It is a separate relation, instead of having a field in campsite_type,
because not all campsite types allow dogs. I could have added a new
field to campsite_type, but then its values it would be meaningless for
campsites that do not allow dogs, and a nullable field is not a valid
solution because NULL means “unknown”, but we **do** know the price —
none.
A separate relation encodes the same information without ambiguities nor
null values, and, in fact, removed the dogs_allowed field from
campsite_type to prevent erroneous status, such as a campsite type that
allows dogs without having a cost — even if the cost is zero, it has to
be added to the new relation.
Besides the dynamic final cart, that was already handled by HTMx, i had
to check the maximum number of guests, whether the accommodation allows
“overflow”, whether dogs are allowed, and that the booking dates were
within the campground’s opening and closing dates.
I could do all of this with AlpineJS, but then i would have to add the
same validation to the backend, prior to accept the payment. Would not
make more sense to have them in a single place, namely the backend? With
HTMx i can do that.
However, i now have to create the form “piecemeal”, because i may not
have the whole information when the visitor arrives to the booking page,
and i still had the same problem as in commit d2858302efa—parsing the
whole form as is would leave guests and options field empty, rather than
at their minimum values.
One of the fieldsets in that booking form are the arrival and departure
dates, that are the sames we use in the campsite type’s page to
“preselect” these values. Since now are in a separate struct, i can
reuse the same type and validation logic for both pages, making my
JavaScript code useless, but requiring HTMx. I think this is a good
tradeoff, in fact.
I have to ask number and age ranges of hosts of guests for all campsite
types, not only those that have price options for adults, children, etc.
because i must compute the tourist tax for adults. These numbers will
be used to generate de rows for guests when actually creating the
booking, which is not done already.
To satisfy the campsite types that do have a price per guest, not only
per night, i had to add the prices for each range in the
campsite_type_cost relation. If a campsite type does not have price
per person, then that should be zero; the website then does not display
the price.
The minimal price for any campsite type is one adult for one night,
thus to compute the price i need at least the campsite type, the dates,
and the number of adults, that has a minimum of one. I changed the
order of the form to ask for these values first, so i can compute the
initial price as soon as possible. To help further, i show the
<fieldset>s progressively when visitors select options.
The “overflow” is for when people want to book plots for more guests
than is permitted, which the system would need to add a new plot to the
“shopping cart”, as it were; not implemented yet.
The ask zone preferences is to whether show the corresponding input on
the booking form, that it was done implicitly when the campsite type had
options, because up until now it was only for plots, but it is no longer
the case, thus i need to know when to show it; now it is explicit.
I use Sortable, exactly like HTMx’s sorting example does[0]. Had to
export the slug or ID of some entries to be able to add it in the hidden
input.
For forms that use ID instead of slug, had to use an input name other
than “id” because otherwise the swap would fail due to bug #1496[1]. It
is apparently fixed in a recent version of HTMx, but i did not want to
update for fear of behaviour changes.
[0]: https://htmx.org/examples/sortable/
[1]: https://github.com/bigskysoftware/htmx/issues/1496
If i parse everything, then the campsite type options no longer have
their minimum values preset, meaning that people have to input each and
every one of them.
Had to change setup_redsys because admins can not read the current
encrypt key, thus it is not possible to `set encrypt_key =
coalesce(…, encrypt_key)`.
Not that it did much sense, anyway, as i was already inside the branch
of the if when encrpty_key is null.
However, it seems that this also affects in the `on conflict` update. I
assume this is because `excluded` is some kind of row of the relation
and has the same restrictions.
user.Language is the language set in the database for the user, that in
the case of guests is always going to be ‘und’, that means Redsys would
select Spanish for everyone.
Every company need to have its own merchant code and encryption key,
thus it is not possible to use environment variables to keep that data,
and i have to store it in the database.
I do not want to give SELECT permission on the encryption key to guest,
because i am going to fuck it up sooner or later, and everyone would be
able to read that secret; i know it would. Therefore, i need a security
definer function that takes the data to encrypt, use the key to encrypt
it, and returns the result; nobody else should have access to that key,
not even admins!
By the way, i found out that every merchant receives the same key, thus
it is not a problem to keep it in the repository.
Since i need that SQL function to encrypt the data, i thought that i may
go the whole nine yards and sign the request in PostgreSQL too, after
all the data to sign comes from there, and it has JSON functions to
create and base64-code an object.
Fortunately, pg_crypto has all the functions that i need, but i can no
longer keep that extension inside the auth schema, because it is used
from others, and the public schema, like every other extensions, seems
more appropriate.
Instead of having the list of currency and language codes that Redsys
uses as constants in the code, i moved that as field to the currency
and language relations, so i can simply pass the lang_tag to the
function and it can transform that tag to the correct code; the currency
is from the company’s relation, since it is the only currency used in
the whole application (for now).
As a consequence, i had to grant execute to currency and the parse_price
functions to guest, too.
To generate the test data used in the unit tests, i used a third-party
PHP implementation[0], but i only got from that the resulting base64-coded
JSON object and signature, using the same that as in the unit test, and
did not use any code from there.
PostgreSQL formats the JSON as text differently than most
implementations i have seen: it adds spaces between the key name and
the colons, and space between the value and the separating comma. The
first implementation used replace() to format the JSON as exactly as
the PHP implementation, so that the result matches, and then tried to do
generate the form by hand using the output from PostgreSQL without the
replace(), to verify that Redsys would still accept my input. Finally,
i adjusted the unit test to whatever pg_prove said it was getting from
the function.
I still have the form’s action hard-codded to the test environment, but
the idea is that administrators should be able to switch from test to
live themselves. That means that i need that info in the redsys
relation as well. I think this is one of the few use cases for SQL’s
types, because it is unlikely to change anytime soon, and i do not need
the actual labels.
Unfortunately, i could not use enumerations for the request’s
transaction type because i can not attach an arbitrary number to the
enum’s values. Having a relation is overkill, because i would need
a constant in Go to refer to its primary key anyway, thus i kept the
same constant i had before for that. Language and currency constant
went out, as this is in the corresponding relations.
In setup_redsys i must have a separate update if the encrypt_key is null
because PostgreSQL checks constraints before key conflict, and i do
not want to give a default value to the key if the row is not there yet.
The problem is that i want to use null to mean “keep the same password”,
because it is what i intend to do with the user-facing form: leave the
input empty to keep the same password.
As now Go needs to pass composite types back and forth with PostgreSQL,
i need to register these types, and i have to do it every time because
the only moment i have access to the non-pooled connection is in the
AfterConnect function, but at that point i have no idea whether the
user is going to request a payment. I do not know how much the
performance degrades because of this.
[0]: https://github.com/ssheduardo/sermepa/blob/master/src/Sermepa/Tpv/Tpv.php
The form is based on the one in the current website, but in a single
page instead of split into many pages; possibly each <fieldset> should
be in a separate page/view. The idea is for Oriol to check the design
and decide how it would be presented to the user, so i needed something
to show him first.
I hardcoded the **test** data for the customer’s Redsys account. Is
this bad? I hope not, but i am not really, really sure.
The data sent to Redsys is just a placeholder because there are booking
details that i do not know, like what i have to do with the “teenagers”
field or the area preferences, thus i can not yet have a booking
relation. Nevertheless, had to generate a random order number up to
12-chars in length or Redsys would refuse the payment, claiming that
the order is duplicated.
The Redsys package is based on the PHP code provided by Redsys
themselves, plus some hints at the implementations from various Go
packages that did not know why they were so complicated.
Had to grant select on table country to guest in order to show the
select with the country options.
I have changed the “Postal code” input in taxDetails for “Postcode”
because this is the spell that it is used in the current web, i did not
see a reason to change it—it is an accepted form—, and i did not want to
have inconsistencies between forms.
jordi fita mas