bug: get command incorrectly reports spaces error; evaluate_content_type uses singular URL prefixes

[jacalata]

Observed behaviour

Running tabcmd get "/workbooks/WorkbookWithoutExtract.twbx" --no-certcheck (no spaces in the URL) produces:

The name of the workbook or view to export cannot include spaces. Use the normalized name of the workbook or view as it appears in the URL.
Exiting...

The workbook exists on the server and the URL contains no spaces.

Investigation findings

The spaces check at get_url_command.py:45 (if " " in args.url) should be False for this input. The secondary spaces check in datasources_workbooks_views_url_parser.py:46 (validate_and_extract_url_parts) is unreachable from the workbook get path — it is only called for view-type URLs. The root cause of the observed error has not yet been fully reproduced in isolation; further debugging is needed to confirm what value args.url actually holds at runtime on Windows.

Confirmed latent bug in evaluate_content_type

get_url_command.py:62-64:

valid_content_types = ["workbook", "view", "datasource"]  # singular
...
if url.find(content_type) == 0:   # matches "workbooks/..." via substring accident

The URL prefixes used in practice are plural (workbooks/, views/, datasources/). The match works by accident because "workbooks/...".find("workbook") == 0, but it is fragile — any URL starting with "workbook" (e.g. a hypothetical "workbookshelf/...") would also match. The correct fix is:

url_prefix_to_content_type = {
    "workbooks/": "workbook",
    "views/": "view",
    "datasources/": "datasource",
}
for prefix, content_type in url_prefix_to_content_type.items():
    if url.startswith(prefix):
        return content_type
Errors.exit_with_error(logger, message=_("tabcmd.error.bad_request.invalid_content_type").format(url))

Files

• tabcmd/commands/datasources_and_workbooks/get_url_command.py — evaluate_content_type (line 57) and spaces check (line 45)
• tabcmd/commands/datasources_and_workbooks/datasources_workbooks_views_url_parser.py — validate_and_extract_url_parts (line 44)