Merge pull request #493 from JustAnotherArchivist/fix-otf-formats

Fix format selection for on-the-fly files

HISTORY.rst

@@ -33,6 +33,7 @@ Unreleased
 - Fixed uploading from a spreadsheet with ``--checksum`` crashing on skipped files.
 - Fixed minor bug in S3 overload check on upload error retries.
 - Fixed various messages being printed to stdout instead of stderr.
+- Fixed format selection for on-the-fly files.
 
 2.3.0 (2022-01-20)
 ++++++++++++++++++

docs/source/cli.rst

@@ -258,7 +258,7 @@ See ``ia help download`` for more details.
 Downloading On-The-Fly Files
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-Some files on archive.org are generated on-the-fly as requested. This currently includes non-original files of the formats EPUB, MOBI, DAISY, and archive.org's own MARC XML. These files can be downloaded using the ``--on-the-fly`` parameter:
+Some files on archive.org are generated on-the-fly as requested. This currently includes non-original files of the formats EPUB, MOBI, DAISY, and archive.org's own MARCXML. These files can be downloaded using the ``--on-the-fly`` parameter:
 
 .. code:: bash
 

docs/source/quickstart.rst

@@ -204,9 +204,9 @@ Or, a list of formats::
 Downloading On-The-Fly Files
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-Some files on archive.org are generated on-the-fly as requested. This currently includes non-original files of the formats EPUB, MOBI, DAISY, and archive.org's own MARC XML. These files can be downloaded using the ``on_the_fly`` parameter::
+Some files on archive.org are generated on-the-fly as requested. This currently includes non-original files of the formats EPUB, MOBI, DAISY, and archive.org's own MARCXML. These files can be downloaded using the ``on_the_fly`` parameter::
 
-    >>> download('wonderfulwizardo00baumiala', verbose=True, glob_pattern='*_daisy.zip', on_the_fly=True)
+    >>> download('wonderfulwizardo00baumiala', verbose=True, formats='DAISY', on_the_fly=True)
     wonderfulwizardo00baumiala:
      downloading wonderfulwizardo00baumiala_daisy.zip: 100%|████| 153k/153k [00:00<00:00, 563kiB/s]
 

internetarchive/item.py

@@ -532,13 +532,13 @@ def get_files(self, files=None, formats=None, glob_pattern=None, on_the_fly=None
         # Add support for on-the-fly files (e.g. EPUB).
         if on_the_fly:
             otf_files = [
-                f'{self.identifier}.epub',
-                f'{self.identifier}.mobi',
-                f'{self.identifier}_daisy.zip',
-                f'{self.identifier}_archive_marc.xml',
+                ('EPUB', f'{self.identifier}.epub'),
+                ('MOBI', f'{self.identifier}.mobi'),
+                ('DAISY', f'{self.identifier}_daisy.zip'),
+                ('MARCXML', f'{self.identifier}_archive_marc.xml'),
             ]
-            for f in otf_files:
-                item_files.append({'name': f, 'otf': True})
+            for format, f in otf_files:
+                item_files.append({'name': f, 'format': format, 'otf': True})
 
         if not any(k for k in [files, formats, glob_pattern]):
             for f in item_files:

tests/cli/test_ia_download.py

@@ -43,6 +43,16 @@ def test_format(tmpdir_ch):
     assert files_downloaded(path='nasa') == {'nasa_archive.torrent'}
 
 
+def test_on_the_fly_format():
+    i = 'wonderfulwizardo00baumiala'
+
+    stdout, stderr = call_cmd(f'ia --insecure download --dry-run --format="DAISY" {i}')
+    assert stdout == ''
+
+    stdout, stderr = call_cmd(f'ia --insecure download --dry-run --format="DAISY" --on-the-fly {i}')
+    assert stdout == f'http://archive.org/download/{i}/{i}_daisy.zip'
+
+
 def test_clobber(tmpdir_ch):
     cmd = 'ia --insecure download nasa nasa_meta.xml'
     call_cmd(cmd)

tests/test_item.py

@@ -237,6 +237,20 @@ def test_download_dry_run(tmpdir, capsys, nasa_item):
     assert {x.split('/')[-1] for x in out.split('\n') if x} == expected
 
 
+def test_download_dry_run_on_the_fly_formats(tmpdir, capsys, nasa_item):
+    tmpdir.chdir()
+    with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
+        rsps.add(responses.GET, DOWNLOAD_URL_RE,
+                 body='no dest dir',
+                 adding_headers={'content-length': '100'})
+        nasa_item.download(formats='MARCXML', on_the_fly=True, dry_run=True)
+
+    expected = {'nasa_archive_marc.xml'}
+    out, err = capsys.readouterr()
+
+    assert {x.split('/')[-1] for x in out.split('\n') if x} == expected
+
+
 def test_download_verbose(tmpdir, capsys, nasa_item):
     tmpdir.chdir()
     with IaRequestsMock(assert_all_requests_are_fired=False) as rsps: