The bundle pipeline can use a bit of a improvement:
Most of the process is moved from npm scripts to rollup.
Add package keys to smooth out the use of bundlers of the end users.
(Rollup by example check amongst others the main, module and browser key
to import the right version of a lib when needed).
Refactor rollup logic to make it more modular.
Add bundled module formats.
Add comments to tsconfig to see more clearly what's possible.
Add reference tag to help @types discovery in test files.
Co-authored-by: Simon Genin (ges) <ges@odoo.com>
Previous code naively handled nested t-set-slots: if a second named
slots was found, it overrode the first.
In this commit, we use a set to make sure that we only use the first
found t-set-slot node. Also, we ignore set-slots defined in a sub
components, because these slots are only relevant to the sub component
itself.
Note that it works as expected because document.querySelectorAll
performs a search depth first, so we will always use the named slots
closer to the parent element, in term of depth.
closes#682
# v1.0.10
Bug fixes!
- qweb fix: scoping issue with t-call in t-foreach
- qweb fix: issue with t-set with a body in a t-call
- qweb fix: handle input value attribute as a property
- qweb fix: allow t-call on arbitrary nodes
- qweb fix: add indeterminate to special input properties
- component fix: allow using t-model with bracketed expression
- component fix: properly validate multiple props
- component fix: issue with higher order component, and t-keys
- component fix: issue with unmounted children that should be destroyed
- component fix: make concurrent renderings more robust in some cases
- component fix: allow using vars with body as props
- observer fix: do not proxify promises
- test infrastructure: stop mocking requestanimation frame
Before this commit, the t-model directive worked well with expressions
such as "state.value", but not with bracketed expression: "state[value]"
(it generated invalid code).
This commit make the t-model smarter by detecting this case, and
properly capturing the base expression and key variable.
closes#694
Input with type="checkbox" have a special property (indeterminate) to
visually display the fact that the input value is non determinate (in my
chrome browser, the checkbox is then drawn with a simple - inside). It
does not actually modify the value of the input, only the way it is
displayed.
So, with this commit, owl will properly set the property, as expected.
closes#713
This is a rarely (if ever) used feature, but according to our qweb
reference implementation, it is possible to use the
t-call directive on an arbitrary html tag, like this:
<div t-call="my.template"/>
It is then interpreted as:
<div><t t-call="my.template"/></div>
So, with this commit, we make sure that the owl qweb implementation
matches that behaviour.
closes#706
Promises are kind of special, and do not behave like usual javascript
values.
For this issue, the problem is that when the observer tries to observe a
promise value, the code will crash with an error like this:
Uncaught TypeError: Method Promise.prototype.then called on incompatible receiver [object Object]
Also, note that it does not make much sense to proxify promise methods
anyway, since they are not (supposed) to be modified.
So, with this commit, we simply consider that promises should be treated
like a primitive value: simply ignored when determining if it should be
proxified
Note that I actually believe that putting promises in a useState is not
a good idea in general.
closes#677
The previous fix (overriding tostring of VDomArray) is actually more
general, and solves the same issue. So, let us simplify the code and
keep the more general solution.
This reverts commit 3bf91afc3f.
Consider this scenario:
- a variable v (with a body) is defined in a template
- it is then passed to a sub component as a prop
- and now, it is t-esc-ed.
Before this commit, the displayed value was [object object], because the
value actually passed to the sub component was a VDomArray (internal
structure used to represent nodelists)
This issue is actually quite a problem in practice, because values in a
templates are translated, but not in attributes. Therefore, using a
t-set directive with a body text content is the proper way to have
translated values at runtime.
We override in this commit the method toString of VDomArray to make sure
it is properly displayed.
Note that we considered changing the way props were generated (by trying
to detect VDomArray, then calling vDomToString), but then the value
would not be able to be used in a t-raw. Also, it is quite elegant to
be able to format the VDomArray only at the end.
closes#670
Before this commit, we artificially replaced in the tests the
requestAnimationFrame by a setTimeout, to actually increase the speed of
the tests. However, this is not really a true replacement. For
example, a real setTimeout can come before or after a real
nextAnimationFrame, depending on when/where it is requested.
Also, this change exposed another problem: the nextTick function did a
setTimeout before a nextanimationframe. This is not a problem when
nextAnimationFrame is replaced by a setTimeout, because then all
expectations holds in owl. However, it is wrong: to get to the next
animation frame, we need to request an animation frame, and THEN wait
with a setTimeout.
closes#729
Because of a "break" statement instead of "continue", the check for valid
props was stopping as soon as it met an optional props, which kind of
invalidate the whole system.
closes#717
Sometimes, HTML is slightly more subtle than what I initially expect.
Rendering some html is simple, we have tags and attributes. However,
once we add behaviour, then the situation is more complex:
<input value="abc"/>
is an input with an INITIAL value of "abc", but the attribute does not
actually represent the CURRENT value of the input, which may be
different if the user did change it.
This is basically the difference between "attribute" and "property".
So, when rendering html with owl, we sometimes want to actually set
the property (current value), instead of the html attribute.
This commit make sure that this is the case for inputs with the "value"
attribute.
closes#722
The body of a t-call directive may be used to define private variables
to the sub template call.
However, the code that handles t-call worked like this:
- compile sub template if necessary
- then compile body of t-call to extract variables
This means that the variables defined in the t-call body were not yet
processed and available in the context. Because of that, when the call
to t-esc is done, there is not internal qweb var, and the code simply
outputs a scope['varname'], which is in our case a VDOMArray, so it is
displayed as [object object]
What this fix does is changing the way t-esc works: if we are in the
context of a sub template, then it assumes that any outside variable may
or may not be a VDomArray, so it needs to check and eventually convert
it to a string, if necessary.
closes#719
Here is a situation that can happen in some complicated case:
1. a parent component is rendered, which includes some children
2. it is then willPatched
3. the sub components are then mounted/willUnmounted
4. because of complicated business logic, this causes the parent
component to be rerendered (before parent "patched" method is called)
5. owl will internally reset its currentfiber to null (but there is a
pending rendering!)
6. subsequent rendering will ignore pending rendering
7. havoc ensues
This is actually one of the reason why modifying a component state in a
willPatch component is actually not a good idea. However, the good news
is that this specific situation can be properly handled: we can simply
make sure that we do not reset currentFiber to null if there is a new
pending rendering.
closes#728
Owl has to manage a lot of interesting situations. One of them is when
a rendering is initiated, which creates a sub component, but then
another rendering starts, which invalidate the previous one, and will
create another sub component. Since the first sub component was not
ever in the DOM, we cannot rely on the vdom patching process to remove
it, so we have to do it manually.
Sadly, this is actually a very tricky situation, since there are other
subtle situations where the code that remove an unmounted widget could
be executed, in particular when the parent component is unmounted, then
remounted, then modified to trigger yet another rendering.
In this commit, we handle this case more carefully by making sure that
the destroyed subcomponent properly configures its pvnode so the patch
process happens as expected.
joint work with the framework team, and in particular LPE for his work on
finding a testcase!
closes#724, #731
Have a hierarchy of A, B, C components where:
```xml
<div t-name="A">
<div>
<B t-key="key1"/>
</div>
</div>
<t t-name="B">
<C t-key="key2"/>
</t>
<div t-name="C">
<div><t t-esc="keys_as_props" /></div>
</div>
```
The subtility of the issues lies in B, which doesn't have its own
concrete DOM element, rather, it borrows it from C.
With the sequence of events:
- change key2
C1 is destroyed and replaced by another instance, and another node.
B1 has its props updated and is patched with the C2's node (CRITICAL)
A1 is patched
- change key1 AND key2
C2 is destroyed
B1 is destroyed
A1 is patched replacing B1 by B2, and their nodes too (which at this point should be C2's to C3's)
Before this commit, at the CRITICAL point, the node representing the component itself
(technically its pvnode) was not updated with the new concrete node provided by B1 patch with C2 node
i.e. it held the previous node still
The second array of steps crashed because at A1 patch, the new B2 node would replace B1, which
was out of the DOM (removed because C1 was destroyed long before),
and therefore without a viable parent to insert B2 node.
After this commit, we update the component's pvnode after the patch which elm had possibly changed
There is no crash anymore for this use case.
Have a t-call nested in a t-foreach nested in a t-foreach
```xml
<t t-name="template">
<t t-foreach="..." t-as="a">
<t t-foreach="..." t-as="b">
<t-call="templateCalled" />
</t>
</t>
</t>
```
Before this commit, the `a` variable was not accessible within the t-call.
That was because the way t-call protected its scope by hiding other protected scope
in this case, the first protected scope for the first `t-foreach` was hidden
After this commit, `a` and `b` are accessible in the t-call, whether the t-call
defines its own variables by `t-set` or not.
Also, as expected from other fixes, there is no leaks of variables defined within a `t-call`
fixes#695
Before this commit, t-raw a string containing `<svg>` did not display
the svg element. This was because of missing namespace in vnode's data.
Now, `vdom.addNS` is called to add the svg namespace when necessary.
With QWeb, we can register globally templates (using the xml tag or
the registerTemplate function). However, these templates, once
compiled, can generate sub template compiled functions. Before this
commit, these sub functions were local to a specific instance.
This means that creating a new QWeb instance and rendering a global
parent template would crash, since it was unable to find the actual sub
function.
This commit fixes the issue: the sub functions are now shared
statically, but with a unique ID, so we do not have issues with sub
functions having a same name in different QWeb instance.
closes#701
# v1.0.8
- qweb fix: do not override t-att-class with class attribute in some cases
- component fix: refuse to mount a destroyed component
- hooks type improvement: useRef is now generic, and can capture the type of a component
- qweb fix: t-esc inside t-call using outside t-set
Before this commit, the generated code was incorrect, and crashed,
when there was a t-esc="abc" in a subtemplate, with t-set="abc"
done outside the subtemplate, with syntax <t t-set="abc">value</t>.
With this commit, we can get the type for the target component for a
ref.
By default, if the generic type is not given, the ref will simply use
the base Component type.
There was some code in qweb to make sure that we support setting class
and t-att-class on the same html element:
<div class="some class" t-att-class="{b: true}">...</div>
But the code did not work in the other direction:
<div t-att-class="{b: true}" class="some class">...</div>
With this commit, we just add the missing if statement
closes#664
This could be done by each application, but it does cost only a few
lines of code, and it helps standardizing the Owl ecosystem.
For example, some library (such as o_spreadsheet) needs to mock side
effects, and Odoo also needs to do that, so this prevents duplicated effort.
closes#686
Before this commit, Env was an indexed type, this means that one could
write env.anything, and it would accept it as a valid type. This is
actually quite dangerous, because we lose the typing advantages for all
keys that are properly defined.
For example, if a component is defined as:
class MyComponent extends Component<Props> {
...
}
Then Typescript will let it use anything from the environment, even if
it is wrong. So, most properly typed Typescript applications should use
instead a sub environment:
interface MyAppEnv extends Env {
someKey: someValue
}
Then, the component should be defined this way:
class MyComponent extends Component<Props, MyAppEnv> {
...
}
Before this commit, any typos in the environment accesses would not be
noticed by typescript.
It is possible that a Component is extended dynamically and if this is
the case, the class that extends it can be anonymous, with property
name=''. If this is the case, current implementation interprets the empty
string to be false so the while loop is terminated without further
scanning the super classes.
In this proposal, we allow anonymous class to be scanned until its
Component ancestor. Basically, the anonymous class assumes the name of
it super.
This new t-set-slot directive is meant to replace t-set when we need to
define the content of a sub slot. All new code should use that
directive.
The old t-set directive is still supported for now, but this should be
removed when we publish Owl 2.0.
Unfortunately, we chose to use the directive `t-set` to define sub slot
contents in a template.
The goal was to reuse a directive for a similar use case: defining sub
template is almost the same as defining a slot content.
Obviously, this introduces a name conflict: the inner content of a
component cannot use t-set t-value anymore (nor t-set with a body
value), since they are interpreted as slot names.
We mitigate the issue here by only interpreting as slot content the
`t-set` statement located immediately below the parent component tag
name and with a body content.
However, a real fix need to introduce an additional directive to resolve
the ambiguity.
Have something like
```xml
<div t-name="Parent">
<t t-call="nodeTemplate">
<t t-set="recursive_idx" t-value="1"/>
<t t-set="node" t-value="root"/>
</t>
</div>
<div t-name="nodeTemplate">
<t t-set="recursive_idx" t-value="recursive_idx + 1"/>
<p><t t-esc="node.val"/> <t t-esc="recursive_idx"/></p>
<t t-foreach="node.children or []" t-as="subtree">
<t t-call="nodeTemplate">
<t t-set="node" t-value="subtree"/>
</t>
</t>
</div>
```
Where we want to propagate a recursion index through recursive t-calls
Before this commit, it did not work as we protected the scope in order
to not leak, in the wronf manner. Namely the protected scope only took
firt level prototype properties of the original scope.
After this commit, this case works as we mark the scope as read only
solves #672
Have
```xml
<body t-name="webclient" />
```
and
```js
const comp = new WebClient();
comp.mount(document.body, {position: 'self'});
```
Before this commit, the body that was there before anything had happened
was *replaced* by the new body node created by the WebClient OWL component
After this commit, we ensure that the element body is the same at reference level
Since ee956a197, the rendering is skipped if the currentFiber is
completed. Unfortunately, cancelled fibers remain set in __owl__,
so when a fiber is cancelled, subsequent calls to render are
skipped.
It would be nice to reset __owl__.currentFiber to null when the
fiber is cancelled, but when trying to do so, a lot of tests fail.
Part of issue #622Closes#665
In this commit, we uses the "!" as suffix to designate an event that
should be captured. This is inspired by the Vue source code.
This solves the issue of communicating additional information to the
underlying virtual dom. However, this will most likely disappear when
we rewrite the vdom as a virtual block system.
closes#650
Simon Genin