Even though patches were submitted that fixed the breakage
that this commit caused to NoZebra searchs, it turns out
this breaks code that uses SimpleSearch - clients of that
routine currently expect ISO2709 blobs, not MARCXML.
No further patches for 3226 will be pushed until a complete,
comprehensive patchset is submitted that (a) includes a
test case for the original bug and (b) demonstrates that
it doesn't break any aspect of bibliographic and authority
searching.
(Bug 3402) Serials issue name not escaped when passed forward to routing slip
The issue descriptor was not being escaped before being passed on to the
routing slip, causing truncation. It's truly impossible to know for sure
what will be there, so uri_escape-ing seemed the best way to ensure that
it gets handed forward and makes it all the way to the printable slip.
Joe Atzberger [Wed, 22 Jul 2009 15:23:16 +0000 (10:23 -0500)]
Cleanup _Findgroupreserve.
Mostly formatting SELECT statement for readability. Also updated POD
and added a comment. This command:
git diff -w C4/Reserves.pm
shows only the POD and comment chnages.
Joe Atzberger [Wed, 22 Jul 2009 14:08:38 +0000 (09:08 -0500)]
3M SIP2 Extensions groundwork and Patron Info popoulation
This includes some initial work for the 3M SIP2 extensions.
It also better populates the Patron object with methods for
a fuller Patron Information Reponse. This is positively affect
EnvisionWare software, as used by NEKLS.
This work was sponsored by the Northeast Kansas Library System.
Bug 3322: Fixes expandedSearchOption in opac. Makes "fewer options" possible.
Copies the logic of catalogue/search.pl to fix the expanded search option in opac. When expanded search is the default, it was impossible to go to the "fewer" options.
Bug 3423 - In Cataloging, fields are not ordered by tag number
In cataloging, depending on installation and how biblio framework has been
created/modified, fields are not displayed ordered by tag number. For example,
in UNIMARC you can have in tab '2':
225
200
210
This patch order fields in tabs by tag and letter in the tag.
Bug 2553: Fixes the alphabetization of the drop-down lists for the Most-Circulated items report.
This patch alphabetizes the descriptions that appear in the Library, Item type, and patron category drop-down lists of the Most-Circulated Items report form.
Bug 2553: Fixes drop-down alphabetization in acquisitions - New order and Receive order.
Fixes alphabetization of library dropdown in New Order, and the shelf location, item type, and collection code in the Receive Items pages. This patch takes care of Joann's original purpose of creating the bug enhancement.
Bug 3408: Tag cloud says login when logged in - fix.
Changed the logic in the html form to have the login prompt only appear if a user is not logged in. Separated the hidemylist functionality from the "show tags from other users" option. Created a link for "hide my tags" or "show my tags" depending on context.
The user can now choose how many tags of other users to show, and can also choose whether to show their own tags or not.
Joe Atzberger [Mon, 20 Jul 2009 20:37:24 +0000 (15:37 -0500)]
Cannot use the same TMPL_VAR to select ALL options.
Only one select can be selected in this case. If preselection of search type is
desired, I recommend a clean implementation w/ jquery instead of adding a conditional
for each select. Please note, that is the only good way to get multiple preselections,
since the same .inc file will be used in each case. So a single variable like
<TMPL_IF NAME="index_bc_selected">
cannot be used, since it would set for ALL (3) advanced search dropdowns.
Joe Atzberger [Mon, 20 Jul 2009 22:47:42 +0000 (17:47 -0500)]
Remove bogus "selected" logic from OPAC adv. search.
The only reason the TMPL_VAR works where it does is because it is part
of the TMPL_LOOP. It doesn't mean anything outside the loops, since
*everything* would be selected with the same variable, indiscriminantly.
Joe Atzberger [Fri, 17 Jul 2009 14:56:31 +0000 (09:56 -0500)]
Add crontab instructrions.
Too many people have been confused about crontab dependent functions
"not working" automatically on their installations. This is probably
because there was no explicit step telling them to schedule the jobs.
The maintainers of the other INSTALL.* docs would be well advised
to insert similar lines in their docs, after confirming proper operation,
since cron/crontab are very system dependent.
When displayed on result page, some UNIMARC biblio
records are marc8 to utf8 converted for whatever
reason! It appears that iso2709 records in Zebra can't
always properly by used to construct MARC::Record.
This patch ask Zebra to return xml record rather than
iso2709 and use it to create MARC::Record (like
C4::GetMarcBiblio). i
The hash keys were made using a base and $item->{date_due}, but this key doesn't exists, so the possibility to have duplicate keys exists, and sometimes the others items were not shown. This patch fix this bug.
When displayed on result page, some UNIMARC biblio
records are marc8 to utf8 converted for whatever
reason! It appears that iso2709 records in Zebra can't
always properly by used to construct MARC::Record.
This patch ask Zebra to return xml record rather than
iso2709 and use it to create MARC::Record (like
C4::GetMarcBiblio).
Garry Collum [Sat, 27 Jun 2009 17:48:18 +0000 (13:48 -0400)]
Bug 3317: Adds author and added-author to email sent from basket (Cart).
Adds the author and added author fields to the email that is sent from the shopping cart. This makes the data sent via email consistent with the data that is viewed in the basket.
There is still a problem with author authority records other than 100 tags appearing correctly in the shopping cart.
Owen Leonard [Tue, 30 Jun 2009 19:56:28 +0000 (14:56 -0500)]
Changes to lists to improve behavior for users with no javascript
This patch moves the creation of some js-based function links to the script so that non-functioning links will not display when javascript is turned off. Other links which require TMPL processing are hidden with CSS and shown with JS.
- my @subf = $field->subfields;
- (defined @subf) or @subf = ();
is not equivalent to
+ my @subf = $field->subfields || ();
as, the results of $field->subields is interpeted in scalar
context, not list context, resutling in @subf containing
the number of subfields, not the subfield data itself, which
engenders the error
Can't use string ("1") as an ARRAY ref while "strict refs" in use
later on. Changing the operator to 'or' instead of '||'
fixes this.
Joe Atzberger [Wed, 8 Jul 2009 21:52:18 +0000 (16:52 -0500)]
Branch auto-selection in additem
Adding a new item should pre-select the library you are logged in at, where possible.
Use GetBranchesLoop. This solves indy branches filtering, and sort order problems.
Set default off of ->{selected}, to establish pre-selected branch.
Fix flags check in onlymine sub, so superlibrarian + other flags still gets everything.
Removed defined(@array) check, since perl -wc warns it is deprecated.
Note new FIXME for botched date processing.
Joe Atzberger [Tue, 30 Jun 2009 14:14:50 +0000 (09:14 -0500)]
LDAP overhaul
Allow replicate and update to be zero.
Break out logic into separate subs.
Do only one bind attempt depending on setting, instead of
necessarily failing first before trying auth_by_bind.
POD added for active directory and to document permutations of
behavior given different conditions. Fixed mistaken debug lines
that called "print STDERR printf ...", i.e. printed the line to output
and "1" to the error log. Added principal_name feature for generating
bind user from Koha userid.
Joe Atzberger [Thu, 2 Jul 2009 03:47:35 +0000 (22:47 -0500)]
Bug 2539. Remove unintegrated "spsuggest" code from search scripts.
The intent of kohaspsuggest is very confused. It wanted to simultaneously
log every search query, and get spelling suggestions from a totally different
database_host:db_user:db_password:database. The implementation seems
never have really made it into 3.0, however, since the only trace of the
tables it wants to query are in misc/migration_tools/22_to_30/phrase_log.sql
The job misc/spellcheck_suggest/make_spellcheck_suggest.pl does not even
target the same remote DB as the syspref tells the search code to do. Instead, it hits the
local DB and, not finding the tables that have no way of existing in a 3.0
install, it CREATES its own tables spontaneously, using MyISAM storage, no less.
This script should be drastically overhauled, or more likely just removed.
Other fixes:
Commented out unconditional warn!
Also reduced bizarre whitespace amounts in catalogue/search.pl and
fixed warning as recommended by perl -wc:
Scalar value @newresults[0] better written as $newresults[0]
John Beppu [Thu, 9 Jul 2009 01:09:00 +0000 (20:09 -0500)]
Minor Fixes for jQuery 1.3.2 Upgrade
I looked through all the relevant Javascript on this site
to find spots that would break in the upgrade from jQuery 1.2
to jQuery 1.3, and I fixed them.
Signed-off-by: John Beppu <john.beppu@liblime.com> Signed-off-by: Galen Charlton <galen.charlton@liblime.com>
Owen Leonard [Mon, 29 Jun 2009 20:00:09 +0000 (15:00 -0500)]
Fix for Bug 2713: Opac detail content overlaps when right column extends too far
Resubmitted. My proposed fix eliminates the OpacNav column on this page. It puts the "search for this title in..." links into a drop-down menu, and it moves the "similar items" display into a tab along with Holdings, Reviews, etc.
Bug 3390 Enabling ZIP compression in pdf generation to reduce file size
Enabling compression gives a ~26% *reduction* over non-compressed Type 3 font embedding.
ie. 4.5 K/pg (compressed graphic) vs. 17.5 K/pg (uncompressed Type 3 font) vs 111 K/pg
(uncompressed graphic).
It also appears that most other applications that export in pdf use compression by
default. (OO Writer, etc.) So this approach appears justified. One could always add code
to allow the user to select embedding mode and compression.
For UNIMARC installations, this patch only works after
updating the "NoZebraIndexes" system preference to
UNIMARC indexes (by default, NoZebraIndexes is being
wrongly set to MARC21 values in UNIMARC installations).
That has also been reported as a Bug, by me (see Bug 2178):
Bug 2178 - "NoZebraIndexes" system preference is being set
to MARC21 values in UNIMARC installations
http://bugs.koha.org/cgi-bin/bugzilla3/show_bug.cgi?id=2178
(bug #3284) fix borrower deletion in independantbranches mode
This patch fix the checks deleting a borrower in independantbranches mode.
Now, we check the user and the librarian are from the same branch, else we deny the deletion(and disable the link to delete).
Garry Collum [Wed, 24 Jun 2009 01:42:05 +0000 (21:42 -0400)]
Bug 3229: Fixes sort-by feature for jquery tablesorter in lists.
This patch adds some variables for the sortfield that are passed through the URL so that the sort-by feature works when the jquery tablesorter is used for a list of titles in opac-shelves.tmpl.
Garry Collum [Sat, 20 Jun 2009 02:42:46 +0000 (22:42 -0400)]
Bug 3347: Inconsistencies with public and private list tables in opac-shelves.tmpl.
This patch fixes inconsistencies with the public and private list tables in opac-shelves.tmpl. It adds a "new list" link to the public list view, and it adds a "sort by" column.
This patch also removes the toggle variables from the tables in opac-shelves.tmpl and Page.pm, and uses the template __odd__ variable for table highlighting if javascript is turned off.
Owen Leonard [Mon, 29 Jun 2009 16:17:49 +0000 (11:17 -0500)]
Removing hard-coded $ symbols from the template, addressing Bug 2547
Pending a system for displaying the correct currency symbol based on the system's default currency all dollar-specific symbols should be removed in favor of no symbol.
Modified fines.pl to not throw warning if --out not specified
It appears, in Ryan's patch, that he wants to throw a warning to the log if
the directory specified in --out is not present. (Further messages will
be given when the open-or-die occurs a few lines later.) However, it was
throwing the warning if --out was not specified at all, which is
undesirable. This patch modifies that bit to check for the presence of
whatever directory is going to be used, either --out, ENV{TMPDIR}, or /tmp.
As before, if the write to the directory fails for any reason--including
its' non-existence--that is handled later, but this message will help
inform the troubleshooter.
Frédéric Demians [Sun, 14 Jun 2009 05:17:40 +0000 (07:17 +0200)]
Bug 3301 - Speed up rebuild_zebra script
With this patch, rebuild_zebra can re-index a whole Koha DB
quickly:
rebuild_zebra -r -b -nosanitize
Biblio (authority) records are dump directly in a file
from marcxml field without beeing transformed into
MARC::Record object and corrected.
DOCUMENTATION:
rebuild_zebra.pl new paramater:
-nosanitize export biblio/authority records directly from DB marcxml
field without sanitizing records. It speed up
dump process but could fail if DB contains badly
encoded records. Works now only with -x and -b
Joe Atzberger [Thu, 18 Jun 2009 16:31:33 +0000 (11:31 -0500)]
Returns reworking to handle empty GetItemIssue
Code cannot rely on issueinformation being populated.
Note there is room for better efficiency to have AddReturn also provide the
itemnumber (where existing) so that GetItemnumberFromBarcode is not called
at both levels. Unfortunately there is discrepancy between this idea (for
efficiency) and the stated purpose of the $iteminformation object returned,
since $iteminformation is specifically the info from the issues table and
MUST be empty when the item was not in fact issued.
Joe Atzberger [Sat, 20 Jun 2009 00:32:42 +0000 (19:32 -0500)]
checkoverdues should not require $dbh
Passing $dbh around is an ancient style that doesn't know
to use C4::Context. C4::Context->dbh is efficient, especially
for modules that already use Context, including almost all C4.
I also internalized $today into the SQL using NOW() in the query
and removed sth->finish. Even though I dislike the return style
that gives the count, then the array, I left it becuase I don't
have time to fix/test all the callers. However, I did convert
it so it doesn't require a $count variable and its own loop.
Joe Atzberger [Sat, 20 Jun 2009 00:32:44 +0000 (19:32 -0500)]
Return amount for CREDITS, POD rework.
The main change here is to get an {amount} field for CREDITS,
not just CHARGES. I also use a fallback to default value (5)
in case noissuescharge has been deleted or set to 0.
POD is reworked to keep the salient chunks
about patronflags in the correct section and format them for
easier reference. Deprecated display layer code.
projects / koha.git / log