The upshot is that despite appearances, whe should NEVER use Text::CSV::Unicode
or any CSV module in non-binary mode. Text::CSV works in binary mode
by using Text::CSV_XS, so the behavior is the same for both (in binary).
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Changed so that issues.issuedate is never modified
during a renewal, since that column records the original
date of the loan. Changed the name and interpretation
of the $issuedate parameter of AddRenewal() to
$lastreneweddate, allowing (e.g.) offline circulation
to set the date of the renewal without changing the
issue date.
As a result of the original bug, issues.issuedate can be
set to NULL for loans that were renewed via the OPAC,
self checkout, or the staff interface when explicitly
renewing a loan. Loans that were renewed by checking
the item out to the same patron will have the issue date
changed to the date of the last renewal.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
CalcFine returned values that mismatched expectations in fines.pl.
fines.pl refactored: added debugging, prevent needless recreation of
Calendar objects by storing them in hash by branch.
Still outstanding problems with fines, including the output of a field
that has no other references in Koha (so is always undef) and the
incorrect description of FinesMode.
Calendar exported "new" erroneously. I also cleaned up the queries to
avoid needlessly compiling additional statement handles.
Please test and consider application to 3.0 maintenance.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
I'm adding some tests for C4::Circulation methods that I'm altering
to allow the offline circulation tool to use C4::Circulation to upload
its data. These test a bit of the old functionality and try to show
that the new functionality does what I think it does.
C4::Circ::Addissue to tests issuedate
these also test AddRenewal.
tests for C4::Circ::MarkIssueReturned
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
switching from 'prove' to 'perl' to run the test scripts when we're preparing
them to be submitted to smolder. Prove adds an additional summary at the end
of the test run which confuses the TAP parser in smolder. It thinks that an
additional test has been run, and it fails it.
Then, I added a '-' before the command in the Makefile. THat's so that even
if 'perl' returns a failing code (because some tess failed), the rest of the
rules are run and a report gets submitted. Othewise, only passing reports
get sent.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
This patch fixes the tests so that they include the new required parameters for longoverdues.pl.
This patch also doesn't include a test script that accidently got in the last one.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
This patch adds a target in the t/Makefile and a script in t/smolder_smoke_signal
to submit the results of test runs to a smolder server. It will allow users to submit
their own test results to a central location by running "make submit" from the t/
directory.
Smolder is a web-based application that aggregates and presents the output of
tests. We have an instance running at:
http://arwen.metavore.com:8000/app/public_projects/smoke_reports/1
More information about using smolder for Koha is available at:
http://wiki.koha.org/doku.php?id=en:development:smoke_reports
I intend to have the test suite run each night automatically and the results
sent to the smolder server. This will help us keep track of the current state
of the the code and possibly catch bugs introduced by changes. I encourge others
to submit their own test results if they want, especially if they are on an unusual
platform or are with a non-standard configuration.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
The tests I wrote for C4::Items::GetItemsForInventory confused the differences
between biblionumber and itemnumber. That wasn't uncovered on my limited test
database, but I uncovered it later.
This fixes that problem by populating a $self->{'items'} list with details of any items
added by KohaTest::add_biblios. Then, tests can probe there for the details of items
they should expect to find when searching.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
This test suite tests the several different ways that we can call C4::Koha::displayServers.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
C4::Koha::get_itemtypeinfos_of was not using plceholders, opening itself up to
potential SQL injection attacks. This patch refactors it to use placeholders to
bind parameters.
I also had to extend C4::koha::get_infos_of to allow us to pass bind parameters into it.
I'm including a test module for C4::Koha::get_itemtypeinfos_of.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
The SQL in C4::Items::GetItemsForInventory wasn't using placeholders and
bind parameters, possibly leaving itself open ot SQL injection attacks. This
patch changes that.
I've also incliuded a test module for C4::items::GetItemsForInventory.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Here are a few improvments to the test suite to make it easier to write some tests
for C4::Items
I extracted "tomorrow" and "yesterday" methods from a test module into the base class
so that they could be used by multiple test modules
Adding callnumber to items added in the test suite.
I recatored KohaTest::add_biblios a bit to remove the manual count of the number of
MARC::Fields that were added.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
&NewOrder did not save the branchcode posted with a new order. This patch adds that param.
Added code to select the branch the order is for in the branch dropdown list on
acqui/orderreceive.pl
Updating POD and tests
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
C4::Context->preference was not using placeholders and was potentially vulnerable to
a SQL injectin attack. This patch refactors the method to use placeholders.
Added some tests for C4::Context.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I also added a SKIP block in case of constructor failure and improved
the layout of existing diag lines.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
patch to C4::Members::ModMember to prevent it from deleting the dateofbirth field when none is supplied.
I also added a KohaTest::random_date method to help generate randomish dates for the test suite.
Added some tests for Member::ModMember. This is an easy method to test, and this bug shows that it
could use some closer examiniation.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
The C4::Date module is deprecated. It is also no longer used anywhere in Koha.
This patch removes it and the associated tests.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
This patch adds the misc/cronjobs/overdue_notices.pl script that is intended to replace
overduenotices.pl, overduenotices-30.pl and overduenotices-csv.pl. It adds messages to
the message_queue to be sent later (by process_message_queue.pl). It also marks borrowers
as debarred if their issues become too overdue.
It is intended to be run from cron nightly with usage something like:
0 2 * * * misc/cronjobs/overdue_notices.pl
C4::Members:
- improved documentation on ModMember
- made ModMember return a useful value (the return value of the database call)
- added a DebarMember method
- adding t/lib/KohaTest/Members/DebarMember.pm to test ModMember
misc/cronjobs/overdue_notices.pl
- designed to replace overduenotices.pl, overduenotices-30.pl, and overduenotice-csv
Changes to C4::Letters:
- EnqueueLetter now lets you pass in to_address and from_address which can override defaults
- _send_message_by_email pays attention to these defaults.
- now handles attachments with MIME::Lite
C4::Overdues
- added GetBranchcodesWithOverdueRules
- added t/lib/KohaTest/Overdues/GerBranchcodesWithOverdueRules.pm to test that.
circ/overdue.pl
- replaced call to obsolete overduenotices-csv.pl with call to overdue_notices.pl
KohaTest:
- added three helper methods: random_phone, random_email, random_ip
- these can be used to populate example records
- you can now pass an optional lengh to random_string
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
When generating the display form of a heading that
happens to (invalidly) have a regular expression
metacharacter as a subfield label, do not crash.
An example of such a heading field is:
<datafield tag="650" ind1=" " ind2="0">
<subfield code="a">Dalziel, Andrew (Fictitious character</subfield>
<subfield code=")">xFiction.</subfield>
</datafield>
The error message associated with the crash is:
Unmatched ) in regex; marked by <-- HERE in m/) <-- HERE / at
/home/koha-pro/kohaclone/C4/Heading/MARC21.pm line 220.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
It was observed that the %thash and @formats variables
were not being properly initalized during a make single-test run.
To ensure initialization, created startup and shutdown
methods to initialize those values as part of the
test object.
I have not yet investigated why the original way of
setting up %thash and @formats did not work.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
t/Labels.t was dependent on a working test database, so I'm moving those
tests into t/lib/KohaTest
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
The t/Items.t tests were actually dependent on the database, so I'm moving them
into t/lib/KohaTest.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
the icondirectories.t test depends on the database, so it needed to be
moved. But, it's really more of an author test since it's testinging the
correct structure of our distribution, so I'm moving it to xt/author.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
the t/Dates.t tests were actually databasase dependent. This patch replaces
t/Dates.t with t/lib/KohaTest/Dates/Usage.pm that relies on a database.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I wrapped the use of the SMS::Send module in an eval to make failures graceful if it
is not present.
I also fixed an error with the number of tests in the SMS::Send tests.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Added support for a new target in the test Makefile
to run a single test module. If you do (for example)
make test-single TEST_FILES=lib/KohaTest/Biblio.pm
only the tests in that module will be run. Unlike
the full test suite as run via 'make test', 'make test-single'
does not clear the test database before running the
tests.
Please note that "TEST_FILES=path/to/test/class.pm" is
required when using 'make test-single'.
Signed-off-by: Andrew Moore <andrew.moore@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
This code is intended to replace current value_builder code in 3.2, but
it does not affect it directly (yet) and is safe to include in 3.0.
This structure will be used to handle more complicated formats, like those
with checkdigits. Please note that "incremental" format is still STRONGLY
recommended because it will always perform the best, and most flexibly.
The desire to include other information (like branchcode) should compel
the proper use of the barcode generator to print the info ON the barcode,
not IN the barcode.
One of the nicer features of this structure is that you are able to
create a new barcode (of the same type) based on any previous Barcodes object.
That means you can create an array of 51 consecutive barcodes like:
my $x = C4::Barcodes->new('annual'); # for example
my @set = ($x);
for (1..50) {
push @set, $x=$x->new;
}
Importantly, this can happen without referencing the database after the
first constructor.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Prior to this patch, ModBiblio() would append
item tags from the previous version of the bib record
to the incoming bib record before saving the results,
even if the incoming bib record already has embedded
item tags.
For example, if a bib is retrieved using GetMarcBiblio() then
saved using ModBiblio(), the caller was obliged
to delete any item tags first to avoid duplication.
ModBiblio() now deletes item tags supplied in the
incoming MARC record. This eliminates the possibility
of duplication, and removes any implication that
ModBiblio() can or should be used to modify item
records - ModItem() should be used for that.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
The add_biblios() routine now reindexes all bibs
in batch instead of waiting for zebraqueue_daemon - this
is moderately faster. A separate set of test
cases for zebraqueue_deamon will be witten.
No documentation changes.
Signed-off-by: Andrew Moore <andrew.moore@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Test cases in KohaTest/Search/SimpleSearch.pm assumed
that they were the first to call add_biblios(); as
this is not necessarily true when the entire test
suite is run, test now counts how many 'Finn Test'
bibs already exist.
No documentation changes.
Signed-off-by: Andrew Moore <andrew.moore@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I changed getitemtypeimagedir to set a default on its argument so that it would not complain if not passed 'opac'.
I improved the documentation on the method.
I edited the t/icondirecotries.t test script to explicitly pass an argument to both getitemtypeimagedir calls.
- and I adjusted one line of whitespace to make similar things look similar
I added a test module for C4::Koha
I added a test module for C4::Koha::getitemtypeimagedir.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I've added methods to to C4::Letters to manage the database table
message_queue. This will let us keep track of messages sent
via email, sms, and rss to patrons. That way, we can show the history,
deal with failures, and reconstruct an RSS feed when needed.
misc/cronjobs/overduenotics.pl has been added. It prepares advance notices
and item due notices and stages messages to be sent in the message_queue
table.
C4::Overdues::Getoverdues now takes two optional arguments to tell it how
old of overdues to fetch.
Also, a C4::Circualtion::getUpcomingDueIssues method was added that
advance_notices.pl uses.
misc/cronjobs/process_message_queue.pl has been added. It sends the email
or SMS messages out of the message queue.
The C4::SMS module didn't work at all, and it has been rebuilt to use
an external perl module from CPAN, SMS::Send.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
This routine retrieves the branch/patron category circulation
rules for a given branch and patron category. The return
value is a hashref containing the following key:
maxissueqty - maximum number of loans across all item types
This will first check for a specific branch and
category match from branch_borrower_circ_rules.
If no rule is found, it will then check default_branch_circ_rules
(same branch, default category). If no rule is found,
it will then check default_borrower_circ_rules (default
branch, same category), then failing that, default_circ_rules
(default branch, default category).
If no rule has been found in the database, it will default to
the built in rule:
maxissueqty - undef
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
You can test getnextacctno like:
perl -e 'use C4::Accounts; print getnextacctno(33), "\n";'
where 33 is a borrowernumber out of the accountlines table. Get that number like:
mysql> select borrowernumber,accountno from accountlines LIMIT 100;
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I had these tests laying around for a while. I just forgot to commit them.
No functional or documentation changes needed.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Defined a function attribute for KohaTest and subclasses
called 'Expensive'. When a test method has that attribute,
it is skipped unless the RUN_EXPENSIVE_TESTS environment
variable is true.
To mark a test method, expensive, do this:
sub test_foo : Tests(4) Expensive { ... }
To mark a whole class and its subclasses expensive,
define a SKIP_CLASS sub (with empty body) with the
Expensive attribute:
sub SKIP_CLASS : Expensive { }
Updated the t/Makefile so that 'make test' runs
nonexpensive tests, while 'make fulltest' runs both
cheap and expensive tests.
Marked KohaTest::Installer test class expensive.
Signed-off-by: Andrew Moore <andrew.moore@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Corrected bug that prevented the 'not' operator
from working consistently - i.e., a search of
'mice not men' would not always work.
Also added test cases for NoZebra, so far focusing on
NZanalyse and adding and deleting bibs. A couple of
the test cases are currently known to fail and
therefore are marked TODO. The tests in question
are to verify that rows in nozebra are removed if
no bib is linked to the relevant word. However, it
looks like such rows are retained, just with
nozebra.biblionubmers set to ''. Is there any
reason to keep these rows?
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
In the course of writing tests for C4::Calendar, I found two bugs and am fixing them here.
One is a documentation bug.
The other prevented C4::Calendar::insert_exception_holiday from working. I was unable
to find anywhere else in the code that depended on the broken behaviour.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I moved C4::Branch::GetBranches to use bind parameters and wrote some tests to demonstrate functionality.
No functional or documentation changes here.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Now handles cases where two SQL scripts in different
directories have the same name, e.g., fr-FR sample_itemtypes.sql.
Bug was masked by existing failure when attempting to load
both sample_itemtypes.sql.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Two new modules to support patron attributes:
- C4::Members::AttributeTypes
OO-module for managing patron attribute types.
- C4::Members::Attributes
Procedural module for retrieving and setting
extended attributes belonging to a patron.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Added tests of loading the DB initialization and
sampple data SQL scripts for every language
and MARC flavour combination defined in
installer/data/{dbms}. Errors that the
new test class should catch include:
* SQL syntax errors in kohastructure.sql
* SQL errors in the sample data
For each language and MARC flavour combination,
all mandatory and optional scripts are loaded.
As of this commit, several errors are reported
by this test class; these will be fixed separately.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Moved routines to clear and create database and manage
zebraqueue_daemon.pl to KohaTest.pm so that they're
available to test classes - needed for installation
and upgrade tests in particular.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Some minor changes to get the test suite working a bit better:
I removed a superfluous method from t/lib/KohaTest.pm.
I made each barcode for the items added in KohaTest.pm unique so that they would actually get inserted.
Then, I removed t/override_context_prefs.pm. If you need that functionality, you're a database
dependent test and should be a module in t/lib.
So, I deleted all of the trivial .t tests that just 'use'd their modules and had no other
tests and replaced them with lib/KohaTest/*pm modules that do a little bit more checking
on those modules.
I removed the references to override_context_prefs.pm in all of the other .t modules.
They all pass now with no override_context_prefs.pm module.
The database_depenedent.pl test script still does not pass entirely. There's a problem with the zebra index
not being reset each time that the tables are truncated. I'll get to that.
no functional or documentation changes here.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I'm adding some functional tests that demonstrate that long_overdue.pl does what we think it does.
These tests don't actually work since override_context_prefs.pm gets in the way of C4::Context::preference. I opened up bug 2088 to address that.
There are no functional or documentation changes due to this patch.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
For test cases to be written soon, added kohastructure.sql (or equivalent)
from the following versions:
2.2.9
3.0-alpha
3.0-beta (note that schema for 3.0-beta2 is the same)
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Removed the categorytable, mediatypetable, and subcategorytable
tables. These three tables appear to be from an incomplete
patch from pre-git that defined these tables to implement
additional search limits, and are not present in the 2.2.9
package. In 3.0, these tables did nothing - the only code that
referred to them are incomplete admin editors that lack HTML
templates and are not referred to anywhere.
The updatedatabase.pl logic to remove these tables is conservative;
the tables are dropped only if they exist and are all empty. Effectively,
this means that they'll be dropped if you've been running a fresh install
of 3.0 from git or the alphas or betas; if you've upgraded your database
from 2.2.9, these tables likely weren't in your database to begin with.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Enhanced the ability of catalogers to specify how
bib and item records should be added, replaced, or
ignored during a staging import.
When an import batch of bib records is staged and commit,
the user can now explicitly specify what should occur
when an incoming bib record has, or does not have, a match
with a record already in the database. The options are:
if match found (overlay_action):
create_new (just add the incoming record)
replace (replace the matched record with the incoming one)
use_template (option not implemented)
ignore (do nothing with the incoming bib; however, the
items attached to it may still be processed
based on the item action)
if no match is found (nomatch_action):
create_new (just add the incoming record)
ignore (do nothing with the incoming bib; in this
case, any items attached to it will be
ignored since there will be nothing to
attach them to)
The following options for handling items embedded in the
bib record are now available:
always_add (add the items to the new or replaced bib)
add_only_if_match (add the items only if the incoming bib
matches an existing bib)
add_only_if_add (add the items only if the incoming bib
does *not* match an existing bib)
ignore (ignore the items entirely)
With these changes, it is now possible to support the following use cases:
[1] A library joining an existing Koha database wishes to add their
items to existing bib records if they match, but does not want
to overlay the bib records themselves.
[2] A library wants to load a file of records, but only handle
the new ones, not ones that are already in the database.
[3] A library wants to load a file of records, but only
handle the ones that match existing records (e.g., if
the records are coming back from an authority control vendor).
Documentation changes:
* See description above; also, screenshots of the 'stage MARC records
for import' and 'manage staged MARC records' should be updated.
Test cases:
* Added test cases to exercise staging and committing import batches.
UI changes:
* The pages for staging and managing import batches now have
controls for setting the overlay action, action if no match,
and item action separately.
* in the manage import batch tool, user is notified when they
change overlay action, no-match action, and item action
* HTML for manage import batch tool now uses fieldsets
Database changes (DB rev 076):
* added import_batches.item_action
* added import_batches.nomatch_action
* added 'ignore' as a valid value for import_batches.overlay_action
* added 'ignored' as a valid value for import_records.status
* added 'status' as a valid value for import_items.status
API changes:
* new accessor routines for C4::ImportBatch
GetImportBatchNoMatchAction
SetImportBatchNoMatchAction
GetImportBatchItemAction
SetImportBatchItemAction
* new internal functions for C4::ImportBatch to
determine how a given bib and item are to be
processed, based on overlay_action, nomatch_action,
and item_action:
_get_commit_action
_get_revert_action
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
[1] When running the database-dependent tests (cd t, make test),
all tables in the test database are dropped prior to running
the installer and test cases. This means that the test
database will start with a clean slate.
[2] It is now possible to specify a single test class to run,
to avoid having to run all of them:
cd t
make test TEST_CLASS=Search
To run all DB-dependent tests, just do the usual
cd t
make test
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
KohaTest->reindex_marc() does a full Zebra reindexing of
bib and authority records and clears zebraqueue. It is
meant for use in test classes that do a lot of bib, authority,
and/or item record changes that would be slowed down by
waiting for zebraqueue_daemon.
Also adjusted list of tables that are automatically truncated.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I've rearranged the two directories where media type icons live.
I've also added two more collections of icons, one from
http://apps.carleton.edu/campus/library/bridge_icons/
and one from liblime (Tina). The first has a license restriction that I added
to the "Licenses" tab on the "About" page.
Then, I've adjusted the Item Types Administration page so that it can deal
with multiple collections of icons.
I also added a test script to verify that the two identical icon directories
are actually identical.
DOCUMENTATION CHANGE: It's possible that we need to add something to the administration
documentation to indicate how you can add more sets of icons if you want. You simply add
directory to koha-tmpl/intranet-tmpl/prog/img/itemtypeimg/ and to
koha-tmpl/opac-tmpl/prog/itemtypeimg. The icons should start showing up.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I'm adding some more warnings in Makefile.PL to indicate that you WILL lose the data
in your testing database.
I've also relaxed the requirement that the teting database have "test" in its name.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
This patch makes the test suite work even when $PERL5LIB isn't set to the install directory.
It replaces $PERL5LIB with '..' in order to help ensure that we're loading the C4 libraries
out of the installation directory.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
* All configuration and run-time files created during
a test suite run now live under t/run
* made sure scripts copied to t/run/bin are executable
* added t/run and t/test-config.txt to .gitignore
Signed-off-by: Andrew Moore <andrew.moore@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
These test libs should have been included in the previous patch.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I'm adding a bunch of additions to the test suite. There is now a test
database that is created, and a test instance of the zebra server and
daemon that are run before the tests in t/database-dependent.pl are
run. This means that you can test things that insert (or destroy)
things in the database.
To use these tests, after you 'make' your koha installation, you can
change to the 't' directory and 'make test' there.
There is now an additional question asked during installation. It asks
whether you would like to run the data-dependent tests. If so, It asks
you for some login information to a test database. I recommend that
you do not not use your production database here. It will delete all
of the data in it.
Things that need improvement:
* I suspect that there are large parts that are not very platform
independent, so they will need to be improved.
* There are some parts of the installer that will probably be
refactored to let this work a little bit better, including being run
directly from a main-level 'make' target at some point.
* Lots more tests to add. Be bold! (see the t/lib directory)
* other tests in t/*.t can possibly benefit from being included in here.
There are two required perl modules for this part of the test suite. I
use:
'Test::Class' => 0.028,
'Test::Class::Load' => 0.02,
They are not listed as requirements in the top-level Makefile.PL
because they're not actually required to install or use Koha, but if
you want to run the test suite, you can install them from CPAN.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I improved the tests a bit for this module so that they at least skip
if there's not enough data in the database to test with.
I was unable to test the actual execution path through the change I actually made.
Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
* IsStringUTF8ish - determine if scalar contains a string in UTF8
* MarcToUTF8Record - convert MARC blob or MARC::Record to UTF8
* SetMarcUnicodeFlag - set appropriate MARC21 or UNIMARC field to
indicate that record is in UTF-8.
Design points of this module include:
* No dependencies on other C4 modules, making it easier to add
more test cases
* All character conversion code in one place
* Single entry point for doing a character conversion on a
MARC record
* Capture of errors and warnings produced by Text::Iconv
and MARC::Charset
* Start of support for guessing the source character set of
a MARC record.
Several functions were moved from other scripts
or modules to C4::Charset:
* C4::Koha->FixEncoding (expanded and renamed
MarcToUTF8Record)
* C4::Koha->char_decode5426
* fMARC8ToUTF8 from bulkmarcimport.pl (renamed
_marc_marc8_to_utf8)
Several batch jobs were adjusted to use MarcToUTF8Record instead of
FixEncoding.
Signed-off-by: Chris Cormack <crc@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Many of the tests were failing or putting warnings
because a valid systempreferences table is
usaully absent by the time 'make test' is run.
Fortunately, only a few modules try to invoke
C4::Context->preference during module initialization,
so added to the test suite override_context_prefs.pm,
which replaces preference() with a sub to return
testing values for three variables: 'dateformat',
'marcflavour', and 'LibraryName'.
Also fixed bug in t/Boolean.t
With this patch and the patch to move the DB-dependent
tests off to the side for the moment, 'make test'
now runs cleanly, at least on Debian.
Signed-off-by: Chris Cormack <crc@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Moved test cases that depend on the DBMS and having
an initialized Koha database to a subdirectory
of t so that they will not be swept up into
the default 'make test'. Goal is to have
these DB-dependent tests runnable either via
a special make target or perhaps from the
web installer.
Signed-off-by: Chris Cormack <crc@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Only functions that do not touch the DB
are currently tested.
Signed-off-by: Chris Cormack <crc@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Input.t was replaced because it tested a now obsolete function.
Input.pm has that function commented out.
Several files were renamed to match their counterparts or
correct misspellingz.
Signed-off-by: Chris Cormack <crc@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Minor adjustments to Dates.pm and associated test. You should be able to run a
perl test that uses Context w/o getting fatalsToBrowser output.
Signed-off-by: Chris Cormack <crc@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Also included is a traditional perl (t/Dates.t) test script.
Signed-off-by: Chris Cormack <crc@liblime.com>
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Create output_html_with_http_headers function to contain the "print $query
->header(-type => guesstype...),..." call. This is in preparation for
non-HTML output (e.g., text/xml) and charset conversion before output in
the future.
Created C4/Interface/CGI/Template.pm to hold convenience functions specific
to the CGI interface using HTML::Template
Modified moremembers.pl to make the "sex" field localizable for languages
where M and F doesn't make sense
in ISO-8859-1.
A new C4::Charset module (tentative name) has been created to guess the
charset of a piece of HTML markup. The CGI programs will be modified to use
this module as they are encountered during translation.
that attempt to modify the library database will _not_ be run unless the
environment variable DoUnsafeDBTests is set to 1. This allows people on
production systems to run the tests without any fear of data corruption, while
developers can run the full suite of tests on a standard sample database.
