EDIT: TL;DR version
I typed this
CREATE INDEX IF NOT EXISTS IDX_FILE_SIZE table_name (file_size);
instead of this
CREATE INDEX IF NOT EXISTS IDX_FILE_SIZE ON table_name (file_size);
Don't do that.
Some silly questions:
Is it a concidence that the offending statement is missing the word ON?
CREATE INDEX IF NOT EXISTS IDX_FILE_FULLPATH_FILE_PARENT_DIR ON table_name (file_fullpath, file_parent_dir);
CREATE INDEX IF NOT EXISTS IDX_FILE_SIZE table_name (file_size); -- missing ON
CREATE INDEX IF NOT EXISTS IDX_TAG_TITLE ON table_name (tag_title);
Somewhere in all the verbiage in your question, did I see the phrase "syntax error"?
Did you try the simple step of running the SQL statements in the sqlite3 command interpreter and seeing which syntax error you were actually getting?
E.g.
SQLite version 3.6.14
Enter ".help" for instructions
Enter SQL statements terminated with a ";"
sqlite> create table foo (bar int, zot int);
sqlite> create index barx on foo(bar);
sqlite> create index zotx foo(zot);
SQL error: near "foo": syntax error
sqlite>
Have you considered perusing TFRRD (The Fantastic Rail-Road Diagram) in the docs?
You wrote: """when I run that command in the smaller script (verifyIndexSmaller), it gives no error. If I then try to run the larger script again, even though the index has been created by the smaller script, I still get the error""".
Have you considered the possibility that you didn't run that command in the smaller script, but actually ran another (fixed!) version of that command?
Do you now understand why S.Lott was trying to get you to cut the waffle and focus on the piece of SQL that was causing the error?
Related
c1.execute(f"create table {sub_table_name1}(srno int, data_name varchar(255), data varchar(255))")
the error is
mysql.connector.errors.ProgrammingError: 1064 (42000): You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '(srno int, data_name varchar(255), data varchar(255))' at line 1
i swear to god i have no idea what is wrong...i'm still a beginner to python and mysql connection and any help is welcome
What is the value of your sub_table_name1 variable at the time you execute that? If it's empty, then I think that error would be expected, because the SQL statement would be:
create table (srno int, ...
MySQL will see that as a missing name. The way it informs you of the error is to tell you the part of the query at the point it got confused:
for the right syntax to use near '(srno int, ...
This means when it saw (srno int, ... it was expecting something else — I'm guessing it was expecting an identifier for the name of the table, but reached the open parenthesis character instead.
Since you said you are new to python (and perhaps to programming), here's a general tip: question your assumptions. You assumed the create table statement had a table name in it. Does it? How would you know? Can you add code to print the SQL statement string after you format it with the variables? Can you add code to print the sub_table_name1 variable before using it in the string? Of course, you should remove such code after you are done debugging.
"The most effective debugging tool is still careful thought, coupled
with judiciously placed print statements."
— Brian Kernighan, "Unix for Beginners" (1979)
That's an old book, but the idea is still true!
I am trying to run a SQL query using psycopg2 in python. The query is supposed to create a set of points (in the {nome1} identifier) for each polygon (in the {nome2} identifier, which was created from the segmentos.json I'm still using to run the for loop) they intersect and save the intersection to a new table (the {nome0} identifier). This bit works fine.
The problem is I want the new {nome0} table to remain droped in case there are no points that intersect the polygon I'm iterating. When I run this code, I get the following error:
SyntaxError: syntax error at or near "IF"
Should I be using a function to run IF statements in psycopg2? What am I missing here?
Thanks in advance!
intersecta="""DROP TABLE IF EXISTS {nome0};
IF ST_Intersection({nome2}.geom,{nome1}.geom)=true THEN
SELECT {nome1}.*, {nome2}.nome INTO {nome0}
FROM {nome2} INNER JOIN {nome1} ON ST_Intersects({nome2}.geom, {nome1}.geom) AND {nome2}.nome=%s;
END IF;
"""
for feature in segmentos.json()['features']:
nomedosegmento=feature['properties']['nome']
nomecompleto=str(feature['properties']['nome']+'_'+tabelagerada)
cur.execute(sql.SQL(intersecta)
.format(nome0=sql.Identifier(nomecompleto),nome1=sql.Identifier(tabelagerada),nome2=sql.Identifier(segmentnome)),[nomedosegmento,])
#print(cur.query)
conn.commit()
I have tried everything and keep getting this error:
pymysql.err.ProgrammingError: (1064, "You have an error in your SQL syntax;
check the manual that corresponds to your MariaDB server version for the right syntax to use near
'INSERT INTO tabSingles (doctype, field, value) VALUES ('Bank Reconciliation', 'a' at line 2")
Expanded query (after python format expansion):
SELECT value INTO #var FROM tabSingles WHERE doctype = 'Bank Reconciliation' AND field = 'bank_account';
INSERT INTO tabSingles (doctype, field, value) VALUES ('Bank Reconciliation', 'account', #var);
DELETE FROM tabSingles WHERE doctype = 'Bank Reconciliation' AND field = 'bank_account';
Can anyone see the problem? Is there some issue with multi-line queries? I have tried the individual lines on the mariadb command line and they appear to work as expected. I've also tried both frappe.db.sql and multisql (thought it meant multiline sql but doesn't). If I comment line 2 out, it also errors on line 3. Sorry to disturb but I've been staring at this for hours and cannot figure it out!
EDIT:
The obvious answer is this, but I'd still like to know why it doesn't like the original query:
UPDATE tabSingles SET field='{new_name}' WHERE doctype='{doctype}' AND field='{old_name}';
For security reasons (mainly SQL injection) MariaDB (and MySQL) servers don't support the execution of multiple SQL statements by default.
For supporting multiple statements execution the client needs to send COM_SET_OPTION command and MYSQL_OPTION_MULTI_STATEMENTS_ON flag to the server, which is not supported by PyMySQL.
Do not try to run more than one statement in a call.
Do use BEGIN and COMMIT.
Do use FOR UPDATE.
You need 5 separate commands:
BEGIN;
SELECT ... FOR UPDATE; -- to keep other connections from messing with the row(s).
UPDATE ...;
DELETE ...
COMMIT; -- do all of the above "atomically"
When I make a MySQL table order, it is created successfully but, when I execute any query against it, it says "error 1064 , syntax error".
When I change the name to orders, it works fine.
But I don't want to change the name. How can I execute our query against the order table?
can you use something like?
select * from `order`
The word order is actually an SQL keyword. You would have the same problem if you tried to use a table called group or select. You can fix it is MySQL by using quotes around it, along the lines of:
select f1, f2 from `order` where blah blah blah ...
However, unless your table will only ever hold a single order (in which case it won't do so for long since the underlying business will soon be bankrupt), you should probably call your table orders.
That solves both your problems, the one you found and the one you didn't :-)
I got here because I was searching for similar solution for SQL CE. There using
order
'order'
"order"
doesn't work.
What worked was:
[order]
Maybe it'll help someone else also.
This should fix the problem:
e.g
mysql>
Create table order(
ID char(5),
QTY(3)
)
;
The following query causes python to crash ('python.exe has encountered a problem ...'
Process terminated with an exit code of -1073741819
The query is:
create temp table if not exists MM_lookup2 as
select lower(Album) || lower(SongTitle) as concat, ID
from MM.songs
where artist like '%;%' collate nocase
If I change from "like" to = it runs as expected, eg
create temp table if not exists MM_lookup2 as
select lower(Album) || lower(SongTitle) as concat, ID
from MM.songs
where artist = '%;%' collate nocase
I am running python v2.7.2, with whatever version of sqlite that ships in there.
The problem query runs without problem outside python.
You didn't write the database system/driver you are using. I suspect that your SQL is the problem. The % characters needs to be escaped. Possibly the db driver module tries to interpret %, and %) as format chars, and it cannot convert non-existing parameter value into a format that is acceptable by the database backend.
Can you please give us concrete Python code? Can you please try to run the same query but putting the value of '%,%' into a parameter, and pass it to the cursor as a parameter?