The script, it seems, is incomplete. It only takes the attachments table and
ticket_change table and migrates them from one .db file to the other, creating
entries in the ticket table as it goes. It creates new ticket numbers as it
goes, I think. That’s what it looks like, at least.
Oh, and we’re using sqlite3. And I found this script online, didn’t write it
all.
Attached is the script.
To get it to work I had to add a column to the attachment and ticket_change
tables to account for the product, defaulting to the Default product.
Seems the information is in the database, but the tickets don’t show up in the
UI.
Patrick
"""
Usage: %s source.db target.db new_component
Copies all tickets, ticket changes and ticket attachments from the source database
to the target database, and updates the 'component' field on each ticket.
* 'component' is replaced and 'milestone' removed from tickets. Customize the
script if you don't / need this behaviour.
* Make backups yourself!
* No attempt is made to preserve ticket IDs from the source.
* Ignores the ticket_custom table - not sure what it's for, it always
seems to be empty.
* Tries to be as generic, but makes some assumptions about the
database schema when accessing columns via indices.
* Expects both databases to be based on the same trac-version, i.e.
have the same schema.
"""
import sys, os
import sqlite3 as sqlite
def main(argv):
try:
source_file, dest_file, new_component = argv
except ValueError:
print("Usage: source.db target.db new_component")
return 1
# connect to databases
source_conn = sqlite.connect(source_file)
source_cur = source_conn.cursor()
dest_conn = sqlite.connect(dest_file)
dest_cur = dest_conn.cursor()
qmarks = lambda seq: ','.join(['?' for r in seq])
try:
# go through tickets in source
tickets = source_cur.execute('SELECT * FROM ticket;')
for ticket in tickets:
# delete the id column - will get a new id
old_id = ticket[0]
ticket = list(ticket[1:])
# reset values of component and milestone rows
ticket[4-1] = new_component # component
ticket[11-1] = None # milestone
# insert ticket into target db
print("copying ticket #%s" % old_id)
dest_cur.execute('INSERT INTO ticket '+
'('+(','.join([f[0] for f in source_cur.description[1:]]))+') '+
'VALUES('+qmarks(ticket)+')',
ticket)
new_id = dest_cur.lastrowid
# parameters: table name, where clause, query params, id column index, table repr, row repr index
def copy_table(table, whereq, params, id_idx, trepr, rrepr_idx):
cur = source_conn.cursor()
try:
cur.execute('SELECT * FROM %s WHERE %s'%(table, whereq), params)
for row in cur:
row = list(row)
row [id_idx] = new_id
print("\tcopying %s #%s"%(trepr, row [rrepr_idx]))
dest_cur.execute('INSERT INTO %s VALUES(%s)'%(table,qmarks(row)), row)
finally:
cur.close()
# copy ticket changes
copy_table('ticket_change',
'ticket=?', (old_id,),
0, 'ticket change', 1)
# copy attachments
copy_table('attachment',
'type="ticket" AND id=?', (old_id,),
1, 'attachment', 2)
# commit changes
dest_conn.commit()
finally:
dest_conn.close()
if __name__ == '__main__':
sys.exit(main(sys.argv[1:]))
On Feb 15, 2015, at 10:18 PM, Olemis Lang <[email protected]> wrote:
> On 2/15/15, Patrick West <[email protected]> wrote:
>> I have about 225 tickets in our new bloodhound installation.
>>
>> I've been able to rescue the old trac database from a server that had
>> crashed.
>>
>> And now I want to migrate all the old trac tickets into our new bloodhound
>> installation.
>>
>> But I'm not finding a way to do this.
>>
>> I tried an old python script that migrates tickets from one trac to another,
>> and it looks like the tickets are in the new db, but I don't see them
>> anywhere in the GUI. I defaulted all the products to DEF.
>>
>
> What does the script do ? Could you please provide more details about
> what you did ?
>
> --
> Regards,
>
> Olemis - @olemislc
>
> Apache(tm) Bloodhound contributor
> http://issues.apache.org/bloodhound
> http://blood-hound.net
>
> Blog ES: http://simelo-es.blogspot.com/
> Blog EN: http://simelo-en.blogspot.com/
>
> Featured article:
>
