| Commit message (Collapse) | Author | Age | Lines |
|
|
|
| |
The database could return rows in differing orders.
|
|
|
|
|
|
|
| |
Adds a new tickbox to the contact editing form that protects the category
and group names from being changed by Open311.
For mysociety/fixmystreet-commercial#1776
|
|
|
|
|
|
|
|
| |
Order the response templates query by external status code so if you
have an update with an external status code the response template
matching that is always used. Previously with the database left to
determine the order it could pick a response template with a matching
state rather than a matching external status.
|
|
|
|
|
| |
Make them take start/end hour arguments, cope if only one given,
optional body, and combine them together in one `fetch` script.
|
|
|
|
| |
New unitary branding, and remove County from name.
|
|
|
|
|
|
|
| |
To avoid issues with autogenerated updates in the bodies back-end
happening quicker that the back-end returns the external_id to us ignore
the check for comment being earlier than the last update if it's the
first comment.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Co-authored-by: Matthew Somerville <matthew@mysociety.org>
Configured to allow:
- updates only from original reporter
- close updates from fetch reports
- use OSM geocoder
|
| |
|
|
|
|
|
| |
If an extra field on a category has a protected flag then do not
overwrite or remove it when populating categories.
|
|
|
|
|
|
| |
If an update with a blank external_status_code is fetched then unset it
otherwise we will always a detect a change in the external_status_code
which might mean phantom updates.
|
| |
|
|
|
|
| |
Lots of false positives from other bodies too.
|
|
|
|
|
| |
By specifying a placeholder in an email template, it can be replaced
by the description returned from the Open311 server.
|
|
|
|
|
|
|
| |
Only parses as multiple groups if cobrand is configured to handle them
to stop issues with existing groups with commas in them.
Groups are parsed as CSV so as to allow commas in group names.
|
|
|
|
|
| |
Rather than have a method in the Cobrand module use the COBRAND_FEATURES
section of the config file.
|
| |
|
|
|
|
|
|
| |
This change ensures that a report fetched via Open311 for a non_public category
will always be marked as non_public even if its Open311 XML doesn’t explicitly
include the non_public tag.
|
|
|
|
|
|
|
|
|
| |
This allows the Open311 endpoint to include ‘private’ in the ‘keywords’ field
of a service definition in order for it to be marked as non_public on FMS.
NB existing categories may be updated to become non_public, but non_public
categories will never have their non_public flag unset by PopulateServiceList.
This is so any existing Open311 categories on FMS don’t suddenly become public.
|
|\ |
|
| | |
|
| | |
|
|/
|
|
|
|
|
|
|
|
| |
On Macs, the Norwegian locale is no_NO, whereas on Debian it is nb_NO:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=206474
This can cause issues running the Norwegian tests on Macs with certain
setups (e.g. on my machine a Homebrew installation of perl and gettext
fails, but MacPorts and system pass). As Swedish is sv_SE in both, the
easiest solution is to switch the tests to Swedish.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Times in the database should be stored in the application server's local
timezone, by e.g. using `current_timestamp`, or by setting that timezone
explicitly before storage (the database columns are all without timezone
so any timezone info is silently ignored). Reports & updates fetched via
Open311 and offline updates were being put into the TIME_ZONE setting if
present, meaning they were stored incorrectly for future usage.
|
|
|
|
|
| |
If an update has a fixmystreet id in it check that it looks like an
integer and if not issue a warning and skip the update.
|
|\ |
|
| |
| |
| |
| |
| | |
The test has to now create a new comment object each time as
`get_cobrand_logged` is cached on the object.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If an update has a fixmystreet_id field then use the contents of that in
preference to the external_id field to match an update to a problem.
This handles the case where a report in a third party system passes
through different types (eg inspection and defect), resulting in the
external id changing. In this case it is sometimes possible to include
the fixmystreet id in each type and hence use that to match things up at
the fixmystreet side.
|
| |
| |
| |
| |
| | |
Otherwise running on a site where one body ID is a substring
of another, comments can get processed for the wrong body.
|
|/
|
|
|
| |
After the change in 1cff2a0e, we need to make sure this is set
so that the relevant fields are not shown in the front end.
|
|
|
|
|
|
|
|
|
|
|
|
| |
Without doing this, a call to e.g. $contact->set_extra_fields(@meta)
in PopulateServiceList.pm, with an unchanged meta that contains some
Unicode values, can write to the database (and cause an unneeded row
in the history table), because the column from the database is UTF-8
decoded, whilst the new text is UTF-8 encoded.
It looks like an attempt was made in filter_from_storage to fix this
issue, but the column comparison for marking a column as dirty takes
place without this being called.
|
|
|
|
|
|
|
|
| |
This adds a question to check whether a flytipping report concerns
flytipping on a road or not. Client-side, it is asked when a road is
selected and destination adjusted according to the answer; server side
it uses the answer to direct any both-tier submissions (from e.g.
non-JavaScript or phone app).
|
| |
|
|
|
|
|
| |
When fetching reports run them through the cobrands
filter_report_description method if it exists.
|
|
|
|
|
|
|
|
|
| |
If there is
<non_public>1</non_public>
tag in an incoming service request then set the created report to
non_public.
|
|
|
|
|
|
|
|
| |
Some of these were occasional failures due to timing issues.
One was due to a hard coded body id meaning the addition of bodies in
other tests meant this was creating two bodies with the same id, which
happened to be used in a test which depended on the bodies having
different ids.
|
|
|
|
|
|
|
| |
- Group wasn’t being set correctly by open311-populate-service-list as cobrand attribute
not being updated for each body.
- Extra metadata was being persisted to the DB every time even if nothing had changed,
causing lots of duplicate entries in contacts_history.
|
| |
|
|
|
|
|
| |
Allows creating scripts that fetch comments for a single body, e.g for
batch updating or because they require special setup.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
* Stop red warning being output by S3 bucket creation failure test
* Make sure 'num_old_reports' is always a number on /around pages
* Stop 'category_extra' being generated when unnecessary
* Prevent warning printed in getservicerequestupdates.t
* Prevent info lines from being output in develop.t
* Remove line of #s in root.t
* And an ID diag in zurich.t
|
|
|
|
|
| |
If an Open311 update has a customer_reference property then add that to
the metadata for the problem.
|
| |
|
|
|
|
|
|
| |
A move between fixed states (presumably from fixed-user to fixed-council)
should not count as a state change for the purposes of generating comment
text from templates.
|