This patch cures occured and makes occurred occur.
Note that I found them while testing bug 11170.
In a follow-up of 11170, I corrected this typo in parcels.tt.
This patch touches update22to30.pl and modborrowers.tt
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Fixes a typo in 2 files.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
The summary is built using the authtypecode selected from the interface.
So when a search is launch on all auth types, the summary is not
correctly built by the BuildSummary routine.
It should get the authtypecode from the authority (call to
GetAuthTypeCode).
To test:
1/ go to authorities/authorities-home.pl
2/ search <something> by authtype personal name
3/ results are displayed with summary
4/ now select the default entry and search again the
results display but without the summary
5/ apply the patch
6/ search default again, now summary shows
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Tested with a UNIMARC database, works as described.
All tests and QA script pass.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch fixes an error caught by t/db_dependent/Acquisition.t, and
adjusts C4::Context::IsSuperLibrarian() to return true if no
userenv is set. This is done on the basis that if no userenv is set,
calls to C4::Context routines are being made from a command-line script,
and if you have access to the command line of a running Koha instance,
you implicitly already have better than superlibrarian access.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
For DUE message (and PREDUE, etc.) there are no check before sending the
message to the message_queue table.
This check avoids to try to send again and again the same message. Now
it is marked as "failed".
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Without the patch a sms notice will remain as 'pending' forever.
With the patch applied, the status is set to 'failed'.
Passes all tests and QA script.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch adds a regression test for verifying that queued
SMS messages meant for patrons who have no SMS alert number
set are marked as failed after the first attempt to send them.
To test:
[1] Run prove -v t/db_dependent/Letters.t. The fourth
test should.
[2] Apply the main patch and run t/db_dependent/Letters.t
again. This time, all tests should pass.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
That's it. A guide box cannot be created if invalid data is passed.
Sponsored-by: Universidad Nacional de Cordoba
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script, includes new unit tests.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Note that I modify the return value. Before this patch, it returned an
empty string or 1. Now it returns 0 or 1.
Test plan:
- same as the original patch
- verify that unit tests pass:
prove t/Context.t
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script, including new tests.
Checked the code and tested superlibrarian behaviour in some places:
moremember.pl:
With IndyBranches only superlibrarian can delete borrowers from
other branches. Accessing the borrower with a direct link.
OK
C4/Members.pm
With IndyBranches only superlibrarian can search for borrowres
from other branches.
OK
tools/holidays.pl
With IndyBranches only superlibrarian can edit holidays for other
branches.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
The method of checking the logged in user for superlibrarian privileges
is obtuse ( $userenv && $userenv->{flags} % 2 != 1 ) to say the least.
The codebase is littered with these lines, with no explanation given. It
would be much better if we had one subroutine that returned a boolean
value to tell us if the logged in user is a superlibrarian or not.
Test Plan:
1) Apply this patch
2) Verify superlibrarian behavior remains unchanged
Signed-off-by: Joel Sasse <jsasse@plumcreeklibrary.net>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Comments on second patch.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
If circulation exports are enabled (by turning on ExportWithCsvProfile),
the table on the checkout page includes three columns of checkboxes --
'renew', 'checkin', and 'export'.
For each loan, the renew and checkout links should behave like radio
buttons, but the state of the export checkbox is meant to be independent
of the renew and checkin checkboxes.
However, if the 'select all' link in the export column is clicked,
active renew checkboxes are toggled off.
The desired behavior is that clicking the select all link in the export
column should only affect checkboxes in that column. This patch
implements this behavior.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script - one line JavaScript change.
Works as described.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Problem:
If you tell gather_print_notices.pl to write output to a location
you do not have write access to, it will silently fail to write the
data, but still mark unsent messages as sent.
Solution:
This patch adds two lines of defense:
1. Check that the location given for the output is writable
2. use "open() or die" instead of just "open()" when writing the
output
The first measure should catch most of the potential errors, but
I guess a directory can be writable, but the open() still can fail
because the disk is full or something similar.
To test:
- Make sure you have some unsent messages in the message_queue table,
that do not have an email adress
- Apply the patch
- Run the script, pointing at a location you do not have access to
write to. Check that the script exits with an appropriate error
message, and that the unsent messages are still unsent. Do this
both with and without the -s option.
- To fake passing the first line of defence, comment out line 62
and put this in instead:
if ( !$output_directory || !-d $output_directory ) {
- Run the script again as above, check you get an appropriate
error and that the message queue is not touched
- Reset line 62 to how it was
- Run the script against a directory you do have access to write to
and check that output is produced as expected and that messages
are marked as sent
- Sign off
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Works as described.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch fixes a problem where a patron could receive duplicate
hold waiting notifications. For example, this could happen if a
circ operator checked in an item more than once and confirmed the
same hold each time.
To test:
[1] Set up a test patron that received hold waiting notifications.
[2] Put an item on hold for the patron, then check the item in
and confirm the hold. Verify that a hold notification is
sent (or inspect the message_queue table).
[3] Check the item in again and confirm the hold again. A duplicate
hold notification will be generated.
[4] Apply the patch.
[5] Repeat steps 2 and 3. This time, only one notification should be
generated.
[6] Verify that prove -v t/db_dependent/Reserves.t passes.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Works as described.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch implements a regression test for verifying that
duplicate hold notifications aren't sent if ModReserveAffect() is
called repeatedly (as might happen if a circ operator accidentally
checks in an item and confirms its hold more than once).
Note that the test depends on the fact that _koha_notify_reserve()
defaults to sending a HOLD_PRINT letter if the borrower has not
specified an email or SMS hold notification.
To test:
[1] Run prove -v t/db_dependent/Reserves.t
[2] The 'patron not notified a second time (bug 11445)' test
should fail.
[3] Apply the main patch and run prove -v t/db_dependent/Reserves.t
again. This time all tests should pass.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch adds a help file to the Renew page found under
Circulation.
To test:
* Go to Circulation > Renew
* Click the help link
* Confirm text and manual link are correct.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Works as described.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Trivial patch to add 'cron-daemon' as dependency for the koha-common
package. 'cron' is usually pulled in any minimal Ubuntu/Debian
install, but in some circumstances (using debootstrap) it might be
absent.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
No entry in debian/control yet, but
according to comments in the file this file is generated
from control.in - so this should be ok.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
There were 2 INSERT in error.
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
I have gone ahead and fixed the typo pointed out by Mathieu:
Endommadgé-> Endommagé
Sample files install without problems, tests look good.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch adds some categories of values in French installer :
- SUGGEST
- OPAC_SUG
- REPORT_GROUP
- LOST
- DAMAGED
SUGGEST and OPAC_SUG are used by Suggestions module.
REPORT_GROUP is used by Reports module.
It also adds a new status for "ETAT" (en commande)
It creates a 995$2 subfield in french frameworks when it did not exist.
It links existing 995$2 subfield with LOST category.
It cleans up the list of authorised values installed with "Lecture
publique" framework :
- some codes are moved in general 1-Obligatoire/authorised_values.sql
(SUGGEST, REPORT_GROUP)
- some are suppressed, because they are also defined in
1-Obligatoire/authorised_values.sql (langue, COUNTRY, statut)
- the code for inserting the ones left is changed (I suppress the `id`
column)
To test :
1. Take a fresh new Koha
2. Install Koha choosing French installer and UNIMARC Lecture publique
3. Check the authorised values are imported
4. Check the cataloguing frameworks are usable :
especially 995 $2 field, which must be mapped with LOST values :
Perdu, Long retard, Perdu et remboursé, Introuvable
you can also check 101$a (language codes), 102$a (country codes)
5. In OPAC, make a suggestion. See if you can select a cause for your
suggestion ("Bestseller" or "'L'exemplaire en rayon est endommagé")
6. In staff interface, manage some suggestions. See if you can select a
cause for rejection or acceptation ("Bestseller", "Budget
insuffisant" etc)
7. In reports, see if you can sort reports according to values of
REPORT_GROUP ("Circulation", "Catalogue", "Adhérents" etc)
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Comments on second patch.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch:
- Makes the new subfield tab show maxlength=9999 as default (instead of
empty-then-zero).
- Updates the help to make exlpicit that 0 or empty defaults to 9999.
- Assumes all the subfields created with maxlength=0 inadvertedly are
meant to mean "no limit" and hence update the database to reflect
that.
To test (this patch and Pablo's):
- Edit a MARC framework, edit some field's subfields.
- Use the 'New' tab to create a new subfield (choose an unused letter).
- See in "More constrains" that the "Max length" field is empty. Leave
it as-is.
- Save the changes (the new subfield).
- Edit the field again, verify that "Max length" is 0.
- Try tu use the framework and the the field/subfield just created
> FAIL
- Apply the patches, upgrade
- Try to use the framework/field/subfield > SUCCESS (0 was converted to
9999)
- Repeat from the begining, "Max length" should show 9999 on the new
subfield tab.
- Leave it empty, it is saved as 9999.
Edit: small typo
Sponsored-by: Universidad Nacionald de Cordoba
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes QA script and tests in t and xt.
Tested:
- deleting an existing subfield
- adding a new subfield with new default 9999
- editing the new subfield, changing value to 8888
- deleting new subfield
- adding new subfield, using 8888 as length
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
The default value for the marc_subfield_structure.maxlenght is 9999
in the DB. Currently the template passes an empty value which is casted to
0 by the CGI.
This simple patch validates the input and converts to the default (9999)
if not defined or 0.
Another approach could be changing the 9999 default and/or treating 0 as
'no-limit'.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Works by defaulting 0 or "" to 9999.
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
When a z39.50 server isn't able to be searched successfully, the yellow
error box came up empty. This patch fixes the problem.
Test Plan:
1) Go to Administration/z39.50 servers
2) Create a fake z39.50 server with a made up address
3) Go to cataloging, search only that server
4) Note the empty yellow alert box
5) Apply this patch
6) Re-run the search, not the alert box has a message in it now
Signed-off-by: Nora Blake <nblake@masslibsystem.org>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Works according to test plan.
When one of the selected servers gives result no dialog
box is shown before and after applying the patch.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
The tests should be executed into a transaction and the SimpleSearch
routine correctly mocked.
Test plan:
Verify that prove t/db_dependent/XISBN.t returns green.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
A unit test fails in t/db_dependent/XISBN.t, the get_xisbn routine, if
ThingISBN is enabled, returns the 3rd biblionumber, not the second one.
Signed-off-by: Chris Cormack <chris@bigballofwax.co.nz>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
bulkmarcimport.pl can crash when searching for duplicates if the 005
field from the incoming or local record is not defined. This patch
fixes it.
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Test plan
1/ Create a record with no 005 field
2/ Try to import it checking for duplicates, notice it crashes
3/ Try with a record with a 005 field, but the one in Koha missing
one, still crashes
4/ Apply patch
5/ No more crash
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Patch fixes the problem described for importing authorities
with the bulkmarcimport.pl when trying to match with existing
records.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch fixes a bug where using the patron editor to add a new
restriction overwrote the first existing one.
Test Plan:
1) Edit a patron, add a restriction
2) Edit the patron again, add a second restriction
3) Note the first restriction has disappeared!
4) Apply this patch
5) Edit the patron again, add another restriction
6) Note the previous restriction is not longer removed
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Tested:
- Adding and removing multiple restrictions from
- the details tab
- the checkouts tab
- the edit patron form
All works as expected.
Patch passes all tests in t, xt, and QA script.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
ModMember supposes the password given in parameter is the
password string, so if it receives the encrypted password,
it will encrypt it again! By simply deleting the password key
from the hash, ModMember leaves the password unchanged.
Test plan:
1/ Create or choose a child patron
2/ Update it to an adult category using the
"Update child to adult patron" link
3/ Try to log in at the OPAC with this patron: It is not
possible, the password has changed
4/ Apply the patch and try again previous steps
Signed-off-by: Mark Tompsett <mtompset@hotmail.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Confirmed the problem and tested the patch fixes it.
Passes all tests and QA script.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
The -munge-config switch has been deprecated for years, and
trying to use it would either not work at all or, if it did "work",
almost certainly damage one's Zebra configuration for Koha.
This patch removes this switch.
To test:
[1] Run rebuild_zebra.pl and verify that no mention is made
of -munge-config.
[2] Run rebuild_zebra.pl to index records in one's test database
and verify that there are no regressions.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Removing a really dangerous option
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Ran rebuild_zebra.pl with various options and confirmed
that data was reindexed successfully.
No regressions found.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
In C4::Items::DelItemCheck, there are two SQL queries: one to check
if item is on loan, the other if item is reserved.
Those two queries use "SELECT * FROM table", fetch the data with
"$var = $sth->fetchrow", and use "$var" as a boolean condition.
This is not correct, SQL query should be "SELECT COUNT(*) FROM table".
As a consequence, it was possible to delete an item without warning to
the operator even if it was waiting on the hold shelf or in transit to
fill a hold.
This patch corrects the SQL queries and sets my ($var) to show that
fetchrow returns an array.
Test plan :
- Set an item A onloan
- Set an item B reserved and the reserve waiting
- Go to items cataloguing : cgi-bin/koha/cataloguing/additem.pl?biblionumber=XXX
- Try to delete item A
=> You get an alert and item is not deleted
- Try to delete item B
=> You get an alert and item is not deleted
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Works, and has the added bonus of being a tiny bit faster.
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes t, xt and QA script tests.
Also tried deleting via batch delete - correct warnings are displayed.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch adds unit tests for two parts of DelItemCheck: checking
if the item is on loan, and checking if it is waiting on the hold
shelf.
To test:
[1] Verify that prove -v t/db_dependent/Circulation/IsItemIssued.t
is successful.
[2] Verify that prove -v t/db_dependent/Reserves.t is *not*
successful -- as it turns out, there was a latent bug where items
waiting on the hold shelf or in transit to fill a hold could still
be deleted without any warning.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Currently, xmlControlfield.js is hard coded to look for XML files for
MARC21:
url: this.themelang + "/data/marc21_field_" + this.tagfield + ".xml",
This patch makes this code use the value from the marcflavour syspref,
as a preparation for making the NORMARC value builders use the XML
technique employed by the MARC21 value builders for 006 and 008.
To test:
- Make sure you have a MARC21 installation
- Set marcflavour = NORMARC
- Go to Cataloguing and start a new record with the default framework
- Open the value builders for 006 and 008 and observe that they still work, showing
the coded values for MARC21
- Apply this patch
- Check the value builders for 006 and 008 and observe that you get a truncated view
with an empty "Select a type of material" dropdown
- Use e.g. the Net console in Firebug to observe requests to
http://localhost/intranet-tmpl/prog/en/data/normarc_field_008.xml
that result in a 404 status
- Set marcflavour = MARC21
- Observe that the value builders for 006 and 008 are now fully working
- 006 and 008 should be the only value builders affected by this change, since
they are the only ones using xmlControlfield.js, but please also verify that
other value builders are still working as expected
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Some libraries would like to be able to search on the sort1 and sort2
fields of patron records.
Test Plan:
1) Apply this patch
2) Add various values for sort1 and sort2 some patrons
3) Browse to members-home.pl
4) Run searches on sort1 and sort2
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch adds language-original to the list of search fields
recognized by QueryParser.
To test:
[1] After doing the tests in the main patch, copy the configuration
file etc/searchengine/queryparser.yaml into place, turn on the
UseQueryParser system preference, and verify that searching on
language-original still works.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
It could be useful to index the original language of a document (i.e.
"fre" for the English translation of a French novel).
This patch renames the Bib-1 use attribute 1095 from
Code-language-original to language-original and uses it to index:
- MARC21 041$h subfield
- UNIMARC 101$c subfield
It adds "language-original" in the list of index in Search.pm.
Test plan :
A. in a MARC21 GRS1 environment
1. Copy Zebra config files (zebradb/biblios/etc/bib1.att,
zebradb/ccl.properties, marc_defs/marc21/biblios/record.abs) from
your source etc/ directory to your main koha etc/ directory
2. Reindex zebra
3. Make some searches, like "language-original:fre"
B. in a MARC21 DOM environment
4. Copy Zebra config files (zebradb/biblios/etc/bib1.att, zebradb/ccl.properties,
marc_defs/marc21/biblios/biblio-zebra-indexdefs.xsl) from your source etc/
directory to your main koha etc/ directory
5. Reindex zebra
6. Make some searches, like "language-original:fre"
C. in a UNIMARC GRS1 environment
7. Copy Zebra config files (zebradb/biblios/etc/bib1.att,
zebradb/ccl.properties, marc_defs/unimarc/biblios/record.abs) from
your source etc/ directory to your main koha etc/ directory
8. Reindex zebra
9. Make some searches, like "language-original:fre"
A. in a UNIMARC DOM environment
10. Copy Zebra config files (zebradb/biblios/etc/bib1.att,
zebradb/ccl.properties, marc_defs/unimarc/biblios/biblio-zebra-indexdefs.xsl)
from your source etc/ directory to your main koha etc/ directory
11. Reindex zebra
12. Make some searches, like "language-original:fre"
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
When item is transfered from items table to deleted items, all fields
must be copies but "timestamp".
This value must be updated to know when the item was deleted.
Test plan:
- Look a an item timestamp :
mysql> select timestamp from items where itemnumber = 2690;
+---------------------+
| timestamp |
+---------------------+
| 2011-09-09 15:30:21 |
+---------------------+
1 row in set (0.00 sec)
- Delete this item in cataloguing module
- Check it is not in items table anymore :
mysql> select timestamp from items where itemnumber = 2690;
Empty set (0.00 sec)
- Look in deleteditems table :
mysql> select timestamp from deleteditems where itemnumber = 2690;
+---------------------+
| timestamp |
+---------------------+
| 2013-12-05 15:33:20 |
+---------------------+
1 row in set (0.00 sec)
=> timestamp as been set to actual date/time
Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Patch set passes koha-qa.pl, works as advertised!
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This is supplementary to the main patch for
bug 6331. Having removed the attribute marc from
items DelItem, we should not try to populate it.
Signed-off-by: Fridolin Somers <fridolin.somers@biblibre.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
There is a difference between "items" and "deleteditems" tables
in "kohastructure.sql"
"deleteditems" has a field "marc" not existing in "items".
This patch removes this obsolete column.
Test :
- after deleting an item, check that the deleted item is properly
stored in deleteditems table
- check that the column marc has been deleted from deleteditems table
Signed-off-by: Mathieu Saby <mathieu.saby@univ-rennes2.fr>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch updates the DBIC schema class for Suggestion
to reflect the dropped default value for the suggesteddate
column.
To test:
[1] Create an empty Pg database and use the deployment script
being worked in in bug 11390. The deployment shoudl
succeed without reporting any errors regarding the
suggestions table.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
The 'default 0' clause got translated as an invalid constant
default of '0000-00-00' when DBIx::Schema is used to deploy
the suggestions table into a Pg database. This patch drops
the default.
To test:
[1] Apply the patch and run the SQL specified in the database
updated.
[2] Verify that the suggestions table no longer has an
explicit default value for the suggesteddate column.
[3] Verify that prove -v t/db_dependent/Suggestions.t
passes.
[4] Verify that installer/data/mysql/kohastructure.sql runs
cleanly in an empty database.
[5] Verify that there are no visible regressions of the
purchase suggestions functionality.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Having a default of 0 on a date seems like a mad thing to do anyway,
so good to get rid of it
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch updates the DBIC schema class for CollectionTracking
to reflect the new name of its primary key column.
To test:
The CollectionTracking class is not currently used, but
if you *really* want to test this, take a look at the following
branch: http://git.librarypolice.com/?p=koha-galen.git;a=shortlog;h=refs/heads/pg
Then, set up a PostgreSQL database, update koha-conf.xml to point to it,
then run pg/deploy and verify that the collections_tracking table is created
in the Pg database.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
'ctId' as a column name conflicts with one of the system
columns that PostgreSQL uses for each table, and consequently
needs to be renamed to enable deploying the schema to a Pg
database. This patch makes this change.
To test:
[1] Apply the patch and run the SQL specified in the database
updated.
[2] Verify that the collections_tracking table no longer has
a ctId column, but now has collections_tracking_id.
[3] Verify that prove -v t/db_dependent/RotatingCollections.t
passes.
[4] Verify that installer/data/mysql/kohastructure.sql runs
cleanly in an empty database.
This patch does not affect user-visible behavior given the fact
that the rotating collections feature is currently disabled.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
This patch removes the legacy Pg schema and MARC framework scripts
as they're out of date. They will be replaced by use of DBIx::Class
to deploy the schema. Loading the sample data and settings will be
accomplished either by making the current scripts in installer/data/mysql
DBMS-independent (or, at least, able to be processed by both MySQL and Pg),
converting them to flat text files and writing code to load them, or a
combination of the two approaches.
To test:
[1] Verify that installer/data/Pg is removed. There is some code
in C4::Installer that refers to that directory, but it cannot
be reached through normal means.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Signed-off-by: Chris Cormack <chrisc@catalyst.net.nz>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Test plan:
On the MARC modification tool:
Add/edit a new action on a field and define a condition on the same
field.
Verify that you get a warning message in red.
See bug 11413 for more information
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Tested adding and editing a template with the same field in the
action and the condition.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
Problem summary: when doing partial receives for the given order -
1) if AcqCreateItem is set to 'ordering', various item data (price,
dateaccessioned, replacementprice, replacementpricedate) are getting
erroneously updated on the wrong (yet to be received == not the ones
being currently received) item records
2) if AcqCreateItem is set to 'receiving', newly received
item records are being created without the aforementioned fields
set to the proper values
This (trivial) patch should deal with both cases, hopefully without
breaking enything else.
To test:
- apply the patch,
- create some orders with 2+ quantity
- test partial & non-partial receives for those orders
- ensure the received item records are getting modified
(for AcqCreateItem set to 'ordering') and/or created (for AcqCreateItem
set to 'receiving') correctly for both partial and non-partial receives
- receiving orders with quantity = 1 / receiving orders in non-partial
mode should be still working fine for 1) & 2) scenarios (i.e.,
AcqCreateItem set to 'ordering' / AcqCreateItem set to 'receiving')
Signed-off-by: Nicole C. Engard <nengard@bywatersolutions.com>
Works as I'd expect now! Awesome patch.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Also: t/db_dependent/Acquisition/
t/db_dependent/Acquisition.t
Created 2 orders with 3 items each for both settings
of AcqCreateItem (on receive, on order) with the patches
applied. No regressions found.
Closed baskets and received shipments for each, with
AcqCreateItem set according to how the order was created.
First recreated the problem without the patches, reloaded
database and confirmed that the patch fixes it.
No problems found.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
