使用Pyodbc查询Vertica时的python缓冲区大小

Question

i am using pyodbc to connect to Vertica DB and then i am using cursor to execute a query but the results of the query include a text field that is huge, the execute runs but it never returns anything, not even an exception. I can see this when i debug and break point at row=cursor.fetchall() It just comes as empty just like this "[]"

If i try with another dataset with less characters in this huge text field then the execute runs and i get data. I am suspecting it has to do with buffer size when retrieving the data but i don't know. Any suggestions on how to do this? Maybe increase buffer size somehow?

Thanks

import requests, pypyodbc, json, pyvertica, pyodbc
from pyvertica import connection
    conn=pyodbc.connect("DSN=HPVertica;UID=muffin;PWD=muffin")
    cursor=conn.cursor()
    query="SELECT field1_int, field2_bigText  FROM someTable WHERE field1_int=103;"

    try:
        cursor.execute(query)
    except Exception as e:
        print('err: {}'.format(e.args[0]))
        print(': {}'.format(e.args[1]))

    row=cursor.fetchall()

Answer 1

I believe this is a known issue in pyodbc, at least I am sure I've seen this complained about before and also some random patches. I just looked at the pyodbc code and it looks like they very recently fixed the issue in this commit a couple weeks ago.

It doesn't appear to be in the latest (as of this post) tag of 3.0.10. You'd either have to try the bleeding edge or wait just a little longer for it to be included in a future tag. (Or patch it yourself based on that commit, I guess).

Answer 2

我使用pypyodbc而不是pyodbc解决了这个问题，不再出现缓冲区错误