])?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Expire_Info/Expire_Day', r'^(0[1-9]|[12][0-9]|3[01])?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Expire_Info/Expire_Year', r'^((19|20|21)[0-9]{2})?\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_Change_Info', r'^[^<\x09]{0,300}\Z'),

+ ('/XML_DIZ_INFO/Program_Info/Program_System_Requirements', r'^[^<\x09]{0,100}\Z'),

+ ('/XML_DIZ_INFO/Program_Descriptions/English/Keywords', r'^[^<\x09]{0,250}\Z'),

+ ('/XML_DIZ_INFO/Program_Descriptions/English/Char_Desc_45', r'^[^<\x09\x0a\x0d]{0,45}\Z'),

+ ('/XML_DIZ_INFO/Program_Descriptions/English/Char_Desc_80', r'^[^<\x09\x0a\x0d]{0,80}\Z'),

+ ('/XML_DIZ_INFO/Program_Descriptions/English/Char_Desc_250', r'^[^<\x09\x0a\x0d]{0,250}\Z'),

+ ('/XML_DIZ_INFO/Program_Descriptions/English/Char_Desc_450', r'^[^<\x09\x0a\x0d]{0,450}\Z'),

+ ('/XML_DIZ_INFO/Program_Descriptions/English/Char_Desc_2000', r'^[^<]{0,2000}\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Application_URLs/Application_Info_URL', r'^http:\/\/.{2,120}\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Application_URLs/Application_Order_URL', r'^((http|https):\/\/.{2,120})?\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Application_URLs/Application_Screenshot_URL', r'^http:\/\/.{2,120}\.(gif|jpg|png)\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Application_URLs/Application_Icon_URL', r'^http:\/\/.{2,120}\.(gif|jpg|png)\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Application_URLs/Application_XML_File_URL', r'^http:\/\/.{2,120}\.(xml|cgi|php|asp)\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Download_URLs/Primary_Download_URL', r'^(http|ftp):\/\/.{2,120}\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Download_URLs/Secondary_Download_URL', r'^((http|ftp):\/\/.{2,120})?\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Download_URLs/Additional_Download_URL_1', r'^((http|ftp):\/\/.{2,120})?\Z'),

+ ('/XML_DIZ_INFO/Web_Info/Download_URLs/Additional_Download_URL_2', r'^((http|ftp):\/\/.{2,120})?\Z'),

+ ('/XML_DIZ_INFO/Permissions/Distribution_Permissions', r'^[^<]{0,2000}\Z'),

+ ('/XML_DIZ_INFO/Permissions/EULA', r'^[^<]{0,20000}\Z'),

+def validate_pad(pad, filename):

+ if not libxml2:

+ return

Wladimir Palant 2014/05/26 10:54:57 I assume that libxml2 is being used due to its XPa

Sebastian Noack 2014/05/26 12:04:53 I just gave etree a try. However their xpath imple

Wladimir Palant 2014/05/26 13:26:53 Why does it have to? Relative lookups should do ju

Sebastian Noack 2014/05/27 12:01:23 I am using mindom (without xpath) now. Also I'm va

+ doc = libxml2.parseDoc(pad)

+ xpctx = doc.xpathNewContext()

+ for xpath, regex in FIELDS:

+ nodes = xpctx.xpathEval(xpath)

+ if not nodes:

+ warnings.warn('invalid PAD file (missing node)\n'

+ 'filename: %s\n'

+ 'XPath: %s' % (filename, xpath))

+ else:

+ value = nodes[0].getContent()

+ if not re.match(regex, value):

+ warnings.warn('invalid PAD file (invalid value)\n'

+ 'filename: %s\n'

+ 'XPath: %s\n'

+ 'value: %s\n'

+ 'regex: %s' % (filename, xpath, value, regex))

« sitescripts/extensions/pad/__init__.py ('K') | « sitescripts/extensions/pad/language.py ('k') | sitescripts/extensions/template/pad.xml » ('j') | sitescripts/extensions/template/pad.xml » ('J')