Fix sub-definitions not being listed #68
Conversation
They were deleted in the parse_examples() function, which would remove quotations, sub-definitions accidentally being treated as such.
Sub-definitions were previously appended to list as: ["Top definition A", "Sub definition ASub definition B", "Top definition B"] Now, they are appended to the "definitions" list as: [["Top definition A", "Sub definition A", "Sub definition B"], "Top definition B"] However, I am not sure if this is the best structure. You may change it if you find a better way.
|
Thanks for this! I'll take a look. Can you fix the failing tests in the meanwhile? |
Previously, if the top definition and its sub-definitions weren't separated by '\n', the code would fail.
|
Done. Everything should be fine now. However, I'll test some more words on my own, to make sure I didn't miss any edge cases. |
|
Thanks so much for this! I just tested it out with my script that uses WiktionaryParser and (with slight modification) it works great. Hope it gets merged soon! 😁 |
|
@Luca1152 @katacarbix while this change works, I'm not sure the definitions list should contain different data types. For your use cases, is it a problem if the sub-definitions are included in the list as top level definitions? @katacarbix what are the modifications you've had to make? Was wondering if the parser could be enhanced to include them
|
|
I added a section to loop through sub-definitions and output them indented. I also put back the colons that I was removing before for formatting. Old: word = parser.fetch(query)
entries = []
for section in word:
for defn in section['definitions']:
for entry in defn['text']:
if entry[:len(query)] != query:
entries.append('('+defn['partOfSpeech']+') ' + entry.rstrip(':'))New: word = parser.fetch(query)
entries = []
for section in word:
for defn in section['definitions']:
for entry in defn['text']:
if type(entry) is list:
entries.append('('+defn['partOfSpeech'])+') ' + entry[0])
for subentry in entry[1:]:
entries.append(' ' + subentry)
elif entry[:len(query)] != query:
entries.append('('+defn['partOfSpeech'])+') ' + entry) |
|
@suyash458 This PR fixes a very serious bug. Is this ever going to get merged and is pip ever going to get a new release? |
Previously, a word that had a layered definition, such as:
would return the following definition:
So the sub-definitions of
A tool with claws or hooks which is used to catch or hold something.weren't listed.This PR makes it so the returned definition becomes:
thus fixing #57.
However, this structure (using arrays to depict a layered definition, where the first element represents the top-level definition) may not be the most ideal one. Not sure.
This causes two of the tests to fail (
grappleandhouse), as they both have such sub-definitions that were previously ignored.