On the same way as late issues, the serial status should be updated to
'claimed' when the issues as exported as csv.
QA note: The updateClaim and UpdateClaimdateIssues subroutine did almost the
same thing, I kick the second on off to centralize the code.
Test plan:
1/ Export some late issues as CSV (serials/claims.pl).
2/ Refresh the page (the export does not do it) and confirm that the
status, the claim date and the claim count have been updated.
3/ Select some others issues, select a notice and send the notification.
Confirm that the behavior is the same.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
While testing this bug I noticed the tests printed useless diags.
This patch removes them for good.
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
When getting the next frequency number, use the real number of days for each month and each year.
Unit tests updated.
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
This guess uses the following algorithm
1/ Get the number of days remaining from first issue of this unit of
time until the end of the unit of time (using 7 days for week, 28 for
month and 365 for year)
2/ Divide it by the number of issues per unit
3/ Take the integer part of the result, this is the number of days to
add to previous serial published date
This patch also try to guess the date of the first issue of a unit of
time. Before that, it was:
- for week: monday of week
- for month: 1st day of month
- for year: 1st day of year
Now it uses firstacquidate.
Example:
firstacquidate = 1970-01-07 (wednesday)
unit = week
issuesperunit = 2 (2 issues per week)
From wednesday to sunday, there are 4 days so we add 2 days to get the
next date.
The next dates should be:
- 1970-01-09 (+ 2 days)
- 1970-01-14 (wednesday of next week)
- 1970-01-16 (+ 2 days)
- ... and so on
Test plan:
1/ Go to numbering pattern creation page (Serials -> Manager numbering
patterns -> New numbering pattern)
2/ In "Test prediction pattern" fieldset, set subscription length to 50
issues (this avoid to compute the next 1000 dates, which can take time)
3/ Fill first issue publication date
4/ Choose the frequency you want to test
5/ click on "Test pattern" button
6/ Check that the dates are correct
7/ Repeat steps 3-7 until you think you have tested enough cases
Interesting frequencies to test:
- 1/day
- 1/week
- 2/week
- 6/week (firstacquidate should be monday or tuesday, or the generated
dates will be the same for all the week)
- 1/month
- 2/month
- 3/month
- 1/year
- 2/year
- 3/year
----------------
Sign off report:
----------------
-Followed the test plan. Unit tests work. Also trying some more strange
frequencies with our librarians:
5/year, 43/year, 10/year.
-Received all serials for a subscription and found no errors.
-Received some serials with frequency 2/month pre-patch, getting publication
date as 1st of every month for each magazine. Having applied the patch,
publication dates started working mid-subscription for new magazines.
-Works well and fine!
Signed-off-by: Olli-Antti Kivilahti <olli-antti.kivilahti@jns.fi>
Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
As C4::Bookseller is no longer imported in C4::Acquisition the tests
fail because of an undefined function call.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
However, this is an unreasonable assumption for a system which
is in use (either lots of testing or production).
TEST PLAN
---------
1) Have a supplier with a late subscription.
2) prove t/db_dependent/Serials/Claims.t
-- will fail
3) apply patch
4) prove t/db_dependent/Serials/Claims.t
-- success
5) run koha qa test tools
Signed-off-by: Paola Rossi <paola.rossi@cineca.it>
Signed-off-by: Jonathan Druart <jonathan.druart@biblibre.com>
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
this patch:
- reintroduces the ISSN column
- fix a wording (already there before the main patch)
- fix the tests if a serial was already in late
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Small change: Removed 2 tabs in claims.tt and fixed order
of ISSN/Issue number column descriptions as they were switched.
Note: The <order> tags are currently not stripped out of the
notice.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
This patch adds a new DB field serial.claims_count
This field already exists for late orders. It makes sense to introduce
it for serial.
Test plan:
0/
a) Does not apply the patch.
b) Remove all your claimissues notices and be sure you have some serial issues
in late.
c) remove email address for the vendor you will use.
d) remove email address for the logged in user.
e) Export claims using the csv export => The selected issues will be
marked as claimed.
f) logout/login (to update the email address).
1/ Apply the patch and execute the updatedb entry.
2/ Go on the Serials > Claims page
3/ Verify that you get a warning message 'No claimissue notice defined'
4/ Verify the vendor list is correct (with the number of serial in late.
You should not get any changes here.
5/ Select one vendor and verify that the issue which was claimed before
has a claim count set to 1.
6/ Verify that you are not able to send notification to the vendor.
7/ Create a claimissue notice.
Something like:
<<LibrarianFirstname>>
<<LibrarianSurname>>
The following issues are in late:
<order><<biblio.title>>, <<biblio.author>> (<<biblio.serial>>)</order>
8/ Go on the Serials > Claims page, the warning message does not appear
anymore.
9/ Select issues. Select a notice. And "Send notification".
You should get an error (no email defined for this vendor).
10/ Add an email for the vendor.
11/ Select issues. Select a notice. And "Send notification".
You should get an error (no email defined for your user).
12/ Add an email address to your user
logout/login
13/ Select issues. Select a notice. And "Send notification".
You should get a happy message: the email has been sent!
14/ The email will contain the order tags if bug 12851 is not
pushed/applied.
Signed-off-by: Paola Rossi <paola.rossi@cineca.it>
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Works as described, some small issues fixed in a follow-up.
Note: If you change the email address of your staff user, you will
have to log out and back in to make the change take effect.
Signed-off-by: Tomas Cohen Arazi <tomascohen@gmail.com>
Show 'Unknown' when planneddate and publisheddate cannot be calculated
Also fixes SQL query in misc/cronjobs/serialsUpdate.pl that was still
using "periodicity != 32" to exclude irregular subscriptions from
results
Test plan:
1) Create a subscription in the serials module. Make sure to choose:
Frequency = Irregular
2) Test the prediction pattern, first publication date is set to
"First issue publication date" field, others will show as
'unknown'
3) Save the subscription
4) Check the created issue - it will show a published date and a
planned date (same as "First issue publication date" field)
5) Receive the issue and check the next generated issue, planned
date and published date should show as 'Unknown'
6) Generate a next issue, planned date and published date should
also show as 'Unknown'
Signed-off-by: Bernardo Gonzalez Kriegel <bgkriegel@gmail.com>
Work as described following test plan.
No koha-qa errors
Signed-off-by: Katrin Fischer <Katrin.Fischer.83@web.de>
Passes all tests and QA script.
Also tested:
- multi receiving generates mulitple issues without dates - 'unknown'
- staff detail page shows the dates empty, which is fine
- OPAC detail page shows the dates empty, which is fine
- serial collection page shows 'unknown' and those issues appear
on the 'manage' tab, as they did in the past
- Editing the issue from the serial collection page leaves the
date fields empty.
- Receving the issue, setting the status to 'Arrived' the Expected on
date is set to 'today' automatically. Date published has to be
entered manually (maybe something we could improve later
- subscription detail > issues tab shows Uknown.
- t/db_dependent/Serials/GetNextDate.t pass.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
These tests are split in two parts: "db dependent" and "db
independent".
GetNextSeq requires DB only when checking for irregularities.
Signed-off-by: Galen Charlton <gmc@esilibrary.com>
