Allow exporting abstract contents to CSV/XLSX

[rppt]

Is your feature request related to a problem? Please describe.
In our model not all reviewers have access to the submissions in indicio. For the review we export all the abstracts to a spreadsheet and ask reviewers to vote in the spreadsheet. Unfortunately, exporting CSV/XLSX from event/N/manage/abstracts/list/ does not allow to include the abstract text in the generated spreadsheet.

Describe the solution you'd like
I would like to be able to see the abstract text in the exported CSV/XLSX when the "Contents" button is pressed in "Customize list" in abstracts list form. I.e. if event/N/manage/abstracts/list/ shows the abstract content, then exporting abstracts to CSV/XLSX will include the abstract text.

For our indico setup I've added the patch below as a temporal solution, but it mangles some characters because I didn't find how to correctly put abstract.description to the generated spreadsheet.

From fbb27612aa86a508fc10179a7ef0adf85749e6bb Mon Sep 17 00:00:00 2001
From: Mike Rapoport <rppt@kernel.org>
Date: Sat, 21 May 2022 22:49:55 +0300
Subject: [PATCH] LPC: add abstract contents to CSV and XLSX exports

---
 indico/modules/events/abstracts/util.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/indico/modules/events/abstracts/util.py b/indico/modules/events/abstracts/util.py
index faea32873f..8d09b5956a 100644
--- a/indico/modules/events/abstracts/util.py
+++ b/indico/modules/events/abstracts/util.py
@@ -70,6 +70,7 @@ def generate_spreadsheet_from_abstracts(abstracts, static_item_ids, dynamic_item
         'score': ('Score', lambda x: round(x.score, 1) if x.score is not None else None),
         'submitted_dt': ('Submission date', lambda x: x.submitted_dt),
         'modified_dt': ('Modification date', lambda x: x.modified_dt if x.modified_dt else ''),
+        'description': ('Content', lambda x: x.description),
     }
     field_names.extend(unique_col(item.title, item.id) for item in dynamic_items)
     field_names.extend(title for name, (title, fn) in static_item_mapping.items() if name in static_item_ids)
-- 
2.34.1

[ThiefMaster]

This change looks good. My guess is that the "mangling" you experienced comes from this sanitization we apply to data that goes into CSV files:

def _prepare_csv_data(data, _linebreak_re=re.compile(r'(\r?\n)+'), _dangerous_chars_re=re.compile(r'^[=+@-]+')):
    if isinstance(data, (list, tuple)):
        data = '; '.join(data)
    elif isinstance(data, set):
        data = '; '.join(sorted(data, key=str.lower))
    elif isinstance(data, bool):
        data = 'Yes' if data else 'No'
    elif data is None:
        data = ''
    data = _linebreak_re.sub('    ', str(data))
    # https://www.owasp.org/index.php/CSV_Injection
    # quoting the cell's value does NOT mitigate this, so we need to strip
    # those characters from the beginning...
    return _dangerous_chars_re.sub('', data)

However, this seems acceptable - if someone wants something that's exactly the same, they can use the Excel (XLSX) export instead.. CSV is not well-suited for multi-line content.

I will open a PR containing your commit and merge it once CI passed.

[rppt]

Thanks!