feat: allow repeated --filter flags for exports

[jacalata]

Motivation

--filter previously accepted one value and required multi-pair filters
to be joined with &. That parsing scheme prevents literal & inside a
value (e.g. Product Name=AT&T gets split). Making --filter
repeatable removes the need for callers to encode delimiters.

Related to #442, which handles the same problem when the filter comes in
via URL syntax rather than the flag.

Behavior change

For users: --filter is now repeatable. Each flag carries exactly
one COLUMN=VALUE pair.

--filter "Region=West&Product=Widget"                # still works (back-compat)
--filter "Region=West" --filter "Product=Widget"     # preferred
--filter "Product=AT&T"                              # new: repeated form allows '&' in value

Back-compat: when only one --filter is present, the value is still
split on &. This means a single flag carrying one filter with & in
its value (e.g. --filter "Product=AT&T" alone) is still ambiguous.
Workaround: pass the filter in the URL query string instead of via
--filter. #442 landed the URL-path handling for literal &, so
tabcmd get "views/View/Sheet.csv?Product%20Name=AT&T%20841000%20Phone"
works correctly (space must be encoded as %20; & may be literal).

Test plan

• pytest tests/commands/test_datasources_and_workbooks_command.py tests/parsers/test_parser_export.py — 34 passed
• Manual e2e: repeated --filter with & in the value returns the
  expected row
• Manual e2e: single-flag &-joined form still parses and filters
  correctly

🤖 Generated with Claude Code

[Copilot]

Pull request overview

Note

Copilot was unable to run its full agentic suite in this review.

Adds support for repeating --filter on the export command while preserving backward-compatible parsing of legacy &-joined filter pairs.

Changes:

• Update export CLI arg parsing to accept multiple --filter flags (action="append").
• Implement back-compat filter splitting when a single --filter contains &.
• Add/extend tests for parser behavior and filter application; update help text to document new usage.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
tests/parsers/test_parser_export.py	Adds tests validating --filter repeatability and default behavior.
tests/commands/test_datasources_and_workbooks_command.py	Adds unit tests for applying parsed filters into TSC request options, including back-compat.
tabcmd/locales/en/tabcmd_messages_en.properties	Expands the --filter help text to document the new repeatable flag semantics and back-compat.
tabcmd/commands/datasources_and_workbooks/export_command.py	Changes --filter arg to append and updates filter parsing logic accordingly.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

Coverage Report

File	Stmts	Miss	Cover	Missing
tabcmd
__main__.py	12	12	12	0%
_version.py	11	11	11	0%
tabcmd.py	15	15	15	0%
version.py	9	5	5	44%
tabcmd/commands
commands.py	10	10	10	0%
constants.py	77	18	18	77%
server.py	135	18	18	87%
tabcmd/commands/auth
session.py	394	50	50	87%
tabcmd/commands/datasources_and_workbooks
datasources_and_workbooks_command.py	157	18	18	89%
datasources_workbooks_views_url_parser.py	142	5	5	96%
delete_command.py	60	16	16	73%
export_command.py	123	21	21	83%
get_url_command.py	127	47	47	63%
publish_command.py	123	28	28	77%
runschedule_command.py	21	7	7	67%
tabcmd/commands/extracts
create_extracts_command.py	42	8	8	81%
decrypt_extracts_command.py	27	2	2	93%
delete_extracts_command.py	37	6	6	84%
encrypt_extracts_command.py	27	2	2	93%
extracts.py	20	2	2	90%
reencrypt_extracts_command.py	27	2	2	93%
refresh_extracts_command.py	48	10	10	79%
tabcmd/commands/group
create_group_command.py	29	5	5	83%
delete_group_command.py	27	2	2	93%
tabcmd/commands/project
create_project_command.py	46	8	8	83%
delete_project_command.py	35	4	4	89%
publish_samples_command.py	30	4	4	87%
tabcmd/commands/site
create_site_command.py	34	5	5	85%
delete_site_command.py	27	2	2	93%
edit_site_command.py	38	2	2	95%
list_command.py	77	12	12	84%
list_sites_command.py	29	2	2	93%
tabcmd/commands/user
add_users_command.py	29	5	5	83%
create_site_users.py	58	11	11	81%
create_users_command.py	59	9	9	85%
delete_site_users_command.py	43	5	5	88%
user_data.py	222	31	31	86%
tabcmd/execution
_version.py	2	2	2	0%
global_options.py	125	8	8	94%
localize.py	66	11	11	83%
logger_config.py	60	6	6	90%
tabcmd_controller.py	42	7	7	83%
TOTAL	2876	454	84%

[Copilot]

🔵 Needs a closer look

The updated filter-path still risks truncating values containing = and can error on ambiguous single-flag & splits unless parsing is made more robust (and covered by a regression test).

Review details

Suppressed comments (2)

Previously missed (2) — in code that hasn't changed since the last review.

tabcmd/commands/datasources_and_workbooks/export_command.py:150

• apply_filters_from_args currently claims repeated flags allow literal '&' or '=' inside values, but the implementation delegates to DatasourcesAndWorkbooks.apply_filter_value, which splits on every '=' and will truncate values like Notes=x=y. Also, the back-compat single-flag split on '&' can yield fragments without '=', which can raise IndexError downstream. Consider parsing COLUMN=VALUE here with split('=', 1) and only treating '&' as a delimiter when every segment contains '=' (otherwise treat '&' as literal and warn) to avoid crashes and preserve values.
  tests/commands/test_datasources_and_workbooks_command.py:217
• Current tests cover '&' in values, but they don't assert the stated behavior that '=' inside a filter value is preserved (e.g. Notes=x=y). Adding a regression test would catch truncation bugs in filter parsing.

• Files reviewed: 4/4 changed files
• Comments generated: 0 new
• Review effort level: Lite