Workaround for that: add a side-effect access of `.places` in the accessor
for `.queue`. Places init is where we kick off BrowserTable. I'm a little
surprised that the queue requires v_f_w…
(Lazy init of DBs is scary mmkay)
On Thu, Aug 27, 2015 at 7:02 AM, Emily Toop <et...@mozilla.com> wrote:
> Hmm, it appears that in _my_ XCode7, at least, applicationDidBecomeActive
> is being called _before_ BrowserViewController.viewDidLoad and therefore
> the tab queue is being created (caused BrowserDB creation) before the
> history tables....
>
> On 27 August 2015 at 14:51, Emily Toop <et...@mozilla.com> wrote:
>
>> Do we use this View: view_favicons_widest?
>>
>> I only ask as it's being created when the app starts, but before we've
>> invoked the HistoryTable which creates the favicons table, which is
>> referenced by the view, which causes the view creation to fail. I can't pin
>> down anything that has changed in this area that could affect the order in
>> which these things occur, as view_favicons_widest is created as part of
>> browser.db creation which always happens first.
>>
>> Can anyone spot what I am missing?
>>
>>
>> On 27 August 2015 at 14:20, Emily Toop <et...@mozilla.com> wrote:
>>
>>> That's what I'm doing. We can't even run the tests at the moment because
>>> the app crashes during initialisation. I'm working on getting that sorted
>>> so we can start running through the tests & resolving anything broken.
>>>
>>> I've added my name to the test suite that I intend to attack first in
>>> the mopad.
>>>
>>> On 27 August 2015 at 14:14, Stefan Arentz <sare...@mozilla.com> wrote:
>>>
>>>> Fantastic. So what is next? Do we keep working in the Swift-2.0 branch
>>>> until we can run the app on a device?
>>>>
>>>> S.
>>>>
>>>>
>>>> On Thu, Aug 27, 2015 at 5:45 AM, Emily Toop <et...@mozilla.com> wrote:
>>>>
>>>>> Today, is a Good Day.
>>>>>
>>>>> We have a cleanly compiling version of a Swift 2.0 port of FxiOS.
>>>>>
>>>>> Thank you everyone who helped out with this.
>>>>>
>>>>> Go forth and prosper my children.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> p.s. - now we just have to get it running 😕
>>>>>
>>>>> On 26 August 2015 at 15:38, Stefan Arentz <sare...@mozilla.com> wrote:
>>>>>
>>>>>> This compiles:
>>>>>>
>>>>>> req.responseString(encoding: nil, completionHandler: { (a, b,
>>>>>> c) in
>>>>>> return handler(a, b, c.isSuccess ?
>>>>>> Result.Success(c.value!) : Result.Failure(c.data, c.error!))
>>>>>> })
>>>>>>
>>>>>> But the forced unwrapping feels wrong.
>>>>>>
>>>>>> S.
>>>>>>
>>>>>>
>>>>>> On Wed, Aug 26, 2015 at 10:33 AM, Stefan Arentz <sare...@mozilla.com>
>>>>>> wrote:
>>>>>>
>>>>>>> I narrowed it down to:
>>>>>>>
>>>>>>> // Yay Swift.
>>>>>>> let stringHandler = { (a, b, c: Result<String>) in
>>>>>>> return handler(a, b, c)
>>>>>>> }
>>>>>>>
>>>>>>> req.responseString(encoding: nil, completionHandler:
>>>>>>> stringHandler)
>>>>>>>
>>>>>>> Which is then changed to:
>>>>>>>
>>>>>>> req.responseString(encoding: nil, completionHandler: { (a,
>>>>>>> b, c) in
>>>>>>> return handler(a, b, c)
>>>>>>> })
>>>>>>>
>>>>>>> but am now stuck on:
>>>>>>>
>>>>>>> StorageClient.swift:466:38: error: cannot convert value of
>>>>>>> type 'Result<String>' to expected argument type 'Result<AnyObject>'
>>>>>>> return handler(a, b, c)
>>>>>>>
>>>>>>> And I'm not sure what that means or how to resolve it.
>>>>>>>
>>>>>>> S.
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Aug 26, 2015 at 1:29 AM, Brian Nicholson <
>>>>>>> bnichol...@mozilla.com> wrote:
>>>>>>>
>>>>>>>> I added some commits to the PR that fix the SchemaTable errors and
>>>>>>>> some other things, which then led to a number of errors in Sync. I
>>>>>>>> think
>>>>>>>> those have all been fixed, though I'm now hitting a segfault in
>>>>>>>> StorageClient.swift, and the error isn't terribly helpful.
>>>>>>>>
>>>>>>>> Dump here: https://pastebin.mozilla.org/8843919
>>>>>>>>
>>>>>>>> Not sure how many targets are left after this, but it does feel
>>>>>>>> like this is close to the end!
>>>>>>>>
>>>>>>>> On Tue, Aug 25, 2015 at 9:38 AM, Emily Toop <et...@mozilla.com>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> I feel I am really close to the end now with this port.
>>>>>>>>>
>>>>>>>>> I'm getting 2 errors around BrowserDB and the way we are declaring
>>>>>>>>> SchemaTable and that is all currently. When these clear, I guess
>>>>>>>>> we'll see
>>>>>>>>> if anything else pops up. If not, we just have to make sure it
>>>>>>>>> actually
>>>>>>>>> works.
>>>>>>>>>
>>>>>>>>> If anyone wants to look at this overnight, please, please go
>>>>>>>>> ahead. The sooner we get this done the better for all.
>>>>>>>>>
>>>>>>>>> IMPORTANT PART 1: You will need XCode 7.0 Beta 6 to run this build.
>>>>>>>>> IMPORTANT PART 2: Do run carthage again when you grab this - I've
>>>>>>>>> had to fork and port a whole pile of our dependencies today in order
>>>>>>>>> to get
>>>>>>>>> this building against XCode 7.0 Beta 6.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On 24 August 2015 at 15:53, Richard Newman <rnew...@mozilla.com>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> I think you're heading in the right direction. I'll take a look
>>>>>>>>>> once I go through the painful steps to repro :D
>>>>>>>>>>
>>>>>>>>>> On Mon, Aug 24, 2015 at 7:14 AM, Emily Toop <et...@mozilla.com>
>>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>>> Just realised I was sending these to a dead address. Here we go
>>>>>>>>>>> again.
>>>>>>>>>>>
>>>>>>>>>>> On 21 August 2015 at 16:16, Emily Toop <et...@mozilla.com>
>>>>>>>>>>> wrote:
>>>>>>>>>>>
>>>>>>>>>>>> I've been working this week on getting Firefox for iOS ported
>>>>>>>>>>>> over to using Swift 2.0 in preparation for work beginning on
>>>>>>>>>>>> V1.0.x and the
>>>>>>>>>>>> release of iOS9 in a few weeks.
>>>>>>>>>>>>
>>>>>>>>>>>> I am most of the way there, but I have now come up against a
>>>>>>>>>>>> compiler crash in the Storage module that I am utterly flummoxed
>>>>>>>>>>>> by.
>>>>>>>>>>>>
>>>>>>>>>>>> Can anyone help me?
>>>>>>>>>>>>
>>>>>>>>>>>> You can find the code for the port at
>>>>>>>>>>>> https://github.com/mozilla/firefox-ios/pull/966
>>>>>>>>>>>>
>>>>>>>>>>>> The compiler crash is
>>>>>>>>>>>>
>>>>>>>>>>>> Command failed due to signal: Abort trap: 6
>>>>>>>>>>>>
>>>>>>>>>>>> It occurs when compiling the Storage module.
>>>>>>>>>>>>
>>>>>>>>>>>> The error output in the build is attached at
>>>>>>>>>>>> XCode7_Compiler_Crash.txt.
>>>>>>>>>>>>
>>>>>>>>>>>> Any help would be appreciated. I thought I had it narrowed down
>>>>>>>>>>>> to BrowserDB, but the further down the rabbit hole I went I
>>>>>>>>>>>> realised that
>>>>>>>>>>>> was a red herring. I suspect it is somewhere in SQLIteLogins
>>>>>>>>>>>> addLogin but
>>>>>>>>>>>> I'll be buggered if I can figure out where.
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> An update: I have managed to change the nature of the compiler
>>>>>>>>>>> crash to one that is a little more informative. See attached error
>>>>>>>>>>> log.
>>>>>>>>>>>
>>>>>>>>>>> In light of this I am going to go through all the
>>>>>>>>>>> classes/structs in the Storage module and ensure that all of our
>>>>>>>>>>> protocol
>>>>>>>>>>> implementations have complete signatures with the correct types.
>>>>>>>>>>>
>>>>>>>>>>> If that leads little insight I'm not sure where to go next.
>>>>>>>>>>>
>>>>>>>>>>> Have filed this compiler crash with apple:
>>>>>>>>>>> http://www.openradar.me/radar?id=4551431268859904
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> mobile-firefox-dev mailing list
>>>>>>>>>>> mobile-firefox-dev@mozilla.org
>>>>>>>>>>> https://mail.mozilla.org/listinfo/mobile-firefox-dev
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> mobile-firefox-dev mailing list
>>>>>>>>> mobile-firefox-dev@mozilla.org
>>>>>>>>> https://mail.mozilla.org/listinfo/mobile-firefox-dev
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> mobile-firefox-dev mailing list
>>>>>>>> mobile-firefox-dev@mozilla.org
>>>>>>>> https://mail.mozilla.org/listinfo/mobile-firefox-dev
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>
_______________________________________________
mobile-firefox-dev mailing list
mobile-firefox-dev@mozilla.org
https://mail.mozilla.org/listinfo/mobile-firefox-dev
