Bulk Data issue

Data Issues
1

Hello
I am trying to read the iXRBL files that you can get from the Companies House page but when I try and use the code that has worked perfectly for all the months before (Except Oct 2024) I am getting fail after fail.
The main error says Failed to process zip: @‘noneType’ object has no attribute ‘groups’
I am sure this is a data fault at creation but Companies house pointed me here.
Does anyone else have this problem for May?
To check the code I decided to re download and process April. Works fine.
I then though I would try and download the data day at a time. Only the first of the month worked. So I know my code works when the file is structured correctly.
Any ideas anyone?
Thanks

2

I have gone back to the basic code

import httpx
from stream_read_xbrl import stream_read_xbrl_zip

# A URL taken from http://download.companieshouse.gov.uk/en_accountsdata.html
if __name__ == '__main__':
    url = 'http://download.companieshouse.gov.uk/Accounts_Bulk_Data-2023-03-02.zip'
    with \
            httpx.stream('GET', url) as r, \
            stream_read_xbrl_zip(r.iter_bytes(chunk_size=65536)) as (columns, rows):
        r.raise_for_status()
        for row in rows:
            print(row)

and run it on the 2025-05-03 file (and changed to https) and it wont even print to the screen. It runs for a about a thousand records then crashes. I dont seem to be able to filter the error as it reports it is in the stream_xbrl_zip file.
The code crashes before I can filter the error.
Thanks

3

Hello
For those who are having the same problem you need to modify the stream_read_xbrl.py file at row 423 onwards.

context_dates = {
    e.get('id'): _get_dates(period)
    for e in document.xpath("//*[local-name()='context']")
    for period in e.xpath("./*[local-name()='period']")[:1]
}

fn = os.path.basename(name)
mo = re.match(r'^(Prod\d+_\d+)_([^_]+)_(\d{8})(?:_[^.]*)?\.(html|xml|zip)', fn)


if not mo:
    raise ValueError(f"Filename does not match expected pattern: {fn}")

run_code, company_id, date, filetype = mo.groups()
allowed_taxonomies = [
    'http://www.xbrl.org/uk/fr/gaap/pt/2004-12-01',
    'http://www.xbrl.org/uk/gaap/core/2009-09-01',
    'http://xbrl.frc.org.uk/fr/2014-09-01/core',
]

core_attributes = (
    run_code,
    company_id,
    _date(date),
    filetype,
    ';'.join(set(allowed_taxonomies) & set(root.nsmap.values())),
)

Specifically I altered
fn = os.path.basename(name)
mo = re.match(r’^(Prod\d+\d+)([^]+)(\d{8})(?:_[^.]*)?.(html|xml|zip)', fn)

if not mo:
    raise ValueError(f"Filename does not match expected pattern: {fn}")

now it doesnt crash. It skips the bad data. I have tested it on the month file of 2025 05 and it worked.
Neil