This file lists specific errors observed in the official IBAN specification. In addition to the specific errors, there persist basic outstanding matters that cause issue for implementers: - Non 1:1 mapping of records to countries (eg: French territories, etc.) - Mixing of free text and hard data in specification fields - Missing clear definition of fields ("Domestic account number example" is in many cases pre-IBAN era, in others it's an IBAN-era BBAN) - Lack of validation of information in fields prior to publishing - Lack of synchronisation between TXT and PDF versions of the registry - Unannounced alterations to TXT publishing format - Use of highly abnormal column-as-record format in TXT registry publishing 2023-04-23 ---------- - Recent TXT registry publication uses broken column-as-record format 2017-08-03 ---------- - In approximately September-October 2016, a 'complete revision' was made to the registry document, after which no change tracking has been published as it was in the past. This makes it very tedious and time consuming to re-check each release against the previous one to keep up to date with changes. In short, we will need to recheck all countries' information on a per-release basis. Current document version changes are sometimes supplied, but as previous document versions are unavailable and releases are made on an irregular schedule, this is nearly useless. 2016-01-31 ---------- - SWIFT released a v63 registry PDF document supposedly including a fix for the BBAN example for Israel with reason "typo", however it still does not match with the specified BBAN format. 2016-01-21 ---------- - Correction for broken example IBANs for multiple countries (checksum failed, length incorrect, etc.) - Correction for broken IBAN length field for Khazakstan (no specified length whatsoever) June 2014 --------- - The TXT registry records of QA and JO do not match those within the PDF and are essentially spurious while looking roughly correct. Dang. March 2014 ---------- - In multiple cases, territories of a major country have a SEPA status that differs from the parent jurisdiction. The IBAN specification as previously released does not include individual records for these territories, and thus cannot convey this important information. - The IBAN registry's old URL is no longer publicly available! Its now ends in a 404, and the apparent new home, located over at http://www.swift.com/products_services/by_type/reference_data/iban_registry_iso13616 does not have any content. The parent portion of this part of the SWIFT website suggests the information may have been moved to a new 'SWIFTRef' site, however that site appears to only peddle paid for directories of BICs and similar. The new location after much searching was found to be http://www.swift.com/products_services/bic_and_iban_format_registration_iban_format_r September 2013 -------------- - Azerbaijan, Brazil, Costa Rica, Palestine, Virgin Islands 'SEPA Country' field in PDF (yes/no) is completely blank February 2013 ------------- - Deployment of unparseable special values such as "Not in use" (FI). - Still(!) missing a registry entry for 2010's 'new' entry of Khazakstan Early 2012 ---------- - Inconsistent record ordering (KW, KZ) - Inconsistent capitalization (DK) - Continued presence of incorrect domestic examples 2011-07-16 ---------- - No information for Kuwait past sixth column - Total absence of information regarding Kazakhstan - '1.00001E+15' instead of a valid BBAN example for Lithuania - Repeated IBAN example in human format instead of IBAN format- specification for UAE - Incorrect domestic example for Bulgaria, Kazakhstan, Latvia, Lithuania, Luxembourg, Macedonia, Mauritius, Romania, San Marino (complete, human-format IBAN instead of domestic example)