def _get_posters_html(movielens_posters_df: pd.DataFrame,
df_item: pd.DataFrame,
item_ids: Iterable[Any],
col_description: str = 'Recommended films:',
titles_provided: bool = False,
image_width: Optional[int] = 500) -> str:
if type(item_ids) == int:
item_ids = [item_ids]
final = pd.Series(dtype='object')
if not titles_provided:
try:
top_movies_titles = [df_item[df_item['item_id'] == x]
['movie_title'].iloc[0] for x in item_ids]
except IndexError as e:
raise IndexError('Ensure user and item IDs start at 1 for MovieLens 100K data:', e)
else:
top_movies_titles = item_ids
for x in item_ids:
url = _get_single_poster_html(movielens_posters_df, x)
final = final.append(pd.Series(url))
final_df = final.to_frame()
final_df.index = top_movies_titles
final_df.columns = [col_description]
return df_to_html(final_df,
image_cols=[col_description],
transpose=True,
image_width=image_width)
def _get_posters_html(movielens_posters_df: pd.DataFrame,
df_item: pd.DataFrame,
item_ids: Union[int, Iterable[Any]],
col_description: str = 'Recommended films:',
image_width: Optional[int] = 500) -> str:
if not isinstance(item_ids, collections.abc.Iterable):
item_ids = [item_ids]
top_movies_titles = [
df_item[df_item['item_id'] == x]['movie_title'].iloc[0]
for x in item_ids
]
final_urls = []
for item_id in item_ids:
url = ''
with suppress((ValueError, TypeError)):
url = movielens_posters_df.query(
f'item_id == {item_id}')['url'].item()
final_urls.append(url)
final_df = pd.DataFrame(final_urls)
final_df.index = top_movies_titles
final_df.columns = [col_description]
return df_to_html(df=final_df,
image_cols=[col_description],
transpose=True,
image_width=image_width)
def test_df_to_html(df_html_test):
expected = """
<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>title</th>
<th>description</th>
<th>link</th>
<th>image</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td><h2>Greg</h2></td>
<td>some text here</td>
<td><a target="_blank" href="https://madeupsite.com">
https://madeupsite.com</a></td>
<td><img src="https://avatars0.githubusercontent.co
m/u/13399445"></td>
</tr>
<tr>
<th>1</th>
<td><h2>Real Greg</h2>
</td>
<td>more text here</td>
<td><a target="_blank" href="https://anotherm
adeupsite.com">https://anothermadeupsite.com</a></td>
<td><img src="https://avatars3.githubusercontent.com/u/31417712"></td>
</tr>
</tbody>
</table>
"""
actual = df_to_html(
df_html_test,
image_cols='image',
hyperlink_cols=['link', 'image'],
html_tags={'title': 'h2', 'image': 'em'},
max_num_rows=None,
)
assert expected.replace('\n', '').replace(' ', '') == (
actual.replace('\n', '').replace(' ', '')
)
def test_df_to_html_with_nonexistent_cols(df_html_test):
with pytest.raises(ValueError):
df_to_html(df_html_test, image_cols='nonexistent_col')
with pytest.raises(ValueError):
df_to_html(df_html_test, hyperlink_cols='nonexistent_col')
with pytest.raises(ValueError):
df_to_html(df_html_test, html_tags={'nonexistent_col': 'mark'})