Another question

[Former user]

Hi,
I'm running the following code on the attached file

with open(synonym_filename,encoding='utf8', errors='ignore') as filehandle:
    taxonset = file_csv.load_synonyms(filehandle, synnamefield='Name', synauthfield='Author',
                                      accnamefield='Accepted_name', accauthfield='Accepted_author', get_info=True)

And I'm getting the error below. I tried debugging a little but I didn't quite understand how to overcome the problem.
Is this a bug or am I missing something?

Thanks

Traceback (most recent call last):
  File "D:\moshe\Dropbox\Dropbox\University\ploidb_src\chromcount\create_schema.py", line 628, in <module>
    log_file = input_tuple[4],authfield=authority)
  File "D:\moshe\Dropbox\Dropbox\University\ploidb_src\chromcount\name_resolve.py", line 160, in write_synonym_results
    accnamefield='Accepted_name', accauthfield='Accepted_author', get_info=True)
  File "C:\Python33\lib\site-packages\taxonome\taxa\file_csv.py", line 179, in load_synonyms
    taxonset.add(tax)
  File "C:\Python33\lib\site-packages\taxonome\taxa\collection.py", line 193, in add
    self._del_qname(taxon.name)
  File "C:\Python33\lib\site-packages\taxonome\taxa\collection.py", line 234, in _del_qname
    raise KeyError(qname)
KeyError: taxonome.Name("Huperzia pulcherrima", "(Wall.) T. Sen & U. Sen")

[Thomas Kluyver]

That looks like a bug. When it adds a taxon to the collection, it adds its names to an index. If its accepted name is already in the index for another taxon, it removes the previous copy of the name, and puts the new copy in place instead. Somehow, it's determining that the name is already in the index, but when it comes to remove it, it's not finding it. I'll have a look into it this evening.

[Thomas Kluyver]

Simplify equality comparison of authorities.

Closes issue #13

→ <<cset 93ab48bb0e7c>>

[Thomas Kluyver]

With a particular pair of authority names, the fuzzy comparison was returning a different result for a==b and b==a. The comparison mechanism seemed like a good idea at the time, but now it seems overly complicated. For the time being, I've replaced it with a much simpler comparison - it may miss some authority matches that it previously found, but it's much more predictable.

[Moshe Einhorn]

Hi Thomas, I work at the same lab with Lior and also encountered this problem. After updating the code from the collection2 branch and running taxonome again on the same data set I get a slightly different error:

2013-06-20 10:21:51,601 ERROR Exception caught while processing C:\temp\names\newzealand40-names.csv C:\temp\names\newzealand40-mapping.csv
Traceback (most recent call last):
  File "D:\moshe\Dropbox\Dropbox\University\ploidb_src\chromcount\create_schema.py", line 733, in <module>
    log_file = input_tuple[4],authfield=authority)
  File "D:\moshe\Dropbox\Dropbox\University\ploidb_src\chromcount\name_resolve.py", line 175, in write_synonym_results
    run_match_taxa (input_taxa,taxonset, tracker=trackers,nameselector=name_selector.NameSelector())
  File "C:\Python33\lib\site-packages\taxonome\taxa\collection.py", line 471, in run_match_taxa
    for taxonpair in streaming_match_taxa(taxa, target, **kwargs):
  File "C:\Python33\lib\site-packages\taxonome\taxa\collection.py", line 457, in streaming_match_taxa
    t2 = target.select(taxon.name, tracker=tracker, **kwargs)
  File "C:\Python33\lib\site-packages\taxonome\taxa\collection.py", line 143, in select
    for n, an, tid in possnames:
ValueError: need more than 2 values to unpack

Regards, Moshe

[Thomas Kluyver]

Thanks, I'd missed a change I needed to make. It should be fixed now, but let me know if not.

[Moshe Einhorn]

From a quick check it seems to be working fine now. Thanks.