restructured API response #4
Conversation
model/node_differ.py
Outdated
| curr_text = curr_text[c["curr"]['offset']:c["curr"]['offset']+c["curr"]['size']].replace("\n", "\\n") | ||
| is_edit_type_found,wikitext,edit_type = is_edit_type(prev_text,c['prev']['type']) | ||
| #check if edit_type in edit types dictionary | ||
| if edit_types.get(edit_type,{}) and is_edit_type_found: |
There was a problem hiding this comment.
I'm not sure the is_edit_type_found variable is having an effect right now. My understanding is that if it's False, you probably don't want to do increment any counts (just continue). If I'm reading the code right, currently if it's False, it still leads to a clause that updates the edit_types dictionary. So maybe instead just add something like: if is_edit_type_found: continue after running the is_edit_type function. Then you don't have to include it in any of the other if/else statements. This applies to each of the insert/remove/change clauses (not just this specific one).
model/node_differ.py
Outdated
| is_edit_type_found,wikitext,edit_type = is_edit_type(prev_text,c['prev']['type']) | ||
| #check if edit_type in edit types dictionary | ||
| if edit_types.get(edit_type,{}) and is_edit_type_found: | ||
| if edit_types.get(edit_type,{}).get('change'): |
There was a problem hiding this comment.
this whole if/else block can be simplified I think to the following: edit_types[edit_type]['change'] = edit_types[edit_type].get('change', 0) + 1
Again, same goes for the corresponding insert/remove clauses higher up in the code
There was a problem hiding this comment.
Thanks for the review, Isaac! I'd implement these changes.
update tree differ to do moves and merge text changes into single blo…
|
Some quick high-level thoughts:
|
That was a mistake. I'd fix and make a commit. |
|
Thanks! Merging |
No description provided.