Entering, or reading back sequences of digits is easier when they’re a
bit more spaced out.
This is because we read words as shapes, but read numbers
digit-by-digit.
So this commit adjusts the tracking of the type to put a bit more space
in for textboxes that are going to accept digits.
We keep seeing Chrome:
- autofilling the sign in form with a phone number in the email box
- autofilling the register form with an email address in the phone
number box
This is because Chrome tries to autofill what it considers to be a
_login_. It detects a login as being:
- a password field
- preceded by a text input field
On the sign in page the password field is preceded by a field which is
supposed to receive an email address. On the register page the password
field is preceded by a field that’s supposed to receive a phone number.
I suspect that this is why it’s missing the two up.
The solution to this seems to be to defeat Chrome trying to be clever,
and make it autocomplete based on the `name` of the fields instead (we
name them sensibly, e.g. `email_address`, `phone_number`). The way to
defeat it seems to be sticking a dummy `input` right before the password
field on the register page. This dummy input is hidden from the page and
from screenreaders, so a user should never know it’s there. but Chrome
will autofill it anyway, with whatever wrong value it wants to.
We do our own server-side validation of things like email address. We
don’t want the browser also trying to do it based on the input type of a
form field. It’s bad because the browser validation message comes up as
a nasty little tooltip under the field.
In pages specific to a service (e.g. dashboard and sub pages) the title
needs to distinguish which service it applies to. This is mainly to give
context to screen reader users who could be managing multiple services.
Implementing this uses template inheritance:
`page_title` includes `per_page_title` includes `service_page_title`
‘GOV.UK Notify’ is inserted into every page title.
Pages that set `service_page_title` get the service name inserted too.
The previous hint text was sort of half way there. Told you that you’d
need your phone handy, but not _why_. This came up multiple times in
user research. Also, the term ‘handy’ is not as familiar to users as
I anticipated.
This commit changes the message to explain exactly what we use your
phone number for.
We’ve seen in research that people can be reticent to give their real
phone number. Telling them that it will be used for something should
help (ie we’re not just collecting it for marketing).
This also rewords the other form hints on this page to be less computery
because we haven’t looked at them in aaaages.
This commit modifies the HTML `<title>` tags for all the pages. It makes two
main changes:
- make the title tag match the `<h1>` of the page, for better or worse
- put the service name after the page title, seperated by an en dash, as per
GOV.UK
Banners should always be the first thing on the page.
Because headers already have padding we don’t want to put padding on the
container.
So banners should also have top padding to distance then from the red bar.
They should also sit in the 3/4 column if the page has side navigation. This
commit adds a new template (`withoutnav_template.html`) which extends
`admin_template.html`. All views then extend one or the other, never the
`admin_template.html` directly. This means that `admin_template.html` doesn’t
have to make decisions about where the flash messages are displayed.
This macro:
- accepts a WTForm form field as a parameter
- renders a form field which follows the GOV.UK Elements patterns, both visually
and in markup terms
It then changes any page which uses either:
- the old, non-WTForms macro or
- the old, WTFforms `render_field` macro
…to use this new macro and removes both of the old ones.
It also adds the option to display hint text above the textbox.
Submit form was
- a confusing name in itself
- not descriptive, because it also has an optional ‘back’ link
This commit also puts this component in as many pages as possible, stripping
out any hard-coded buttons and links.
It replaces any usage of buttons for ‘back’, because these are links, not
buttons (ie they don’t change any data).
…or how to move a bunch of things from a bunch of different places into
`app/static`.
There are three main reasons not to use Flask Assets:
- It had some strange behaviour like only
- It was based on Ruby SASS, which is slower to get new features than libsass,
and meant depending on Ruby, and having the SASS Gem globally installed—so
you’re already out of being a ‘pure’ Python app
- Martyn and I have experience of doing it this way on Marketplace, and we’ve
ironed out the initial rough patches
The specific technologies this introduces, all of which are Node-based:
- Gulp – like a Makefile written in Javascript
- NPM – package management, used for managing Gulp and its related dependencies
- Bower – also package management, and the only way I can think to have
GOV.UK template as a proper dependency
…speaking of which, GOV.UK template is now a dependency. This means it can’t be
modified at all (eg to add a global `#content` wrapper), so every page now
inherits from a template that has this wrapper. But it also means that we have a
clean upgrade path when the template is modified.
Everything else (toolkit, elements) I’ve kept as submodules but moved them to a
more logical place (`app/assets` not `app/assets/stylesheets`, because they
contain more than just SASS/CSS).
